...

URL

by user

on
Category: Documents
2

views

Report

Comments

Description

Transcript

URL
Monitoring Effective Use of Household Water
Treatment and Safe Storage Technologies
in Ethiopia and Ghana
by
Matthew M. Stevenson
B.A. Chemistry
Amherst College
Submitted to the Department of Civil and Environmental Engineering
in partial fulfillment of the requirements for the degree of
Master of Engineering in Civil and Environmental Engineering
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
September 2008
© 2008 Matthew M. Stevenson. All rights reserved.
The author hereby grants to MIT permission to reproduce and to distribute publicly
paper and electronic copies of this thesis document in whole or in part in any medium
now known or hereafter created.
Signature of Author: ______________________________________________________
Matthew M. Stevenson
Department of Civil and Environmental Engineering
September 8, 2008
Certified by: _____________________________________________________________
Susan Murcott
Senior Lecturer of Civil and Environmental Engineering
Thesis Supervisor
Accepted by: ____________________________________________________________
Dr. Daniele Veneziano
Chairman, Departmental Committee for Graduate Students
2
Monitoring Effective Use of Household Water
Treatment and Safe Storage Technologies
in Ethiopia and Ghana
by
Matthew M. Stevenson
Submitted to the Department of Civil and Environmental Engineering on
September 4, 2008 in Partial Fulfillment of the Requirements for the Degree of
Master of Engineering in Civil and Environmental Engineering
ABSTRACT
Household water treatment and storage (HWTS) technologies dissemination is beginning
to scale-up to reach the almost 900 million people without access to an improved water
supply (WHO/UNICEF/JMP, 2008). Without well-informed and effective use as
intended, these promising technologies will not be deployed to maximum advantage.
Successful scale-up thus requires monitoring and evaluation (M&E) of behavioral
indicators to achieve safe water and improved health. This thesis offers a consistent
framework for the operational monitoring of Effective Use of a set of eight HWTS
technologies including dilute bleach solution, Aquatabs, solar disinfection (SODIS), cloth
filters, the ceramic pot filter, the biosand filter, PUR and associated safe storage
practices.
During late 2007, key members of the WHO-hosted International Network to Promote
Household Water Treatment and Safe Storage (“The Network”) who are involved with
M&E of HWTS systems were contacted. A literature search on monitoring efforts
involving the eight HWTS followed. The author traveled to Ethiopia and Ghana during
January 2008 to investigate multiple HWTS implementations and field-test preliminary
monitoring methods as part of that process. Interviews were conducted with HWTS
Network partners and the users of their HWTS products, household water quality testing
was conducted, and documents on usage and monitoring were collected and compiled.
A framework for operational monitoring of Effective Use behaviors at the household was
developed through these efforts. The framework consists of a set of Monitoring
Observations specific to each technology, comprised of the five categories of Treatment,
Safe Storage, Maintenance, Replacement Period, and Physical Inspection, as well as a set
of common Water Quality Monitoring paramaeters. Field methods for measuring
turbidity, residual free available chlorine, and E.coli as an indicator of microbiological
water quality are described that require minimal training, time, and equipment and that
are cost-effective (US $3.60 for a complete set of household tests).
Keywords household water treatment, safe storage, behavior, monitoring, water quality
Thesis Advisor: Susan Murcott
Title: Senior Lecturer of Civil and Environmental Engineering
3
4
ACKNOWLEDGEMENTS
I wish to dedicate this work to my mother, Mara Speiden Stevenson, for whom I took on
this degree and have endeavored to finish in her absence.
Thank you to all who have seen me through this year. To Susan for your patience and
guidance, and to the Network for your contributions and vigilant service, I thank you.
5
6
TABLE OF CONTENTS
1. Introduction ________________________________________________________ 15
1.1 The Need for Safe Drinking Water at the Point of Use ________________________ 15
1.2 Development of Household Water Treatment and Safe Storage Technologies _____ 16
1.3 International Network to Promote Household Water Treatment and Safe Storage_ 16
1.4 The World Health Organization Guidelines for Drinking-Water Quality, 3rd Ed__ 17
1.4.1 Health-Based Targets ________________________________________________________ 17
1.4.2 Water Safety Plans __________________________________________________________ 18
1.4.3 Applying the GDWQ to HWTS ________________________________________________ 19
1.5 Consistent, Sustained and Effective Use ____________________________________ 20
1.6 Monitoring and Evaluation Indicator Compendium __________________________ 21
1.7 Thesis Purpose and Scope-“Effective Use” __________________________________ 21
2. Methods − Interviews and Field Trips ___________________________________ 23
2.1 Interviews and Correspondence with Network Members ______________________ 23
2.2 Expert Review of Effective Use Sections ____________________________________ 23
2.3 Field Trips −Ethiopia and Ghana _________________________________________ 24
2.3.1 Ethiopia___________________________________________________________________ 24
2.3.2 Ghana ____________________________________________________________________ 25
2.3.3 Interviews _________________________________________________________________ 25
2.3.4 Household Visits____________________________________________________________ 26
3. Methods − Water Quality Monitoring____________________________________ 27
3.1 Turbidity
__________________________________________________________ 27
3.1.1 Field Turbidity Measurement __________________________________________________ 27
3.2 Microbial Indicators ____________________________________________________ 29
3.2.1 Microbial Quantification Methods ______________________________________________ 30
3.3 Chlorine Disinfection____________________________________________________ 32
3.3.1 Chlorine Residual Measurement ________________________________________________ 33
3.3.2 Disinfection Potential with Turbidity ____________________________________________ 34
3.4 Portable Water Testing Laboratory _______________________________________ 35
4. Effective Use Write-ups of HWTS Technologies ___________________________ 37
4.1 Safe Storage __________________________________________________________ 38
4.1.1 Safe Storage Effective Use Brief for HWTS-Treated Water __________________________ 39
4.1.2 Monitoring Observation ______________________________________________________ 39
4.1.2.1 Safe Storage ___________________________________________________________ 39
4.1.2.2 Maintenance ___________________________________________________________ 40
4.1.2.3 Replacement period______________________________________________________ 41
4.1.2.4 Physical Inspection ______________________________________________________ 41
4.1.3 Water quality monitoring _____________________________________________________ 41
4.1.4 Discussion _________________________________________________________________ 42
4.1.4.1 Settling _______________________________________________________________ 43
4.2 Sodium Hypochlorite Solution ____________________________________________ 45
4.2.1 Sodium Hypochlorite Solution Effective Use Brief _________________________________ 45
7
4.2.2 Monitoring Observation ______________________________________________________ 46
4.2.2.1 Treatment _____________________________________________________________ 46
4.2.2.2 Safe Storage ___________________________________________________________ 47
4.2.2.3 Maintenance ___________________________________________________________ 47
4.2.2.4 Replacement period______________________________________________________ 47
4.2.2.5 Physical Inspection ______________________________________________________ 48
4.2.3 Water quality monitoring _____________________________________________________ 48
4.3 Aquatabs
__________________________________________________________ 50
4.3.1 Aquatabs Effective Use Brief __________________________________________________ 51
4.3.2 Monitoring Observation ______________________________________________________ 51
4.3.2.1 Treatment _____________________________________________________________ 51
4.3.2.2 Safe Storage ___________________________________________________________ 52
4.3.2.3 Maintenance ___________________________________________________________ 52
4.3.2.4 Replacement Period______________________________________________________ 53
4.3.2.5 Physical Inspection ______________________________________________________ 53
4.3.3 Water quality monitoring _____________________________________________________ 53
4.3.4 Discussion _________________________________________________________________ 54
4.4 SODIS
__________________________________________________________ 56
4.4.1 SODIS Effective Use Brief ____________________________________________________ 57
4.4.2 Monitoring Observation ______________________________________________________ 58
4.4.2.1 Treatment _____________________________________________________________ 58
4.4.2.2 Safe Storage ___________________________________________________________ 59
4.4.2.3 Maintenance ___________________________________________________________ 59
4.4.2.4 Replacement period______________________________________________________ 59
4.4.2.5 Physical Inspection ______________________________________________________ 59
4.4.3 Water Quality Monitoring_____________________________________________________ 60
4.4.4 Discussion _________________________________________________________________ 61
4.4 Cloth Filter
__________________________________________________________ 63
4.5.1 Cloth Filter Effective Use Brief ________________________________________________ 63
4.5.2 Monitoring Observation ______________________________________________________ 64
4.5.2.1 Treatment _____________________________________________________________ 64
4.5.2.2 Safe Storage ___________________________________________________________ 65
4.5.2.3 Maintenance ___________________________________________________________ 65
4.5.2.4 Replacement Period______________________________________________________ 65
4.5.2.5 Physical Inspection ______________________________________________________ 66
4.5.3 Water Quality ______________________________________________________________ 66
4.6 Ceramic Pot Filter ______________________________________________________ 67
4.6.1 Ceramic Pot Filter Effective Use Brief ___________________________________________ 67
4.6.2 Monitoring Observation ______________________________________________________ 69
4.6.2.1 Treatment _____________________________________________________________ 69
4.6.2.2 Safe Storage ___________________________________________________________ 70
4.6.2.3 Maintenance ___________________________________________________________ 71
4.6.2.4 Replacement period______________________________________________________ 73
4.6.2.5 Physical Inspection ______________________________________________________ 73
4.6.3 Water quality monitoring _____________________________________________________ 74
4.6.3.1 Sampling Procedure _____________________________________________________ 75
4.7 Biosand Filter__________________________________________________________ 77
4.7.1 Biosand Filter Effective Use Brief ______________________________________________ 77
4.7.2 Monitoring Observation ______________________________________________________ 78
4.7.2.1 Treatment _____________________________________________________________ 78
4.7.2.2 Safe Storage ___________________________________________________________ 79
4.7.2.3 Maintenance ___________________________________________________________ 80
4.7.2.4 Replacement Period______________________________________________________ 82
8
4.7.2.5 Physical Inspection ______________________________________________________ 82
4.7.3 Water quality monitoring _____________________________________________________ 82
4.7.3.1 Sampling Procedure _____________________________________________________ 85
4.7.4 Discussion _________________________________________________________________ 85
4.7.4.1 Recontamination in storage units ___________________________________________ 86
4.7.4.2 Training materials pertaining to safe storage with biosand filter ___________________ 87
4.7.4.3 Storage unit cleaning frequency ____________________________________________ 88
4.8 PUR
__________________________________________________________ 89
4.8.1 PUR Effective Use Brief______________________________________________________ 90
4.8.2 Monitoring Observation ______________________________________________________ 91
4.8.2.1 Treatment _____________________________________________________________ 91
4.8.2.2 Safe Storage ___________________________________________________________ 91
4.8.2.3 Maintenance ___________________________________________________________ 92
4.8.2.4 Replacement Period______________________________________________________ 92
4.8.2.5 Physical Inspection ______________________________________________________ 92
4.8.3 Water Quality Monitoring_____________________________________________________ 93
4.8.4 Discussion _________________________________________________________________ 94
5. Determination of Effective Use from Monitoring Visits _____________________ 95
5.1 Kale Heywet Church Biosand Filter Program _______________________________ 95
5.2 Sample Effective Use Monitoring Checklists ________________________________ 97
5.3 Discussion of Effective Use Monitoring Results _____________________________ 101
6. Discussion_________________________________________________________ 103
6.1 Monitoring and Evaluation______________________________________________ 103
6.2 Field Interviews _______________________________________________________ 104
6.3 Best Practices for Field Monitoring _______________________________________ 105
6.4 Common Threads in Household Monitoring _______________________________ 106
6.5 Technology−Specific Observations _______________________________________ 109
6.5.1 Pretreatment ______________________________________________________________ 109
6.5.2 Maximum turbidity for use with the biosand filter _________________________________ 109
6.5.3 Dosing volume and pause times for the biosand filter ______________________________ 110
6.5.4 Consistent use of PUR and other consumable HWTS ______________________________ 110
6.5.5 Ceramic Pot Filter __________________________________________________________ 111
7. Conclusion ________________________________________________________ 113
References ________________________________________________________ 117
Appendix A: Behavior and Sustained Use Questionnaire _______________________ 127
Appendix B: Fieldtrip Interviews ___________________________________________ 129
Appendix C: Household Monitoring Reports _________________________________ 145
Appendix D: Portable Laboratory Testing Addendums_________________________ 187
Appendix E: Effective Use Monitoring Checklists _____________________________ 191
Appendix F: Usage Instructions per Technology ______________________________ 207
Appendix G: PSI PHAST for use with Waterguard ____________________________ 227
9
10
LIST OF FIGURES
Figure 1 WHO Framework for Safe Drinking-Water
................................................ 17
Figure 2 Consistent water treatment
......................................................................... 20
Figure 3 Comparison of NTU and TU............................................................................. 28
Figure 4 DelAgua Turbidity Tube ................................................................................... 28
Figure 5 Various Safe Storage Containers ................................................................... 43
Figure 6 CDC Settling Pictorial ................................................................................... 44
Figure 7 PSI Nigeria Waterguard Label ...................................................................... 46
Figure 8 Aquatabs ............................................................................................................ 50
Figure 9 SODIS Usage Pictorial ....................................................................................... 58
Figure 10 (a) CDC Cloth Filter Usage Schematic; (b) GWEP Filter in Ghana................ 64
Figure 11 Potters for Peace CWP Maintenance Poster.................................................... 71
Figure 12 Typical square concrete household biosand filter unit. ................................... 77
Figure 13 Log10 concentrations of E.coli throughout BSF treatment and use .............. 86
Figure 14 BSF with Safe Storage from Machakos, Kenya ........................................... 87
Figure 16 PUR Usage Instructions printed on back of packet......................................... 91
Figure 17 Effective and Ineffective Use among Kale Heywet Church BSF Users ......... 96
Figure 18 Example Monitoring Checklist for Household 3 of the KHC BSF Users....... 97
Figure 19 Example Monitoring Checklist for Household 7 of the KHC BSF Users....... 99
LIST OF TABLES
Table 1 Regional Diarrheal DALYs ................................................................................ 15
Table 2 Risk Levels from E.coli ...................................................................................... 30
Table 3 Bill of Quantity for 25-Household Water Testing Kit........................................ 35
Table 4 Biosand filter Effective Use metrics................................................................... 84
Table 5 Water quality in BSF households after BSF intervention ................................. 86
Table 6 Water Quality Results for Kale Heywet Church Biosand Filter Users............... 96
Table 7 Sample Household Monitoring Data Format for KHC BSF Users .................. 101
11
12
LIST OF ABBREVIATIONS
AED
AWD
BOD
BSF
CAWST
CFU
CT
CWP
DALY
EAWAG
E.coli
FAC
GDWQ
G-Lab
GWEP
GWSC
HDPE
HIP
HWTS
JMP
KHC
KWAHO
M&E
MDG
MF
MIT
MPN
NGO
NTU
O&M
PF
PFP
PHAST
PHW
POU
PSI
QMRA
RADWQ
SANDEC
SNNPR
SODIS
TC
TSS
TU
U5
UNICEF
USAID
WHO
WSP
Academy for Educational Development
Acute Watery Diarrhea
Burden of disease
Biosand Filter
Center for Affordable Water and Sanitation Technology
Colony forming unit
Ceramica Tamakloe
Ceramic Water Purifier
Disability Adjusted Life Year
Swiss Federal Institute for Aquatic Science and Technology
Escherichia coliform
Free Available Chlorine
Guidelines for Drinking Water Quality
Global Entrepreneurship Lab, a course at Sloan School of Management
Guinea Worm Eradication Project
Ghana Water and Sewerage Corporation
High-Density Polyethylene
Hygiene Improvement Project
Household Water Treatment and Storage
Joint Monitoring Program
Kale Heywet Church
Kenya Water for Health Organization
Monitoring and Evaluation
Millennium Development Goals
Membrane Filtration
Massachusetts Institute of Technology
Most Probable Number
Nongovernmental organization
Nephelometric turbidity unit
Operations and Maintenance
3M Petrifilm
Potters for Peace
Participatory Hygiene and Sanitation Training
Pure Home Water
Point of Use
Population Services International
Quantitative Microbial Risk Assessment
Rapid Assessment of Drinking Water Quality
Department of Water and Sanitation in Developing Countries at EAWAG
Southern Nations, Nationalities and Peoples Region
Solar Disinfection
Total Coliform
Total Suspended Solids
Turbidity units
Children under five years of age
United Nations Children’s Fund
United States Agency for International Development
World Health Organization
Water Safety Plan
13
14
1. Introduction
1.1 The Need for Safe Drinking Water at the Point of Use
In 2008 alone, 1.5 million people will perish due to the ravishes of diarrheal diseases,
most of them before their fifth birthday (JMP, 2008). This figure has decreased
significantly from the five million deaths per year throughout the 1970s, thanks in large
part to vigilant implementation of lifesaving technologies such as Oral Re-hydration
Therapy and expansion of borehole, piped distribution networks, and other improved
water supplies in rural and urban localities, respectively, throughout poor- and middleincome countries. Despite these gains, however, death from dehydration due to diarrhea
is still an unacceptably large problem, with impacts disproportionately affecting the poor.
Diarrheal diseases in high income countries account for only 418 deaths or 0.2% of the
total burden of disease (BOD) as calculated in total Disability Adjusted Life Years
(DALYs)1, while in low- and middle-income countries diarrheal disease accounted for
1.6 million deaths and 3.8% of DALYs in 2001 (Lopez, 2006). The countries of SubSaharan Africa shoulder the lion’s share of diarrheal disease burden, as shown in Table 1.
Table 1 Regional Diarrheal DALYs
Data from
2001
(Lopez, 2006)
DALY U5*
DALY Total
Low/Middle Income
Latin America and
Countries
Caribbean
(1000s of DALYs)
(1000s of DALYs)
53,000
1,888
58,700
2,632
4.2% of total BOD**;
#6 in rank of total BOD
*U5 refers to children under five years of age.
** BOD refers to Burden of Disease
Sub Saharan Africa
(1000s of DALYs)
20,707
22,046
6.4% of BOD, #4 in
continental BOD
East Asia and
Pacific
(1000s of DALYs)
7,017
8,782
Fecal-oral transmission of diarrheal diseases accounts for 85% of all preventable DALYs
worldwide due to their significant effect on the population under five years of age.
Throughout the past twenty years, a few influential reports on whether the vector path is
mostly waterborne or water-washed have produced differing directions in policy and
budgetary planning. From a health-based perspective, the best option for securing safe
water for domestic use is the same that is available to over 99% of high-income country
dwellers: clean piped water consistently available within the household. Water at the
household tap eliminates both of the contamination routes identified by Cairncross et al.
(1996), namely ‘public domain’ contamination at the source (including un-safe sources,
and the processes of filling and transporting) and ‘domestic domain’ contamination
1
Using a cost effectiveness analysis, the health benefits of an intervention are measured in Disability
Adjusted Life Years (DALYs) in order to compare diverse waterborne health outcomes ranging from brief
self-limiting disease to fatal episodes. DALYs incorporate both a disability weight associated with the
outcome (a measure of severity of disease/ disability on a scale of 0 to 1, with 1 symbolizing death) as well
as the duration of the outcome’s effect in years. The disability weight given to diarrheal diseases is 0.105
in Disease Control Priorities in Developing Countries (Jamison, 2006). Given this weight, one child’s death
accounts for 30 DALYs (Jamison, 2006). DALYs allow health benefits and cost to be compared across a
variety of interventions (Havelaar, 2003).
15
within the household (through handling, storage and use). Both of these contamination
pathways must be dealt with in order to consistently reduce the likelihood of diarrheal
disease. However, in low-middle income countries, the capital expenditure required for
the large infrastructure projects necessary to treat and pipe water is often unavailable.
1.2 Development of Household Water Treatment and Safe
Storage (HWTS) Technologies
In response to the logistical and financial constraints inherent in providing piped or other
“improved” supplies to the people of developing countries, a new set of household
technologies has been developed and disseminated to many places in the developing
world during the past fifteen years. While these methods are employed in the home and
can be less costly both in capital expenditure as well as achieving similar health impacts
as improved source interventions, they require proper and consistent implementation, use,
and maintenance, to achieve effect.
Such products include safe storage containers as distributed by the Centers for Disease
Control (CDC) for use in their Safe Water System (SWS), dilute bleach-based
chlorinating solutions, solid tablet chlorine disinfectants such as Medentech’s Aquatabs,
solar disinfection techniques such as SODIS, simple cloth filters as used in the Guinea
Worm Eradication Program, ceramic pot filters such as those promoted by Potters for
Peace, scaled-down slow sand filters such as the biosand filter, and sachets of solid
flocculent and disinfectant such as Proctor and Gamble’s PUR™. Among the many
HWTS technologies, these are the technologies that will be researched in this thesis.
While various HWTS technologies also exist to treat specific chemical constituents such
as arsenic and fluoride, these technologies will not be covered in this thesis. All of these
HWTS techs are in the scale-up stage throughout the world and are encountering
constraints based on distribution, user acceptance, effective use of the products, training
methods, sustainability, etc.
1.3 The International Network to Promote Household Water
Treatment and Safe Storage
One hundred and seventeen organizations currently comprise the World Health
Organization-hosted International Network to Promote Household Water Treatment and
Safe Storage. This inter-disciplinary public-private partnership brings together leading
proponents of HWTS from government, industry, academic and non-profit sectors. Until
now, efforts to monitor and evaluate (M&E) HWTS implementation and scale up have
been largely restricted to individual organization’s initiatives. Information on M&E
methods, targets, indicators, tools and results are few and exist mainly in unpublished
literature. While transfer of information is one key constraint to scale-up efforts, there
has been little coordination within the Network towards a common set of M&E methods,
targets, tools and indicators. In order to improve the implementation and scale up of
HWTS, the Network needs to share information and experiences, and this thesis
endeavors to develop one opportunity for information sharing on M&E.
16
1.4 The World Health Organization Guidelines for DrinkingWater Quality, 3rd Edition
This thesis provides a common framework for monitoring HWTS, building on the
structure of the WHO 3rd Edition Guidelines for Drinking-Water Quality (GDWQ) to
derive monitoring frameworks for a range of core HWTS technologies. In the 3rd Edition
GDWQ, the WHO lays out a comprehensive framework for ensuring safe drinking-water,
comprised of these requirements:
• Well established health-based targets
• Systems that are properly constructed, managed and operationally monitored
• Establishment of an independent system for surveillance monitoring.
Figure 1 WHO Framework for Safe Drinking-Water (WHO, 2004)
1.4.1 Health-Based Targets
The health-based targets used by the WHO provide a thorough method with which to
ensure drinking water quality. Four types of health based targets are outlined. These
targets are arranged from the general to the specific, as described below.
Health Outcome Targets
Health outcome targets are specified reductions in prevalence of a given waterborne
disease or water-related condition in places with high existent burdens. When disease
burden attributable to water-related disease is high (i.e., given an emergency situation,
endemic exposure, chemical contamination, etc.), changes in prevalence of such diseases
through a given treatment intervention can be measured. If epidemic conditions do not
exist, analysis based on exposure estimates and dose-response relationships is conducted
in the form of quantitative microbial risk assessment (QMRA) in order to determine
tolerable levels of risk in a given population.
17
Water Quality Targets
Many normally occurring components of natural waters as well as anthropogenic
chemical pollutants have been shown to be acutely toxic, carcinogenic or otherwise
harmful. Chemical contamination of water is of the utmost concern for health-based
monitoring. The short and long term acceptable concentrations of over 125 chemical
contaminants has been characterized in the WHO 3rd Edition Guidelines, as based on
consumption levels of drinking water. These recommendations are called “water quality
targets” (WQTs) and they aid in the determination of necessary treatment measures.
Performance targets
Performance targets refer to intended reductions of microbial concentrations between the
feed water and the finished drinking water product. Indicator microbes are measured as
proxies for groups of pathogens and are reported as presence/absence, absolute risk, or
percent/log reduction from influent. Measurement of more than one indicator is often
needed to show different sources of contamination. Developing performance targets
relies on knowledge of tolerable disease burden in conjunction with severity of disease
outcomes and dose-response relationships for a given pathogen or target microbe (WHO,
2004).
Specified technology targets
The regulation of small water treatment systems at the household or community level is
hindered by lack of monitoring and oversight. Once these systems are in place, local
governments and implementing organizations often lack the capacity to develop
functional maintenance and monitoring programs, diminishing the prospects for proper
management or effective treatment. Through developing specified technology targets,
the WHO notes that national governments can aid community-scale organizations by
developing standards and recommendations concerning applicability, implementation,
and operation of smaller systems. Because the testing of compliance to these targets is
resource intensive, national training protocols and adequate support systems can be
developed in order to ensure better results. The WHO 3rd Edition Guidelines set no
specified technology targets but rather recommends that this work take place at a national
level.
1.4.2 Water Safety Plans
Health-based targets are very useful in the formulation of Water Safety Plans (WSPs), the
WHO recommended methodology for ensuring safe provision of water (see Figure 1
WHO Framework for Safe Drinking-Water (WHO, 2004)).
WSPs consist of a
comprehensive system assessment at the outset of the project, encompassing a thorough
investigation of hazard identification and health-based targets. Next, control measures
are designated to deal with the hazards laid out in the system assessment. A means of
operational monitoring is identified to ensure that each control measure is operating
adequately. Finally, management plans are established for routine maintenance,
upgrading or replacing the system, and for operation under normal as well as during
hazardous conditions. Through these steps, the WSP identifies hazards to health and sets
a plan in motion to adequately deal with those hazards over the lifetime of the treatment
system.
18
1.4.3 Applying the GDWQ to HWTS
The framework of health based targets, WSPs and operational monitoring proposed by
the WHO GDWQ provides a clear model with which to derive implementation protocols
for any given treatment system. Thus, while performance targets are usually applied to
microbial contamination in piped supplies, these targets are also useful metrics for
monitoring the performance of HWTS technologies. This thesis will designate specific
microbial, turbidity, and free available chlorine (FAC) performance targets based on
existing literature and expert input for each HWTS. Analogous to specified technology
targets, they are intended for use in monitoring and evaluation programs and are for
review by national authorities. Feed water quality will be generalized with a focus on
fecal contamination and sediment load in the derivation of such targets, with the caveat
that these recommendations may warrant adaptation to site-specific contexts. Such
specific performance targets will form the basis of the microbial, turbidity, and FAC
guidelines recommended in this thesis.
The WHO recommends that governing and operating authorities lay out WSPs for smallscale systems because individuals and communities often lack the capacity to do it
themselves. This thesis will use information collected from implementing organizations
and experts in order to lay out preliminary WSP-style frameworks for operational
monitoring as they apply to specific HWTS technologies. The system assessment for
most of these self-contained treatment systems has already been conducted by the
designer and/or implementing agency.
Treatment processes constitute control measures that are designed into HWTS
technologies to avert potential dangers from raw source water. While lab-based testing
proves the potential treatment characteristics of the technologies, effectiveness of
treatment in the home lacks rigorous study. Various agencies have developed operational
monitoring techniques to validate the performance of control measures in the field, yet
these tests lack the precision of lab-based testing. As of yet, many of the HWTS systems
are missing a WSP-style analysis in the field, and the literature and experience involving
monitoring and evaluation has not been collected and analyzed in a common framework.
Collection, development and standardization of information concerning HWTS
monitoring and evaluation is one end product of this thesis.
Operational monitoring consists of both physically inspecting contamination prone areas
and using a regime of microbiological, turbidity, and residual chlorination testing to
validate treatment controls. An emphasis is placed on monitoring during implementation
in order to catch shortfalls of construction and/or training and operation (Baker, 2007).
Management plans consist of a body of literature and promotional material with info on
training methods, proper treatment, maintenance, replacement period, and technological
alternatives. Effective Use can also be operationally monitored through inspection and
informal interview in the house. The WSP framework will be revised in order to
standardize the M&E of HWTS, as laid out in the Effective Use Write-ups given in
chapter 5.
19
1.5 Consistent, Sustained and Effective Use
At the June 2005 3rd international meeting of the WHO-hosted Network in Bangkok,
Susan Murcott proposed an extension of the WHO GDWQ framework to include HWTS.
She presented indicators for all four of the WHO’s health-based targets described above
in section 1.4 and proposed three additional targets as they pertain to HWTS: (5)
Behavioral Outcomes, (6) Coverage, Use and Sustained Use, and (7) Financial Targets
(Murcott, 2005). Murcott also reported on the research of MIT Master of Engineering
student Robert Baffrey, whose field work involved an investigation of the M&E methods
of the eleven organizations implementing HWTS in Kenya (Baffrey, 2005). Murcott and
Baffrey developed an extensive survey, which, in a shortened form, was posted on the
Network website with responses analyzed and reported in Murcott (2006) and at the
Network meeting in London of the same year.
At the Bangkok Network meeting in 2005, Figueroa led a lunch-break discussion that
sought to define and measure proper storage and household management (serving) of
water. Most notably, this discussion led to defining Consistent Water Treatment as a
household with treated water on hand everyday and all of whose members drink that
water everyday, as developed in Figure 2 (Figueroa, 2005).
Definition
Measurement
(i) Household has treated
water for drinking every
day. Treatment may or
may not occur every
day. Frequency of
treatment will depend on
type of technology used
and number of
household members.
(ii) All members in the
household drink this
treated water.
Three measurements are suggested.
Preferably get the three of them if time
and resources allow. From total
households in study area:
(i) Number of households that report
having treated water for drinking in
the house.
(ii) Number of households that show
treated water in the house.
(iii) Number of households with a
negative test for E.Coli in their
treated water, OR positive test for
chlorine residual among those using
chlorine-based technology.
Figure 2 Consistent water treatment
Data source
Household-based
data; preferably
population based
survey.
Data will include:
(i) self-reported
information;
(ii) direct
observation at
end of survey
(iii) tests for water
safety
(Figueroa, 2005)
The USAID Hygiene Improvement Project (HIP) Agency for Educational Development
(AED) has also actively contributed to the discussion of M&E indicators of HWTS. At
the USAID-HIP-AED E-conference in January, 2007, Orlando Hernandez proposed three
alternative indicators for measuring the behavioral outcomes target: (1) Volume of sales
of HWTS products, (2) Number of liters of water treated, and (3) Percentage of
households practicing effective household water management (Hernandez, 2006).
Hernandez (2008) has compiled the first multi-system HWTS monitoring survey.
Specifically designed for use with AED’s work in Ethiopia, his was the first survey to
monitor behaviors of the suite of HWTS covered in this document. His survey and
definition of Effective Use were used to aid the Effective Use Write-ups of this thesis.
20
1.6 Monitoring and Evaluation Indicator Compendium
In September 2007, Susan Murcott, Orlando Hernandez, and Boni Matigbay, the
Network Secretariat, formed a working group with MIT students to create a compendium
of best practices concerning the additional targets and indicators proposed in order to
expand the 4 healht-based targets described in the WHO GDWQ 3rd Ed. The
compendium idea came about since the Network did not seem ready to adopt common
metrics concerning these additional M&E targets.
The MIT team is comprised of two engineering graduate students and four MIT Sloan
School students, under the supervision of Senior Lecturer Susan Murcott of the Civil and
Environmental Engineering Department. The MIT Sloan team (Udit Patel, Shivani Garg,
Geeta Gupta and Eswar Mani) focused on analyzing financial and commercial indicators,
presenting their final report in February, 2008. Kate Clopeck of the Technology and
Policy Program and the Department of Urban Studies and Planning will spend two years
researching behavioral indicators pertaining to Adoption and Sustained Use, including a
rate of adoption indicator. Matt Stevenson investigated the target of Effective Use, as
presented in this thesis. Susan Murcott, Orlando Hernandez, and Boni Matigbay
provided input through meetings, teleconferences, draft reviews, and collaboration with
regard to developing survey tools and generating a compendium of M&E tools.
1.7 Thesis Purpose and Scope-“Effective Use”
The intention of this thesis research is to develop a set of categories with which to assess
the “Effective Use” of a core group of household water treatment and safe storage
technologies (HWTS). “Effective Use” is defined as the proper operation of HWTS
technologies in the home, as instructed by the implementing organization, resulting in the
production and storage of safe water in order to limit exposure to a variety of waterborne
diseases. Two broad categories were developed to check the characteristics of Effective
Use through monitoring in the home. Monitoring Observations refer to specific
observations to make in the households of HTWS users, including the five categories of:
(1) Treatment, (2) Safe storage, (3) Maintenance, (4) Replacement period, and (5)
Physical inspection. Water Quality Monitoring includes specific measurements of
turbidity, chlorine residual, and/or microbial water quality for each HWTS technology.
While every HWTS technology has its own unique features pertinent to monitoring and
evaluation, the intent of this thesis is to provide a common framework across multiple
HWTS technologies, fulfilling the needed first step towards the standardization of
common metrics for behavioral indicators of HWTS. This practical set of categories will
be compiled in brief and then described in detail in the Effective Use Write-ups in
Chapter 5, with an associated monitoring checklist for each technology included in
Appendix E: Effective Use Monitoring Checklists of this document.
This standardized framework for monitoring and evaluation, drawn from global “best
practices” will provide a valuable resource for those implementing HWTS within the
WHO-hosted Network and around the world. With common monitoring and evaluation
tools, results can be compared across HWTS systems and implementing organizations
leading to effective handling of the barriers to scale-up.
21
22
2. Methods − Interviews and Field Trips
2.1 Interviews and Correspondence with Network Members
Preliminary contact was made with a selected group of Network members in order to gain
a clearer picture of the existing frameworks, tools and indicators developed for
monitoring and evaluation of HWTS, as well as for the author of this thesis to refine the
metrics of Effective Use of the various technologies before going into the field.
Interviews were conducted using the questionnaire developed by Kate Clopeck and Matt
Stevenson (Appendix A: Behavior and Sustained Use Questionnaire). During this period,
the author contacted Derek Baker of CAWST in early December, 2007 as an expert on
cost-effective operational monitoring. A very fruitful discussion ensued, with Baker
presenting material about CAWST’s current monitoring and evaluation projects in Haiti
and Lao PDR, as well as key parameters for operating and monitoring the main HWTS
covered in this document. Joe Brown of the University of Alabama was contacted later
in December, 2007 and provided information on the methods, failures and error analysis
of doing field-based health impact studies, as presented in his PhD thesis (Brown, 2007).
As the research progressed into Effective Use metrics for each of the given technologies,
the author contacted a new set of Network members with specific questions about
monitoring their given technologies. In May 2008, Rob Quick of the Centers for Disease
Control (CDC) and Eric Fewster of Bushproof were interviewed by phone concerning
specific monitoring techniques proposed in the Effective Use write-ups for their
respective Safe Water System (sodium hypochlorite solution and safe storage) and the
biosand filter. Philip Downs of the Carter Center’s Guinea Worm Eradication Project
(GWEP) was contacted by phone in July, 2008 concerning field practices for training and
monitoring cloth filter use. These conversations presented the researcher with a large
volume of current research concerning their respective methods of monitoring and
evaluating HWTS. Similarly, Joe Moran of Medentech concerning Aquatabs, Jeff Albert
of Aquaya and Greg Allgood of P&G concerning PUR™, and Danielle Lantagne of the
CDC concerning both the Safe Water System and the ceramic pot filter were contacted
with specific technical and monitoring methodology questions during May and July,
2008. These exchanges validated and improved the Effective Use Write-ups as well as
provided up-to-date literature on the various systems.
2.2 Expert Review of Effective Use Sections
Following revision of their first drafts, the various Effective Use Write-ups were sent out
to the key Network members most involved with a given technology. In response, useful
contributions were made by Regula Meierhofer of EAWAG, Matthias Saladin of
Fundación SODIS, Ron Lentz of CAWST, Paul Edmondson of Medentech, Tom Mahin
of the Massachusetts Department of Environmental Protection, and Greg Allgood of
Procter and Gamble. The various Effective Use Write-ups were greatly aided by the
review of these experts.
23
2.3 Field Trips −Ethiopia and Ghana
During January 2008, the author spent 10 days in Ethiopia followed by 17 days in Ghana
to research existing HWTS implementations in those countries. These two countries
were chosen as field trip destinations due to their emerging use of a range of HWTS.
Annual per capita rural water procurement expenditures by the Ethiopian government had
fallen to US $9 in 1995, with no signs of rising (Shenkut, 1995). Partially in response to
the inability to serve the rural poor, in recent years, the government of Ethiopia has
helped to promote new treatment technologies. In October, 2007, the Ethiopian
government hosted a country meeting of the WHO-hosted Network, showing a
willingness to revamp their water sector to be more inclusive of their rural population. In
June, 2008, the Ghanaian government hosted the 3rd International Network Symposium in
Accra, Ghana. A number of presentations were made at these meetings, which this
author has used for background. With the interest in household water quality and
treatment growing in these countries, many long-term programs currently operating, and
multi-million dollar investments being made by the Hilton Foundation, World Vision,
and USAID in Ethiopia in particular, these two countries offered appropriate field sites to
investigate Effective Use in scaling-up of HWTS.
2.3.1 Ethiopia
The Federal Democratic of Ethiopia lies in the Horn of Africa with an area of 426,000
square miles (roughly twice the size of Texas) and a population of 77 million people,
making it the second most populous country in Africa, after Nigeria. Ethiopia is bordered
by Sudan, Eritrea, Djibouti, Somalia and Kenya. Ethiopia’s long-standing isolation from
surrounding economies and the Western World stalled the installation of infrastructure
projects throughout the country. The citizens of Ethiopia have recently been forced
through thirty years of harsh military rule, drought, civil war and internal land conflict,
especially in the northern and eastern border areas, with severe impacts on both the
populous and state infrastructure. With 80% of the population in the small-scale
agricultural sector concentrated in the highlands (above 1500m), Ethiopia is a land of
dense population and intense farming practices. In 1900, 40% of the land was forest
cover. Today, 3% remains, representing one of the world’s fastest rates of deforestation.
With a harsh topography of fertile highlands abutted by steep slopes falling to arid
lowlands and the hot rift valley, installation of dams for water storage is difficult. Rivers
flowing out of the highlands are so silt-laden that they require treatment even for
irrigational use. With a weak drilling sector and historically weak institutions such as the
Water Works Construction Authority (EWWCA) and the Water Resources Development
Authority (WRDA), water resource development is difficult and limited in scope (Abate,
1994). Many of these authorities have been reorganized in the past few years to fall
under the Ministry of Water.
63% of rural inhabitants get their drinking water from unimproved sources, exposing
them to increased likelihood of diarrheal disease as long as they remain without treatment
(RADWQ, 2007). PSI’s TRaC survey in 2006 showed that among caregivers of children
under age fourteen in Addis Ababa and SNNP Region, 53% stored water in narrow
mouthed containers with lids yet only 3.8% had used their sodium hypochlorite solution
Watergaurd (PSI, 2007). Additionally, 50% of water storage containers in the household
24
were leaking or otherwise unsanitary, with 15% used for storing liquids other than
drinking water as well (RADWQ, 2007). Much work is left to be done for HWTS in
Ethiopia.
2.3.2 Ghana
Located on the southern coast of West Africa, Ghana is a nation of 22 million people in a
climatically varied, yet flat land of 92,500 square miles, slightly smaller than the state of
Oregon. In 1957, Ghana gained its independence from Britain, making it the first
independent sub-Saharan country in otherwise colonial Africa. Rainfall is seasonal in
Ghana, with two rainy seasons in hot and humid southern Ghana and one rain throughout
the north. The author carried out interviews among organizations in the national capitol,
Accra, in southern Ghana, and in Tamale, the regional capitol of Northern Region, as
well as conducted household monitoring visits in their outlying communities. He also
traveled to Bolgatanga in the Upper East region to witness an emergency distribution of
ceramic pot filters.
The infant mortality rate during 2007 was 54 deaths per 1,000 live births (About, Inc.,
2007). While the country average shows slightly better neonatal health than the
surrounding countries, the Northern Region of Ghana had 154 deaths of children under
five years of age for every 1,000 live births. Suffering greatly from diseases like malaria,
yellow fever, schistosomiasis, and meningitis as well as high disease burdens from waterrelated diseases such as bacterial and protozoal diarrhea, hepatitis A, and typhoid fever,
Ghanaians attain a life expectancy of 59 years (World Fact book, 2007).
2.3.3 Interviews
Meetings with various businesses and NGOs by Stevenson and the G-Lab Sloan business
group were organized and held jointly in both countries. Meetings consisted of the GLab team asking a set of questions, as laid forth in their Final Report (Patel et al., 2008),
followed by Stevenson gaining information on both Effective and Sustained Use by
utilizing the framework developed in Appendix A: Behavior and Sustained Use
Questionnaire. Attending these interviews provided this author with in depth knowledge
of the relationships between supply chains, business operations and Effective Use.
Stevenson interviewed several organizations independently in Ghana after the G-Lab
team departed.
Synopses of the eight formal interviews conducted while in Ethiopia and Ghana are
compiled in Appendix B: Fieldtrip Interviews. The synopses introduce the interviewee
and provide an overview of the HWTS program investigated. Following this intro, the
notes list lessons learned specifically concerning training methods, Effective and
Sustained Use, and monitoring and evaluation activities. In addition, the synopses
provide a list of materials collected and give reference to any associated field visits.
While Appendix C: Household Monitoring Reports contains compiled notes on the field
interviews, a few consistent concepts noted throughout the interviews are reviewed in the
Discussion Chapter, under Field Interviews.
25
2.3.4 Household Visits
Where possible, Stevenson prearranged with the program managers of the interviewed
organization to visit the users of the various technologies in their homes. Prior to
commencing the trip and in preparation to making household visits to users of the HWTS
technologies in both Ethiopia and Ghana, the author compiled an interview questionnaire
for use in the house. The interview questions were drawn from the work of Peletz
(2007), Baffrey (2005), and Hernandez (2008), as well as through collaboration with
Clopeck. Since no formal surveying was done by the author, the interview format was
informal and technology/context specific, with notes written up in shorthand in the field
and formalized as field notes in Appendix C. Water samples were taken and analyses
were undertaken by the author during these visits. The goal of the interview framework
was to gain insight into appropriate questions to be posed on one-time household visits.
With this in mind, the author raised questions on an informal basis without asking the
same questions in every household visited. The number of households visited for any
given implementation was small given the time and logistical constraints inherent in
visiting rural settings on a short field study. The author visited houses outside Debre Zeyt
town in Oromiya region, Ethiopia, east of Addis Ababa using the biosand filter under the
guidance of the Kale Hewyet Church, seven households in total. In Northern Region
Ghana, four houses in Kpanvo using the HydrAid design of the biosand filter under
Osman Mumuni’s implementation for International Aid were visited. Monitoring of five
households was witnessed using the Kosim ceramic pot filter distributed by Pure Home
Water for UNICEF to flood-affected victims in Upper East Region, Ghana. Six
households using the ceramic pot filter as distributed by Enterprise Works were visited in
a peri-urban area outside of Accra, Ghana. The results of these household visits are
summarized in Appendix C: Household Monitoring Reports under the given
implementation, and are utilized directly in the recommendations for Monitoring
Observations in the Effective Use Write-ups. No households using PUR, Aquatabs or
other chlorine products were visited by this author, as the organizations implementing
those technologies did not monitor users of recurrent-use products and had no specific
clients to contact. There were no agencies to visit who were implementing SODIS in
either Ethiopia or Ghana, however, the author is familiar with KWAHO’s monitoring
operations of their SODIS campaign in Kibera, Nairobi as referred to in the SODIS
Effective Use Write-up and received very helpful advice from SODIS experts.
A standardized formal survey of ~50 households using the ceramic pot filter, branded the
Kosim by Pure Home Water in Northern Region, Ghana, was conducted during January
2008 by Kate Clopeck. The author accompanied Clopeck for 2 days (10 household
visits) of surveying. A good deal was learnt through this process as to effectual
household monitoring methodology as well as specific survey questions and observations
that could yield quantifiable and/or replicable answers during single household visits.
The water quality results of this survey are partially presented in Appendix C on pages
160-163.
26
3. Methods − Water Quality Monitoring
3.1 Turbidity
Suspended sediment causes many critical interactions in the natural environment. Too
much suspended sediment will kill fish and prevent photosynthesis of algae, whereas too
little mud transported during the flood season can make fertile river valleys go barren. In
the context of water treatment, turbidity has the potential to block the UVA light needed
for disinfection by the sun, to transport adsorbed chemicals and pathogens directly to the
user, to cause negative odor and aesthetics, to incur sedimentation and blockages in pipes,
and even to negate the effects of chlorination.
There are a number of ways to measure particulate and dissolved matter in waters. Color
makes the first aesthetic impression, and tells a great deal about processes occurring
within the water. Dark brown translucent waters contain organic matter, harboring and
shielding microbes from disinfection treatment. Light brown or reddish opaque waters
contain mineral particles, and necessitate physical filtering or flocculation. These
suspended and colloidal particles have differing physical properties, including turbidity,
attenuation cross sections and average particle diameters.
Turbidity is a very simple and useful optical measurement. Nephelometric turbidity units
(NTU) are measured by a device called a nephelometer which has supplanted a reagent
specific method called Jackson turbidity (JTU). NTU, JTU, and optical clarity measure
optical effects (refraction and attenuation, respectively) instead of mass concentration of
particulates in water, which can be measured by total suspended solids (TSS). NTU and
optical clarity provide a proxy to mass concentration that is suitable at the low turbidities
of water that people drink.
3.1.1 Field Turbidity Measurement
NTU does not have a direct environmental interpretation like that of visual clarity (beam
attenuation), as measured by a simple device called a turbidity tube (Davies-Colley,
2001). The turbidity tube used in the field by this author (DelAgua Ltd) is measures the
depth of water where lack of clarity occurs, not unlike the familiar Sechi-disk. In the
turbidity tube method, one fills a specifically designed clear plastic tube with water until
the target (a cross or circle) at the bottom of the tube just disappears from sight. Proper
operation necessitates holding the middle of the tube at arms length and allowing time for
gas bubbles to settle out before taking a measurement. Beam attenuation is very
important in evaluating techniques like SODIS, whose bacterial inactivation depends on
sun-derived UVA light directly interacting with the microbes. One benefit of using the
turbidity tube over nephelometry is that it takes in situ measurements, because turbidity is
likely to change significantly within a few hours during storage and transport. For full
operating procedures of the DelAgua Turbidity Tube, see Appendix D.
The turbidity unit (TU) of the turbidity tube, a metric designed to mimic NTU, can be
read from the side of the tube, according to the depth of attenuation. Visual clarity
measurement using the naked eye is accurate. However, despite an attempt made to
27
calibrate TU on the tube to actual NTU measurements, TU and NTU are measuring
different light properties (attenuation and scattering, respectively) and the calibration is
not very comparable. By applying a t-Test quantitative analysis to field TU
measurements with subsequent lab NTU measurements on the same sample, Losleben
found that there is a significant difference between corresponding values of TU and NTU
and consequently the range of values is very large, despite decent R2 values as given by
the regressions on Figure 3 (Losleben, 2008).
Comparing Nephelometric Turbidity Units (NTU)
to Turbidity Units (TU)
Turbidity Tube (TU)
450
400
field data
350
MIT lab
Linear (MIT lab)
300
Linear (field data)
250
MIT lab
200
y = 0.9156x - 26.25
R2 = 0.9433
150
100
field data
50
y = 1.4938x - 122.1
R2 = 0.7259
0
0
50
100
150
200
250
300
350
400
Turbidimeter (NTU)
Figure 3 Comparison of NTU and TU
(Losleben, 2008)
Figure 4 DelAgua Turbidity Tube
(Photo: Fitzpatrick, 2008)
The human eye can easily detect turbidity of greater than 5 NTU, and thus the WHO
recommends that waters fall below 5 NTU for aesthetic acceptance (WHO, 2004). In
terms of field monitoring, the recommended turbidity tube method has a lower resolution
of 5 TU, which is more or less comparable to 5 NTU, and is the maximum recommended
turbidity for drinking water in Ghana (GSB, 1998). Water falling below the
characterization capabilities of our measurement technique is thus physically acceptable
to the user and qualifies for chlorine treatment of a single dose of either Aquatabs or
liquid hypochlorite. Sampling the lower turbidities often requires more treated water
than is available or feasible to take from the household (about 3 cups), as using the tube
contaminates treated water. Despite its limitations, the turbidity tube is well suited to
operational monitoring of HWTS, with turbidity ranges <5, 30-40, and >100NTU easily
discernible using the tube. As the correlation between TU and NTU is not strong and the
precision of TU is limited given the turbidity tube method, this thesis refers to NTU as a
more precise measure than TU.
The DelAgua turbidity tube costs about US $100 when shipping is included from
England and portable electronic Nephelometers can cost hundreds of dollars. Therefore,
28
home-made turbidity methods are well suited to low-budget monitoring campaigns, as
long as the methods can be calibrated. For example, user training for the SODIS program
includes the EAWAG-proposed method in which a 0.5-liter bottle is filled and stood
upright atop a newspaper headline. If the large black print can still be read, the water is
less than 30 NTU and suitable for treatment. If not, pretreatment through settling in a
separate container or flocculation is warranted until the newspaper headline test is passed.
Many SODIS users are non-literate, and may not have ready access to newspapers. In
this case, place one’s hand behind the bottle, and if your fingers are still visible when
looking through the bottle horizontally, enough UVA will pass through and the water is
suitable for treatment. These techniques can be particularly important for SODIS
monitoring programs, where specific turbidities are impossible to recommend due to the
variability and condition of available PET bottles. Likewise, people need measurement
capabilities in the home to conduct effective treatment with SODIS and other
technologies.
3.2 Microbial Indicators
The standard method for measuring microbiological performance of water treatment
processes involves a percent or log reduction in the concentration of a microorganism
between the influent and the effluent of the process. By testing the same water sample
before treatment and again after treatment for microbial counts, treatment efficiency can
be deduced. Such a test typically entails an indicator organism, such as total coliforms.
Total coliform (TC) bacteria comprise a diverse array of aerobic and facultatively nonaerobic, gram-negative, non-spore forming bacilli that readily grow at 35-37 degrees
Celsius given a variety of media broths (WHO, 2006).
In terms of diseases stemming from contact with contaminated water, index
microorganisms are used for their ability to identify likelihood of fecal contamination.
Fecal contamination is directly inferred through the presence of two widely used index
organisms, Escherichia coliform (E.coli) and thermotolerant coliforms, which are both
coliforms of direct fecal origin and part of the total coliform family. Thermotolerant
coliforms are culturable at higher temperatures that are lethal to other coliforms (44.5°C),
whereas E.coli can be cultured at body temperature (35° C). Both thermotolerant
coliforms and E.coli cannot grow outside of the body, and thus infer direct fecal
contamination of the water tested.
Monitoring and evaluation programs for household water treatment technologies need to
take into account not only technology performance through treatment but also the
likelihood of diarrhea (water safety) at the point of use for proper program evaluation.
E.coli and thermotolerant coliforms are usually present in too small numbers per the
normal 100 ml sampling to record statistically significant reductions through treatment,
and thus are not good indicator organisms of treatment efficiency. Likewise, certain
organisms falling under the umbrella of total coliforms can grow heterotrophically on a
variety of substrates outside of the human body, and thus can not serve specifically as
index organisms for fecal contamination. Separate index and indicator organisms are
thus needed to concurrently assess technology performance and absolute risk. E.coli
comprise >99% of the target index organism fecal coliforms, and can be grown at 35-37
29
degrees from the same lactose-based media as total coliforms. E.coli is the recommended
index organism, and total coliforms are the recommended indicator organism for
monitoring HWTS systems (WHO, 2004).
3.2.1 Microbial Quantification Methods
Two of the most common methods for quantification of total coliforms and E.coli include
membrane filtration and most probable number (MPN) techniques. With a dedicated lab
setup and trained technician for these time-intensive methods, a small number of accurate
counts are obtainable. MPN has the added advantage that it can take accurate counts with
turbid waters. If more data at less resolution is desirable within a given timeframe, 3M
Petrifilms provide a much simpler method without the need for extensive personnel
training or even an incubation oven, eliminating expensive equipment and lab space as
well. At a detection limit of 100 E.coli per 100 ml, however, the Petrifilm method can
only detect high absolute risk from E.coli (see Table 2 Risk Levels from E.coli below)
(Metcalf, 2006).
In order to surmount the inspecificity of 3M Petrifilm as well the resource intensive
membrane filtration and MPN methods, IDEXX has developed the Colilert method. This
method is widely used, and provides an affordable alternative that can gain resolution
comparable to membrane filtration methods below 100cfu/100ml (Jacobs, 1986). This
simple method requires no added lab setup than that of the 3M Petrifilm, and like the
Petrifilm can be incubated on the body and run completely in the field. Colilert refers to
a family of coliform testing products, yet the specific product referred to in this paper is
the simplest and cheapest of the Colilert methods, called the 10 mL pre-dispensed
Colilert MPN Tube. These tubes come with the growth medium already dispensed in the
sampling tubes, which last 15 months at 2-25°C on the shelf. This product requires a
10mL sample and can detect presence of E.coli down to 10 CFU per 100ml, quantifying
low absolute risk as per the 1997 WHO GDWQ. See Appendix D for the operating
procedure for 10 mL pre-dispensed Colilert MPN Tubes. The limits of detection of
E.coli by the combined Colilert and 3M Petrifilm method are shown below, in Table 2
Risk Levels from E.coli.
Table 2 Risk Levels from E.coli
Risk Level
Conforms
Low
Intermediate
High
Very High
E.coli in sample
(CFU per 100ml)
<1
1-10
10-100
100-1000
>1000
Colilert MUG+
- (Below detection)
+
+
+
# Blue Colonies
on 3M Petrifilm
0
0
0
1-10
10
(WHO, 1997; Metcalf, 2006)
Table 2 illustrates the WHO 1997 risk-based categories. At <1 E.coli CFU per 100ml,
the risk to the user from drinking water is negligible. At <10 CFU per 100ml sample,
WHO characterizes risk of waterborne disease as “low,” although diarrheal disease often
results from drinking this type of water (Metcalf, 2008). Using the 10 mL pre-dispensed
30
Colilert MPN Tube, low risk would be quantified as a negative result for the 10ml
undiluted sample used for the Colilert MUG test. With at least one CFU per 10ml
Colilert sample, “intermediate” or higher levels of risk is assessed but cannot be
quantified unless multiple Colilert tubes are used per sample (increasing cost ~US $1.40
per Colilert sample). Using 10 Colilert MPN Tubes can yield results on the order of tens
of CFU per 100ml through use of a most probable number method. This specificity is
lost at low counts on a 3M Petrifilm, given a one milliliter sample size. Only very high
risk waters can be quantified to the hundreds of CFU per 100ml using 3M Petrifilms, and
at this point the danger of contracting diarrheal disease is “high.”
Dilutions are not needed to quantify WHO E.coli risk levels with either of these methods,
negating any need for sterile lab equipment other than the sampling bags and disposable
pipettes. This is beneficial not only because the 3M Petrifilm method lacks
reproducibility at higher dilutions, but it also allows community members to conduct tests
themselves and facilitates community education units on microbial contamination of
water (Levy, 2007; Metcalf, 2008).
Measurement of the treatment efficiency achieved by an HWTS system requires multiple
visits and is fraught with challenges and potential innaccuracies. One needs to sample
the raw water at the time of addition to the filter, account for the volume displacement in
order to know when that water will exit the tap, and then undertake a subsequent trip to
the household to sample and test the treated water. Such testing is out of the scope of
most operational monitoring frameworks in terms of time, money, and intrusiveness, and
has only rarely been conducted academically. Using existent raw water in the home or at
the source during a monitoring visit as a proxy for the water fetched and used in a given
filter also incurs major uncertainties. As noted in the Safe Storage Write-up, up to 0.5
log reductions were recorded due to transport and settling, depending on source load
(Levy, 2007). While percent reductions in TC cannot be quantified on one-time
monitoring visits, if multiple visits to a given home are possible, better data can be
gleaned from usage. Taking five inlet and five outlet samples from a single filter over the
course of a week, for example, can show trends in reductions and absolute risk from
E.coli, as well as discount outliers (Lentz, 2008).
Using the 3M Petrifilm method is useful to know if the water is of intermediate or high
risk. However, if chlorine residual exists, turbidity recommendations are met for the
given treatment process, and/or Effective Use is judged through monitoring observations,
testing treated water with the Petrifilm method may not be warranted. As low risk (<10
E.coli per 100ml) is the microbial judgment of Effective Use standardized for all of the
technologies in this thesis, the Colilert method is always needed in order to make this
judgment for a given household’s system. Simultaneous testing with the 3M Petrifilm
method will more than double the overall cost of that microbial test, and has to be judged
on an individual sampling basis if funds are limited (see Table 3 Bill of Quantity for 25Household Water Testing Kit). However, when testing a system such as the biosand
filter which can be sampled directly after treatment (at the spout) as well as from treated
water in safe storage, the Petrifilm method’s total coliform results can yield much more
specificity on recontamination through storage, as numbers of TC are often more than
31
one hundred times greater than those of E.coli. In this way, the Petrifilm can help
determine handling efficiency through post-treatment resurgence of the indicator total
coliforms.
These tests are specific for E. coli, because they contain a substrate for the Betaglucuronidase enzyme produced by E. coli, but not by other coliform bacteria. The tests
yield striking results within 12-18 hours, MUG + fluorescent blue in Colilert, a blue
colony with a gas bubble on the Petrifilm. Petrifilms and Colilert tubes can be incubated
on the body, such as in the pocket or under the belt on the small part of the back. To
incubate, place up to 8-10 Petrifilms together between two cardboard pieces and wrap
together with a rubber band. The cardboard protects the Petrifilm from bending, yet
allows sufficient heat penetration. Similarly, a sock can be used to hold the Colilert tubes
close to the body without risk of them breaking. The E.coli and TC cultured by these
methods is non-toxic, and safe to humans. Sleep with them at night and results can be
obtained by the following morning. 3M needs to be sealed for moisture after opening
(masking tape), but need not be stored in a refrigerator. The Petrifilm expiration date is
for food service regulations and can be extended if properly stored (Metcalf, 2008).
Colilert tubes need to be kept sealed against moisture. For complete operating
procedures of Colilert and Petrifilm, see Appendix D.
3.3 Chlorine Disinfection
Chlorine in water most often exists in the form hypochlorous acid (HOCl Æ H+ + OCl-),
just as it does in dilute bleach solution. The long-known disinfection potential of
chlorine occurs from this weak acid’s ability to pass through both the polar and non-polar
regions of a cell membrane in its non-protonated and protonated forms, respectively.
Once inside the cell, hypochlorite’s acute toxicity kills the organism. After about 30
minutes of contact time with water, a certain amount of chlorine is used up through
interactions with bacteria and sediment particles. This amount is known as the chlorine
demand. If dosed correctly, a certain concentration of residual free available chlorine
(FAC) is left after disinfection. Free available chlorine concentration is the amount of
chlorine as hypochlorous acid (in the +1 oxidation state) per liter. The residual
disinfection capacity that is thus left over to take care of subsequent recontamination is
another advantage of chlorine disinfection.
In the absence of direct health impacts, an upper limit of 5.0 mg/L residual FAC is a
conservative guideline set by the WHO to assure adequate disinfection while providing
acceptable taste levels (WHO, 2004). Measuring the low end of FAC values is more
important to monitoring proper use of chlorine disinfection. Under the current WHO
Guidelines on Drinking Water Quality 3rd Edition, water vendors are required to provide
0.5 mg/L residual FAC after 30 minutes contact time (WHO, 2004). In an attempt to deal
with the realities of home storage due to intermittent municipalities, the CDC developed a
method that incorporates storage time into the dosing method and may be useful to the
implementers of safe storage and POU treatment campaigns. In order to avoid adverse
tastes in the water, the CDC recommends that a maximum of 2.0 mg/L FAC is present
after 30 minutes of contact time in the water. Sodium hypochlorite solution, Aquatabs,
and PUR are all meant to provide 2.0 mg/L FAC after 30 minutes of contact time. After
32
24 hours (the assumed average residence time in storage), the CDC stipulates that not less
than 0.2 mg/L FAC remain (CDC, 2005).
3.3.1 Chlorine Residual Measurement
There are many products and methods that provide varying degrees of accuracy in
measuring FAC. While some methods are expensive and time intensive, others are
cheap, easy and durable. Four of the possible methods are reviewed here.
Free available chlorine is generally unstable in aqueous solution, sensitive to direct
sunlight as well as agitation. Appropriate measurement methods must take place quickly
and easily at the household during a monitoring campaign. The simplest method is that
of a DPD test strip for free chlorine and total chlorine. One such product, HACH
“AquaChek” is a simple strip that suffers slightly from color interpretation differences
among individuals and has a lower limit of resolution of 0.5 mg/L, such that it can not
accurately quantify low residual FACs (often there is <0.5 mg/L in treated water). For
example, 6 of the 37 households showing FAC when tested during follow up visits in
Swanton’s Kosim and Aquatab study had FAC levels <0.2 mg/L (Swanton, 2008). While
these FAC levels would lack quantification by this Hach Aquachek DPD chlorine test
strip method, they would show presence or absence of chlorine nonetheless. Rob Quick
of the Centers for Disease Control’s Safe Water System says that measuring presence or
absence of FAC is the most useful metric for looking at behavior change, so the DPD test
strip method is recommended for its simplicity, cost-effectiveness, and timely results
(Quick, 2008).
As for other commonly used methods, color wheels (e.g. HACH Cat. No. 21290-00) are
generally imprecise among individual testers at low mg/L FAC, and titrators (see HACH
Method 8210) require a good deal of lab setup and time that would be impractical for a
mobile monitoring program. The most accurate, yet most expensive method is that of the
digital colorimeter (Hach Cat. No. 58700-00). This unit costs around US $400 without
reagents, has high accuracy at low mg/L FAC, is durable, battery powered and water
proof, and only takes about 3 minutes to get an accurate reading down to low FAC levels.
The pH of natural waters has a large impact on the effectiveness of liquid hypochlorite
chlorine treatment. Only one third as much of the FAC is protonated at pH 8 as at pH 7
(HOCl has a pKa of 7.46), and protonation is the key to traveling through cell walls and
consequently disinfection. The special properties of sodium dichloroisocyanurate
(NaDCC) in Aquatabs negate some of the pH sensitivities inherent in using dilute liquid
bleach (Clasen & Edmondson, 2006). The WHO guidelines for residual FAC apply up to
nearly pH 8 for liquid hypochlorite. pH considerations need to be taken into account at
the outset of any chlorine disinfectant implementation. To ensure Effective Use with
waters above pH 8, double dosing may need to be encouraged in trainings such that
higher residual FAC levels are achieved to compensate for the accompanying
disassociation. While a simple pH strip test in situ would suffice for a monitoring agent
to test water at the household, users of liquid hypochlorite would not be able to test their
own pH and thus they cannot be held responsible for ineffective treatment due to high
pH. Although microbial testing of FAC-positive water samples is most likely to turn up
33
negative, and is not even recommended by Rob Quick (2008) of the CDC Safe Water
when FAC is present, microbial testing of FAC-positive water is recommended in this
thesis in order to confirm adequate dosing to counteract the effects of high-pH and
turbidity. If microbial analyses fail the WHO-categorized low-risk metric despite having
residual FAC, a check of pH can be done to see if dosage needs to be adjusted. When
sampling potentially chlorine-treated water for microbial analysis, it is necessary to
neutralize FAC. This can be achieved by dosing the water with sodium thio-sulphate, as
commonly available in powder form pre-dosed in sterile sampling bags2.
3.3.2 Disinfection Potential with Turbidity
Slightly turbid waters ma be highly biologically contaminated and can have a very high
chlorine demand. The Sphere Project set out to produce a document of minimum
necessary standards for emergency response zones, which came out in 2004 under the
title “Humanitarian Charter and Minimum Standards in Disaster Response.” In this book,
the authors stipulated that there must exist in disinfected waters not less that 0.5 mg/L
residual FAC after 30 minutes of contact time with turbidity less than 5 NTU (Sphere,
2004). This is less stringent than the 0.1 NTU recommended maximum turbidity for
chlorine disinfection by the WHO (WHO, 2004). Both the Sphere and WHO turbidity
specifications, however, are of lower turbidity than the surface or other unimproved
source waters for which HWTS chlorination products were designed to treat. In fact, if
Effective Use was based on chlorination without filtration only at turbidities below 5
NTU, many useful applications of Aquatabs and liquid dilute bleach would be out of the
question, especially in emergency situations. Using liquid hypochlorite with no prefiltering, Crump found a 17% reduction in diarrheal incidence in waters averaging 55
NTU after treatment. In the same study, a 25% reduction was noted for waters treated
with both a flocculant and disinfectant with an average post-treatment turbidity of 8
NTU, still above the WHO and Sphere specifications (Crump et. al., 2004). Despite
having relatively high turbidities (30+ NTU), direct chlorine treatment can incur
substantial health benefits. The usage information on these products, however, requires
double dosing of visibly dirty water (>5NTU) and water from sources falling outside the
UNICEF/WHO Joint Monitoring Program classification of “improved” in order to ensure
the required residual FAC.
2
An exemplary product is 100 ml Stand-Up Whirl-Pak® Thio-Bags®, Product Number: B01402WA, US
$22 per box of 100 bags).
34
3.4 Portable Water Testing Laboratory
Table 3 Bill of Quantity for 25-Household Water Testing Kit
Quantity
Product
25
10 mL predispensed Colilert
MPN Tubes
E. coli count
Petrifilms
4-oz Stand-Up
Whirl-Pak® Thio
-Bags®,
1ml sterile plastic
pipettes
“Aquacheck”
Chlorine Foil
Singles
25
25
25
25
4
1
1
1
Cardboard strips
Plastic spreader
for Petrifilm
Battery-operated,
long wave UV
lamp
Turbidity Tube
Manufacturer
Part number
IDEXX,
Westbrook,
Maine
3M, St. Paul,
MN
Nasco,
Modesto, CA
Hach,
Loveland, CO
W200
Cost per
unit US$
1.50
Cost/25
HH US$
37.50
6414
1.20
30.00
B01402WA
0.22
5.50
0.15
3.75
0.53
13.25
27939-44
Cost of consumables: 3.60
0
3M, St. Paul,
Included with 0
MN
Petrifilms
Spectronics
15
Corp.,
Westbury, NY
DelAgua
90
Cost of hardware:
Total cost for 25 full samples:
US $90
0
0
15
90
US $105
US $195
Table 3 is adapted from the portable laboratory developed by Robert Metcalf, Professor
of Biological Sciences, California State University, Sacramento. This chart shows that
once the hardware is purchased, the cost of consumables for a full set of all three tests is
US $3.60 per sample. Most houses will not need a complete test, especially if they are
not using a chlorine product. Similarly, if Effective Use is assumed through observation
and knowledge of a clean source, high risk levels of E.coli need not be measured and use
of the 3M Petrifilm is not necessary for that household. However, for systems such as the
Biosand, treated water directly from the spout as well as treated water in safe storage
needs to be tested, incurring greater costs for an extra Colilert sample. Other methods for
measuring turbidity may be applicable that would negate the need to purchase the
DelAgua turbidity tube, as explained in this chapter and the SODIS Effective Use Writeup, greatly lowering up-front costs. If ordering Petrifilms or Colilert tubes in small
volumes or from overseas, shipping will become another significant proportion of the
cost and must be factored in. A section for reporting water quality monitoring results is
included in each of the Effective Use Checklists, as provided for each technology in
Appendix E.
35
36
4. Effective Use Write-ups of Household Water Treatment and
Safe Storage Technologies
“Effective Use” is defined as the proper operation of HWTS technologies in the home, as
instructed by the implementing organization, resulting in the production and storage of
safe water in order to limit exposure to a variety of waterborne diseases.This chapter
recounts the steps needed to perform Effective Use for the eight HWTS systems selected.
Each of the Effective Use Write-ups in this chapter provides in depth information about
treatment, safe storage, maintenance, and replacement period for a given technology in
the form of a monitoring framework. The framework develops a set of monitoring
observations and water quality tests as two independent methods of evaluating Effective
Use in the home. The reduced Effective Use Briefs are intended as the core addition of
this thesis to the compendium of indicators for the Network, and are appropriately
referenced and researched within the body of the Effective Use Write-up. For a more
explicit household monitoring survey for each technology, please refer to the Effective
Use Monitoring Checklist forms compiled in Appendix E.
Safe storage does not have an Effective Use Monitoring Form because there is no
treatment associated with it. Settling occurs pre-HWTS treatment and thus is not part of
the safe storage of HWTS-treated water. As an integral part of HWTS, however, safe
storage will be defined and included explicitly for each technology as one of the
categories of Monitoring Observation.
Some categories overlap, especially when maintenance refers to cleaning the safe storage
unit because it is built into the treatment technology. When noting hygiene, consistent
use or various other aspects of the HWTS system that fall outside the four categories of
treatment, safe storage, maintenance and replacement period, they will be contained in
the “physical inspection” category, with direct observations to make note of included.
There are many types of observations used in these monitoring frameworks, including
inspection and testing by the monitor, self-reporting by the user, prompted questions and
judging hygiene traits through proxy observations, among others. These observations
have been organized by content as they apply to Water Quality Monitoring and the five
categories included in Monitoring Observations. Organizing the paper on the basis of
content rather than type of observation is based intuitively upon how monitoring visits in
the home naturally proceed.
Measuring Effective Use assumes that the system in the household monitored is
operational and that water treated by this system is currently available both for
consumption in the household and for testing by the monitoring agent. Without treated
water available for consumption or testing, inconsistent use can be assumed, and the
reasons for this should be noted before moving onto the next household. Figueroa’s
definition will be used throughout this text when referring to “consistent use,” and will be
measured in part by her proposed metrics of having treated water on hand during
monitoring visits and/or showing residual chlorine when tested (Figueroa, 2005).
37
4.1 Safe Storage
Contamination of water often occurs in the household through handling practices, such
that improved sources often cannot guarantee provision of safe water (Wright, 2004).
Household water treatment techniques treat water that has become contaminated both at
the source as well as through domestic handling, with the goal of reducing contamination
to levels of low microbial risk, as defined by the WHO (WHO, 2004). Once treated, the
practice of safe storage is needed to retain safe water quality. Safe storage vessels are
especially designed to eliminate sources of recontamination by keeping foreign and dirty
objects (e.g., hands, ladles) out of the system. Used only for storing and dispensing
treated water, they are especially effective in conjunction with proper hygiene and
cleanliness.
Monitoring Effective Use of safe storage practices involves the observation of two
categories: proper hardware and proper practice. Hardware refers to the vessel used to
store water. With HWTS such as the CWP and SODIS, the hardware is self contained.
Other treatment techniques require additional hardware to enable the practice of safe
storage. Practices involve the use and maintenance of the safe storage containers, as well
as other hygienic measures taken in order to limit recontamination of the water after
treatment.
Three types of safe storage have been identified as pertains to this thesis:
1. Safe storage of untreated source water
2. Safe storage methods that are built into HWTS technologies
3. Separate safe storage post-HWTS treatment
In this document, safe storage will refer to specific practices related to each of the HWTS
reviewed. Apart from the process of settling, the first category of pre-HWTS safe storage
will not be specifically researched. Thus, when referring to safe storage, post-HWTS
treatment storage (types 2+3) is inferred. As safe storage within itself is not considered
adequate treatment of unsafe water, safe storage will not have an Effective Use
Monitoring Form of its own, as the other HWTS technologies have as compiled in
Appendix E, but rather safe storage will be included as a category in each HWTS
Effective Use Brief and Monitoring Form. Similarly, Water Quality Monitoring will
refer to the treated water contained in safe storage containers related to each HWTS
process.
38
4.1.1 Safe Storage Effective Use Brief for HWTS-Treated Water
Safe Storage Effective Use Brief
Monitoring Observations
Safe Storage 1. Container is used only for treated water.
2. Lids are kept on tight, and only opened for addition or pouring of
treated water.
3. Design incorporates a tap or a small sealable opening for pouring.
4. Vessel is clean, leak-free and in good condition.
5. Located indoors, out of the sun, off of the floor, in a stable position out
of reach of animals and small children.
Maintenance 1. Inner and outer surfaces as well as tap are cleaned and disinfected with
bleach or detergent using treated water on a regular basis.
2. Soap or disinfectant used to clean storage unit can be produced by user.
Replacement 1. None specified other than by the manufacturer.
Period
Water Quality Monitoring
Turbidity of <10NTU is ideal to slow settling or biofilm growth.
Turbidity
Bacteriological quality is <10 E.coli /100ml or no greater than that from the
Microbial
associated treatment process.
Testing
4.1.2 Monitoring Observation
When promoting HWTS technologies that do not have residual disinfection potential
after treatment (for example, SODIS, biosand and ceramic filters), safe storage practices
need special attention during training and monitoring to encourage Effective Use because
they provide the only protection against post-treatment contamination. In household
monitoring visits by this author, safe storage and hygienic use of products often led to
failing the “Effective Use” judgment as based on observational monitoring
characteristics. Although safe storage is explicitly noted in brief for each technology in
their Effective Use Write-up, the following set of safe storage techniques apply to all
technologies in order to best ensure safe water outcomes and reduction in diarrheal
disease.
4.1.2.1 Safe Storage
The most important aspect of using a safe storage container effectively is ensuring that it
is used solely for safe storage. Thus, a dedicated appropriate storage container must be
procured by the user separate from the container used to collect water. A proper training
program will focus on separate containers for fetching and for storage, and monitoring
should ensure such use. Hygienic conditions are also necessary when using storage units,
and training needs to focus on limiting hand to water contact, dipping ladles into the unit,
and to keep a hard cover on the unit at all times other than when adding or decanting
treated water. CDC recommends that a label with usage instructions be included on
every marketed safe storage device. Such a label should instruct on proper filling,
disinfecting, hygienic measures to ensure safety of stored water, periodic cleaning, as
39
well as suggested applications of treated water (drinking, hand washing, cleaning
utensils, and rinsing fruit and vegetables) (CDC, 2001).
The recommended design features of safe storage units were developed by the CDC in
their SWS Manual and have been included here:
1. Appropriate shape and dimensions with a volume between 10 and 30 liters so that
it is not too heavy, fitted with handles to facilitate lifting and carrying, with a
stable base to prevent overturning. If possible, a standard sized container should
be used because then dosing can be standardized. 20 liter vessels have worked
well in earlier studies. If children often carry water, the vessel will have to be
smaller or the child will need to collect water in a smaller container and pour it
into the safe storage container.
2. Durable material, resistant to impact and oxidation, easy to clean, lightweight, and
translucent. High-density polyethylene (HDPE) is often the most appropriate
material that is readily available. HDPE should be specially treated with
ultraviolet absorbers, or exposure to sunlight over time will damage the plastic
and vessels will crack.
3. An opening large enough to facilitate filling and cleaning but small enough that
even a child cannot easily insert a hand with cup or other utensil to dip out water.
The inlet should be fitted with a durable screw-on lid, preferably fastened to the
container with a cord or chain. A diameter between 6 to 9 cm is optimal.
4. A durable spigot or spout for pouring that is resistant to oxidation and impact,
closes easily, and can discharge approximately one liter of water in about 15
seconds.
5. Instructions for use of the container, disinfection of contents, and cleaning the
interior, permanently affixed to the container on material that does not deteriorate
when wet or moist.
6. A certificate that indicates the container complies with requirements of the
Ministry of Health or an equivalent appropriate authority.
(CDC, 2001)
4.1.2.2 Maintenance
Cleaning of safe storage units on a regular basis is necessary to reduce the likelihood of
contamination associated with storage. Cleaning must include the inside of the unit, the
outside, the tap, lid, and associated surfaces. One method for proper cleaning of safe
storage units was prepared by the CDC in their SWS Manual, as follows (CDC, 2001):
•
•
•
•
•
•
•
Pour 1-2 liters of water into container
Add double the usual dose of sodium hypochlorite (e.g., 2 capfuls instead of one)
Add detergent
Add hard rice grains or gravel
Agitate vigorously
Pour out solution
Rinse
40
If bleach is not available, disregard that step and continue. A cloth or sponge can be used
in place of abrasives.
When monitoring in the household, ask the interviewee the last time he or she cleaned the
storage container.
• Is there a biofilm or settled solids on the inside of the container?
• Can the water caretaker produce soap and other articles used to clean the
container?
4.1.2.3 Replacement period
Users need to replace the safe storage unit if it is cracked or leaking, or if the tap is
broken. High density polyethylene (HDPE) left in the sun can deteriorate in a matter of a
few years, but if cared for properly can be expected to last for 5-10 years of service. A
specific replacement period is not given, as deterioration and manner of wear can vary
significantly based on the design, material, and environmental conditions.
A
recommended replacement period can be determined by the manufacturer of a given
storage unit, if applicable.
4.1.2.4 Physical Inspection
During household monitoring visits, inspect the storage container as noted below.
First, note the design of the safe storage vessel.
• If being used with a dosage-dependent disinfectant, is the vessel a standard and
appropriate volume?
• Does the vessel have a tap or ability to pour for dispensing?
• Is the opening smaller than a hand (6-9cm) and covered securely with a clean hard
lid?
• Does the vessel conform to the characteristics of a safe storage unit as defined in
the 6 steps laid out in the CDC SWS Manual (see 4.1.2.1 Safe Storage)?
Location of the vessel within the home is important to pathogen re-growth and
recontamination. Is the vessel:
• Inside?
• Out of direct sunlight?
• Off of the floor?
• Stably situated?
• Out of reach of animals and small children?
Hygienic habits can also be teased out of direct observation of the storage conditions.
• Is the unit visibly dirty or leaking?
• Is a dedicated clean cup associated with the vessel for drinking?
• Is a bar of soap associated with the vessel for hand washing?
4.1.3 Water quality monitoring
Turbidity in stored water should be less than that recommended by the preceding
treatment process, however turbidities of greater than 10 NTU are likely to incur biofilm
41
and sedimentation in the container, requiring vigilant cleaning. High levels of turbidity,
sedimentation and/or biofilm within the container could show that it is being used as a
settling basin or for collection and should be inquired about, in order to ensure proper
usage.
Effective Use is achieved if water in safe storage is of the same (or better) quality than
the treated water before it enters the storage unit. While minor increases in total
coliforms during post-treatment storage are normal in systems without residual
disinfection, increases in E.coli counts denote fecal contamination and show improper
handling of the stored water. Monitoring for such recontamination can be difficult as
treatment and subsequent storage do not take place concurrently. Thus, stored water is
often incomparable to recently treated water due to source and temporal variation.
Likewise, asking the user to treat the water at the time that the monitoring staff conducts
their visits is likely to induce bias into results. Whether or not reductions can be made
note of between treatment and storage, absolute levels of greater than 10 E.coli per 100
ml of HWTS treated water in safe storage containers constitutes ineffective usage of the
HWTS system, including storage.
4.1.4 Discussion
Safe-storage forms a key component of the Center for Disease Control’s Safe Water
System, which distributes the container pictured on the left in Figure 5 Various Safe
Storage Containers below. To its right is the Oxfam container distributed in emergencies.
Both containers feature a durable and easy to clean high density polyethylene shell, small
sealable opening for daily filling, large sealable opening for periodic cleaning, and a
spigot for hands-free dispensing. All of the Potters for Peace style ceramic pot filters
incorporate a similar closed storage unit made of a bucket with a lid and spigot into their
various designs. See the Kosim filter third from the left below, with a polypropylene
storage unit as marketed by Pure Home Water in Northern Ghana. While often the
cheapest option, with no large opening for cleaning and no tap, commonly available 20
liter plastic jerrycans are of limited value as safe storage units, failing many of the criteria
in the CDC SWS Manual (CDC, 2001). Traditional clay pots as used all over the world
have been modified to include a spigot, narrow opening, and a hard tap (on the right,
below). These various models allow for evaporative cooling of the water, they maintain
adequate levels of free available chlorine and show that safe storage containers can be
produced locally and cheaply (Ogutu, 2001).
42
Figure 5 Various Safe Storage Containers
(CDC, Murcott, Murcott, WHO)
Many studies have been performed using safe storage as a primary HWTS intervention.
Two key studies are discussed here. In her PhD work in Northern Coastal Equador, Levy
monitored users during collection and transport of water, and through storage and use,
taking water samples and testing for E.coli at each juncture. While greater than 0.5 log
reductions in E.coli concentration due to die-off and settling were witnessed on average
during transport home from quantifiably high-risk sources, half of the samples
experienced a 0.2 log increase in E.coli concentration during domestic use, consistently
recontaminating to high-risk levels. Noting the variance in Wright’s 2004 meta-analysis
of post-source contamination, Levy concludes that source conditions will dictate how
much reduction or contamination occurs between source and household. Similarly, in
Pakistan, Jensen found that the amount of contamination at the domestic level always
hovered at about 100 E.coli per 100ml, independent of source-level contamination
(Jensen, 2002).
The work of both authors provides evidence of the benefits of narrow-necked containers.
Domestic levels of contamination were on average 30% lower in the narrow-necked clay
pots as compared to otherwise similar wide-necked containers in Pakistan (Jensen, 2002).
Levy also showed a positive correlation between having an opening that was too small to
place a dipper or hand inside and lowered E.coli counts (Levy, 2007).
Both authors showed that when dealing with low risk-level waters, safe storage has
positive impacts. Jensen claims that protection at the domestic level (i.e., safe storage) is
only important if water quality is <100 E.coli per 100ml. For water that is of WHOdesignated low microbial risk at the source or as treated through some type of HWTS,
safe-storage helps to ensure that recontamination does not occur at the domestic level and
thus is a central part of HWTS (Levy, 2007).
4.1.4.1 Settling
Households drawing their drinking water from surface water sources are likely to be
affected by high levels of turbidity, and pre-settling is an important treatment step in
these instances. Settling should be performed in a separate container from the safe
storage container so as to prevent a biofilm from growing in the safe storage unit.
43
Decanting settled water into a safe storage unit requires a cloth filter so as to prevent
resuspension of dirt and microbial contamination in the safe storage container (Roberts,
2001). Settling can greatly increase runtimes between successive cleanings for ceramic
as well as sand filters. Similarly, waters with lower influent loads result in lower
absolute risk levels among the treated water of biosand and ceramic filtration, given
existent treatment efficiencies (Brown, 2007). Settling may also be used to reduce
turbidity to the required levels for solar treatment or chlorine disinfection. Settling and
die-off are unlikely to consistently bring contaminated water into the low- or mediumrisk WHO designations for microbial quality (Levy, 2007; Wright, 2004). Because
settling is not meant to be performed in safe-storage containers and is not a dependable
treatment technique, it will not be included in the general framework for Effective Use of
safe storage as laid out below.
1
2
3
3
Transport and settling
container
4
5
6
Dedicated safe storage unit
Figure 6 CDC Settling Pictorial
(CDC, 2001)
44
4.2 Sodium Hypochlorite Solution
Chlorine treatment of centrally treated water dates back to the early 1900s with proven
health benefits. Promotion and marketing of household chlorine products by the Centers
for Disease Control (CDC) and the Pan American Health Organization date back to the
mid-1990s. Using a relative risk reduction of 0.49, Clasen calculated that household
chlorine use costs only US $53 per DALY averted, making it the most cost effective of
all the HWTS. With no infrastructure investment necessary and only US $0.66 per
person treated per year, chlorine solution is also among the most affordable, easiest to
produce, and most widely available forms of HWTS (Clasen, 2007).
4.2.1 Sodium Hypochlorite Solution Effective Use Brief
Monitoring Observations
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
manufacturer’s specifications, without prompting from the monitor:
1.1. Add a single dose to clear water of the correct volume.
1.2. Double dose for water that is visibly dirty and/or from an
unimproved source, following filtering through a clean folded cloth.
1.3. Shake thoroughly and let sit for 30 minutes prior to drinking.
2. Pretreatment is recommended for turbid waters.
1. Separate containers for fetching and disinfection/storage are used,
Storage
visible, clean, and have no leaks.
2. The volume for treatment as specified on the hypochlorite product is
easily measurable in the safe storage container.
3. Safe storage container for treated water is located indoors, out of the
sun, off of the floor, in a stable position and out of reach of animals and
small children.
4. Design of safe storage unit incorporates a tap or a small sealable
opening for pouring.
5. Lids are kept on tight, and only opened for addition or pouring of
treated water.
Maintenance 1. Regularly scheduled cleaning of the storage unit.
2. Soap or disinfectant used to clean storage unit can be produced by user.
Replacement 1. Expiration date as specified by manufacturer or distributor on bottle.
Period
1. Water bottles for use during travel or school are clean and producible to
Physical
the interviewer if consistent use is claimed outside the home.
Inspection
2. Unexpired sodium hypochlorite solution sufficient for at least ten
treatments is in stock and easily accessible if consistent use is claimed.
3. A dedicated clean cup is associated with the safe storage unit.
Water Quality Monitoring
Free available chlorine presence is shown if treatment is claimed.
Chlorine
Residual
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial
Testing
45
4.2.2 Monitoring Observation
4.2.2.1 Treatment
As a consumable product, there is often little ability to run trainings for users at the
outlet/street vendor level. Therefore, easily interpretable instructions for use of sodium
hypochlorite solution need to be included on the bottle. The Society for Family Health,
partner to Population Services International (PSI) in Nigeria, prints the following label
for their Waterguard (1.0% sodium hypochlorite solution) product in English:
Figure 7 PSI Nigeria Waterguard Label
(POUZN, 2007)
Each bottle is listed with a batch number and expiration date, along with the mailing
address of the manufacturer and producer (see Figure 7 PSI Nigeria Waterguard Label).
Aside from dosing instructions and product warnings, labels should promote uses of
chlorine-treated water other than just drinking, including washing hands and dishes,
rinsing fruit, and house cleaning (Lantagne & Gallo, 2008).
A single dose of chlorine solution in the suggested volume is adequate for clear water
from improved sources. Double dosing is advisable if the water is visibly dirty (at least 5
NTU), providing 4.0 mg/l total chlorine for treatment and leaving at least 0.5 mg/L free
available chlorine (FAC) after 0.5 hours as recommended by the World Health
Organization (WHO). These use and dosage directions are easily tailorable to specific
countries’ literacy rates, languages and typical storage units. See Appendix F: Sodium
Hypochlorite Solution Usage Instructions for examples of PSI labels from Kenya,
Madagascar, and Ethiopia.
46
During a household interview, ask the user to demonstrate her/his treatment techniques,
without any further prompting.
• Are they able to follow the instructions?
• If the water is visibly turbid, ask if any attempt at pretreatment is made (e.g.,
letting stand for sedimentation, or pre-filtering through cloth or other filter)?
• Before checking chlorine levels, ask them if and when this water was treated?
• How much water was treated at that time, and how much solution was used to
treat it?
• Do the two previous claims match up with the free available chlorine results?
4.2.2.2 Safe Storage
Safe storage is a necessary component of the sodium hypochlorite HWTS system. While
the Safe Storage Effective Use Write-up has much greater detail, the following safe
storage characteristics are important to note in the home of sodium hypochlorite users.
Upon entering the house for a monitoring visit, ask the user to take you to where the
drinking water is stored.
• Is a dedicated safe storage container in use, separate from the container used for
fetching water?
• Is the volume for treatment as specified on the hypochlorite product easily
measurable in the given safe storage container?
• Does the design of the safe storage unit incorporate a tap or a small sealable
opening for pouring, such as to eliminate recontamination by the introduction of
dirty objects for dipping such as ladles, cups or hands?
• Is the safe storage unit kept out of direct sunlight, as the sun quickens degradation
of residual chlorine and speeds re-growth of bacteria?
• Is the lid to the unit kept on tight, and only opened for addition or pouring out of
treated water?
• Is the unit clean and free of leaks, situated indoors, off of the floor, in a stable
position and out of reach of animals and small children?
4.2.2.3 Maintenance
Minimal maintenance is required with the use of sodium hypochlorite solution, as the
chlorine residual is effective at sterilizing containers. CAWST recommends cleaning the
storage unit at least once a week for any chlorine product (Adams, 2007). Dilute bleach
solution provides an excellent cleansing agent for use in cleaning of storage units (see
4.1.2.2 Maintenance for safe storage cleaning instructions using dilute bleach).
Inspection of the safe storage unit’s cleanliness is necessary. When in the house, ask to
see the soap or disinfectant used to clean the unit, and question the user as to when the
last time it was cleaned.
4.2.2.4 Replacement period
NaOCL is minorly unstable in liquid solution, and PSI prints an expiration date of one
year after production on its product to ensure adequate treatment. Witnessing expired
Waterguard or similar product in the household is a good sign that the user is disusing the
product or hoarding the product for special occasions (e.g., sick children or cholera
outbreaks) instead of using the product consistently. Similarly, lack of a minimal
47
chlorine presence shows that claims of consistent or active use are suspect. Another
question to ask in this vein is whether family members carry treated water or chlorine
solution while traveling. Ask the family to present water bottles, if traveling with water
is claimed.
4.2.2.5 Physical Inspection
Liquid chlorine products are consumables, and should be used on a daily basis. PSI’s
market-based strategy in Ethiopia develops stable distribution networks with visible and
well-positioned outlets, as needed for widespread and sustainable use of their liquid
chlorine products. The volume equivalent of at least 10 treatments of hypochlorite
solution must be present in the house in order to help ensure Consistent and Sustained
Use of the product.
When conducting monitoring, ask the user to retrieve their hypochlorite product, if any is
in stock.
• Is the product easily accessible, suggesting daily use is probable?
• Is sufficient supply of unexpired product present (at least ten doses)?
• If not in stock, how long have they gone without treatment and why?
Asking for a glass of water is often very informative, especially if you ask with the intent
of drinking it.
• Did the user act hygienically while getting the water?
• Alternatively, did they wash out the glass with untreated water and then dip it into
the storage unit without washing their hands? Observe behavior, as monitoring
visits can induce bias among the user’s habits.
• Is there a dedicated drinking cup or bar of soap near the storage container? These
will show a high level of hygiene, and suggest that recontamination is less likely.
4.2.3 Water quality monitoring
Quick (2008) of the CDC contends that presence/absence of FAC is the most useful
metric of behavior change with household chlorination using sodium hypochlorite
solution. Using a DPD FAC test strip, any pinkness or other indication of FAC indicates
current treatment with liquid chlorine solution and this is satisfactory to the chlorine
requirement. The WHO guidelines say that FAC levels of greater >5.0mg/L can incur
negative taste perceptions, as well as higher levels of carcinogenic disinfection
byproducts (WHO, 2006). An FAC of >5.0 mg/l witnessed in households is most often
evidence of improper dosing.
Higher FAC levels are needed if the pH of the stored water is greater than 8. However,
unless the users were instructed to dose accordingly given a chlorination implementation
in a place of naturally high pH, failure to have higher FAC is not perceived as ineffective
use of the product on their part, but rather ineffective promotion of chlorine treatment.
Thus, pH measurements are not necessary and the FAC presence/absence still holds.
Effectiveness of disinfection can further be confirmed with microbial testing results.
48
WHO 3rd Edition Guidelines specify that waters over 5 NTU are not suited for chlorine
treatment (WHO, 2004). If no other treatment option exists, chlorination of turbid water
will help to disinfect the water regardless of moderate turbidity (Quick, 2008). Double
dosing achieves acceptable residual FAC in waters that are visibly turbid (>5NTU) and/or
from an unimproved source. If raw water is measured to be >50NTU at the household,
diminishing microbial reductions from chlorination are likely and pretreatment is
warranted.
Chlorine treatment has the potential to completely eliminate E.coli counts in treated
drinking waters, and the low risk category of <10 E.coli/100 ml should be expected from
treatment with even moderately turbid waters, as tested from household storage samples
(Quick, 2008). Low E.coli counts have been found to correlate well with the existence of
residual FAC (Arnold, 2007).
49
4.3 Aquatabs
Aquatabs are a specifically formulated and branded solid form of sodium
dichloroisocyanurate (NaDCC). This product is produced by Medentech in Ireland under
strict pharmaceutical regulations and comes in many sizes for different treatment
regimes. As a household water treatment and safe storage (HWTS) product, Medentech
produces a 67mg NaDCC tablet which treats twenty liters of clear water. This specific
67mg product will only be referred to in this text. NaDCC produces the same active
disinfection ingredient as other chlorine products, but has a few advantageous properties
compared to sodium hypochlorite. NaDCC is stable in Aquatabs form as a solid, making
storage, handling, shelf life, and transport much easier than with liquid bleach. In
solution, NaDCC produces HOCl as a disinfectant, but withholds half of the potential
free chlorine in a stored, inaccessible form until its use is demanded. This is especially
useful to work around the pH sensitivities inherent in dilute liquid bleach. Aquatabs have
acidic constituents that lower pH and increase effective disinfection as well. Aquatabs
are hard to produce and cost a bit more than dilute bleach per health impact for all of
these material benefits (Clasen, 2006).
(Photo: Swanton, 2008)
Figure 8 Aquatabs
50
4.3.1 Aquatabs Effective Use Brief
Aquatabs Effective Use Brief
Monitoring Observations
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
Medentech, without prompting:
a. 1 tablet per 20 liters of clear water
b. 2 tablets for 20 liters visibly turbid water
c. Let sit 30 minutes before consumption.
2. Pretreatment is recommended for turbid waters
Safe Storage 1. Two separate 20 liter containers for fetching and disinfection/storage are
used, visible, clean, and have no leaks.
2. Safe storage container for treated water is located indoors, out of the
sun, off of the floor, in a stable position and out of reach of animals and
small children.
3. Design of safe storage unit incorporates a tap or a small sealable
opening for pouring.
4. Lids are kept on tight, and only opened for addition or pouring of
treated water.
Maintenance 1. Regularly scheduled cleaning of the storage unit.
2. Soap or disinfectant used to clean storage unit can be produced by user.
Replacement 1. Product expires 5 years after date of manufacture, as printed on Aquatab
sleeve.
Period
1. Water bottles for use during travel or school are clean and producible to
Physical
the interviewer if consistent use is claimed outside the home.
Inspection
2. At least one sleeve of ten non-expired tablets is in stock and easily
accessible.
3. A dedicated clean cup is associated with the safe storage unit.
Water Quality Monitoring
If raw water is ≥80 NTU, pretreatment should be witnessed or emphasized.
Turbidity
Free available chlorine presence is shown if treatment is claimed.
Chlorine
Residual
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial
Testing
4.3.2 Monitoring Observation
4.3.2.1 Treatment
Instructions for treatment with Aquatabs are included on the sleeve of ten tabs as sold at
the outlet. Medentech prints the following information on the ten-tab sleeves:
“NaDCC 67mg Use one tab to treat 20 litres of clear water in a jerrycan.
If the water is dirty, filter it first with cloth then treat with two tabs.
Close your jerrycan and wait for 30 minutes before use. Do not swallow
the tablet. Medentech, Ireland. Distributed by Precision dx Ltd.”
51
On the reverse side, batch number and expiration date are listed. All the information is
printed for every two tabs on the ten-tab sleeve. (The specific sleeve used for
transcribing these instructions was manufactured by Medentech for their Ghanaian
distributor, Precision dx Ltd.)
Double dosing 20 liters with two 67mg Aquatabs is advisable if the water is visibly dirty
(at least 5 NTU), providing 4.0 mg/l total chlorine for treatment and leaving at least 0.5
mg/L free available chlorine (FAC) after 0.5 hours as recommended by the World Health
Organization (WHO). A study conducted by the Tanzanian Ministry of Water and
Livestock Department found that treatment of 47 NTU shallow well water in Tanzania
with 500 E.coli CFU/100 ml resulted in complete reductions to zero plate counts and
conformity with WHO standards from initial raw water counts of 20,000 total coliform
and 500 E.coli (Mjengera, 2005). In their document “Emergency and HWTS use of
Aquatabs,” Medentech advises users to pre-treat raw waters of turbidity above 80 NTU
with methods of settling, filtration, or flocculation before treating with two 67mg
Aquatabs in 20 liters of water.
4.3.2.2 Safe Storage
Although Aquatabs provide residual disinfection throughout use, safe storage practices
are a necessary component of the Aquatabs HWTS system. While the Safe Storage
Effective Use Write-up has much greater detail, the following safe storage characteristics
are important to note in the home of sodium hypochlorite users. One key storage
observation is whether the storage unit is placed in direct sunlight. Although the half-life
of free available chlorine (FAC) exposed to sunlight is increased by an order of
magnitude with Aquatabs over that of sodium hypochlorite solution due to stabilization
with cyanuric acid, direct sunlight on storage vessels will eventually drive out residual
chlorine, eliminating residual disinfection and is to be avoided (Kuechler, 2004).
Upon entering the house for a monitoring visit, ask the user to take you to where the
drinking water is stored.
• Is a dedicated safe storage container in use in which can easily be measured 20
liters, separate from the container used for fetching water?
• Does the design of the safe storage unit incorporate a tap or a small sealable
opening for pouring, such as to eliminate recontamination by the introduction of
dirty objects for dipping such as ladles, cups or hands?
• Is the lid to the unit kept on tight, and only opened for addition or pouring out of
treated water?
• Is the unit clean and free of leaks, situated indoors, off of the floor, in a stable
position and out of reach of animals and small children?
4.3.2.3 Maintenance
Minimal maintenance is required with the use of Aquatabs, as the chlorine residual is
effective at sterilizing containers. CAWST recommends cleaning any safe storage unit
prior to initial treatment and at least once a week for any chlorine product (Adams, 2007).
Even without performing the added task of regular cleaning, if the vessel is covered with
52
a hard lid and residual chlorine is maintained, then the storage unit should remain in a
suitable condition (Edmondson, 2008).
4.3.2.4 Replacement Period
As NaDCC is stable in solid form, Aquatabs have a shelf life of 5 years, regardless of
storage humidity and sensitive only to extreme heat. Household possession of expired
Aquatabs is a potential sign that disuse or hoarding may be taking place. If Aquatabs are
found to be expired, local distributors’ supplies might need to be checked for being past
their expiration dates.
4.3.2.5 Physical Inspection
Aquatabs are a consumable product, and are intended to be used on a daily basis.
Medentech’s sourcing of in-country for-profit distributors is a strategy positioned to
develop stable distribution networks with visible and well positioned outlets, as needed
for consistent and sustainable use of Aquatabs. At least one sleeve of ten non-expired
Aquatabs must be present in the house and preferably partially used in order to help
ensure Consistent and Sustained Use of the product. Checking household stocks and
expiration dates is necessary in a monitoring campaign. Another useful check to ensure
consistent use is the presence of any chlorine (free or total) in “treated” water. Lack of a
minimal chlorine presence shows that claims of recent treatment, correct treatment, or
consistent use are suspect. Another good question to ask is whether family members
carry treated water or Aquatabs sleeves while traveling. Asking the family to present
water bottles in order to back up their answers to consistent use while traveling can lend
be informative.
Notice the level of hygiene implicit in the water handling habits of the given user. Users
can be prompted to fetch a glass of water to aid in this endeavor. A dedicated clean cup
associated with the safe storage unit shows a decent level of hygienic practice.
4.3.3 Water quality monitoring
Despite WHO regulations that waters should be under 5 NTU for regular chlorination, no
upper limit to turbidity is set for Aquatabs, based upon the various studies done by
Medentech. However, treating water above 80 NTU is likely to have diminished results,
necessitating pretreatment. If turbidity is visible or measured as greater than 5 by the
monitor, two tabs should have been used to treat the water.
Disinfection with NaDCC is the sole control measure of Aquatabs. The WHO (1993)
stipulates that at least 0.5mg/L FAC remains after 30 minutes contact time. As long as
0.2 mg/L FAC exists in water 24 hours after treatment, sufficient residual disinfection
potential exists (CDC, 2005). Assuming that unreasonable recontamination has not
occurred (this can often be loosely confirmed through physical observation of user
habits), using a DPD FAC test strip, any pinkness on the Free Chlorine test indicates
treatment with Aquatabs and this is satisfactory to the chlorine requirement.
Effectiveness of disinfection can further be confirmed with microbial testing results,
although CAWST claims that microbiological testing is only needed if no free chlorine
can be measured (Adams, 2007).
53
While free chlorine is very successful at inactivating bacteria in clear water (~4 log
removal), cryptosporidium and Mycobacterium have shown resistance to disinfection. In
water of >10 NTU, 1.8-2.8log reductions in bacteria have been noted through chlorine
disinfection (Schlosser et al., 2001). Simple monitoring campaigns as described here
have limited ability to accurately quantify log reductions from raw to treated water given
the time delay between fetching, treating, storing and using water in the home, and thus
measuring concentrations of E.coli in treated waters provides a potential additional water
quality measurement beyond chlorine residual. Almost all of Medentech’s collected
literature concerning field microbial testing reports E.coli counts of <1 CFU/100ml.
Molla reported 84% of the households surveyed out of 50 households provided with
Aquatabs for a month had no shows of fecal coliform (Molla, 2005). Given such low
showings of E.coli in the field, measurement of less than 10 E.coli CFU/100ml shows
that total treatment worked as intended, verifying that the control measure was correctly
implemented by both the user and the technology and that low risk to the user has been
achieved (Moran, 2008).
4.3.4 Discussion
In both Ethiopia and Ghana, Medentech has paired with a single distributor, giving this
company sole-rights to import and sell Aquatabs. Aquatabs are produced by Medentech
under strict pharmaceutical regulation in Ireland and sold under distribution agreements
to national companies with a strong track record in related consumable goods. Once
imported in bulk by the distributor, the sleeves are repackaged with that company’s logo
and user instructions are reprinted in the given language. Medipharm is the distributor in
Kenya, and their packaging is shown in Appendix F: Aquatab Usage Instructions, along
with usage instructions from Medentech. Medentech also works with Population
Services International (PSI), the not-for-profit social marketing organization, in countries
throughout Sub-Saharan Africa and South Asia.
Easier to dose than liquid chlorine, only the dirtiness and volume of water needs to be
judged to use Aquatabs. In a study conducted in Tanzania, 70% of FAC measurements
taken at the household were within WHO limits of 0.5 to 5.0 mg/l, showing a high level
of accuracy in dosing. 27% of results were reported as below 0.5 mg/l, some of which
would have qualified as correctly dosed, depending on the time elapsed after treatment
(Medentech, 2006). FAC levels higher than the inherent 2 mg/l dosing given the 67mg
tablet have often been recorded in treated water (Swanton, 2008). Such a result is
attributed to using a full tablet on less than 20 liters of water, which may be a common
practice when less than 20 liters of water are available for drinking and treatment. The
upper limit of 5.0 mg/l FAC in drinking water set by the WHO as a guideline value is
recommended for lifetime consumption. Over-dosage is not a problem on a short-term
basis (WHO, 2006). The guideline values of NaDCC that has been derived is well above
the recommended maximum dosage of 8.5 mg/L (using two tabs in 20 L) (Edmondson,
2008). Aquatabs with NaDCC have been found preferable to similarly dosed sodium
hypochlorite solution (HOCl) in a number of field-based taste tests (De Angelis, 1998).
Potential overdosing leading to high levels of disinfection by-products such as
trihalomethanes is limited with Aquatabs due not only to the ease with which it is dosed,
54
but also the reduced production of such byproducts by NaDCC as compared with liquid
bleach (Macedo, 1997).
Training is often minimal when Aquatabs are sold to the consumer at the kiosk or
pharmacy, so ease of use is an important feature of this product. No mixing is needed
with 67mg Aquatabs. The effervescent nature of Aquatabs allows the FAC to distribute
itself homogenously throughout the storage vessel without the need for introducing
foreign dirty objects for stirring or spillage from shaking. Neither in training materials or
usage instructions is a time limit for use of treated water specified with Aquatabs. Given
the recommended dosing, treated water meets the WHO and Center for Disease Control
standards of 0.5 mg/L FAC after 30 minutes contact time and >0.2 mg/L FAC after 24
hours, respectively, as shown in field studies from multiple countries. For example, an
average of 0.79 mg/L FAC was shown after 2 days in Vietnam (Chau, 1996). Joe Moran
of Medentech, Ireland claims that no timeline for consumption is recommended because
behaviors concerning usage of treated waters are not controllable by the distributors of
the product, and people are expected to use the water as they would normally do despite
recommendations to the contrary. Such a lack of stipulation is not unreasonable, as the
average time to use water is daily. Upon questioning the distributors in both Ethiopia and
Ghana, this lack of a stipulation was confirmed by their non-use of such a guideline in
training and dissemination.
A Brazilian Government study showed a 44.5% reduction in stool parasites over the
course of a one year Aquatab intervention among 618 participants (Ministério da Saude,
1996). Such reductions represent great quantitative evidence of the health benefits of
using Aquatabs on a regular basis. Through monitoring as laid out in the preceding
paragraphs, Effective Use of Aquatabs can be maintained and improved such that quality
of treatment increases and greater individual health benefit ensues.
55
4.4 SODIS
Solar disinfection (SODIS) is a point of use water treatment method that disinfects
through a combination of direct radiative inactivation, indirect photolytic degradation,
and moderate pasteurization with increased temperatures. SODIS treatment has been
shown to inactivate bacteria, viruses, and protozoa including cryptosporidium and giardia
(Wegelin, 1994; Mendez-Hermida, 2005). SODIS was originally investigated at the
American University of Beirut during the 1970s as an efficient way to disinfect water for
use with oral rehydration therapy (Acra, 1984). A serious and prolonged effort to study
and promote SODIS has been undertaken by the the Swiss Federal Institute for Aquatic
Science and Technology (EAWAG) since 1991. SANDEC, the Department of Water and
Sanitation in Developing Countries at EAWAG has contributed a great deal of research
concerning microbiological efficiency, health impact assessment, and material testing as
well as international advocacy, collaboration, and training of SODIS programs (Wegelin,
1994). SODIS has also been studied by Masters of Engineering students at MIT in
Nepal, Haiti, and Ghana (<http://web.mit.edu/watsan) as well as a number of other
academic institutions.
56
4.4.1 SODIS Effective Use Brief
SODIS Effective Use Brief
Monitoring Observations
1. User demonstrates correct knowledge of treatment, without prompting:
Treatment
a. Fill clean bottles with raw water and close lid tightly.
b. Place the bottles on a corrugated iron sheet or on the roof, and in
a place with continuous direct sunlight throughout the day.
c. Leave in direct sun from morning to dusk. If ≥50% overcast,
leave out for 2 days.
2. Use of clean and clear PET bottles that are ≤5 liters in volume and not
heavily scratched.
Safe Storage 1. SODIS treatment bottles provide post-HWTS treatment safe storage,
and thus need to have no leaks and be kept clean, stored in a safe
location out of reach of animals and small children, with lids kept on
tight.
Maintenance 1. Clean the bottles with soap and a bottle brush if available if you observe
the formation of algae on the inner side of the bottle.
2. Soap or disinfectant used to clean the bottles can be produced by user.
Replacement 1. Replace bottles when heavily scratched, opaque, or leaking.
Period
2. Treated water is available, and if weather conditions permit, water is
Physical
currently being treated.
Inspection
3. A dedicated clean cup is associated with the safe storage unit.
Water Quality Monitoring
If when one’s hand is placed behind a full bottle laying horizontally and the
Turbidity
fingers are still visible, then the turbidity requirement is satisfied and water
can be adequately treated. Pretreatment to reduce turbidity is needed if
fingers cannot be seen.
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial
Testing
57
4.4.2 Monitoring Observation
4.4.2.1 Treatment
Treating water with SODIS is straight forward, although there are a few key aspects to
keep in mind. Below is the schematic developed by SANDEC and published in their
SODIS Manual.
(Meierhofer, 2002)
Figure 9 SODIS Usage Pictorial
These directions are clear and simple. After making sure that the bottle is clean, fill the
bottle ¾ full with water, close, and shake it for 20 seconds to enhance aeration.
Completely fill the bottle, and seal tightly. During the morning, place the bottles on a
firm darkened or reflective surface, preferably a clean corrugated iron roofing sheet that
is raised off of the ground. When placing the bottles, ensure that they will be exposed
58
directly to sunlight for the entire day. If over 50% cloud cover persists, leave the bottles
out a second day (http://www.sodis.ch/Text2002/T-Howdoesitwork.htm). Retrieve the
bottles at dusk and the water is ready for consumption. SODIS is not recommended for
rainy days.
Leaving a few inches of airspace and shaking the bottles vigorously prior to sun exposure
has been recommended by a number of agencies, including the Global Research Institute
as in the attached instructions (See Appendix F: SODIS Usage Instructions). SODIS
deactivates microbial contaminants partially through the creation of reactive oxygen
species by indirect photolysis. Matthias Saladin of the Fundación SODIS in Bolivia no
longer recommends shaking the bottles, as natural waters have the requisite 3 mg/L of
dissolved oxygen and adequate agitation occurs through pouring into the bottle (Saladin,
2008). He also recommends putting the bottles out for the whole day, as most SODIS
users do not possess clocks. Accordingly, he proposes a simpler five-step usage
framework that can be viewed at www.fundacionsodis.org/en.
4.4.2.2 Safe Storage
Ensure safe storage practices by using the SODIS treatment bottles themselves as safe
storage containers. SODIS treatment bottles provide post-HWTS treatment safe storage,
and thus need to have no leaks and be kept clean, stored in a safe location out of reach of
animals and small children, with lids kept on tight. Secondary safe storage containers are
not recommended because SODIS treatment does not provide any residual post-treatment
disinfection potential, unlike the various chlorine products.
4.4.2.3 Maintenance
Proper maintenance requires regular cleaning of the bottles. KWAHO explicitly
recommends cleaning of the bottles prior to the first usage. Although the bottles are
subject to the daily disinfection process, cleaning bottles with brushes and soap is
necessary from time to time to remove algae that may grow on the inner surface, as the
film formed impedes UV-A transmittance. The extent of algae growth is partially
dependent on the quality of the local source water.
4.4.2.4 Replacement period
Usage and exposure to the environment tends to incur scratches that can block a large
percentage of the UV-A disinfection potential, and heavily scratched or opaque bottles
need replacement. Non-sealable or leaking bottles need to be replaced for sanitary
reasons, as well. No firm timeline is recommended for bottle replacement, as it will be
situation dependent.
4.4.2.5 Physical Inspection
Household monitoring of SODIS needs to be conducted on days without rain in order to
directly witness use. When visiting a home, ask to see bottles undergoing treatment:
• Are an adequate number of bottles (2 per person) currently being treated
(Meierhofer, 2002)? Current treatment can help ensure claims of consistent use,
an important component of reducing the likelihood of diarrheal disease.
59
•
Are bottles undergoing treatment lying on their side, positioned in direct sunlight
throughout the day, and on a clean surface off of the ground? Some regular users
do not place bottles outside for treatment everyday, although they should have
treated water on hand (Saladin, 2008).
Within the house, ask the person in charge of water treatment (if available) to explain
how they treat the water, making sure to note the directions listed in the Treatment
section. While in the house, inspect the storage and hygiene conditions.
• Are the bottles made of PET, less than 5 liters in volume, clean and not heavily
scratched with all labels removed? Leaks can also be assessed.
• Are bottles clean on both the inner and outer surfaces?
• Does the user have a suitable system of bottle rotation that can allow for bottles to
be exposed for two days depending on the weather while providing sufficient
drinking water for the household?
• Is treated water available for consumption in the house?
• Do users carry the treated bottles to work or school, incurring consistent use?
• Is a clean cup present for individual drinking use?
4.4.3 Water Quality Monitoring
Turbidity reduces the transmittance of UV-A radiation, and therefore it is recommended
to pretreat water of turbidity greater than 30 NTU. Testing of turbidity can be achieved
with a Turbidity-tube by the monitor. Pre-settling before addition to the SODIS bottle
can be encouraged, but results will vary. User training for the SODIS program should
include the EAWAG-proposed method in which a 0.5-liter bottle is filled and stood
upright atop a newspaper headline. If the large black print can still be read, the water is
less than 30 NTU and suitable for treatment. If not, pretreatment through settling in a
separate container or flocculation is warranted until the newspaper headline test is passed.
Many SODIS users are non-literate, and may not have ready access to newspapers. In
this case, instruct users to place one’s hand behind the bottle, and if your fingers are still
visible when looking through the bottle horizontally, the water is suitable for treatment.
This technique needs confirmation, yet has the added advantage of confirming light
transmittance through both the water and bottle width (Saladin, 2008). These techniques
can be particularly important for monitoring programs, avoiding the wastage of water
necessary to fill the Turbidity Tube. Measurements of turbidity in a household’s SODIS
bottles need to be taken throughout varying climatic seasons in order to fully judge
effective pretreatment and applicability.
Exposing natural waters that contain nutrients to sunlight and enhanced temperatures
creates conditions under which many bacteria can multiply. While not producing a sterile
water, SODIS treatment has been shown to achieve the intended die-off of pathogenic
microorganisms, as shown through SANDEC’s multiple results of zero fecal coliforms
after treatment (EAWAG/SANDEC, undated). Accordingly, reductions in total coliforms
need not be monitored with the use of SODIS. Simple monitoring programs as laid out
here have very little ability to accurately quantify log reductions from raw to treated
water given the time delay between fetching, treating, storing and using water in the
60
home, and thus absolute numbers of E.coli in treated waters will be measured. The goal
of SODIS is to incur low microbial risk as defined by the WHO, and thus waters treated
effectively should result in <10 E.coli per 100 ml of sample when tested.
4.4.4 Discussion
With minimal hardware cost both to the user and the implementing agency, SODIS is a
very cheap method with the potential for great health impact. Randomized controlled
studies have been conducted that show reductions in diarrheal disease comparable to
many of the other HWTS. In one such study, Rose reported a reduction of 40% among
100 users and a high acceptance rate among female users (Rose, 2006). Clasen reports a
cost of US $61 per DALY averted, putting SODIS on par with the most cost effective
HWTS intervention, dilute bleach solution (Clasen, 2007). SODIS has the added
advantage of being a self-contained safe storage unit that is available worldwide.
Reusing a large number of water bottles can reduce the burden of rubbish accumulation,
keeping scarce land free from debris in crowded urban dwellings.
SODIS treatment requires sustained incident solar radiation of 500 W/m2 for 5 hours for
adequate microbial inactivation (EAWAG, 1997). While semi-arid regions between
latitudes 15°N and 35°N have ideal solar activity throughout the year, the majority of
developing countries lie between 35N and 35S and often have adequate sunshine for
SODIS treatment as well (Tech Note 5).
The effects of turbidity and bottle type on UV-A transmittance within the bottle have
been well studied. As to be expected, UV-A radiation is reduced through absorbance and
dispersion as it travels through water. Only 50% of incident UV-A makes it to a depth of
10 cm in water of 26 NTU, prompting SANDEC to recommend a 30NTU upper limit of
turbidity for SODIS treatment, as well as containers that are at most 10 cm in depth (Tech
Note 7). In terms of material recommendations, translucent polyethylene (PE) bags have
been shown to inhibit UV-A transmittance less than bottles made of glass or polyethylene
terephthalate (PET), a polyester. However, PET bottles can transmit an acceptable 70%
or more of incident UV-A light and are much more available. Chemically, both PET and
polyvinyl chloride (PVC) contain additives such as UV-stabilizers. While these
stabilizers are largely immobile and pose minimal health danger, PET contains much less
stabilizers than PVC and is thus preferable (Tech Note 2). The plasticizers used in PET,
di(2-ethylhexyl)adipate (DEHA) and di(2-ethylhexyl)phthalate (DEHP) are also of
possible concern. SANDEC has shown that SODIS treated water contains concentrations
of these plasticizers on the order of 1 to 3 logs below the WHO guideline values.
Similarly, acetaldehyde and formaldehyde concentrations posed little health risk
(http://www.sodis.ch/Text2002/T-PETBottles.htm). Based on their greater durability,
availability and suitable chemical properties, only clear PET bottles of less than 5 liters in
volume have been recommended for use in SODIS applications (Saladin, 2008).
Temperatures above 50°C are lethal to many organisms, including cholera, giardia cysts,
and schistosomas eggs over the course of an hour (EAWAG/SANDEC, undated). In
combination with UV-A radiation, synergistic treatment effects occur at temperatures
above 50°C, resulting in increased treatment potential. SANDEC has developed a
61
reusable paraffin-based sensor that is placed inside the bottle and melts at 50°C,
indicating that a water temperature of 50°C has been reached. However, SODIS is
effective also at water temperatures below 45°C due to the effect of UV-A radiation only
(bacteria, viruses and cysts of Giardia and Cryptosporidium are disinfected). Due to
ongoing research on synergistic effects of UVA radiation and heat inactivation,
Meierhofer’s and Metcalf’s recommendations, and increased burden on the user, the use
of a temperature indicator is not necessary. Decreasing treatment times based on
synergistic effects is not recommended. Similarly, SODIS proponents no longer
recommend painting one side of the bottle black, but rather recommend placing the
bottles on corrugated zinc-plated iron roofing in order to increase reflection, heat, and
sanitary conditions (Baffrey, 2005).
62
4.4 Cloth Filter
Cloth filtration is an ancient water treatment technique, dating back to at least 500 BCE
in India, yet in recent decades has been found to be particularly effective against specific
pathogens with large carrier hosts.
4.5.1 Cloth Filter Effective Use Brief
Cloth Effective Use Brief
Monitoring Observations
1. Fasten cloth to water storage vessel and tighten string, using same side
Treatment
up each time (for Guinea Worm Cloth).
2. Fold sari cloth at least 4 times and wrap tightly around rim of storage
vessel inlet.
3. Filter all water that is fetched immediately at source or upon returning
home from the source.
4. Use filtered water for all domestic water uses.
5. Always use manufactured cloth filters with the same side up.
Safe Storage 1. Maintain separation of filtered water from non-filtered water.
Maintenance 2. Rinse off filter after each use, with a final rinse of cloth filtered water
and then leave cloth in the sun for decontamination.
3. Clean cloth filter with soap regularly.
4. Soap or detergent used to clean cloth filter can be produced by user (if
applicable).
Replacement 1. Replace filters when visible tearing or holes occur.
Period
1. User stores cloth filter in a safe and accessible place.
Physical
Inspection
2. Cloth filter is clean and without tears or holes.
3. User correctly describes or enacts use and cleaning.
4. User knows where to get a new cloth filter (if bought or distributed).
Two current cloth filter applications include use of the sari cloth in Bangladesh to combat
cholera and the use of cloth filters in the Guinea Worm Eradication Program (GWEP) in
Africa. In the mid 1990s, Huq showed that 99% of cholera parasites (those bound to
planktonic copepods) were removed by quadruple-folded sari cloth in Bangladesh (Huq,
1996). Since then, there has been considerable press coverage of this seemingly simple
intervention, with Dr Claire-Lise Chaignat, coordinator of the World Health
Organization's global taskforce on cholera control claiming "The method can save
thousands of lives during massive epidemics, particularly those of children under the age
of five" (BBC, 2003). In a 133,000 person study conducted over three years in
Bangladesh, Colwell found a 48% reduction in cholera incidence accompanied with a
reduction in severity of the cases. She also claimed high cultural acceptability and 90%
correct usage among the intervention group. Mothers in the study often self-reported
lower disease burden within their own households, which has positive implications for
effective and sustained use of cloth filtration (Colwell, 2003).
63
Similar to that of cholera, the vector of dracunculiasis (guinea worm) is a water-borne
cyclops that can be filtered by cloth. The Carter Center distributed 1.4 million filters
from Jan. 2003 to June 2007 in Ghana alone, and in combination with vigilant rural
monitoring and chemical treatment of affected water sources with Abate®, the cloth filter
intervention has helped to incur a 91% reduction during the peak transmission season in
the first quarter of 2008 as compared to 2007 (GW Wrap-up, 2008). The reduction in
guinea worm incidence is so pronounced that the GWEP expects world-wide eradication
within the next few years. Cotton cloth filtration is also a component of the PUR
flocculation/disinfection system, as described in the PUR Effective Use Write-up.
4.5.2 Monitoring Observation
4.5.2.1 Treatment
Household use of the cloth filter requires simple training, as demonstrated in the
following schematic for sari cloth or similar homemade cloth filters. This pictorial
illustration was developed by the Centers for Disease Control (CDC) and published in
their Safe Water System Handbook.3 Older sari cloths had an effective pore size of 20μm
when folded 4 to 8 times, as recommended by Colwell and shown in the diagram below
(Colwell, 2003). In Bangladesh, where Colwell’s study took place, women fetch water
by dipping their kalash water containers into streams and standing water, such that tightly
covering the inlet is very important to ensure that targeted contamination does not bypass
filtration, also shown in the diagram below.
Figure 10 (a) CDC Cloth Filter Usage Schematic; (b) GWEP Filter in Ghana
(CDC SWS Handbook; Murcott, 2007)
3
http://www.cdc.gov/safewater/manual/sws_manual.pdf
64
The following instructions are specific to use of the Vestegaard Frandsen S.A. 30 inch,
120 micron pore size nylon and cloth filter unit as employed by GWEP, Ghana (Murcott,
2007):
• Fasten cloth to wide-mouth storage vessel and tighten string.
• Pour source water into the center of the cloth
• Allow the poured water to pass through the cloth before adding more.
• Do not exceed the capacity of the cloth to strain the water.
Important to the use of the Vestegaard guinea worm cloth filter, pictured on the right
above, is that it is used with the same side-up every time so as to prevent contamination
from previous uses.
While training on appropriate usage of cloth filters is straightforward, vigilant monitoring
is often needed to promote the sustained and consistent use of the filter every time water
is fetched, and to ensure that filtered water is used for all domestic water uses, not just
cooking and drinking (Aikhomu, 2000).
4.5.2.2 Safe Storage
Filtering all of the water brought to the house and/or maintaining separation of filtered
and unfiltered water will ensure adequate safe storage of cloth-filtered water. If all water
is immediately filtered at the source or following transport to the home, the difficulty of
maintaining separation of filtered and non-filtered water through safe storage is greatly
diminished. As other microbial contamination from waterborne bacteria, viruses or
protozoa is not removed by cloth filtration, safe storage does not ensure the quality of
cloth filtered water. Secondary treatment is often required to reduce the likelihood of
diarrheal disease.
4.5.2.3 Maintenance
Maintaining a filter has two goals, to prevent both tears and re-contamination. To
prevent cloth filters from tearing, it is important that they are kept in a safe and clean
location that is easily accessible for daily use. Vigilant user-inspection for small holes
goes hand-in-hand with awareness of tearing. Keeping the filters off of the ground is a
key training lesson in the prevention of recontamination. The user should also be
informed of where to find a replacement cloth filter, if the current one tears or is spoiled.
While cleaning techniques are tailorable to the material and make of the cloth filter, it is
important to rinse them off after each use, with a final rinse of cloth filtered water and
then leave them in the sun for decontamination (Colwell, 2003). Occasional cleaning
with soap/laundering of sari cloths is recommended. Cloth filters do not clean the water
of bacterial contamination, and thus regular cleaning of storage units is useful to keep
biofilm from accumulating as a result of nutrient-rich treated waters. As previously
mentioned, additional treatment for other microbial contamination may be needed as well
4.5.2.4 Replacement Period
Aikhomu et al. (2003) found that distributed cloth filters often do not last a full
transmission season. The Carter Center Guinea Worm Eradication Project’s community
65
volunteers’ vigilant monitoring campaign (which visits each household at least once per
week) advises that it is important not only to quickly report and treat individual cases of
guinea worm, but also to inspect and replace faulty cloth filters, providing a visible
contact for whom community members can turn to when their filters are spoiled. Such a
large network of constant monitoring resulted in 100%-witnessed Effective Use of the
cloth filter in the 50 or so households visited by this author near Tamale, Ghana in
January, 2008. Multiply-folded sari cloths, while harder to inspect than Vestegaard’s
guinea worm cloth filter, adds multiple layers of protection and is easily replaceable in
Bangladesh when worn out or torn. In fact, older sari cloths are recommended as their
effective pore size is smaller due to moderate wear.
4.5.2.5 Physical Inspection
When monitoring at the household or water source, ask the person who fetches and/or
treats water to see their cloth filter. Where do they store the filter? Does the storage area
provide adequate protection from tearing? If the filter is readily accessible and clean, this
suggests possible consistent use. Personally inspect the filter. Pin prick holes in
manufactured filters may not be perceived as problematic, yet they represent highly
increased likelihood of disease transmission. Ask the user to show you how they operate
the cloth filter, if possible. Do they effectively cover the inflow of the water storage
container? Ask them the last time the cloth was cleaned and the method of cleaning to
make sure that previous training and common knowledge match up (Hernandez, 2008).
4.5.3 Water Quality
Other than the specific pathogens targeted by the cloth filter intervention (e.g., cholera
and guinea worm vectors), which are very difficult to measure, no noticeable changes in
water quality are likely to occur. While subsequent treatment will most likely be needed
to make cloth-filtered water microbiologically safe to drink, straining through a cloth is
often very important to other treatment methods such as PUR. Turbidity reduction is
typically unnoticeable, and thus no turbidity measurement is needed. Cloth filters will
not affect E.coli concentrations, so indicator bacteria testing as called for in other
interventions is irrelevant here. Visually, the water should be free from large suspended
debris if filtration is claimed by the user, although the naked eye cannot see particulates
on the order of 100μm or less.
66
4.6 Ceramic Pot Filter
The ceramic filter is a porous flower pot-shaped device that holds 8 liters of water when
filled. The ceramic pot filter element sits in a self-contained safe storage unit of 20-45
liters with a lid to form an enclosed single unit, referred to here at the ceramic water
purifier (CWP). While working with the Central American Research Institute for
Industry (ICAITI) in the early 1980s, Dr. Fernando Mazariegos developed the original
design for a colloidal silver impregnated ceramic pot-shaped filter. Many organizations
have embraced the filter since then, including Ron Rivera’s Potters for Peace, which has
helped greatly to disseminate the filter to over 1.5 million people in 21 countries (Rivera,
2008).
With effective pore sizes that range from 0.6-3 μm as necessary for sufficient flowrate,
typical ceramic filters from Nicaragua allow small numbers of E.coli (2 μm long by
0.5 μm wide) and other bacteria to pass through (Lantagne, 2001). To help solve this
problem, Mazariegos painted the underside of the ceramic pot with colloidal silver in
order to inactivate these bacteria. Brown witnessed 99.5% E.coli reductions in his lab
tests using filters with and without the addition of colloidal silver (Brown, 2007). These
findings compared quite well with results from field testing in Cambodia, in which 80
households with ceramic water purifiers CWPs from an International Development
Enterprises (IDE) distribution the previous year were monitored thrice over a four year
period, resulting in 98% average reduction in E.coli counts. This study recorded a 46%
reduction in diarrhea prevalence among 203 users and thus showed the ability of users to
effectively manage their drinking water through use of the CWP (Brown, 2007). While
viruses are smaller than 0.2 microns, and thus are not targeted by the ceramic pot filter,
Brown found 1-2 log removal of the indicator MS2 bacteriophage in laboratory trials
(Brown, 2007). With a cost of US $6-25 worldwide and health impacts similar to other
treatment options, the health-based cost effectiveness of ceramic filters compares well
with other HWTS.
4.6.1 Ceramic Pot Filter Effective Use Brief
Ceramic Pot Filter Effective Use Brief
Monitoring Observations
1. Water is added to the CWP every day.
Treatment
2. Ceramic pot is frequently topped off in order to achieve faster flow rate.
3. Ceramic pot is not overfilled. 3-5cm below the brim is the maximum
recommended fill level to prevent spillage over the lip and into storage.
4. Storage unit is not filled above the bottom of the ceramic pot.
5. Lid for the CWP is kept in place except when being filled.
6. Proper installation is witnessed, including:
6.1. Raised above the ground to about table height
6.2. Sits level on a stable base that is large enough to accommodate it
6.3. Located out of direct sunlight and out of reach from young children
and animals.
6.4. Tap is not resting on any nearby object and does not leak.
67
Safe Storage
Maintenance
Replacement
Period
Physical
Inspection
7. Turbid waters undergo settling for at least one hour before ceramic
filtration
1. The CWP includes a closed safe storage unit with a tap that should keep
treated water safe if the ceramic pot remains in place throughout use as
directed and the storage unit is regularly cleaned.
2. If possible, check to see if clay particles have settled in the storage unit.
These are likely to be from the ceramic pot itself, yet infer infrequent
cleaning (i.e., improper maintenance) of the storage unit. If found, the
monitor should ask when the last time the storage unit was cleaned.
3. Secondary storage is not recommended without chlorine disinfection to
retain microbiological quality of treated water. Safe storage
characteristics and effective use of the chlorine product should be noted
if secondary storage is used.
1. Cleaning of the ceramic pot is needed when a significant reduction in
flowrate occurs. Conversely, cleaning can be regularly scheduled with a
frequency determined by source water quality.
2. To clean the ceramic pot, scrub the inside with a hygienic brush and
rinse with filtered or boiled, cooled water. Never use soap or
disinfectant with the ceramic pot itself.
3. Regular cleaning of the safe storage unit, tap and lid with filtered or
boiled water and soap or chlorine disinfectant is necessary.
4. Ask the user when the last time the CWP was cleaned, and make sure
she/he has a sound scheduling mechanism for cleaning.
5. Ceramic pot, storage unit and tap are clean with no visible leaks or
cracks.
1. No expiration period suggested. Replace filter when cracked or broken.
1. There is water in the storage unit and the ceramic pot is partially full or
at least damp infers active use.
2. A clean cup that is used only for drinking is associated with the CWP.
3. Water bottles for use during travel or school are clean and producible to
the interviewer if consistent use is claimed.
4. User demonstrates hygienic method when asked to add or fetch water to
the CWP.
5. Instructional material is displayed with the CWP, if provided during
purchase or installation.
Water Quality Monitoring
Treated water is expected to be clear (<5NTU) unless influent is >100NTU
Turbidity
from source, which requires settling before treatment.
Free available chlorine presence in secondary safe storage if chlorine
Chlorine
treatment
is claimed.
Residual
Microbial testing shows <10 E.coli CFU/100 ml of treated water from
Microbial
storage unit(s).
Testing
68
4.6.2 Monitoring Observation
4.6.2.1 Treatment
Installation
Effective use of the filter starts with proper installation. As opposed to the biosand filter,
where the user takes part in construction and installation with careful oversight by
technicians, the CWP is often left to the user to install after instruction. Important
installation instructions are as follows, and can be taught through group trainings or by
having a salesperson deliver the filter directly to the home. Prior to assembly, wash all
pieces of the system using boiled water, and disinfectant or soap for the non-ceramic
parts, taking note of the techniques described in the Maintenance section below. A single
Aquatab is provided by Pure Home Water (PHW) for an initial disinfection during
installation. Assemble the unit in proper fashion, as demonstrated on stickers, posters or
through training sessions, making sure the tap does not leak. Place the unit on a stable,
flat base, raised to table height above the ground (~75 cm) in a safe yet convenient
location indoors, out of the sun where animals and young children can not access it.
Once in place, run enough water through the new filter to decrease the taste of clay in the
water to acceptable levels if necessary, but otherwise the filter is ready for use. All of
these aspects of setup can lead to longer and safer use of the CWP, and may be observed
directly through operational monitoring at the home. Following installation, day to day
operation of the filter is straightforward, with only a few key points to mention.
Turbidity
Rivera recommends straining turbid waters through a cloth tied around the edge of the
unit while adding water for filtration (See Appendix F: Ceramic Pot Filter Usage
Instructions for the pictorial schematic from Potters for Peace). Pure Home Water’s
training materials specifically instructs users to let turbid water settle in their primary
storage units for at least one hour prior to adding to the CWP (See Appendix F: Ceramic
Pot Filter Usage Instructions for the pictorial schematic from PHW). Despite achieving
better efficiency or larger log reductions with high influent turbidities, higher resultant
turbidities and contamination levels have been noticed in filtered waters coming from
more turbid sources (i.e., a 98% reduction of 1000 E.coli per 100ml represents
intermediate risk water) (Peletz, 2008). Elevated source waters turbidities reduced ~50%
before treatment through settling in large clay urns in Northern Region, Ghana with
residence time on the order of a few days before undergoing ceramic filter treatment
(Johnson, 2007). Without this pretreatment, turbidity was higher than 10 TU from a
number of filters in Swanton’s study despite otherwise effective use (Swanton, 2008).
High turbidities require more cleaning, and thus greater wear on the filter, more burden
on the user, more chance for breakage during cleaning, and consequently more chance for
recontamination in the storage unit through ineffective cleaning. While CWP’s are suited
to a variety of source waters, pre-treatment settling of highly turbid waters is needed for
improved performance of this technology. Measuring of turbidity to aid in discerning
whether the filters are being used effectively will be covered later in the Water quality
monitoring section.
69
Flowrate
Pressure from elevation head is the mechanism that drives filtration in this system. If
users understand this, then they will know to keep the filter filled for faster filtration.
This author witnessed users in Ghana waiting until the filtration had finished before
adding more water, and these same users complained of the filter not providing enough
water for the family. In the effort to improve flowrate by keeping the filter full, it is
important that the user not overfill the pot, as overflowing water can make its way around
the narrow lip of the ceramic pot and into the safe storage unit without filtration, reducing
filtration effectiveness by half. Baumgartner et al. recommend not adding water to the
above 3-5 cm from the top of the ceramic pot (Baumgartner, 2007). When adding water,
use a clean cup or calabash to add small amounts of water at a time. To avoid
contamination from overfilling, and possibly cracking or knocking over the filter, never
add water from a large bucket or from any height to the filter. These are fragile units, and
breakage is a major reason for discontinued use. Other than for filling and cleaning the
filter, the lid to the unit must stay on at all times. A missing lid signifies poor hygiene
and improper use.
Water Level
Regarding water level, on of the CWP sold by Pure Home Water in Northern Region,
Ghana, a usage sticker is placed on the safe storage unit with clear line drawn at the level
of the bottom of the ceramic pot in order to discourage users from filling the storage
above that line (Pure Home Water, 2008). For a visual representation of this sticker, see
Appendix F: Ceramic Pot Filter Usage Instructions. Allowing the storage unit to fill
above the bottom of the pot can negate the head gradient within the filter unit, leaving
water inside the filter walls for unnecessarily long times. With a flowrate of 1-2.5 liters
per hour when new, these filters do not produce a lot of water in short amounts of time.
Witnessing excess water in storage during a household visit may be a sign that the
household tends not to drink the water from the filter.
4.6.2.2 Safe Storage
The CWP includes a closed safe storage unit with a tap that is a well regarded feature of
the unit, and if the ceramic pot remains in place throughout use as directed and the
storage unit is regularly cleaned, the self-contained storage unit should keep treated water
safe. However, the storage units have been noted to be prone to recontamination through
improper use, and careful attention to them must be given during household visits
(Brown, 2007).
If there is not much water in the ceramic pot, lift the pot and check to see if the inner
surface of the storage unit (typically a 20 liter bucket) is clean. Clay particles settled in
the storage unit are likely from the ceramic pot itself, and infer infrequent cleaning (i.e.,
improper maintenance) of the storage unit. If settled clay particles are found, the monitor
should ask when the last time the storage unit was cleaned. The recommended regular
interval is one to two weeks, and should coincide with cleaning and disinfection of the
tap, lid, and associated outer surfaces of the CWP.
70
Secondary storage is not recommended without chlorine disinfection to retain
microbiological quality of treated water. If secondary containers are used for storage,
safe storage characteristics of the container and effective use of the chlorine product
should be noted.
4.6.2.3 Maintenance
The following maintenance protocol was developed by Potters for Peace, and is in use
throughout the world among production facilities that were initiated by Ron Rivera’s
trainings (PFP website, 2007):
Figure 11 Potters for Peace CWP Maintenance Poster
In Brown’s study area in Cambodia, where source water averaged under 10NTU, average
cleaning rate was 2.3 times per week (Brown, 2007). Similar high rates of cleaning were
claimed in Northern Region Ghana during surveying in January, with much greater
turbidities (averaging >100NTU). Unlike with the biosand filter, cleaning the CWP does
not disrupt treatment and high rates of cleaning are admissible. The system should be
cleaned when the flow-rate is reduced or stops, or when any plastic part becomes visibly
dirty (Pure Home Water, 2008).
Basing cleaning frequencies as needed on reduced flow rates seems impractical, given the
average 0.5 liter per hour flowrates for old, clean filters as measured by Swanton
(Swanton, 2008). With the biosand, the flowrate is quick (recommended 36 liters per
hour (CAWST, 2007)) and visibly drops in flow when cleaning is needed. In contrast
with the CWP, there is no visible flowrate as the unit is self contained. However,
71
flowrates can be discerned from the dripping rate emanating from a filtering CWP.
Regular cleaning of the ceramic pot may be easier than making observations based on
flowrates. Nevertheless, multiple cleanings per week were claimed and correspondingly
clean units were witnessed in surveying visits in Ghana, showing that users can
effectively clean their units without following a prescribed schedule. While disinfection
of the storage unit and especially the tap on a regular basis is warranted as well, little
information exists that looks at storage practices in conjunction with survival of fecal
bacteria on the timescale of a week (a typically recommended maintenance interval).
Thus, we can recommend weekly disinfection of the storage unit and tap as a seemingly
conservative yet unfounded recommendation. A full cleaning every week might be too
much to ask of women in the household, especially without more evidence from the field.
Further research into rates of contamination and regrowth in storage units of ceramicfiltered water is needed
When cleaning the CWP, the user must use two hands and lift the empty ceramic pot
filter by the rim without touching the outside of the ceramic element. The plastic lid to
most of these systems can also provide a suitable surface on which to place the ceramic
pot, both lying upside down, given that it has been cleaned with sterilized water. Never
placing the filter on the ground or touching the under-surface of the ceramic pot are key
training points to prevent post-treatment contamination! Many people witnessed in
northern Ghana used dirty water to rinse of drinking cups as well as to clean their storage
units. In this vein, PHW specifically highlights that a dirty cup negates the benefits of
using the filter in their training materials. While using filtered water to clean the cup is a
clear lesson, coordinating the use of filtered water with soap in the cleaning and then
rinsing of all parts of the system during maintenance can be tricky. Having salespeople
demonstrating these cleaning techniques directly during house-to-house follow-up
monitoring to a distribution of filters to flood affected victims in the Upper-East region of
Ghana was much appreciated by the residents, with a complete reinstallation often
necessary to fix the leaky and fragile taps. Using only treated water (filtered, boiled
and/or chlorine disinfected) during cleaning processes is imperative, and cannot be
overstressed.
When conducting monitoring visit, the monitor can ask the user to describe or
demonstrate their techniques:
• When was the last time they cleaned their filter?
• Do they only brush the inside of the ceramic pot? Can they show the brush or
cloth used to clean the ceramic pot? Is it appropriate and hygienic?
• Is treated or boiled water and soap or chlorine solution used to clean the storage
unit and tap and available upon request? Using dirty water for cleaning is a
common faulty practice witnessed among users, and should be checked.
• Is the basis of the ceramic cleaning schedule logical (either due to flow rate
stoppage or on a frequent time basis) and the filter clean inside (visibly bright
ceramic)?
• Are the storage unit and tap cleaned on a regular basis with treated or boiled water
and soap or bleach disinfectant?
72
4.6.2.4 Replacement period
As noted in the bottom right column of Figure 11 above, a lifespan of one and a half
years is expected of the filter by Potters for Peace. Similarly, Brown found that filters
had an average lifetime of two years (2% per month reduction in original number of
filters over 44 month study). However, Brown found that filters perform adequately up
to four years, with breakage of almost all filters by 48 months limiting his observation
(Brown, 2007). Lantagne measured effective treatment after 5+ years of use (Lantagne,
2001a). Microbial removal efficiency does not degrade over time, thus recommendations
on a one or two year replacement are unwarranted if the filter remains unbroken (Brown,
2007). Due to the instruction at the outset of the Cambodia program to discontinue use
after one or two year lifetime, 5% of people stopped using the filter after a few years of
operation despite proper functioning of the filter, incurring greater cost and/or greater
likelihood of disease to those users. With high breakage rates, a sustainable program will
need pay close attention to the supply chain so that when filters break, the users know
where to quickly and cheaply get a new filter. Given an average two year lifetime for the
ceramic element, the plastic safe storage unit should last through many replacement
ceramic pots.
4.6.2.5 Physical Inspection
Many implementing organizations hand out training materials such as those shown in
Appendix F: Biosand Filter Usage Instructions. If these are included in the distribution
of filters, it is important to witness them positioned along with the filter, especially if
multiple users are involved with the filter.
When in the home conducting an interview, be sure to ask the user if they always drink
filtered water. Do they carry treated water to work or school, incurring consistent use? If
so, can they produce the bottles used? Consistent use of the filter implies current use,
which means maintaining the system in good working order and having the storage unit
at least partially full and the filter at least damp. Depending on local climatic conditions,
filters take about 3 days to dry out completely, such that dry filters are not currently in
use. A clean drinking cup associated with the BSF is recommended to limit
recontamination. Noting hygiene practices when asking the user for a glass of water can
be informative as well.
To recapitulate the main things to inspect during a household monitoring visit, take note
that the CWP unit israised off of the ground and situated on a stable base, installed out of
the sun and rain, and inaccessible to animals and small children. Make sure that the
spigot is visibly clean, does not leak, and has nothing exerting pressure on it, as it will
tend to break. These measures will provide a decent level of hygienic practice as well as
make breakage less likely. These measures will provide a decent level of hygienic
practice as well as make breakage less likely.
73
4.6.3 Water quality monitoring
Johnson found 99.7% reductions in E.coli among rural households with highly turbid
source water in Northern Region, Ghana (Johnson, 2008). While Brown measured 98%
average reductions in his field trial in Cambodia, the range of reductions was very large.
Using a method of sampling from untreated water and comparing the microbial results to
the treated water from the safe storage unit, Brown recorded 99.9+% reductions from
many filters, yet 17% of filters tested yielded increases in contamination (Brown, 2007).
While many samples underwent large amounts of recontamination through improper
handling and cleaning of the storage units, the method of sampling was also faulty,
unable to take into account temporal differences between the treated water and the source
water. As noted in the Safe Storage section, up to 0.5 log reductions were recorded due
to transport and settling of source water, and given the slow flow rate of a ceramic filter,
it is unlikely that untreated water in the household closely resembles the source water
actually used for the treated water in the safe storage unit (Levy, 2007). Without a
controlled monitoring campaign using multiple visits over the course of a few days,
accurately recording reduction efficiency is challenging. Thankfully, this work has
already been done in the laboratory and in multiple field studies, and is not necessary in
operational monitoring campaigns as described here. Microbial testing is useful as a
proxy indicator for risk of diarrheal disease to the user, however.
Median E.coli per 100 ml for treated water from n=203 in Cambodia over 3 sampling
rounds was <10 CFU/100ml, showing that household use of ceramic filters can achieve
the proposed microbial target of Effective Use. With 66% of samples containing less
than 10 E.coli per 100 ml, our metric is a stringent measure of microbial water quality. It
may be too stringent in the case of highly contaminated influent waters, despite Effective
Use practices. While measuring E.coli counts of an untreated sample from storage or
from water that is currently undergoing treatment (in the top of the filter at the time of the
visit) will not yield reproducible information on reduction efficiency, knowledge of
influent contamination levels can often yield information on whether treated water
quality of <10 or <100 E.coli per 100 ml represents Effective Use (in either case, E.coli
should not show up on Petrifilms). Improperly treated water carried very high E.coli
loads, lying far outside the normal or acceptable limits in Brown’s study. Likely due to
post-contamination or faulty filters, 14% of users had >100 E.coli per 100 ml. Brown
claims that recontamination through cleaning and handling is a large threat to the
effectiveness of the CWP (Brown, 2007).
The ceramic pot filter has been applied to waters of all turbidity levels, attaining
significant reductions in turbidity. More than 75% of treated water samples taken in
Brown’s study had less than 2NTU after treatment from an average of 10NTU in source
water. In Ghana, where turbidities often are in the 100-1000 NTU range, the ceramic pot
filter brought turbidities down to non-detectable levels (<5NTU) in 22 out of 24 samples
taken (Swanton, 2008). When these same filters were sampled a week later, the two
households with higher turbidities in their treated water during the first round had nonvisible amounts of turbidity in their treated water yet other users had visible amounts of
turbidity, ruling out cracks in the filter, yet exemplifying inconsistent Effective Use
through lack of pretreatment settling of the highly turbid water (Swanton, 2008). During
74
the rainy season in Northern Ghana in 2007, the average turbidity value of the dams tested
was 690 TU, with a median of 300 TU (Foran, 2007). With settling in large clay urns
prior to filtration, Johnson consistently found >50% decreases in turbidity. After presettling, the average 190NTU waters (n=33) were brought down 92% to 11NTU (n=19)
post treatment, showing Effective Use despite treated water having visible turbidity
(Johnson, 2007).
Visible turbidity in treated waters (>5NTU) can result from treatment of very high
turbidity waters (>100NTU) without constituting ineffective use. Letting water from
high turbidity sources settle before filtration is recommended for Effective Use. If source
water is found to be >100NTU when tested, storage of water for settling should be
enquired about and directly observed in the home to ensure effective treatment.
However, source water of moderate turbidity that still has visible turbidity following
treatment could signify a cracked or faulty filter. Enquire of the user how fast the
flowrate is, and if they say it is high (>3 liters/hour), check for cracks and/or replace the
filter. Measuring 5-10NTU water in the turbidity tube can require up to 3 glasses of
water, and this water should not subsequently be used for drinking or bacterial indicator
testing. Under scarce water conditions, a visual check of turbidity can suffice while at
the same time allowing the monitor to check hygienic practices as the user fetches a glass
of water.
Chlorine should not be noticed within the ceramic filter or storage unit, but can provide
added treatment and necessary protection from recontamination in secondary storage
units. In fact, unless chlorination is involved, secondary storage of ceramic pot-filtered
water is not recommended. Due to the slow flow rate of the filters, obtaining sufficient
volume of water for standard treatment of chlorine (10 or 20 liters) is difficult, given
demands on drinking water (Swanton, 2008). If chlorine treatment for secondary storage
is claimed, checking presence/absence of free available chlorine will indicate effective
treatment.
4.6.3.1 Sampling Procedure
1. When sampling treated water from the tap of the ceramic water purifier (CWP)
system’s storage unit, make a visual check of the turbidity level of the treated
water. If turbidity is very visible in the sample bag, first take a turbidity
measurement if sufficient volume of treated water exists. Second, check the
condition of the ceramic pot. Are there any cracks or problems? Third, check the
quality of the influent source. If the source turbidity is >100NTU, turbidities
>5TU in treated water are expected. Ask the user if they practice settling of the
water? Does this check out with what you can see?
2. If source water is of visibly poor quality, it is best to test its bacteriological quality
using Petrifilms, if possible. Water samples can be drawn from primary storage
units or directly from water currently undergoing filtration within the pot. These
results can aid in the judgment of effective turbidity and bacterial treatment of the
water using the CWP.
3. If safe storage of treated water exists outside of the CWP safe storage unit, take a
sample for microbial testing. If chlorination is claimed by the user, take a
75
presence/absence measurement of free residual chlorine of this stored sample and
ensure that the sampling bag contains sodium thiosulphate or an equivalent
compound to neutralize the effects of free chlorine. In any case, effective safe
storage would dictate that the quality of the stored water is better or the same as
the water from treatment.
76
4.7 Biosand Filter
Originally developed by David Manz at the University of
Calgary during the mid 1990s, the biosand filter is a
household version of the slow sand filters that have been used
at the municipal level throughout the world since their
invention by a London architect, James Peacock, in 1791. The
continuing use of slow sand filters in the London water works
has helped to control cholera and other waterborne diseases
since the mid 1800s to the present.
Figure 12 Typical square concrete household biosand filter unit.
(CAWST, 2007)
4.7.1 Biosand Filter Effective Use Brief
Biosand Filter Effective Use Brief
Monitoring Observations
1. Water is added daily to the filter.
Treatment
2. Uses separate containers to pour dirty water and store filtered water.
3. Adds water slowly and with the diffuser plate in place.
4. Never adds bleach into the intake of the filter.
5. No one touches the spout of the filter with anything unless cleaning it.
6. Uses the filtered water for as many tasks as possible.
7. The lid to the filter is in place, diffuser plate intact, unobstructed clean
spout and smooth sand bed at water depth of 4-6 cm.
8. Flow rate < 0.6 L/min when full of water.
9. Proper installation of biosand filter is witnessed, including:
a. Sitting flat on firm ground.
b. Out of direct sunlight.
c. Out of reach of animals
d. No visible leaks or cracks.
10. Pretreatment is recommended for turbid waters (>100 NTU)
Safe Storage 1. A dedicated safe storage container is used to catch and store the treated
water from the spout of the BSF.
2. Safe storage container is located with the BSF indoors, out of the sun,
off of the floor, in a stable position and out of reach of animals and
small children.
77
Maintenance
Replacement
Period
Physical
Inspection
3. Design of safe storage unit incorporates a tap or a small sealable
opening for pouring.
4. The safe storage container has a lid that is kept on tight, and only
opened for addition or pouring of treated water.
1. Cleaning schedule is not prescribed but is determined by significant
reduction in flowrate. Less than one cleaning per week helps to ensure
proper biological treatment.
2. User demonstrates “swirl and dump” method successfully:
a. Harrow with a wooden stick or spoon
b. Decant muddy water
c. Refill water (after replacement of diffuser plate)
d. Check flowrate and repeat if necessary.
3. User cleans the spout and storage unit with treated water and soap or
chlorine solution each week.
4. Soap or disinfectant used to clean storage unit can be produced by user
1. No replacement period suggested.
1. Water bottles for use during travel or school are clean and producible to
the interviewer if consistent use is claimed outside the home.
2. User demonstrates hygienic method when asked to add water to filter
and fetch a glass of water.
3. A dedicated clean drinking cup is associated with the safe storage unit.
4. Instructional material is displayed, if provided during purchase or
installation.
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU).
Turbidity
Free available chlorine presence in storage if chlorine treatment is claimed.
Chlorine
Residual
Microbial testing shows <10 E.coli CFU/100 ml in water from both running
Microbial
spout and storage unit.
Testing
4.7.2 Monitoring Observation
4.7.2.1 Treatment
Shown below is the Samaritan’s Purse version of standard operating procedures for the
biosand filter, which covers the main points relating to Effective Use4.
1. ONLY pour water in the filter with the diffuser basin in place - failing to do this will
damage the filter.
2. ALWAYS use two buckets: one to pour in dirty water and one to collect filtered water.
If only one bucket is used, the dirty bucket will contaminate the filtered water.
4
For greater detail, see Appendices H and J of the CAWST document “Installation Operation &
Maintenance Manual: Biosand Water Filter,” Version 2007-01, as included in this document’s Appendix B:
Biosand Filter Usage Instructions. As presented here are revised with the aid of Ron Lentz of CAWST,
August, 2008.
78
3. NEVER attach anything to the spout, such as a longer pipe, a hose or a valve.
4. ALWAYS use filtered water for as many tasks as possible: drinking, cooking, cleaning
food, cleaning clothes, washing children, and feeding animals. Using the filter for all
your water needs will contribute to better health.
5. NEVER put bleach in the water before pouring it into the filter and NEVER pour
bleach directly into the filter - this will damage the filter.
6. ALWAYS pour the water into your filter SLOWLY.
7. NEVER move the filter once it has been installed - unless it is an emergency.
Moving the filter will cause water to come out more slowly. If moved, the filter must be
placed in a level position before using.
8. ALWAYS keep the lid on the filter when not in use.
9. DO NOT touch the spout of the filter unless cleaning it - keep animals and children
away.
(Earwaker, 2006)
The following is a list of specific physical attributes which the monitoring agent should
check at households with mature filters (in use for more than one month since start-up) to
ensure effective treatment procedures, as adapted from “HWTS Technologies: Key
Operating Parameters” (CAWST, 2007):
• The filter housing does not leak
• No tap or hose is attached to the spout
• The diffuser plate is in place, clean, and effectively preventing sand disturbance
• The filter is installed out of the sun and rain, near the kitchen, and away from
animals
• The spout is clean
• Flow rate is not more than 0.6 liters per minute when the filter is full of water
• Sand is level and sits 5cm below standing water level
While examining the filter, ask the user how often the filter is filled. For effective
treatment, the filter needs addition of water every day so that anoxic conditions are
avoided in the lower sand layers. The filter is intended to be used intermittently such that
residence time within the filter is sufficient for effective treatment. This is distinctly
different from the ceramic filters, which operate most efficiently when kept filled.
4.7.2.2 Safe Storage
Utilization of a clean storage unit that is covered when treatment is not occurring is very
important to the BSF system, yet is often overlooked during implementation. Current
designs of the BSF often do not allow for the addition of a permanent, closed safe storage
unit with a tap at sufficient levels above the floor. A dedicated, closed and hygienically
handled safe storage unit is imperative for Effective Use.
Chlorine treatment is
recommended by CAWST within safe storage units of BSF treated water in order to
maintain a residual protection against recontamination through use (CAWST, 2007). See
the 4.7.4 Discussion section for more in depth information on safe storage with the BSF.
79
While the Safe Storage Effective Use Write-up has much greater detail on safe storage
within the home, the following safe storage characteristics are important to note along
with a biosand filter.
• Is a dedicated safe storage container in use, separate from the container used for
fetching water?
• Does the design of the safe storage unit incorporate a tap or a small sealable
opening for pouring, such as to eliminate recontamination by the introduction of
dirty objects for dipping such as ladles, cups or hands?
• Is the safe storage unit kept out of direct sunlight, as the sun speeds re-growth of
bacteria?
• Is the lid to the unit kept on tight, and only opened for addition or pouring out of
treated water?
• Is the unit clean and free of leaks, situated indoors, off of the floor, in a stable
position and out of reach of animals and small children?
4.7.2.3 Maintenance
While the use of biosand filters is straightforward, maintenance requires proper training
and execution commensurate with increased user responsibilities. The following
cleaning steps are recommended2 (CAWST, 2007; Lentz, 2008):
Swirl and Dump
• Remove the lid to the filter;
• Add 4 liters of water to the top of the filter
• Remove the diffuser plate
• “Swirl” an appropriate tool such as a wooden stick or spoon around in the top
layer of sand at least 5 times. You will disturb the surface of the sand but do not
mix the surface layer below the top 5 cm of sand. The water above the sand will
become dirty.
• Scoop out dirty water with small container (i.e. cup or cut open plastic bottle)
Avoid scooping out sand.
• Discard the contaminated water outside the house in an appropriate location such
as a soak pit or garden
• Repeat this until all the water has been removed from the filter
• Smooth and level the sand surface
• Replace diffuser
• Add 20 liters or 5 gallons of water and replace lid
• Check flow rate
• Repeat if flow rate is still low (less than 0.6 liters per minute)
• Wash your hands with soap and clean water
Cleaning of the top sand layer in this way is only needed when the flowrate is reduced to
an unacceptable minimum. Slower flow means cleaner water, and cleaning the unit on a
schedule or too frequently disrupts effective treatment, (see 4.7.3 Water Quality
Monitoring section below). Cleaning as needed based on flowrate is an essential
maintenance lesson. On the other hand, regular cleaning and disinfection of the outlet
80
spout and the safe storage unit is necessary to limit likelihood of diarrheal disease, using
either chlorine solution or soap. Earwaker noted that 60% of users in his study of Kale
Hewyet Church’s biosand implementation in Ethiopia clean the spout without soap or not
at all, representing significant likelihood of post-treatment contamination (Earwaker,
2006).
Wet harrowing, as described above can alternatively be done with the palm of the hand
and gentle kneading of the fingers, incurring less damage to the schmutzdecke while
achieving similar results. Cleaning techniques that require the removal of sand, however,
are strongly discouraged and outdated. Removing sand is unnecessary because most of
the physical particles that cause the reduction in flowrate are trapped in the top few
centimeters of sand. The process of removing and replacing the sand creates air pockets
and cracks in the filter bed, as well as unnecessarily disturbs the schmutzdecke (Fewster,
2004). Removal and replacement of sand was taught during the pilot scale distribution of
biosand filter in Ethiopia by Kale Hewyet Church in the late 1990s and resulted in
recurring losses of sand among users. Before addition to the filter during installation,
sand is sifted to the appropriate grain size and thoroughly washed. Often sourced from
outside the communities, replacing lost sand was impossible for many users and Kale
Hewyet Church spent a good deal of money and time replenishing sand and eventually
retraining all of the users. Losing sand changes the pause depth (the resting height of
water above the top of the sand layer), which is designed to be between 4cm and 6cm in
order to facilitate optimal oxygenation to the schmutzdecke, one of the key innovations to
allow intermittent flow. With similar effects to the loss of sand, placing pipes, hoses or
valves on the outlet of the filter can change the pause depth and kill the schmutzdecke,
compromising the microbial treatment properties and placing the user in danger.
There are currently two schools of thought as to how to clear the turbidity entrained in the
upper layers of a biosand filter: stirring gently using a clean tool such as a spoon or stick
down to at most 5cms, or using the flat palm of ones hand to gently stir up the trapped
dirt5. While the tool stirring technique is likely to suspend more solids into the water,
causing greater lengths of time between cleanings, the flat palm method disturbs the
schmutzdecke much less and thus biological treatment is likely to stay more constant.
Though both are used widely throughout the world as maintenance techniques for the
biosand filters, there has yet to be any study as to which method actually causes more
degradation to the treatment efficiency. Doing such a study could also yield information
on the max frequency of cleaning possible to maintain sufficient treatment, and from this
could be back-calculated a maximum suitable turbidity for treatment by the BSF. Before
recommendation of the more apt technique is made, research must be conducted. Until
that point, available data seems to say that both methods are suitable for cleaning the
BSF.
5
Using the index finger down to the second knuckle, as previously proposed by CAWST and Samaritan’s
Purse, among others, places users in contact with untreated water and biologically active media. Placing
the finger into the sand is to be avoided in order to limit possible infections in open sores on the hands
(Lentz, 2008).
81
When conducting monitoring at a household, ask the user to describe their maintenance
techniques to you:
• When was the last time they cleaned their filter? This should not be within the
last week, in order to achieve maximum filter performance. Cleaning should be
performed when filter rate is too slow.
• Are the storage unit and spout cleaned on a regular basis with BSF treated or
boiled water and soap or bleach disinfectant?
4.7.2.4 Replacement Period
Unlike consumable HWTS products, biosand filters have no expiration date. With proper
operation and maintenance, the sand should not have to be replaced during the lifetime of
the concrete (20 – 40 years) or plastic filter housing (2-5 yrs) (Lentz, 2008). Sand may
occasionally need to be added to maintain the standing water level at 5 cm or less.
Biosand filters typically outlast other HWTS hardware installments and achieve higher
rates of Sustained Use as well (Sobsey, 2007). For example, the rate of Sustained Use
was 85% after five years for Kale Hewyet Church distribution in Ethiopia, a number they
continue to claim even after almost ten years of use (Earwaker, 2006). Despite a few
leaks due to construction and breakages during transport, concrete biosand filters have
been effectively used for nine years in some households in Ethiopia. The sand in these
filters has never been replaced, although sand has been added to many units due to the
aforementioned outdated cleaning method. Consequently, there is no set date for the
replacement of biosand filters.
Depending on their construction, the various plastic models may fatigue or degrade after
many years. Some agencies suggest a product life of five years for plastic biosand filters,
but this is dependent on the type of plastic used (HDPE, PP, other), mode of manufacture,
and other variables. Keeping biosand filters out of the sun is very important to
preventing degradation as well as for Effective Use, preventing algae from forming in the
standing water layer.
4.7.2.5 Physical Inspection
Many implementing organizations hand out training materials such as those shown in
Appendix F: Biosand Filter Usage Instructions. If these are included in the distribution
of filters, it is important to witness them positioned along with the filter, especially if
multiple users are involved with the filter.
Ask the user if they always drink filtered water. Do they carry treated water to work or
school, incurring consistent use? If so, can they produce the bottles used? A clean
drinking cup associated with the BSF is recommended to limit recontamination. Noting
hygiene practices when asking the user for a glass of water can be informative as well.
4.7.3 Water quality monitoring
The main barrier that the biosand provides against diarrheal disease is its ability to reduce
fecal bacterial contamination. Measuring reductions in indicator organisms is a common
metric of treatment efficiency. Samaritan’s Purse set a target reduction of 95-97% in
total coliform count between raw and treated waters for their co-implementation with the
82
Kale Hewyet Church in 2001 (Earwaker, 2006). For accurate measurements of treatment
efficiency, one must sample the raw water at the time of addition to the filter, perform an
accounting for the volume displacement in order to know when that water will exit the
tap, and then undertake a subsequent trip to the household to sample and test the treated
water. Such testing is out of the scope of a simple monitoring program in terms of time,
money, and intrusiveness. Using existent raw water in the home or at the source during a
monitoring visit as a proxy for the water fetched and used in the treatment of water to be
collected from the biosand filter also incurs major uncertainties. Jenkins notes up to 0.4
log differences in treated water quality as a result of the length of residence time within
the filter (maximum 12 hours), and similarly 0.3 log discrepancies for the amount of
water added at one time (Jenkins, 2008). As noted in the Safe Storage Effective Use
Write-up, up to 0.5 log reductions were recorded due to transport and settling, depending
on source load (Levy, 2007). These results render percent reductions from one time
monitoring visits with error bars on the scale of the anticipated treatment efficiencies! If
multiple visits to a given home are possible, better data can be gleaned from usage.
Taking five inlet and five outlet samples from a single filter over the course of a week,
for example, can show trends in reductions and absolute risk from E.coli, as well as
discount outliers.
The most useful measurement in terms of effective treatment is to get a proxy of general
treatment through measurement of absolute E.coli levels from the outflow of the unit, and
if water is stored, to get a representative sample from the storage unit. Noting the level of
recontamination from storage can show the effectiveness of safe storage techniques, and
whether training and usage is appropriate. With average E.coli reductions of 93%,
E.coli-quantifiably low risk as per WHO Guidelines was found by Stauber among 55% of
treated waters by the 55 biosand filters in use in Bonao, Dominican Republic, with an
average of <5 E.coli/100ml among all samples (Stauber, 2006). Similarly, in the WEDC
Monitoring paper from Machakos Kenya, Fewster reports that 70% of households had
less than 10 E.coli/100ml (Fewster, 2004). Effective Use is thus measured by these
authors as <10 E.coli per 100 ml sample, and such a level should be measured in both the
treated water from the tap of the unit as well as any water in the associated storage unit.
Turbidity is an important variable in the use of biosand filters. As filters clog with debris
and slower flow rates occur, better treatment takes place through finer straining,
increased residence time, and less system pressure exerted through greater head loss.
Likewise, less frequent cleaning has been associated with significant improvement in
turbidity and microbial reductions (Jenkins, 2008). With high influent turbidity, filter
run-times are reduced and maintenance is more frequent, incurring greater exposure to
microbial contamination following each cleaning as the schmutzdecke reestablishes itself.
Pre-implementation, raw waters used in treatment need to be analyzed for turbidity levels
during all climatic seasons, and CAWST recommends the use of biosand filters in areas
with a maximum raw water turbidity of less than 50–100 NTU (CAWST, 2007).
Seasonality of raw water quality and pre-treatment methods need to be directly measured
and/or questioned of the user, as both can have significant effects on turbidity.
Maintenance schedules should be based on the time at which flowrate reduces to
83
unacceptable levels, as determined by the user. Higher influent turbidities require more
frequent maintenance. There is no maximum time between cleanings.
Biosand units are very effective at reducing turbidity. Despite ineffective use and
influent turbidities commonly over 300 NTU, samples of treated water tested by the
author in both Ethiopia and Ghana in January, 2008 almost never had visible turbidity.
The technicians trained by Fewster and MEDAIR of both Machakos, Kenya and
Maintirano, Madagascar, take <5NTU to be Effective Use (Wiesent-Brandsma, 2004;
Fewster, 2008). In the 2000 Machakos survey, <5NTU was as stringent a measure of
Effective Use as the microbial testing, with similar percentage of households passing the
<10 E.coli per 100ml Effective Use metric as shown in the Table 4 Biosand filter
Effective Use metrics derived from the monitoring data of the MEDAIR Machakos filters
in 1999 and 2000 (Mol, 2000).
Table 4 Biosand filter Effective Use metrics
Effective Use metric
n
Number failing effective use
% practicing effective use
<10 E.coli
per 100 ml
153
37
76
<5NTU
<10NTU
124
32
74
124
12
90
Both <5 NTU and
<10 E.coli/100 ml
124
9
93
Although higher turbidities are associated with 50% higher E.coli results, turbidity
measurement is not a good indicator of microbial water quality on an individual basis,
with only 1 in 4 results of >5NTU correlating with samples of >10 E.coli/100ml.
Similarly, despite the uniformity of measurements of <5NTU treated water from the
filters in Maintirano, Madagascar, microbial results were outside of the low risk category
(see Table 2 Risk Levels from E.coli), with high influent turbidity and E.coli loads
(Wiesent-Brandsma, 2004). While turbidity measurement cannot suffice as a cheap
substitute for microbial testing, <5NTU is recommended as an independent measurement
of effectively treated water by the BSF.
Chlorine treatment is recommended by CAWST within storage units of treated water in
order to maintain a residual protection against dirty storage units and recontamination
through use (CAWST, 2007). In such cases, measuring free chlorine levels is applicable.
Chlorine should never be added to water before biosand treatment, as chlorine can
deactivate the biologically active sand layers.
During the first few weeks of operation following installation, the biologically active
schmutzdecke has not fully formed and microbiological treatment is only expected to
have 30-70% removal efficiency through physical straining (CAWST, 2007). However,
the level of treatment through a BSF in the process of ripening is still better than the raw
water, and users can thus be instructed to use the treated water immediately after
installation. As for secondary barriers to help protect users during this initial period, Eric
Fewster recommends that boiling biosand-treated water should be recommended if
boiling is already a common practice within the community (Fewster, 2008). The same
can be said of chlorine treatment. While monitoring for proper use and retraining in the
home can be very important within the first few weeks, microbial testing for Effective
84
Use of the filter is not warranted during this period and would be recommended a month
or more after installation. A recommended procedure for taking all of the water quality
measurements is as follows:
4.7.3.1 Sampling Procedure
1. Before taking a sample from the filter, grab a sample from the storage unit (if
water remains in it). If chlorination is claimed by the user, take a
presence/absence measurement of free residual chlorine of this stored sample.
2. In preparation to taking a sample of treated water, fill the filter to a consistent
level that can yield an appreciable flow rate. This level can be a specified volume
added to a filter that is not currently filtering water (at rest), or a known depth of
water above the schmutzdecke, such as using the diffuser plate as a reference
depth. Filling the filter to the top will not be possible at all households due to
water availability and is not recommended for normal use to achieve maximum
efficiency (Baumgartner, 2007).
3. Once you have added water, take a flow rate measurement using a container with
known volume. The flow rate should not exceed 0.6 liters/minute when the top
reservoir is full of water. Remaining under the 0.6 liters/minute will help to
ensure that adequate treatment is taking place. Use the water collected in the flow
rate test to take a turbidity measurement, if sufficient volume exists.
4. Before taking a microbial sample, evaluate operating conditions as the primary
indicators. If the diffuser plate is broken, the filter body is leaking, or the sand is
too shallow, then the filter is not working properly and microbial testing is useless
(Lentz, 2008). If operating conditions are adequate and taking a sample for
microbial analysis is warranted, take a sample from both the filtering spout as
well as from any treated water in storage in order to analyze user contamination
during storage.
4.7.4 Discussion
The BSF has shown impressive treatment results in laboratory testing. Palmateer
undertook an extensive study of the effects of the intermittent slow sand filter on a
variety of chemical and biological contaminants, using Manz’s original square concrete
intermittent slow sand filter design (MISSF). Palmateer reports that 100% of Giardia
cysts and 99.98% of Cryptosporidium oocysts were removed when spiked with 10-100
times normal environmental pollution levels (Palmateer, 1999). Elliott found that
Echovirus reductions were within a range of 1 to 4.3 log and with mean reductions of 2.1
log after 30 days. Bacteriophage (viruses that infect bacteria) reductions were much
lower, ranging from zero to 1.3 log10 (95%) with mean reductions of 0.5 log (70%).
Viral reductions by the BSF are thus expected to differ substantially depending upon the
virus encountered (Elliott, 2008). The first rigorous health impact field study of the
biosand filter, as conducted in the Dominican Republic by Stauber showed 47%
reduction in diarrhea among the intervention group, placing the biosand on par with the
other household water treatment and safe storage (HWTS) interventions studied herein,
with better potential for Sustained Use through increasing efficiency over time and robust
design (Stauber, 2007).
85
4.7.4.1 Recontamination in storage units
Storage of filter-treated water has a large potential to become recontaminated. The
storage unit of the ceramic water purifier (CWP) greatly prohibits recontamination by
creating a closed system. With current designs of the biosand filter, however, the storage
unit is most often left open and without a spout or narrow mouth as is recommended for
safe storage. In training materials and household usage, safe storage practices were
largely overlooked in the implementations of the BSF, as witnessed by this author in
Ethiopia and Ghana. While a limited number of accurate membrane filtration samples of
the BSFs in Kpanvo village, Ghana, did not find recontamination in storage, the results
are statistically insignificant. The results presented in Stauber’s PhD thesis in 2007
provide convincing statistical evidence of fecal recontamination occurring during storage,
as found in the following table.
Table 5 Water quality in BSF households after BSF intervention (Stauber, 2007)
Storage brought a 79% reduction with an average low risk (<10 E.coli per 100ml) of
treated water at the BSF outlet down to a 53% reduction for overall treatment, resulting in
intermediate risk for the majority of the 500+ users sampled. Likewise, Sobsey’s study
from Cambodia on the BSF shows consistent recontamination in storage over the course
of five monitoring visits on a large number of filters (a subset of n=1365), per month for
5 months longitudinal study, as shown in Figure 13 below.
Figure 13 Log10 concentrations of E.coli throughout BSF treatment and use (Sobsey, 2006)
86
A redesign of existing BSF units may be necessary to accommodate the safe storage unit
needed. One such design has been produced and implemented through the efforts of
Bushproof in Machakos, Kenya. Using the original square concrete BSF design, it
includes a closed safe storage unit with a tap. However, it will be impossible to elevate
BSFs in this fashion in many circumstances because of the weight of the unit as well as
the height needed to lift the water. Despite an appropriate design, the storage is visibly
dirty and the top is loose, failing two key monitoring observations in this instance.
Figure 14 BSF with Safe Storage from Machakos, Kenya (Baffrey, 2005)
Storage units need to be covered at all times, necessitating a downspout from the filter
outlet that connects directly into the storage unit. Safely elevating the heavy units
currently in use will present a challenge. Thus, if no tap can be installed on the storage
unit, then the unit should have a narrow mouth and be able to pour, so as not to incur
direct handling of water within the unit. Unless the storage unit is elevated to practically
engulf the spout of the BSF, chancing contact with the spout and contamination,
however, narrow mouthed storage units can be ineffective due to the tendency of the flow
out of the BSF to change its exit angle and fall in varying places depending on pressure
head.
4.7.4.2 Training materials pertaining to safe storage with biosand filter
In the materials distributed by CAWST, International Aid and the Kale Hewyet Church
as collected in Appendix F: Biosand Usage Instructions, instructional material for the
BSF leaves out cleaning of the associated storage units. Similarly, the SODIS materials
collected from EAWAG only inform the user to clean the bottles before the initial use
and not on a regular basis (see pictorial in SODIS Effective Use section). Training
materials as well as labeling instructions associated with chlorine disinfectants often
leave out cleaning of storage units, although this is recommended by manufacturers and
distributors (see pictorial in Sodium Hypochlorite Solution Effective Use section as well
87
as Appendix G). Likewise, very few of these materials and programs adequately stress
the separation of raw and treated water. Many users were witnessed to clean their storage
units and drinking cups using untreated source water with neither soap nor disinfectant
directly before treating or drinking water.
4.7.4.3 Storage unit cleaning frequency
The recommendations for cleaning frequency of storage units as based on the frequency
of filter cleaning found in the BSF and CWP Effective Use sections are based on using
turbid source water without pretreatment. However, in the case of the first household
visited in the Kale Heywet Church field sight in Ethiopia, the mother interviewed
pretreated her source water by filtration in the riverbed, significantly decreasing turbidity
such that she claimed not to have cleaned the storage unit in 4 months. If she was taught
to clean her storage unit when cleaning her filter, this might explain why her storage unit
was so dirty at the time of monitoring despite a well-functioning BSF. 26.3% of
households visited in Earwaker’s study of the KHC BSFs reported that they wait until
slow flow to clean the filter, although regular cleaning was the norm. Moreover, about
half of the users interviewed cleaned the filter more than once a week, under high
turbidity conditions (Earwaker, 2005). Cleaning the safe storage unit on a weekly basis
as recommended in the Effective Use sections for the BSF and CWP is a conservative
estimate of the likelihood of recontamination through use occurring within a week and is
not based on field trials or evidence. Recommended frequency of storage unit cleaning
will depend on design of the unit, as well as the expected treatment efficiency of the
HWTS system and effective safe storage practices, and will vary between
implementations. Further research on rates of contamination within storage units is
needed for all of the HWTS in varying situations to determine if cleaning of the storage
units should be scheduled, tied to the frequency of filter cleaning, or done by inspection
of the unit by the user.
88
4.8 PUR
PURTM Purifier of Water is the brand name given to
Procter & Gamble’s combined flocculent and
disinfectant product. PUR works to treat source waters
with a wide range of turbidity and pathogen load,
providing a regulated dose of iron sulfate (352mg ferric
iron) to remove suspended matter such as protozoa,
viruses, sediment, humic matter, and Giardia and
Cryptosporidium oocysts, as well as calcium
hypochlorite to kill bacteria and other pathogens. Other
ingredients include bentonite, sodium carbonate,
polyacrylamide flocculant, and potassium permanganate,
chemicals generally used in municipal water treatment
that together achieve four major processes: precipitation,
coagulation, flocculation and disinfection (Reller, 2003).
PUR is the only mass-produced sachet combining these
chemicals in solid form, and has been marketed
successfully in many countries by PSI and others, as well
as used by UNICEF, Americares, Samaritan’s Purse,
Figure 15 PUR
World Vision, CARE, and others in emergency
situations ranging from cholera outbreaks in Ethiopia to
flooding following the tsunami of 2005 to the earthquake aftermath in Pakistan in 2005 to
flooding in Myanmar. One sachet treats ten liters of water and come in strips of 12. Two
strips of 12 provide 240 liters of water, enough for a typical household for three weeks in
emergency situations (Aquaya, 2008).
89
4.8.1 PUR Effective Use Brief
PUR Effective Use Brief
Monitoring Observations
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
Proctor and Gamble, without prompting:
1.1. Add: Cut open one packet and add contents to ten liters of water.
1.2. Mix: Stir aggressively for 5 minutes and let sit for 5 minutes; if
non-flocculated after the wait, stir again until floc falls out.
1.3. Filter: Poor water into clean storage container through a clean and
dry cotton cloth that is free of holes.
1.4. Drink: Wait 20 minutes to drink. Do not consume if yellow.
2. Complete consumption of the ten liters of treated water should occur
within 24 hours.
Safe Storage 1. Two separate, dedicated 10 liter containers for fetching/flocculation and
disinfection/storage are used, visible, clean, and have no leaks.
2. Safe storage container for treated water is located indoors, out of the
sun, off of the floor, in a stable position and out of reach of animals and
small children.
3. Design of safe storage unit incorporates a tap or a small sealable
opening for pouring.
4. Lids are kept on tight, and only opened for addition or pouring of
treated water.
Maintenance 1. Rinse off the cloth filter after each use, with a final rinse of cloth filtered
water and then leave cloth in the sun for decontamination.
2. Regular cleaning of cloth filter with soap.
3. Regular cleaning of the treatment and storage containers with soap or
disinfectant.
4. Soap and/or disinfectant used to clean storage unit and cloth filter can
be produced by user.
Replacement 1. Product expires 3 years after date of manufacture, as is printed on sachet
Period
1. Water bottles for use during travel or school are clean and producible to
Physical
the interviewer if consistent use is claimed outside the home.
Inspection
2. The household contains a supply of unexpired sachets for consistent use.
3. A dedicated clean cup is associated with the safe storage unit.
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU)
Turbidity
Free available chlorine presence is shown if treatment is claimed.
Chlorine
Residual
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial
Testing
90
4.8.2 Monitoring Observation
4.8.2.1 Treatment
Information on appropriate treatment using PUR is drawn from selected promotional
material specific to each implementing organization’s training methods. These materials
come in a variety of languages and can be quite detailed (see Appendix F: PUR Usage
instructions). The schematic below is printed on the backside of PUR sachets in English.
Other information on the packet includes the brand name, dosage information, weight,
manufacturing date and subsequent expiration date, precautions against ingestion of the
powder, manufacturer and trademark information, and ingredients
Figure 16 PUR Usage Instructions printed on back of packet
A yellow color may result from waters that are heavily laden with detergents or oils, and
this water is not suitable for consumption, as noted in step 4 of Figure 16. Communityscale or individualized trainings are often utilized as part of implementations and/or
social marketing campaigns in order to ensure correct use of the product and provide
face-to-face training on how to correctly carry out the instructions included on the
product package itself. Face-to-face instruction is very important in places where literacy
rates are low or where regional dialects are more heavily used than national languages.
Training sessions often stress that only treated drinking water should be consumed, that
safe storage is important to keep the water potable, and that hand washing is an important
part of diarrhea prevention. Free available chlorine (FAC) levels can fall under the
Center for Disease Control (CDC) recommended 0.2 mg/L after 24 hours, and therefore
usage should occur within that time (Aquaya, 2008). Filtered floc should be disposed of
in the latrine or bushes away from children and animals. Environmental studies and
assessment has shown no environmental concerns with floc disposal (Allgood, 2008).
4.8.2.2 Safe Storage
Safe storage is a necessary component of the PUR HWTS system. While the Safe
Storage Effective Use Write-up contains much greater detail, the following safe storage
characteristics are important to note in the home of PUR users. Upon entering the house
for a monitoring visit, ask the user to take you to where the drinking water is stored.
91
•
•
•
•
•
•
Is a dedicated safe storage container in use, separate from the container used for
fetching and flocculation of water?
Is 10 liters easily measurable in the fetching/flocculation container?
Does the design of the safe storage unit incorporate a tap or a small sealable
opening for pouring, such as to eliminate recontamination by the introduction of
dirty objects for dipping such as ladles, cups or hands?
Is the safe storage unit kept out of direct sunlight, as the sun quickens degradation
of residual chlorine and speeds re-growth of bacteria?
Is the lid to the unit kept on tight, and only opened for addition or pouring out of
treated water?
Is the unit clean and free of leaks, situated indoors, off of the floor, in a stable
position and out of reach of animals and small children?
4.8.2.3 Maintenance
According to Dr. Greg Allgood, the Director of the Children’s Safe Drinking Water
Program at Procter and Gamble, no system maintenance is needed with the use of PUR
other than cleaning of the buckets used to treat the water and washing of the filter cloth.
PUR provides 2.0 mg/L of total chlorine to 10 liters of water, with 90% of samples
showing greater than 0.5 mg/L FAC after 30 minutes of contact time, conforming to
World Health Organization (WHO) and CDC standards for health and taste, respectively
(WHO, 1993; CDC, 2005). Although sterile cloths and containers are not needed
because the residual disinfection potential noted is at a maximum at the time when
straining and transfer to storage take place, people are trained to wash the filter cloth
between usages and to maintain clean storage containers (Allgood, 2008).
4.8.2.4 Replacement Period
PUR has a shelf life of 3 years. Household possession of expired PUR is a potential sign
that disuse or hoarding may be taking place. If PUR is found to be expired, local
distributors’ supplies might need to be checked for being past their expiration dates.
4.8.2.5 Physical Inspection
Direct physical observation is a strong tool in the roster of operational monitoring
techniques. When first entering the home, witness the hardware associated with
treatment. Ask the user to demonstrate her/his treatment techniques, without any further
prompting. Ask to see the straining cloth used and make sure it is clean, of minimal pore
size and not ripped. Asking for a glass of water is often very informative, especially if
you are willing to drink it. Did the user act hygienically while getting the water, or did
they wash out the glass with dirty water and then dip it into the storage unit without
washing their hands? These are two different behaviors with potentially very different
outcomes.
As with all HWTS products, consistent use is very important to reduce the incidence of
illness from water-borne pathogens. To ensure consistent use, there are a few simple
correlations to make note of. During monitoring, purchasing habits can be asked of the
user, with emphasis on regularity of purchasing. The expiration date for a PUR packet is
3 years after the date of production. Thus, old or expired packets are a good sign that the
92
individual is hoarding rather than using PUR on a daily basis. Another useful check to
ensure consistent use is the presence of any chlorine (free or total). Lack of a minimal
chlorine presence shows that claims of active use are suspect. Another good question to
ask in this vein is whether family members carry treated water or PUR packets while
traveling. In order to confirm consistent use, ask the family member to present PUR
packets in stock for daily use as well as clean water bottles for use when traveling.
4.8.3 Water Quality Monitoring
The main advantage of PUR over other HWTS products is its use of ferric sulfate as a
primary coagulant. Ferric sulfate is one of two main control measures of the PUR
product. Flocculation polymers and a bit of clay fill out the PUR sachet mixture in order
to enhance the coagulation process. Flocculation is needed because the suspended
particles that make up measurable turbidity harbor pathogens and needlessly consume
FAC, making disinfection unviable. However, PUR is also useful in non-turbid water as
it can flocculate out cryptosporidium and giardhia oocysts that are resistant to chlorine
disinfection. According to Norton, of 100 samples of Bangladeshi pond water ranging
from 6-92 NTU, upon treatment with PUR, 97% fell below 5 NTU as recommended by
WHO for effective disinfection and general potability (WHO, 2004). Measuring
turbidity to be less than 5 NTU is an appropriate operational monitoring method to see if
PUR was used effectively such that coagulation is occurring properly and adequate
disinfection can take place.
Disinfection with sodium hypochlorite is the second powerful control measure used in
PUR, forming a system of multiple barriers within this single product. 2.0 mg/L total
chlorine is provided in demineralized water (Allgood, 2008). The original disinfection
takes place following flocculation and straining into the storage unit. The WHO (2006)
stipulates that at least 0.5 mg/L FAC remains after 30 minutes contact time at a pH less
than 8. As pH goes above 8, less and less of the full FAC becomes available for
disinfection. Thus, if testing FAC directly after treatment, pH should be measured to
make sure that treatment falls below pH 8. Measurement of FAC during a monitoring
program, however, will most likely not occur directly after treatment. As long as
0.2mg/L FAC exists in water of at most 24 hours age, sufficient residual disinfection
potential exists (CDC, 2005). PUR was designed to provide such a residual
concentration given a range of up front disinfection and residual recontamination.
Assuming that unreasonable recontamination has not occurred (can be loosely confirmed
through physical observation of user habits), using an HACH FAC test strip, any
pinkness on the Free Chlorine test indicates treatment with PUR and this is satisfactory to
the chlorine requirement. Effectiveness of disinfection will further be confirmed with
microbial water quality testing results.
In a laboratory setting, PUR is very effective in removing bacteria (7 log removal),
viruses (4 log) and parasitic cysts like Giardia and Cryptosporidium (3 log)
(http://www.psi.org/our_programs/products/Pur.html). These log removals meet US
EPA standards for water purifiers and PUR Purifier of Water is approved for use in the
US for emergency water treatment (Allgood, 2008). The large concentrations of
pathogens needed to measure 7 log removal of bacteria usually do not exist in natural
93
waters, and most studies from the field do not look at percent or log removal of bacteria,
but rather report absolute numbers of E.coli present after treatment. Souter’s analysis is
consistent with such reductions, in which he found no E.coli among 320 samples of PURtreated water in developing countries (Souter, 2003). Such low numbers may imply
highly Effective Use among the households visited, but more realistic estimates lie in
Reller’s study from Guatemala in 2003. Forty-eight percent of households in this study
conformed to WHO Guidelines for safe potable drinking water of <1 E.coli CFU/100ml,
as per Table 2 Risk Levels from E.coli. Given such low results for E.coli in the field,
measurement of less than 10 E.coli per 100ml shows that treatment was effectively
administered (Reller, 2003).
4.8.4 Discussion
PUR is intended to be used on a daily basis, and thus has to be restocked regularly within
the household. As with all other consumable products, a stable distribution network with
visible and well positioned outlets is needed to enable widespread and consistent use of
PUR. Population Services International (PSI) in Ethiopia attempts this by pushing the
product throughout the country via sales representatives with their own vehicles while
simultaneously establishing a vast distribution network through offering competitive
margins to distributors, wholesalers and retailers through the sale from PSI at the cost of
production. While the outlets that sell more do better business for themselves, PSI also
targets the outlets in remote regions in line with their social marketing imperative. Even
PSI’s own implementation throughout Ethiopia is currently limited to the accessible (i.e.,
large) roads, and does not often reach remote areas as transport is expensive and limiting
to these regions. In order to provide access in these remote areas, P&G and PSI work
with a network of NGOs including local nurses associations, World Vision, CARE, Aga
Khan Foundation, and others.
94
5. Determination of Effective Use from Monitoring Visits
This chapter provides a case study on how to use the Effective Use Monitoring Checklists
(see Appendix E for forms covering the suite of technologies). Developed from the
Effective Use Briefs and the recommended Water Quality Methods, these Effective Use
Monitoring Checklists provide a standardized, user-friendly method with which to
conduct rigorous household operational monitoring evaluations and can be easily tailored
to individual organization’s needs. In this chapter, this framework is applied to
monitoring data collected from users of the biosand filters distributed by the Kale Heywet
Church (KHC) near Addis Ababa, Ethiopia during January, 2008 (see Appendix B for
program information and Appendix C for field notes on each of the households visited).
Following a pictorial presentation and water quality data to familiarize the reader with the
KHC implementation, sample Effective Use Monitoring Checklist forms are filled out for
the two households pictured below. Results for all of the households visited in the form
of a mock program evalauation are presented at the end of the chapter and discussed.
5.1 Kale Heywet Church Biosand Filter Program
Starting with a pilot biosand project of 700 filters in 1999, Kale Heywet Church (KHC)
scaled up their operations over the past six years to provide 8000 filters. With consistent
funding by Samaritan’s Purse of Canada, the BSF program at KHC employs a large field
staff that stays in touch with their users and can respond to problems quickly.
Located a few hours east of Addis Ababa in the Ethiopian highlands, the community of
Filtino received many of their biosand filters from Kale Heywet Church (KHC) and
Samaritan’s Purse’s original pilot scale implementation in 1999, with filters working well
since then and consistent community involvement of the technicians at the nearby
factory/field office. A largely denuded countryside, the river and irrigation ditches that
serve as water sources for BSF users have very turbid water (from 200-1000 TU
measured in-house) and are of pH 8.5-9 (basic volcanic soils).
The water quality tests conducted among KHC BSF users during January are presented in
Table 6. In terms of judging Effective Use through microbial water quality testing, the
Petrifilm method used had a minimum level of detection of 100 E.coli/ 100 ml, and was
thus unable to judge microbial water quality, as Effective Use is <10 E.coli/100ml for
most of the HWTS systems covered. Future work will continue to use the 3M Petrifilm
method along with the 10 ml pre-dispensed Colilert MPN tube system as proposed in
Chapter 3.
95
Table 6 Water Quality Results for Kale Heywet Church Biosand Filter Users
Unfiltered
Treated
Storage
E.coli/
T.coli/
E.coli/
T.coli/
E.coli/
T.coli/
Turb
100ml
100ml
100ml
100ml
100ml
100ml
TU
HH1 1/5/08
500
2000
<100
100
clear*
HH2 1/5/08
2000
14000
<100
25000
clear*
HH3 1/5/08
500
14000
<100
<100
<100
1400
clear*
HH4 1/12/08
12
<100
20000
<100
<100
700
3600
HH5 1/12/08
12
5000
18000
100
6200
600
10400
HH3 1/12/08
30
<100
18000
<100
<100
<100
2400
HH6 1/12/08
6
200
5000
14000
<100
700
1000
1900
-HH7 1/12/08
7
1000
29000
<100
1900
100
2600
*“Clear” means that visually there was no visible turbidity; assumed turbidity of <5NTU because
measurement using the turbidity tube was not possible due to minimal amount of sample available;
Source for HH 1-2 was a turbid river >100m away (no water quality data available). Source for HH3HH8 was an open, flowing irrigation ditch of Turbidity >500TU possessing 4000 E.coli/100ml and 22000
T.coli/100ml, as tested during the second sampling visit.
HH
HH3:
Date
FloRt
L/hr
Turb
TU
~clear*
1000
500
Effective Use
Intermediate/low microbial risk in
water taken from storage
HH7:
Ineffective Use
High microbial risk in water taken
from storage
Figure 17 Effective and Ineffective Use among Kale Heywet Church BSF Users
Pictorial Analysis: Note the elevated and dedicated safe storage unit in household three
(HH3), with a separate small mouthed clean jerrycan for fetching water, tile floor, and
visible presentation of KHC’s maintenance poster and sticker. Note the improper
placement of filter in household seven (HH7), located in a goat pen, accessible to
96
animals, with holes in the thatched roof that allow direct sunlight onto the filter housing.
Despite neither household showing a dedicated drinking cup for their filter, HH3 showed
much better hygienic procedure when fetching water. The lack of a designated safe
storage unit and unhygienic conditions in the picture of HH7 show that despite seemingly
effective microbial treatment occurring with both filters, recontamination occurred both
through observation and water quality testing in HH7 and not HH3 (see KHC Water
Quality Results in Table 6).
5.2 Sample Effective Use Monitoring Checklists
The following Figures 18 and 19 present as much data as was taken at the households
pictured above.
Figure 18 Example Monitoring Checklist for Household 3 of the Kale Heywet Church BSF Users
Biosand Filter Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
Matt Stevenson
Filtino community, Oromia region, Ethiopia
---------, mother of the household and BSF caretaker
HH3
January 12th around 12:00 PM
____________________
_____________________
Notes: Well kept house with clean tile floor and CGI roof. Has had BSF 8 years and likes
it very much.
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist
1. Water is added daily to the filter.
Treatment
2. Uses separate containers to fetch/pour dirty water and
store filtered water.
3. Adds water slowly with the diffuser plate in place.
4. Pretreatment is claimed for turbid waters
(>100NTU).
5. The spout is unobstructed and clean.
6. Smooth and level sand bed at water depth of 4-6 cm.
7. BSF is sitting flat on firm ground.
8. The lid to the filter is in place and clean.
9. System is out of direct sunlight.
10. System is out of reach of animals.
11. Filter has no visible leaks or cracks.
12. Filter flowrate is ~0.6 L/min.
13. Dedicated safe storage unit is used.
Storage
14. Design of safe storage unit incorporates a tap or a
small sealable opening for pouring.
15. The safe storage container has a lid that is kept on
tight except for adding or pouring treated water.
(Yes/No/ NA)
Yes
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
97
Yes
16. Safe storage container is located with the BSF
indoors, out of the sun, off of the floor, in a stable
position and out of reach of animals and small
children.
17. Safe storage unit is visibly clean.
Yes
18. User uses and demonstrates “swirl and dump” cleaning method:
Maintenance
18.1. Adds ~4 liters of water to the top of the filter
Yes
18.2. Scoops out dirty water with small container,
Yes
levels sand and replaces diffuser plate.
18.3. Fills with water and repeats the process if flow
Yes
rate is still slow.
19. Filter cleaning schedule is determined by significant
Yes
reduction in flowrate.
20. BSF cleaned less than once a week.
Yes
21. User cleans the spout and storage unit with treated
NA
water and soap or chlorine solution each week.
22. Soap or disinfectant used to clean storage unit can be
NA
produced by user.
23. Water bottles for use during travel or school are clean
No
Physical
and producible to the interviewer if consistent use is
Inspection
claimed outside the home.
24. User demonstrates hygienic method when asked to
Yes
add water to filter and fetch a glass of water.
25. A dedicated clean drinking cup is associated with the
Yes
safe storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
92%
Notes: Displays KHC maintenance materials on wall above filter; Generally hygienic usage
of the system, despite lack of a dedicated clean drinking cup.
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU).
Turbidity
Chlorine Residual Free available chlorine presence in safe storage if chlorine
treatment is claimed
Microbial testing shows <10 E.coli CFU/100 ml in water from
both running spout and storage unit.
( Yes/No/ NA)
Yes
NA
<100 E.coli/
100ml
Notes: Moderate recontamination shown through storage practices, although unable to
deduce microbial Effective Use metric by just using the Petrifilm.
Microbial Testing
Sample from
running spout
Sample from
storage of
treated water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
--**
-24 hr Colilert (Yes/No)
Yellow? Fluoresces?
---
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
0
0
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
0
14
# E.coli/
100ml
<100
# E.coli/
100ml
<100
Risk
Level
Low/Int
Risk
Level
Low/Int
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
98
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Take a sample of treated water from the storage unit for microbial analysis (if available). If
chlorine treatment is claimed in stored water, test for presence of chlorine residual while at
Procedure
the household and use a Sodium Thiosulphate sampling bag for transporting sample to
laboratory. Keep the sample out of the sun and start microbial tests within 6 hours.
2. Fill the BSF to a consistent level (not to the top).
3. Check the turbidity of the filtering water if it is visible and sufficient volume exists.
4. While taking a sample for microbial analysis from the pouring BSF spout, take a flow rate
measurement by counting seconds until 100ml is full in the Whirlpak bag.
*NA indicates that the question was not asked at the time of interview
**Not using Colilert at this time
Figure 19 Example Monitoring Checklist for Household 7 of the Kale Heywet Church BSF Users
Biosand Filter Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
Matt Stevenson
Filtino community, Oromia region, Ethiopia
---------, mother of the household and BSF caretaker
HH7
January 12th around 2:30 PM
____________________
_____________________
Notes: BSF kept in goat pen, with children drinking out of the spout directly. Has had the
BSF for 9 years. Materially poorer than her neighbors and less educated.
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist
1. Water is added daily to the filter.
Treatment
2. Uses separate containers to fetch/pour dirty water and
store filtered water.
3. Adds water slowly with the diffuser plate in place.
4. Pretreatment is claimed for turbid waters (>100NTU).
5. The spout is unobstructed and clean.
6. Smooth and level sand bed at water depth of 4-6 cm.
7. BSF is sitting flat on firm ground.
8. The lid to the filter is in place and clean.
9. System is out of direct sunlight.
10. System is out of reach of animals.
11. Filter has no visible leaks or cracks.
12. Filter flowrate is ~0.6 L/min.
13. Dedicated safe storage unit is used.
Storage
14. Design of safe storage unit incorporates a tap or a small
sealable opening for pouring.
(Yes/No/ NA)
Yes
No
Yes
No
No
NA
Yes
Yes
No
No
Yes
Yes
No
No
99
15. The safe storage container has a lid that is kept on tight
No
except for adding or pouring treated water.
16. Safe storage container is located with the BSF indoors,
No
out of the sun, off of the floor, in a stable position and
out of reach of animals and small children.
17. Safe storage unit is visibly clean.
No
Maintenance 18. User uses and demonstrates “swirl and dump” cleaning method:
18.1. Adds ~4 liters of water to the top of the filter
No
18.2. Scoops out dirty water with small container,
No
levels sand and replaces diffuser plate.
18.3. Fills with water and repeats the process if flow
No
rate is still slow.
19. Filter cleaning schedule is determined by significant
Yes
reduction in flowrate.
20. BSF cleaned less than once a week.
Yes
21. User cleans the spout and storage unit with treated
No
water and soap or chlorine solution each week.
22. Soap or disinfectant used to clean storage unit can be
No
produced by user.
23. Water bottles for use during travel or school are clean
No
Physical
and producible to the interviewer if consistent use is
Inspection
claimed outside the home.
24. User demonstrates hygienic method when asked to add
No
water to filter and fetch a glass of water.
25. A dedicated clean drinking cup is associated with the
No
safe storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
32%
Notes: Removes sand to clean BSF; Washes a cup with unfiltered water when asked for a
cup of water;
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU).
Turbidity
Chlorine Residual Free available chlorine presence in safe storage if chlorine
( Yes/No/ NA)
No, 10NTU
--
Microbial Testing
No***
treatment is claimed
Microbial testing shows <10 E.coli CFU/100 ml in water from
both running spout and storage unit.
Notes: Treatment is not working correctly; unsafe levels of recontamination occur in
storage; Ineffective Use noted through both observation and water quality testing.
Sample from
running spout
Sample from
storage of
treated water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
NA
NA
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
NA
NA
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
0
4
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
1
29
# E.coli/
100ml
<100
# E.coli/
100ml
100
Risk
Level
Low/Int
Risk
Level
High
100
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Take a sample of treated water from the storage unit for microbial analysis (if available). If
chlorine treatment is claimed in stored water, test for presence of chlorine residual while at
Procedure
the household and use a Sodium Thiosulphate sampling bag for transporting sample to
laboratory. Keep the sample out of the sun and start microbial tests within 6 hours.
2. Fill the BSF to a consistent level (not to the top).
3. Check the turbidity of the filtering water if it is visible and sufficient volume exists.
4. While taking a sample for microbial analysis from the pouring BSF spout, take a flow rate
measurement by counting seconds until 100ml is full in the Whirlpak bag.
*NA indicates that the question was not asked at the time of interview
**Not using Colilert at this time
***Microbial quality was marked as a failure if either the treated water from the spout or the stored water failed the
Effective Use metric.
5.3 Discussion of Effective Use Monitoring Results
Table 7 Sample Household Monitoring Data Format for Kale Heywet Biosand Users
Treatment
Storage
H 1 2
H *
3 4
5 6
7 8
9 1
0
1 1
1 2
1 1
3 4
1 1
5 6
1
2
3
4
5
6
7
%
y
y
y
y
y
y
y
1
0
0
y
y
n
y
y
y
y
y
y
y
1
0
0
y
y
y
y
y
y
n
8
6
y
y
y
y
y
y
y
1
0
0
y
y
y
y
y
y
n
8
6
y
n
y
n
n
n
n
2
9
y
y
y
y
y
y
y
1
0
0
y
y
y
y
y
y
n
8
6
y
n
n
n
n
n
n
1
4
y
y
-
y
y
y
y
y
y
y
1
0
0
y
n
y
y
y
y
n
7
1
y
y
y
y
y
y
y
1
0
0
n
n
n
n
n
n
n
0
y
y
y
y
y
y
n
8
6
Maint-enance
1 1
7 8
a
y y
n y y
y n
y y
y n
n n
7 5
1 0
1
8
b
y
n
n
n
1
8
c
y
y
n
y
n
n
5
0
Physc
Inspct
Monitoring
Observation
1
9
2 2
0 1
2
2
2 2
3 4
2
5
% of criteria
passed
y
n
y
y
y
y
y
y
y
y
y
1
0
0
-
n
n
n
y
y
n
3
3
y
n
y
n
n
n
n
2
9
92
52
88
70
81
65
32
69
y
n
y
y
y
y
n
7
1
y
n
y
y
y
n
n
5
7
*Numbers refer to the line items on the Biosand Filter Monitoring Checklist (see Figures 18 and 19)
HH
Monitoring
Observation
%
Water Quality
Turbidity
1
2
3
4
5
6
7
92
52
88
70
81
65
32
Yes
Yes
Yes
Yes
Yes
Yes
No
Microbial
from Spout
--<100 E.coli/100mL*
<100 E.coli/100mL*
No
<100 E.coli/100mL*
<100 E.coli/100mL*
Microbial
from Storage
<100 E.coli/100mL*
No
<100 E.coli/100mL*
No
No
No
No
*Could not judge Effective microbial treatment due to limit of resolution of Petrifilm
101
The results of this small data set show interesting positive correlations between the two
complementary methods of Effective Use operational monitoring: Monitoring
Observations and Water Quality Monitoring. With an average 73% adherence to
Effective Use monitoring observations for the seven households visited, household 7 had
the lowest monitoring observation score (32%) and passed neither of the water quality
tests. Households 1 and 3, on the other hand, had an average 90% monitoring
observation score, and failed neither water quality test. The noted agreement between
monitoring observations and water quality testing suggests that they can act as
reaffirming independent checks of Effective Use.
One of the water samples taken from the spout of the BSF was in the high risk category
(HH5, see Table 6) and one of the treated water samples failed the turbidity test (HH7).
With only two among the seven BSF users failing the treatment water quality measures,
treatment was not where the largest lapses in Effective Use occurred. The Treatment
category of Table 7 is filled with markings of correct action (note the high percentages
along the bottom row corresponding to Treatment, averaging 86%) correlating with five
out of the seven passing the water quality checks on treatment. Safe storage and handling
(shown by Physical Inspection in this case) had much lower percentages of correct action
(54% and 40%, respectively), and consequently five out of 7 households measured high
risk from E.coli in their storage containers. While this discussion highlights a few trends,
with larger data sets many more accurate correlations could be drawn to aid the
implementing organizations to judge failure and successes in terms of attaining Effective
Use in their HWTS programs.
102
6. Discussion
Throughout about 40 household visits made by this researcher during January, 2008, it
was observed that numerous users were successful in meeting various criteria for
Effective Use as laid out for the individual HWTS technologies, including correct
treatment, safe storage, maintenance and water quality. The existence of appropriate
training and/or monitoring programs was found to be one apparent cause of this success.
When users failed the observational analysis, it was often due to hygiene or storage
practices. Ineffective use all around was noticed in rare cases, which is a testament to the
success of the HWTS implementations. A few of the most common results are recounted
below.
6.1 Monitoring and Evaluation
During household monitoring visits, the author witnessed many of the technicians,
salespeople and community representatives correcting the actions of the HWTS users
upon direct observation of their usage techniques. Involved with all stages of
implementation, these people were intimately familiar with the training, technologies,
and various aspects of implementations as well as with many of the users themselves.
Their prior experience with the users and implementations allowed them to make
appropriate recommendations on Effective Use to the users. Such monitoring was
witnessed, for example, in Shak Ibrahim’s retraining users during follow-up monitoring
to the UNICEF/Pure Home Water distribution of CWPs to flood affected areas in Ghana
(see Appendix C).
This type of constructive operational monitoring can occur throughout implementations
and can contribute to substantial gains in % of households practicing Effective Use.
Operational monitoring is distinguished here from verification monitoring as described in
the WHO GDWQ 3rd Ed. that is often most useful for project evaluation and can benefit
from having independent third party monitors. If operational monitoring is conducted by
independent monitoring agents, however, it may lose the ability to correct improper use,
limiting the overall effects of the operational monitoring campaign.
Monitoring observations as they pertain to Effective Use are normally correctable on the
spot. While measurement of turbidity and residual FAC can automatically confirm claims
of consistent use, appropriate dosing, and well functioning systems, microbial water
quality monitoring acts primarily as a pass/fail metric and is not an active operational
monitoring technique unless a second visit is planned and made to the households or
community, in order to explain the results of the water quality testing. With a simple
microbiology lesson to the community, Bob Metcalf has used the Petrifilm and Colilert
results to inform community members (or HWTS users, in our case) of their own
treatment efficacy and water quality. While adding extra cost, this second visit to the
community or household could act as retraining and be incorporated into budgets at the
outset.
Consistent use of HWTS technologies is not well understood or documented (Figueroa,
2005). Household operational monitoring of Effective Use only provides a snapshot of
103
treatment and does not prove that people are drinking HWTS water on a consistent basis.
Without consistent use, maximum health benefits may not be realized. Moreover,
environmental conditions often greatly change the needs of HWTS throughout the year.
Use of SODIS may be infeasible during the rainy season not only because of lack of solar
radiation but due to the increase in turbidity of surface water source, necessitating
filtration or use of PUR. Consistent and Sustained Use of HWTS may thus not be
technology specific, but more generally apply to multiple water management techniques
(Meierhofer, 2008).
6.2 Field Interviews
While Appendix C: Household Monitoring Reports contains compiled notes on the field
interviews, a few consistent concepts noted throughout the interviews are reviewed
below.
Regular follow-up and long-term monitoring efforts were expensive but had the potential
to support Effective Use, as witnessed in a few of the organizations (i.e., with the Kale
Heywet Church and the Carter Center’s Guinea Worm Eradication Project (GWEP)).
With other implementations, however, monitoring and evaluation was not included
and/or was often the last thing on the funding list. M&E was often not put into budgets
upfront such that available funding was used up in other ways and M&E was never
conducted. This occurred with International Aid’s installation of biosand filters in
Kpanvo community near Tamale, Ghana. Most of the monitoring funds were used up in
the microbial testing of the filters a few days after installation, before the schmutzdecke
had fully formed, optimal treatment results were measurable, or maintenance techniques
could be analyzed. While this example singles out preliminary monitoring without
proper foresight, improperly thought-out monitoring of HWTS is not uncommon.
For their consumable products, PSI in Ethiopia and the Medentech distributors in both
Ethiopia and Ghana, namely EtMedix and Precision, respectively, had no programs
allocated in their budgets for household monitoring. While the Aquatabs groups claimed
that use of the product was simplified by the dosing method, with proven ability of
Effective Use and health benefit, the managers of these groups seemed to rely on the
assumption that because people bought the product they would use it properly. With
these agencies neither collecting user information (name, address, etc.) nor visiting the
households themselves, the author was unable to field-test the assumption of Effective
and Sustained Use of the products that were purchased. However, PUR’s initial entrance
into South American markets as a commercial product only showed 5% consistent use
following free distribution and user-claimed health impacts, shadowing doubt on the “if it
is bought it is used” assumption (Luby, 2008). Instead of household behavioral
monitoring, these three groups emphasized instead financial/commercial targets such as
monitoring through the supply chain, by responding to customer and distributor
complaints and tracking stock turnover.
When funding was used up following mass distributions of HWTS, such as with
Enterprise Works’ subsidized distribution of ceramic water purifiers (CWPs) near Accra,
Ghana, behavioral monitoring and evaluation efforts became reliant upon voluntary
104
activity by the community liaisons and salesmen and seemed unlikely to retain support of
Enterprise Works. In contrast, developing a system of community volunteers that helps
with initiation of distribution within the communities and who routinely checks usage of
filters as done with the Carter Center GWEP and Pure Home Water encouraged
community involvement in the project.
Organizations such as Oxfam and UNICEF have distributed PUR and sodium
hypochlorite solution (Waterguard, as produced by PSI Ethiopia) in acute watery diarrhea
(AWD, a.k.a. cholera) outbreaks in Ethiopia in recent years. While trainings are
generally held in large community meetings for these products (see Appendix G for PSI’s
training materials), and use incurs visible health benefits as claimed by the users
themselves, Sustained Use is often not noticed following the implementation. When
hoarding of PUR was noticed in households following an Oxfam distribution in southern
Ethiopia, the project switched to source protection as proper use was “not able to be
monitored in the household” (see interview with Gladys Inzofu in Appendix B).
Likewise, Tsegaye Gebre, Program Manager of Kale Heywet Church expressed doubt as
to whether private organizations would be able to appropriately monitor the use of
biosand filters. Given that his program spends about US $30 on the hardware for each
filter, the main cost is in training, support staff, and continued monitoring of the users,
even after 10 years of use. The total cost of each filter thus comes to about US $100,
with Tsegaye expressing that the high degree of Effective Use and Sustained Use
witnessed in his program is due in great part to the behavioral monitoring work that is so
costly, and would not likely be supported by a for-profit venture.
6.3 Best Practices for Field Monitoring
Below are four examples of “best practices” observed during field monitoring:
1. The Carter Center Guinea Worm Eradication Project (GWEP) working together
with the Ghanaian Ministry of Health has an extensive system of community
volunteers in place for weekly monitoring of each household using their cloth
filters. While a conversation with Philip Downs, Assistant Director of GWEP for
the Carter Center in Washington, D.C. during July 2008 has suggested that filters
are replaced more often than needed, rather than based on breakage rates as
assessed by community volunteers, >95% Effective Use of the cloth filters was
witnessed throughout the surveying of 56 joint Pure Home Water Kosim and
GWEP cloth filters by Kate Clopeck. This achievement is no doubt due to
appropriate training, vigilant monitoring by local entities and active replacement
of damaged filters.
2. Pure Home Water’s (PHW) sale of the Kosim ceramic water purifier (CWP) is
structured both through a salesperson from PHW and a community liaison from
within the village. The salesperson provides training and technical support, while
the community liaisons follow up with households to take new orders, monitor
problems and secure replacement filters. In addition to continued user familiarity
with the community representative, each Kosim filter storage unit has a sticker
105
with PHW’s name, phone number and address as well as with a set of detailed
pictorial and text instructions (PHW, 2008).
3. The joint PHW and UNICEF distribution of 5000 Kosim CWPs to flood affected
residents of the Upper East Region, Ghana, had an ambitious program for a
monitoring follow-up visit to 1 out of 5 houses who received a filter. This M&E
program was planned into the initial funding for the distribution. A four-person
PHW survey team covered a representative sample of the communities involved
in the distribution. Planned and funded M&E programs are essential, as is
flexibility in their execution. This is a good example of household monitoring of
HWTS taking place within the confines of an emergency situation.
4. The Kale Heywet Church (KHC) biosand filter project has been consistently
funded by Samaritan’s Purse of Canada since its inception in 1999. Sustained
funding has allowed KHC to plan a long-term monitoring and evaluation
campaign with which to help enforce Effective Use amongst its users, keeping
contact between users and KHC’s technicians throughout this timeframe.
6.4 Common Threads in Household Monitoring
A few themes of HWTS use common to many of the households visited in both Ethiopia
and Ghana are examined here.
1. Effective Use was notably hampered by user’s day to day responsibilities. In
particular, pregnant mothers and those with newly born children were often
unable to care for their biosand filters (BSFs) and ceramic water purifiers
(CWPs). At Household 2 at the field location of the Kale Heywet Church BSF
implementation (see Appendix C), a woman with a newborn was relying on
children to fetch water for her. While she knew of the practice of riverbed
filtration and knowledge of proper maintenance, the stresses of being pregnant
prevented her from completing these tasks. Her storage unit was very dirty and
without a lid, yet fecal contamination was not too great, showing proper treatment
with the BSF, an amazingly robust device! However, from measuring the total
coliform counts in the treated water, recontamination had clearly occurred in the
storage unit and the water that this user and her newborn were consuming was in
the WHO category of high risk of waterborne disease (see Table 2). Because no
one else was able to maintain the system, the HWTS system that was designed to
protect a mother and her children had failed them at this most crucial juncture,
when health was at a premium to the mother and her baby. Similarly, in the
Kpanvo village near Tamale, Ghana, a woman interviewed who was an owner of
both the CWP and BSF had stopped using the CWP altogether during the last
stretch of her pregnancy, as she too was relying on others to fetch water for her
and admits that she was not able to properly manage her water at that time. With
mothers as the sole attendants to the HWTS, safe water is not guaranteed
whenever the mother is pregnant or otherwise predisposed and without assistance
from another HWTS caretaker within the household, which could add up to weeks
106
out of the year. Similarly, sole caretaking as witnessed among men who purchase
the Kosim from Pure Home Water in Northern Ghana resulted in no great
improvement in access to clean water for the household, as some of these men
locked their Kosim in their room for their sole personal use. Inclusive training of
multiple users may be warranted for HWTS use, for in sharing the system, more
people are likely to learn about proper water management and obstacles to
Effective Use may be less likely to develop.
2. The frequency of cleaning by users of both BSF and CWP is higher than was
anticipated. When asking about cleaning frequency during household interviews,
it was very hard to get clear answers. During water testing of the Kpanvo BSFs,
many of the users reported cleaning their BSF every 3 days. For the 16 samples
tested using membrane filtration, results of treated water in safe storage as well as
freshly treated during the monitoring visit all turned up negative for E.coli or with
low risk (<10 E.coli per 100ml).
In kpanvo and other settings, users had a hard time remembering the last time
they cleaned their filters. People genuinely may not have been able to remember
the last time they cleaned the unit because they clean it so often with the high
turbidities encountered. Depending on the outcome of microbial water quality
testing, such avid maintenance based on flow rate may be positive or counter
productive. If the maintenance is scheduled (such as is warranted with the
cleaning of the CWP’s and BSF’s safe storage units) and the user can not
genuinely remember when they last cleaned the unit, then their scheduling
mechanism is not working, and ineffective use is suspect.
3. One of the limitations of single-visit unannounced household visits is their
inability to truly engage the interviewee. People often do not feel comfortable
enough with strangers (especially foreigners) in their homes to answer certain
questions. People may get the idea that they are supposed to answer a certain
way, and thus answers to questions about frequency of cleaning or hygiene habits
will not yield accurate answers. Similarly, people may be unwilling to answer
questions about their family’s or their own health, as these are private questions.
While in my monitoring of about 40 households during the trip, it appeared that
my own presence during the household visits contributed to causing all but one of
the users to not answer questions pertaining to diarrhea prevalence, or flat out
reject the possibility of their children having diarrhea in the foreseeable past, due
to the wonders of their great HWTS product! When foreigners are left outside of
the interview and translators from the region conduct the interviews, more
positive answers are found (Greene, 2008). However, even these results are
suspect as a health impact study of Aquatabs in Ethiopia reported growth in
waterborne disease prevalence throughout the first two weeks of the
implementation, showing people’s reluctance to give truthful answers until
monitors were known on first-name bases.
107
4. Emergency implementations of HWTS have limited ability to garner Consistent
or Sustained Use among users. Henock Gezahegn of PSI complained of the
inability to gain customers following emergency distributions, despite the instant
and meaningful health gains witnessed during the use of their products. This
“emergency product” mentality disrupts PSI’s advertising of PUR and
Watergaurd sodium hypochlorite solution as a “lifestyle product.” Following a
cholera outbreak when Watergaurd or PUR is distributed free of charge to the
user, people may come away thinking that these products are only needed during
cholera scares and thus they stop using the product and may tend to hoard a
supply for the next time emergency conditions resurface. The goal of preventing
the emergency through proper water management is thus lost. Outside of a statedeclared emergency, people may be ignorant of the threat to their health posed by
their water supply. So, despite their visible health impacts, neither Consistent nor
Sustained Use occurs, hampering Effective Use. Such was the case with the
Oxfam distribution described by Gladys Inzofu (Appendix B), which led Oxfam to
switch to a strategy of source protection. Often in remote areas far away from the
main roads, distribution networks are not easily established in emergency zones,
such that when emergency organizations such as UNICEF and Oxfam that
distribute consumable disinfectant HWTS products free of charge declare the
emergency over and move onto the next project, even the users who may wish to
continue HWTS use are left without outlets from which to buy the products nor
knowledge with which to order the product. This facet is especially important for
the CWP, for while Effective Use is witnessed following emergency distribution,
replacement must be readily available when breakage occurs in order to maintain
Sustained Use. Pure Home Water deals with this by posting their name, address,
and telephone number on their CWP units. Despite lack of product info, the
recipients of emergency aid or otherwise freely distributed HWTS are often
unaware of the donating agency. In the case of the joint UNICEF/Pure Home
Water distribution of 5000 CWPs to flood affected areas of Ghana, the name
UNICEF, though recognizable to the people involved, was never uttered.
Subsequent to PHW workers reporting on this lack of name recognition, UNICEF
took it upon themselves to add UNICEF stickers to the CWPs. In another
example, people who received BSFs from International Aid in Kpanvo village,
Ghana, were unaware of both the parent agency and the impetus behind the
distribution, claiming that the BSF was given to them “by the white man” (namely
Carl Allen, the Peace Corps coordinator who provided major assistance during
installation the filters in Kpanvo). Without adequate monitoring programs to
follow these hastened and/or emergency implementations, new users will not
know where to turn for replacement, with usage questions or when their HWTS
has problems, greatly hindering Effective and Sustained Use. Such problems are
often solved via direct purchasing of the products, community involvement during
project planning, proper labeling with contact information, and appropriate
monitoring set out prior to implementation.
5. The separation of raw and filtered water was not well understood among users of
all of the technologies witnessed. With SODIS, CWP, BSF, and cloth filters, in
108
which there is no residual protection offered to HWTS treated waters, this is of
special concern. Despite proper use and maintenance of HWTS systems, nonhygienic handling of treated water, including hands in storage units and washing
drinking cups with source water was one of the most commonly observed reasons
for not achieving Effective Use based on the author’s observational monitoring.
Revision of materials and training methods needs to include routine maintenance
of storage units for all HWTS, stressing the separation of untreated and treated
water, as well as using a dedicated clean cup for drinking.
6.5 Technology−Specific Observations
6.5.1 Pretreatment
Various pretreatment techniques achieve better treatment efficiency and lengthen times
between cleaning for a variety of HWTS techniques. In Northern Region, Ghana, settling
in primary storage units or the container used for fetching water from the source brought
turbidity consistently below 10NTU for the CWP, within the definitions of Effective Use
(Swanton, 2008). Riverbed filtration was promoted by the training program of Kale
Heywet Church, and Household 1 (Appendix C) had the best Effective Use witnessed for
the program, reducing 1000NTU source water to almost clear before use. Pretreatment
has the effect of reducing frequency of cleaning for both the BSF and CWP as well as
storage units, decreasing potential ineffective use. In regions of seasonably high turbidity
that can threaten the viability of certain HWTS techniques, settling and prefiltration may
have the potential to bring turbidity down to levels suitable for Aquatabs, sodium
hypochlorite solution, or SODIS, let alone lower the absolute risk level achieved through
use of these HWTS. Settling, riverbed-sand filtration, alum coagulation and other
pretreatment techniques need to be investigated and promoted to reduce the likelihood of
diarrheal disease in conjunction with the use of HWTS.
6.5.2 Maximum turbidity for use with the biosand filter
The biosand filter (BSF) can be a commercially viable product, as proven by technicians
who produce the filters for under US $10 as a side job in Machakos, Kenya, using a
cylindrical concrete design (which saves on material and labor costs). As expressed by
Tsegaye Gebre during an interview in January, 2008, the common belief was that BSFs
cannot be sold due to the large amount of follow up needed to ensure proper use.
The biosand filter is often looked to as an HWTS product for use with raw waters of low
and constant turbidities. CAWST recommends that biosand is used for raw waters with
turbidity <50–100NTU (CAWST, 2007). During the dry seasons as witnessed in
Ethiopia and Ghana during January, 2008, however, despite a high frequency of cleaning
and lack of settling or other pre-treatment, the BSF showed an ability to reduce high
turbidities as well or better than the CWP, consistently bringing turbidities to below
5NTU. Using BSFs with high turbidity waters showed promising results during field
testing in Ethiopia and Ghana during January, especially in conjunction with pretreatment
techniques of settling and river-bed filtration (see Appendix C). Further study under high
turbidity conditions is needed to confirm or remove the CAWST recommendations of 50100 NTU for influent waters to the BSF.
109
6.5.3 Dosing volume and pause times for the biosand filter
Recent research has shown that the BSF removes viruses with an efficiency of less than
90% (Stauber, 2007). Jenkins’ found high variability in virus removal rates with the
BSF, averaging 0.50 log removal with a standard deviation of 0.46 log. Viral, bacterial
and turbidity reduction is markedly greater with long pause times between refilling as
well as adding smaller amounts of water at a given time (Jenkins, 2008). These results
agree with those found by Baumgartner, who showed that pause times longer than 12
hours using dosing volumes equivalent to the pore space within the sand matrix resulted
in the best treatment. Baumgartner also showed that pause times greater than a day (36
hours) showed a decrease in treatment efficiency (Baumgartner, 2007). Sobsey
recommends not exceeding the pore volume when adding water, adding once in the
morning and once at night. Once the research into pause time and pore replacement
volume dosing is formally presented, new methods of use must be investigated.
Scheduled operation frameworks would require more training upon inception and
possible retraining of current users, and the potential to follow the framework also has to
be addressed.
6.5.4 Consistent use of PUR and other consumable HWTS
While Consistent Use is assumed by PSI Ethiopia of the repeat purchasers of PUR and
Watergaurd, as reported by their retailers, very little monitoring of Constitent, Sustained
and Effective Use of consumable products has been conducted. As demonstrated in the
initial commercial distribution of PUR in Guatemala in 2003, commercial indicators
(e.g., % of repeat customers) do not necessarily demonstrate Effective or Sustained Use
of the product. The original study recorded a 39% reduction in diarrhea. Yet, of the 462
households surveyed after 6 months, only 18% of the houses deemed “active repeat
users” through surveying results had residual FAC! Moreover, only 16% of households
had at least one sachet in the house and 12% had purchased PUR within the last two
weeks, usually only buying 4-5 sachets, which would not allow them to practice
Consistent Use as per the Figueroa definition. Only 5% of total deemed “active repeat
users” despite the large health impacts witnessed by users just a few months earlier
(Reller, 2003). While the price of PUR was high (US $0.14 per sachet), and Procter and
Gamble decided to distribute PUR as a subsidized or free product from then on, this study
has worrying implications for Consistent Use of consumable HWTS.
The minimum recommended quantities of consumable HWTS products available at the
time of a household monitoring visit, as laid out in the Physical Inspection sections of
their Effective Use Write-ups, are intended as guidelines for demonstrating Consistent
Use for these particular products, in conjunction with residual FAC present at the time of
monitoring. These recommendations are not based on monitoring data, as little exists,
and thus might suggest overly ambitious reserve quantities of product to be present in the
household. Specific monitoring programs need to adjust minimum quantities of their
consumable products present within the household to their specific circumstances and
monitoring results. Both of these metrics should be seen to grow over time as the
behavior change of Consistent Use is adopted. Likewise, if intermittent use of
consumable HWTS is found to be the norm through future monitoring efforts (i.e., if
110
hoarding of consumable disinfectants witnessed following emergency programs is the
norm), health impact assessment on intermittent use may be warranted.
The intention of the Effective Use monitoring frameworks laid out for sodium
hypochlorite solution, PUR and Aquatabs is to provide both commercial and non-profit
agencies a low cost and efficient means of monitoring for free chlorine, microbial water
quality, and overall Effective Use. If users are known and documented during
distributions or at the kiosk during sales, monitoring visits can be arranged and Effective
Use can be judged with consumables and even in emergency situations. This would
provide two new avenues for HWTS monitoring where currently many assumptions exist,
yet little factual evidence.
6.5.5 Ceramic Pot Filter
The main burden of waterborne diseases falls on children, especially those under the age
of five. A single ceramic pot typically produces enough water for a family of five. As
witnessed in Northern Region Ghana, only certain people in the household are seen to use
the Kosim, in some instances limited just to the husband who purchased it. Since
children are target end users of HWTS in the goal of reducing waterborne disease burden,
then Effective Use of the ceramic water purifier (CWP) cannot be based on a single pot
filter for an entire family in regions of large family sizes or elevated water consumption.
However, purchasing and operating more than one filter is also a hindrance to Effective
Use, so expectations for number of filters per family have to be reasonable and one filter
is better than none.
111
112
7. Conclusion
Monitoring campaigns can lead to various improvements in the given distribution of
HWTS. Following an evaluation of their pilot biosand filtration project that showed loss
of sand due to their method of cleaning, the Kale Heywet Church changed its training
protocol to teach a wet harrowing technique to the users in their scale-up project of
almost 10,000 biosand filters. Similarly, refinements of the technologies and their
distribution have been made to HWTS following operational monitoring.
High percentages of users practicing Effective Use of HWTS filtration technologies have
been documented here, with vigilant monitoring campaigns associated with higher
percentages of effectively used systems. Simplified household monitoring frameworks
and associated field techniques for measuring water quality have been presented here
with the intention of providing useful tools for organizations to conduct operational
monitoring and gather data on their customers’ usage. Through vigilant monitoring at the
household, groups are able to increase the Effective Use of their HWTS during and after
implementation.
The importance of this document will ultimately lie with the utilization of the Effective
Use Briefs and Monitoring Checklists by members of the WHO-hosted Network on
HWTS and others. Its inclusion in the MIT compendium of behavioral and commercial
indicators, to be prepared by Kate Clopeck in 2009 will provide a body of work which
organizations throughout the Network and the world can use to operationally monitor
their implementations, both inexpensively and in real time.
Health-impact based cost effectiveness of HWTS compares well with that of improved
sources yet requires significantly less capitol than the piped water systems that are
ultimately the most desirable solution (Clasen, 2006). Many parts of this world,
however, are decades away from receiving piped distribution networks with a clean and
reliable supply of water, and HWTS provide an alternative approach in the goal of greater
access to safe water. They require low capital investment, little infrastructure other than
a suitable distribution network, and can promote self-sustaining business models.
The 2008 WHO/UNICEF/JMP Progress Report recognizes that the quality of source
water may not reflect the quality at point of use. Source quality may thus not be as
strongly associated with changes in diarrhea occurrence. There is “increasing evidence
that simple, low-cost interventions at the community level are capable of improving the
microbial quality of domestically stored water and of reducing the associated risks of
diarrhea disease” (WHO, 2006). Both quantity and quality of drinking water have to be
ensured in order to improve health. Although HWTS technologies do not improve access
to larger quantities of safe water, they can ensure the safety of water at the point of use.
HWTS can work hand-in-hand with improved sources to maintain quality of water to the
home where it is needed most.
113
The beauty of HWTS technologies are that they puts the control over family health back
into the family’s hands, so that households are not left without access to clean water from
stalled or unsuccessful larger scale government and donor water projects. Needless to
say, HWTS goes hand-in-hand with continual development of water-services
infrastructure, source improvement, and effective treatments such as oral rehydration
therapy. Meanwhile, self-empowerment is the key to this intervention. The overall
impact is in the hands of the user, and yet important work is yet to be done to ensure that
people are able to use these technologies effectively.
114
115
116
References
Abate, Z. (1994). Water resource development in Ethiopia. Reading, UK: Ithaca Press.
About, Inc. (2007). Geography and Map of Ghana. The New York Times Company.
<http://geography.about.com/library/cia/blcghana.htm>. Accessed 9 Dec. 2007.
Acra, A., Raffoul, Z., & Karahagopian, Y. (1984). Solar Disinfection of Drinking Water and Oral
Rehydration Solutions. Paris: UNICEF.
Adams, M., & Baker, D. (2007). Monitoring and Evaluation Indicators: Household Water Treatment.
Power Point presentation. Calgary: CAWST.
Aikhomu, S.E., Brieger, W.R., & Kale, O.O. (2000). Acceptance and Use of Communal Filtration Units in
Guinea Worm Eradication. Tropical Medicine and International Health, 5 (1), 47-52.
Albert, J. (personal communication, May, 2008)
Allgood, G. (personal communication, May, 2008)
Aquaya, (2008). Standard Operating Procedure for the Deployment of Procter & Gamble’s PUR Purifier of
Water in Emergency Response Settings. http://www.psi.org/Pur-emergencyrelief/resources/aquayaSOP.pdf Accessed May 22, 2008.
Arnold, B., & Colford, J. (2007). Treating Water with Chlorine at Point-of-Use to Improve Water Quality
and Reduce Child Diarrhea in Developing Countries: A Systematic Review and Meta-Analysis. American
Journal of Tropical Medicine and Hygiene, 76(2), 2007, 354–364.
Baffrey, Robert M. N. (2005). Development of Program Implementation, Evaluation, and Selection Tools
for Household Water Treatment and Safe Storage Systems in Developing Countries. Masters thesis,
Massachusetts Institute of Technology, Cambridge, MA: MIT.
http://web.mit.edu/watsan/docs_theses_kenya.html Accessed September 4, 2008
Baker, D. (personal communication, December, 2007)
Baumgartner, J., Murcott, S., & Ezzati, M. (2007). Reconsidering ‘appropriate technology’: the effects of
operating conditions on the bacterial removal performance of two household drinking-water filter systems.
Environmental Research Letters, 2.
BBC (2003). Cloth filter could cut cholera deaths. London: BBC. 1/14/2003.
Brown, J. (personal communication, December, 2007)
Brown, J. (2007) Effectiveness of Ceramic Filtration for Drinking Water Treatment in Cambodia. Ph.D.
Dissertation. Chapel Hill: University of North Carolina.
Cairncross, S., Blumenthal, U., Kolsky, P, Moraes, L., & Tayeh, A. (1996) The public and domestic
domains in the transmission of disease. Tropical Medicine and International Health, 1, 27–34.
CAWST (2007). Installation Operation & Maintenance Manual: Biosand Water Filter. Version 2007-01.
Calgary: CAWST..
CAWST (2008). Poster 27: Disinfection Black English. Calgary: CAWST.
117
CAWST (2008). Poster 21, Latin America English: Biosand Filter Maintenance. Calgary: CAWST.
CDC (2001). Safe Water Systems for the Developing World: A Handbook for Implementing HouseholdBased Water Treatment and Safe Storage Projects. Atlanta: Centers for Disease Control and Prevention.
CDC (2005). Chlorine Dosing Report: Kathmandu Valley, Nepal, Janakpur, Nepal and Biratnagar and
Surrounding Areas, Nepal. Atlanta: CDC, PSI:.
Centers for Disease Control and Population Services International (2005). Chlorine dosing report:
Kathmandu Valley, Nepal, JanakPur, Nepal and Biratnagar and surrounding areas, Nepal. Atlanta: CDC.
Central Statistical Agency [Ethiopia] and ORC Macro. (2006) Ethiopia Demographic and Health Survey
2005. Addis Ababa, Ethiopia and Calverton, Maryland, USA: Central Statistical Agency and ORC Macro.
Chau, P., Duc, L., & Tap, T. (1996). Measurement of the Effectiveness of Drinking Water Purification
using Aquatabs, Pantocid Tablets, and a Water Filter Box. Hanoi, Vietnam: Department of Sanitation and
the Environment, Institute of Military Medicine.
Clasen, T., & Edmondson, P. (2006). Sodium dichloroisocyanurate (NaDCC) tablets as an alternative to
sodium hypochlorite for the routine treatment of drinking water at the household level. International
Journal of Hygiene and Environmental Health, 209, 173-181.
Clasen, T.; Cairncross, S.; Haller, L.; Bartram, J., & Walker, D. (2007). Cost-effectiveness of water quality
interventions for preventing diarrhoeal disease in developing countries. J Water Health; 5(4), 599-608.
Clasen, T., & Haller, L. (2007). Water Quality Interventions to Prevent Diarrhoea: Cost and CostEffectiveness. Geneva: WHO.
Clasen, T., Roberts, I., Rabie, T., Schmidt, W., & Cairncross, S. (2006). Interventions to improve water
quality for preventing diarrhoea. (A Cochrane Review). The Cochrane Library, 3. Oxford: Update
Software.
Clasen, T. (2006). "Household Water Treatment for the Prevention of Diarroeal Disease." PhD thesis,
University of London, London School of Hygiene & Tropical Medicine, London.
Clasen, T., Schmidt, W., Rabie, T.,, & Cairncross, S. (2007). Interventions to improve water quality for
preventing diarrhoea: systematic review and meta-analysis. British Medical Journal, London.
Colwell, R., Huq, A., Islam, M., Aziz, K., Yunus, M., Khan, N., Mahmud, A., Sack, R., Nair, G.,
Chakraborty, J., Sack, D., & Russek-Cohen, E. (2003). Reduction of cholera in Bangladeshi villages by
simple filtration. PNAS, 100(3), 1051–1055.
Crapper, D. (2007). Q3 presentation. Addis Ababa: PSI.
Crump, J., Otieno, P., Slutsker, L., Keswick, B., Rosen, D., Hoekstra, R.M., Vulule, J., & Luby, S. (2005)
Household based treatment of drinking water with flocculant-disinfectant for preventing diarrhoea in areas
with turbid source water in rural western Kenya: cluster randomised controlled trial. BMJ 2005, 331,478-.
Cruz, O. (1996). Evaluation of the product sodium dichloroisocyanurate in water for consumption by
communities with basic sanitation and a pattern of water-borne disease. Ministério da Saude, Rio de
Janeiro, Brazil.
Davies-Colley, R. J., & Smith, D. G. (2001). Turbidity, Suspended Sediment, and Water Clarity: A
Review. Journal of the American Water Resources Association, 37(5), 1085.
118
De Angelis, J., & Neves. V. (1998). “DESINFECÇÃO DE ÁGUAS COM SUBSTÂNCIAS À BASE DE
DICLOROISOCIANURATO DE SÓDIO.” AIDIS.
Dejachew, G. (2002) “Evaluation of household biosand filters in Ethiopia.” Loughborough, UK: WEDC.
Downs, P. (personal communication, July, 2008)
Earwaker, P. (2006). Evaluation of household biosand filters in Ethiopia. Masters thesis, , Silsoe, UK:
Cranfield University.
EAWAG/SANDEC SODIS Technical Note Series: 1-17.
EAWAG/SANDEC (1997). SODIS News No. 1.
Edmondson, P. (personal communication, July, 2008)
Elliott, M., Stauber, C., Koksal, F., DiGiano, F., & Sobsey, M. (2008). Reductions of E. coli, echovirus
type 12 and bacteriophages in an intermittently operated household-scale slow sand filter. Water Research
42 (2008 ) 2662 – 2670.
Esrey, S., Habicht, J., Latham, M., Sisler, D., & Casella, G. (1988). Drinking water source, diarrheal
morbidity, and child growth in villages with both traditional and improved water supplies in rural Lesotho,
Southern Africa. Am J Public Health; 78: 1451-1455.
Fewster, E., Mol, A., & Wiessent–Brandsma, C. (2004). The long term sustainability of household biosand filtration. 30th WEDC International Conference, Vientiane, Lao PDR.
Fewster, E. (personal communication, May, 2008)
Fewtrell, L., Colford, J. (2004) Water, Sanitation, and Hygiene: Interventions and Diarrhoea; A Systematic
Review and Meta-analysis. Lancet Infect Dis. 5(1), 42-52.
Figueroa, M. E. (2005). Definition of behavior-related variables for household water treatment and storage.
Unpublished notes from a lunchtime meeting at the WHO Network on HWTS Symposium. Bangkok: June,
2005.
Green, V. (2008) Household water treatment and safe storage options for Nothern Region Ghana:
consumer preference and relative cost. Masters Thesis. Cambridge, MA: MIT. Accessed September 4,
2008. http://web.mit.edu/watsan/docs_theses_ghana.html
GSB (1998). Water quality – requirement for drinking water. Ghana Standards 175 Part 1:1998.
Havelaar, A., & Melse, J. (2003). Quantifying public health risks in the WHO Guidelines for DrinkingWater Quality: a burden of disease approach. Report 734301022/2003 RIVM www.rivm.nl.
Hernandez, O. (2006). Measuring behavioral outcomes when promoting household water treatment and
storage. Discussion paper prepared for the January 22-26, 2007, E-Conference Hosted by the
USAID/Hygiene Improvement (HIP).
Hernandez, O. (2008). Household survey: Baseline questionnaire. Learning by Doing Initiative,
Implemented by WSP and the USAID Hygiene Improvement Project. Unpublished document. Academy
for Educational Development: Washington, D.C.
Hernandez, O. (personal communication, April, 2008)
119
.
Howard, G., Bartram, J. (2003). Domestic Water Quantity, Service Level and Health. WHO, Geneva.
Huq, A., Xu, B., Chowdury, M., Islam, M., Montilla, R., Colwell, R. (1996). “A Simple Filtration Method
To Remove Plankton-Associated Vibrio cholerae in Raw Water Supplies in Developing Countries.”
Applied and Environmental Microbiology. p. 2508–2512 Vol. 62, No. 7.
Huttly SR, Blum D, Kirkwood BR, Emeh RN, Okeke N, Ajala M, Smith GS, Carson DC, DosunmuOgunbi O,& Feachem RG (1990) The Imo State (Nigeria) Drinking Water Supply and Sanitation Project,
2. Impact on dracunculiasis, diarrhoea and nutritional status. Trans R Soc Trop Med Hyg. 84(2),316-21.
Jacobs. (1986). IDEXX Summary: Comparison Of Membrane Filter, Multiple-Fermentation-Tube, and
Presence-Absence Techniques for Detecting Total Coliforms in Small Community Water Systems. Applied
and Environmental Microbiology, p.1007-1011.
Jamison, D. T., Breman, J. G., Measham, A. R., Alleyne, G., Claeson, M., Evans, D. B., Jha, P., Mills, A.,
& Musgrove, P. (2006). Disease Control Priorities in Developing Countries. Oxford University Press;
World Bank, Washington DC.
Jenkins, M., Tiwari, S., & Darby, J. (2008). Intermittent Slow Sand Filters for Household Use in
Developing Countries: Factors Affecting Bacterial, Virus, and Turbidity Removal. Abstract as submitted
by author Jenkins through personal communication, June 2008. Poster as presented at WHO Annual
HWTS Conference, June 2-5 2008, Accra, Ghana.
Jensen, P., Ensink, J., Jayasinghe, G., van der Hoek, W., Cairncross, S., & Dalsgaard, A. (2002). Domestic
transmission routes of pathogens: the problem of in-house contamination of drinking water during storage
in developing countries. Tropical Medicine and International Health 7 (7), 604–609.
Johnson, S., Peletz, R., & Murcott, S. (2008). Results from Household Ceramic Filter Evaluation in
Northern Ghana. Unpublished paper submitted for presentation at the 33rd WEDC International
Conference, Accra, Ghana, 2008.
Kikkawa, I. (2008). Modification of a Biosand Filter in the Northern Region of Ghana. Masters Thesis,
Cambridge, MA: Massachusetts Institute of Technology.
http://web.mit.edu/watsan/docs_theses_ghana.html Accessed September 4, 2008.
Kuechler, T.C. (2004). The Great CYA Debate. Presented at the 2004 International Pool & Spa Expo. Las
Vegas, NV.
KWAHO (2005) Poster on the use of SODIS.
Lantagne, D. S. (2001) Investigation of the Potters for Peace Colloidal Silver Impregnated Ceramic Filter –
Report 1: Intrinsic Effectiveness. Allston, MA: Alethia Environmental. Available at <http://www.edccu.org/pdf/report1-final.pdf>. Accessed August 10, 2008.
Lantagne, D. S. (2001a) Investigation of the Potters for Peace Colloidal Silver Impregnated Ceramic Filter
– Report 2: Field Investigations. Allston, MA: Alethia Environmental. Available at
<http://pottersforpeace.org/wpcontent/uploads/alethia-report-2.pdf>. Accessed August 10, 2008.
Lantagne, D. S. Quick, R., & Mintz, E. (2006). Household water treatment and safe storage options in
developing countries: a review of current implementation practices. Woodrow Wilson International Center
for Scholars' Environmental Change and Security Program.
http://www.wilsoncenter.org/waterstories/Household_Water_Treatment.pdf Accessed August 10, 2008.
120
Lantagne, D., & Gallo, W. (2008). Safe Water for the Community: A guide for establishing a communitybased safe water system program.” Edition 1, January 2008. Atlant: Centers for Disease Control,
Lantagne, D. (personal communication, July, 2008)
Lentz, R. (personal communication, July, 2008)
Levy, K. (2007). Environmental Drivers of Water Quality and Waterborne Disease in the Tropics with a
Particular Focus on Northern Coastal Ecuador. Berkeley: University of California.
Lopez, A. D., Mathers, C. D., Ezzati, M., Jamison, D. T., Murray, & Christopher J. L. (2006). Global
Burden of Disease and Risk Factors. Oxford University Press; World Bank, Washington DC.
Losleben, T. (2008) Pilot Study of Horizontal Roughing Filtration in Northern Ghana as Pretreatment for
Highly Turbid Dugout Water. Masters Thesis, MIT: Cambridge, MA.
http://web.mit.edu/watsan/docs_theses_ghana.html Accessed September 4, 2008.
Luby, S. et al. (2008) Difficulties in Bringing Point-of-Use Water Treatment to Scale in Rural Guatemala.
Am. J. Trop. Med. Hyg., 78(3), 382–387.
Macedo, J. (1997), Determinacao de trihalomethanes em aguas de abastecimento publico e industria de
alimentos. Dissertacao (Doutorado em Ciencia e Tecnologia de Alimentos) – Univesidade Federal de
Vicosa, Vicosa, MG.
Maertans, M., & Buller, A. (2005) Kale Heywet Ethiopia Household Water and Sanitation Project
Evaluation. Calgary, Canada: Samaritan’s Purse International Relief,
Mahin, T. (personal communication, August, 2008)
Medentech (2006). Tanzania: The Aquatabs Experience. Wexford, Ireland.
Medentech (2006). Aquatabs: Technical Report. Wexford, Ireland.
Meierhofer, R., & Wegelin, M. (2002). Solar Water Disinfection: A Guide for the Application of SODIS.
SANDEC Report No 06/02. Duebendorf, Switzerland.
http://www.sodis.ch/files/SODIS_Manual_english.pdf
Meierhofer, R. (personal communication, July, 2008)
Méndez-Hermida, F., J.A. Castro-Hermida, E. Arez-Mazás, S.C. Kehoe, & Kevin G. McGuigan. (2005).
Effect of batch process solar disinfection (SODIS) on the survival of Cryptosporidium parvum oocysts in
drinking water. Applied Environmental Microbiology 71(3), 1653–1654.
Metcalf, R. (2006). Point of Use Water Testing in Developing Countries. Sacramento, CA: California State
University,
Metcalf, R. (2008) Water Testing/Solar Water Pasteurization in Kenya and Tanzania. Sacramento, CA:
California State University,
Metcalf, R. (personal communication, June, 2008)
Metcalf, R. Undated. E.coli Count Petrifilm operating instructions.
Metcalf, R. Undated. Colilert MPN Test operating instructions.
121
Metcalf, R. Undated. Microbial water quality field reporting format.
Mjengera, H. (2005). Testing the Effectiveness of Aquatabs in Disinfecting Drinking Water. Dar es
Salaam, Tanzania: Water Quality Laboratories Unit, Ministry of Water and Livestock Department,
Mol, A. (2000). Evaluation/Final Report: Family Bio-Sand Filtration Project in Machakos District, Kenya.
Nairobi: MEDAIR, June 1999-September 2000.
Molla, N.A., Hossain, A., Edmondson, P., & Shipin, O. (2005) Pilot study on the effect of an intervention
using sodiumdichloroisocyanurate tablets (Aquatabs) for drinking water treatment in Dhaka, Bangladesh.
Moran, J. (personal communication, May, 2008)
Murcott, S. (2006). Implementation, Critical Factors and Challenges to Scale-Up of Household Drinking
Water Treatment and Safe Storage Systems. Background Paper on HWTS for the Electronic Conference
May 12-22, 2006, Hosted by USAID / Hygiene Improvement Project (HIP).
Murcott, S. (2005). Behavioral and financial targets behavioral and financial targets in implementing,
scaling up, in implementing, scaling up, monitoring and evaluating monitoring and evaluating household
water treatment household water treatment and safe storage technologies and safe storage technologies.
Annual Meeting of International Network to Promote HWTS, Quito, Ecuador.
Murcott, S. (2007). Guinea Worm Cloth Filter: Household Water Treatment and Safe Storage Product and
Implementation Fact Sheet. http://stellar.mit.edu/S/project/hwts-network/materials.html#topic3/ Accessed
June, 2008.
Muriuki, G. (2007). Powerpoint Presentation at the WHO Stakeholder Forum on Household Water
Management in Addis Ababa, October 2007. Nairobi: Medipharm.
Norton, D. M. (2003). Field Trial of a Flocculent-Disinfectant Point-of-Use Water Treatment for
Improving the Quality and Microbial Safety of Surface Pond Water in Bangladesh. 10th Asian Conference
on Diarrhoeal Diseases and Nurtition (ASCODD), Dhaka, Bangladesh.
Oates, P., Shanahan, P., & Polz, M., (2003). Solar disinfection (SODIS): simulation of solar radiation for
global assessment and application for point-of-use water treatment in Haiti. Water Research, 37, 47–54.
Ogutu P, Garrett V, Barasa P, Ombeki S, Mwaki A, &Quick R. (2001) Seeking Safe Storage: A
Comparison of Drinking Water Quality in Clay and Plastic Vessels. American Journal of Public Health.
91 (10),1610–1611.
OXFAM – DELAGUA (2000) Portable Water Testing Kit Users Manual. (abridged web download
version) Revised and updated 4th edition. www.delagua.org Accessed August, 2008.
Palmateer, G., Manz, D., Jurkovic, A., McInnis, R., Unger, S., Kwan, K., & Dutka, B. (1999). Toxicant
and Parasite Challenge of Manz Intermittent Slow Sand Filter. Environmental Toxicology 14: 217-225.
Patel, U., Garg, S., Mani, E., & Gupta, G. (2008) Global Household Water Treatment & Safe Storage
Monitoring and Evaluation Project. G-Lab Final Report. MIT Sloan School of Management: Cambridge,
MA.
Peletz, R. (2006). Cross-Sectional Epidemiological Study on Water and Sanitation Practices in the
Northern Region of Ghana. Thesis. Cambridge, MA: Massachusetts Institute of Technology.
Potters for Peace (2008) <http://pottersforpeace.org/>. Accessed 9 May 2008
122
POUZN (2007). Best practices in Social Marketing Safe Water Solution for Household Water Treatment:
Lessons Learned from Population Services International Field Programs. Research Report No. 2. USAID
http://www.psi.org/resources/pubs/usaid-wwd.pdf Accessed February, 2008.
Pragnya, A. Y. (2005). "Appropriate Water Treatment for the Nyanza Province of Kenya." Masters thesis,
Massachusetts Institute of Technology, Cambridge, MA. Accessed September 4, 2008.
http://web.mit.edu/watsan/docs_theses_kenya.html
PSI (2006) Diarrhea DALYs Ethiopia 2006. Addis Ababa: PSI.
PSI (2007) DALYs prevented using SWS 2007. Addis Ababa: PSI.
PSI (2007) D3: Doing Development Differently: Annual Report 2006-2007. Addis Ababa: PSI.
PSI (2007) Sales of Waterguard 2006 & 2007. Addis Ababa: PSI.
PSI (2008) PUR Usage Instructions in 4 Languages. Addis Ababa: PSI.
PSI Research Division (2004) PSI Behavior Change Framework “Bubbles”: Proposed Revision.
Washington, DC: PSI.
PSI Research & Metrics. (2007). Ethiopia (2006): Maternal and Child Health TRaC Study among
Caregivers of Children Fourteen Years and Younger in Addis Ababa and SNNPR. First Round. Addis
Ababa: PSI.
Pure Home Water-Ghana, (2008) Ceramic Pot ("Kosim") Filter Training Manual.
http://stellar.mit.edu/S/project/hwts-network/materials.html#topic6
Reller, M. E. et al., (2003 ). “A Randomized Controlled Trial of Household-Based Flocculant-Disinfectant
Drinking Water Treatment for Diarrhea Prevention in Rural Guatemala.” American Journal of Tropical
Medicine and Hygiene, 69(4), pp. 411–419
Rivera, R. (2008). “Ceramic Weapons of Mass Bacterial Destruction: produced locally to
help make water potable at the household level, worldwide.” Accessed 24 July, 2008.
<http://www.changemakers.net/en-us/node/5662>.
Roberts, L., Y. Chartier, O., Chartier, G. Malenga, M. Toole, & H. Rodka. (2001). Keeping clean water
clean in a Malawi refugee camp: a randomized intervention trial. Bulletin of the World Health Organization
79, 280-287.
Rose, A., Roy, S., Abraham, V., Holmgren, G., George, K., Balraj, V., Abraham, S., Muliyil, J., Joseph, A.,
& Kang, G. (2006). Solar disinfection of water for diarrhoeal prevention in southern India. Arch. Dis.
Child.; 91,139-141.
Quick, R. (personal communication, May, 2008)
Safe Water Systems for the Developing World: A Handbook for Implementing Household-Based Water
Treatment and Safe Storage Projects. CARE/CDC Health Initiative, Atlanta.
http://www.cdc.gov/safewater/manual/sws_manual.pdf Accessed December, 2007.
Saladin, M. (personal communication, July, 2008)
Samaritan’s Purse, (2001) “BioSand Household Water Filter.” 4rth Ed., Calgary, Canada.
Schlosser, O., Robert, C., Bourderioux, C., Rey, M., & de Roubin, M.R., (2001). Bacteria removal of
inexpensive portable treatment systems for travelers. Journal of Travel Medicine 8, 12-18.
123
Shenkut, M. (1996). Experiences with Slow Sand Filtration in Ethiopia: Advances in Slow Sand and
Alternative Biological Filtration. Ed: Graham, N. J. D. Wiley and Sons, England.
Sobsey, M. (2007). Household Water Treatment for the Developing World by Biosand Filters: Recent
Developments from the Lab and the Field. Powerpoint Presentation delivered at WHO HWTS Network
meeting in Addis Ababa, October, 2007.
Souter, P. F. (2003). Evaluation of a new water treatment for point-of-use household applications to
remove microorganisms and arsenic from drinking water. Journal of Water and Health, 01.2. IWA
Publishing.
Stauber, C., Elliot, M., Koksal, F., Ortiz, G., DiGiano, F, & Sobsey, M. (2006) Characterisation of the
biosand filter for E. coli reductions from household drinking water under controlled laboratory and field
use conditions. Water Science & Technology 54 (3), 1–7. IWA Publishing.
Stauber, C. (2007). The Microbial and Health Impact of the Biosand Filter in the Dominican Republic: A
Randomized Controlled Trial in Bonao. PhD Dissertation. Chapel Hill: UNC.
Swanton, A. (2008). Evaluation of the Complementary Use of the Ceramic (Kosim) Filter and Aquatabs in
Northern Region, Ghana.” Masters Thesis, Cambridge, MA: Massachusetts Institute of Technology..
http://web.mit.edu/watsan/docs_theses_ghana.html Accessed September 4, 2008
Sphere Project (2004). Humanitarian Charter and Minimum Standards in Disaster Response. Geneva: The
Sphere Project,
Tadesse, D. et al. (2006) Rapid Assessment of Drinking-Water Quality in the Federal Republic of
Ethiopia: Country Report. Addis Ababa: Federal Government of Ethiopia
World Health Organization and United Nations Children’s Fund Joint Monitoring Programme for Water
Supply and Sanitation (JMP). (2008) Progress on Drinking Water and Sanitation: Special Focus on
Sanitation. New York: UNICEF, Geneva: WHO.
Wegelin, Martin, S. Canonica, K. Mechsner, T. Fleischmann, F. Pesaro, & A. Metzler. (1994). Solar water
disinfection: Scope of the process and analysis of radiation experiments. Aqua (Oxford) 43(4), 154–169.
WELL (1998), Guidance manual on water supply and sanitation programmes. Loughborough, UK: WEDC.
WHO, (1997) Guidelines for Drinking Water Quality, Vol. III, Surveillance and Control of Community
Supplies. 2nd Ed. Geneva: WHO.
WHO, (2004). “WHO Guidelines for Drinking Water Quality,” 3rd edition. Geneva: World Health
Organization. <http://www.who.int>.
WHO Collaborating Center for Research, Training and Eradication of Dracunculiasis. (2008). Guinea
Worm Wrap-Up #180.
WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation (JMP), (2004). Meeting the
MDG drinking water and sanitation target: a mid-term assessment of progress. Geneva WHO.
Wiesent-Brandsma, C. (2004). Evaluation of SSF Project in Maintirano, Madagascar. Antananarivo:
Medair.
World Factbook (2007) Country Comparison: Life Expectancy at Birth. Washington, DC: CIA.
<http://www.indexmundi.com/g/r.aspx?c=gh&v=30>. Accessed 9 Dec. 2007.
124
Wright, J., Gundry, S., & Conroy, R. (2004). Household drinking water in developing countries: a
systematic review of microbiological contamination between source and point-of-use. Tropical Medicine
and International Health, 9(1), 106–117.
125
126
Appendix A:
Behavior and Sustained Use Questionnaire
HWTS Monitoring and Evaluation Project
Behavior and Sustained Use Questionnaire
Kate Clopeck and Matt Stevenson
Interest is strong among various Network partners to develop and widely share M&E
tools. Until now, efforts to systematically monitor and evaluate (M&E) household water
treatment and safe storage (HWTS) implementation and scale up have been largely
restricted to individual organization’s initiatives and information on M&E methods,
targets, indicators, tools and results are few and exist mainly in unpublished literature. In
A new initiative to expand that preliminary work, called the “HWTS M&E Project,” is a
collaboration between USAID’s Hygiene Improvement Project (HIP), the WHO Network
Secretariat and a seven-person MIT team comprised of Masters of Engineering and Sloan
School of Management MBA faculty and students, who will identify and share the M&E
targets, indicators, tools and results applied by organizations engaged in HWTS
implementation and scale up.
More information on the “HWTS M&E Project” can be found at:
http://web.mit.edu/watsan -> “HWTS M&E Project”
Pre-Interview
1. Gather background information and business description of organization
2. If possible, gather background on contact being interviewed
3. Visit organization website (if available)
Questionnaire:
Introductions (5 Mins)
ƒ Interviewer introductions
ƒ Interviewee introduction
ƒ Walk through agenda and provide quick overview of purpose
Product Questions ( > 5 Mins)
ƒ Can you briefly describe your product?
ƒ Do you have a technical data sheet? (If yes, could you please send us this information
as an e- copy or hard copy?)
Behavioral Questions (15 minutes)
ƒ How do you define “Effective Use” of your product?
ƒ How do you ensure that your product is being used effectively by households?
ƒ How do you measure the outcomes of this work?
127
ƒ
ƒ
ƒ
ƒ
Have you performed any health impact studies? (if yes, could you please send us this
data or any relevant report)
Have you ever performed any water quality testing of the HWTS product in user
households? (if yes, what water quality measures have your tested, what test methods
have you used, could you please send us this data or any relevant report)
Do you provide customers with a step-by-step guide on product assembly, operation
& maintenance or other general information that is provided to households who
obtain your product? Do you have this as a hand-out, written on the product itself or
what? Could you please provide an e-copy or hard copy). Is this always provided or
only on request.
Does the product you disseminate have a replacement period/expiration date. If yes,
how is that information communicated to the users?
Coverage and Sustained Use Questions (15 minutes)
ƒ What is your target population?
o How was this determined?
o How many houses have you reached do far?
ƒ How was measured? (sales vs follow up visits)?
ƒ Total sales volume?
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
ƒ
How do you measure coverage and sustained use of the product by your target
audience?
Do you distinguish between types of users (frequent/infrequent, correct/incorrect?)
Do you have any other way of measuring coverage?
How is the household drinking water treatment product delivered to the target group?
Do you visit that group at the time of dissemination?
Do you do follow-up visits for service visits?
Do you follow-up for monitoring and evaluation?
o How often? (1 month?-ROA, 1 year?)
o What do you check for? (Methods?)
o How many households?
o How do you pick the households? (if not all costumers)
o Could you please send any data?
o How many employees are dedicated to follow-up visits. How much time?
o If no follow-up visit, do you have any other way of measuring sustained use?
Do you rely of self-reports of efficacy, staff monitors, village volunteers, other?
o Could you please send us data or reports of monitoring?
o Any comments/concerns with self-reporting?
How many units of your product are needed to supply safe water for 1 year for one
household?
What training materials do you have for your product?
Do you consider coverage/sustained use a metric for measuring the success of your
product? Of your organization?
Why do you think coverage and sustained use is so hard to measure?
128
Appendix B: Fieldtrip Interviews
Persons Interviewed
Organization
Page
Ethiopia:
Tsegaye Gebre
Henock Gezahegn
Menassie Kifle & Kassa
Gladys Inzofu
– Kale Heywet Church
– Population Services International
– EtMedix / Medentech
– Oxfam Consultant
130
132
135
136
Ghana:
Jesse Jones Agbanya & Ebenezer Aidoo
– Precision dx
Abaazan Peter Adagwine, Shak Ibrahim, Peter Alhassan – Pure Home Water
Mumuni K. Osman
– International Aid
Atsu Titiati
– Enterprise Works
137
139
141
143
129
Tsegaye Gebremehin
Kale Heywet Church
January 5th 8th 12th 2008
Addis Ababa and Oromia Region, Ethiopia
Interviewee’s Role and Organization
• Business Manager for water and sanitation program (includes POU, hygiene
training unit, and drilling operations) at Kale Heywet Church
Implementation Background
• Pilot biosand project of 700 filters in 1999; scale up with 8000 filters 3 years
later
• Tsegaye describes demand for BSF as from small groups of people from
within the community who are in contact with users of BSF and other HWTS
technologies
• Funded by Samaritan’s Purse and using the rectangular concrete BSF
• Employs a large field staff, with community education team,
construction/installation team, and monitoring team with field office/factory
in rural areas of implementation and sufficient vehicles for staff to quickly
travel to site locations
Training
• Using Life-Water 5-step PHAST program with community health workers
• KHC BSF program spends ~$30-35 on construction, and 100$ total for a
given filter with KHC health package
• KHC emphasizes joint hygiene training and proper use (health package) over
the commercial benefit; claims the operation could not be self sufficient
without Samaritan’s Purse
• When the program found that they were losing sand through cleaning
techniques which removed sand, they gathered the users for a health and
maintenance meeting for group re-training to the wet harrowing cleaning
method. They first taught the 2nd knuckle finger stir, then taught flat palm
technique with later users, (see BSF Effective Use Maintenance section).
• Tsegaye finds that education package is more important than making a
distinction between non-profit and for-profit BSF ventures
Effective and Sustained Use
• No expiration period is proposed; Tsegaye views the BSF as a typical wat/san
infrastructure project with a normal 10-20 year lifespan
• James Webster of Cranfield University of Silsoe, UK (thesis advisor to Paul
Earwaker, who wrote a masters thesis analyzing the KHC BSF
implementation in 2005) found that 88% of pilot study BSFs were still
operating 2 years after implementation, as written in his proposal for further
scale up with KHC
130
•
•
•
•
•
KHC monitoring staff questions users at household about their use habits
(“which method is best?”) and asks for demonstrations of usage and cleaning
Samaritan’s Purse has worked with KHC to develop treatment goals of 95%
reduction and 10 E.coli per 100ml in treated water
KHC sees sustained individual use because of ownership incurred through
POU product (people gave labor and committed to making a latrine)
Tsegaye claims that program is effective because good will of the church is
perceived and trusted by the communities and individuals involved
Empowering the community by including it in the manufacturing and post
sales education processes helps to boost adoption and sustained use
Monitoring and Evaluation Activities
• Monitoring team conducts 1 month (see Earwaker Appendix) and 3 month
visits to the households, and sometimes 6 month visits.
• The staff comprises of one technician and one local community representative
per 100 community members
• Community members are encouraged to pay the technicians 2-3 birr (0.220.33USD)
• KHC keeps in touch with users, and is alerted to the few problems that occur
• 3 Water tests in last 3 yearsÆfound 90% reductions in E.coli and total
coliforms
• Stays in touch/keeps presence in community and claims to hear through the
grapevine if systems are not functioning appropriately (although we witnessed
some that were not functioning well unannounced), and technicians then make
household visits as needed
Field Visit Notes
• Two rounds of field visits involving 8 BSF users in total, conducted on
January 5 and 12, with Monitoring Observations and Water Quality Results
written up in Appendix C: Household Monitoring Reports
Materials Collected
• WHO presented evaluation
• Household handout from pilot intervention (see Appendix F: BSF)
Dejachew, G. (2002) “Evaluation of Household BioSand Filters in Ethiopia.” WEDC,
Loughborough, UK.
Maertans, M., Buller, A. (2005) “Kale Heywet Ethiopia Household Water and Sanitation
Project Evaluation.” Samaritan’s Purse International Relief, Calgary, Canada.
Samaritan’s Purse, (2001) “BioSand Household Water Filter.” 4rth Ed., Calgary, Canada.
131
Henock Gezahegn <[email protected]>
Population Services International (PSI)
January 9th 2008
Addis Ababa, Ethiopia
Interviewee’s Role and Organization
• Program Manager and Business Strategist for PSI Ethiopia
Implementation Background
• PSI is the sole distributor for PUR in Ethiopia (branded Wuha-telel, literally
translated “water-clarifier”)
• PSI’s Wahu-agar is a dilute bleach solution that is produced in a factory in
Addis Ababa and provides 1.97ppm free available chlorine (FAC) to twenty
liters of water with a single dose (one capful). Translated in this document as
Watergaurd, the literal translation of Wahu-agar is “water-partner”, implying
PSI and their product are the friend of the user, changing the paradigm of the
government as the sole provider of water.
• Waterguard is sold with full product cost recovery on a commercial model
through a large and defined distribution network, through their own traveling
salesmen and to international aid organizations for free distribution in
emergencies, with the PSI ETH office setting the price at every step along the
way. PSI’s costs aside from the physical Waterguard product are funded
through USAID and other partnering agencies.
• PSI increases efficiency of distribution by not overstocking retailers, using a
rule of stocking to a maximum of 1.5 times the rotation volume per month
(see G-Lab Report Appendix J).
Training
• Does community trainings with materials included in Appendix G: PSI
Participatory Hygiene and Sanitation Training Materials, which are
translated into Somali (not shown) for use in Muslim areas of the Eastern part
of Ethiopia
• Trainings are conducted by salespeople
• Includes written and pictorial usage materials on all products, as shown in
Appendix F: PUR and Sodium Hypochlorite Solution Usage Materials
Effective and Sustained Use
• PSI is marketing to change behavior (through increasing awareness and
availability), so effective and sustained use metrics are important to them
• TRaC survey is conducted annually to look at behavioral determinants of
peoples decisions on whether or not to use the range of PSI’s products in order
to stem diarrheal disease incidence and the prevalence of worms. The TRaC
survey directly measures sustained and “frequent” use of PSI’s products,
including Waterguard and PUR.
132
Monitoring and Evaluation Activities
• PSI does not do household monitoring. This seems to be the norm among
promoters of consumable HWTS products.
• PSI monitors through distribution channels in order to gauge complaints and
problems with the product
• PSI monitors batch # and expiration date by tracking their stock in order to
know their distribution/find the efficient outlets as well as to allow for product
recall.
• PSI is currently using 1 year expiration in order to achieve critical rotation in
the early stages of promotion, but PSI has convinced the CDC to extend
Waterguard’s expiration to 18 months, and is attempting to extend to 28
months as only a 5% decrease in the 1.5% hypochlorite solution is noticed
within the 28 month timeframe because of addition of 0.1% NaOH (pH 11.9
stabilizes HOCl)
• PSI undertook a rapid assessment in the form of TRaC survey in 2006. From
this survey, they can judge which behavioral constructs are determinant to the
use of the product using segmentation tables and user/non-user ratios so that
they can prioritize their marketing activities. One main finding of the TRaC
survey is that self-efficacy and social norms are the largest determinant of
purchasing and using PSI’s various health products.
• PSI uses TRaC to judge social capitol, which is the capacity to change
behavior: how often do users talk about the product? ÆFound to be more
often in urban. Do users feel that they know how to use the product? Do they
believe the product is good? Do they recommend it to their friends?
• PSI does not conduct health impact studies because effects of SWS, PUR,
ORS and bed nets are already proven and documented
Materials Collected
• PHAST-style Watergaurd Training Materials Æ See Appendix G
• Gov. of ETH 2006 Rapid Assessment of Drinking Water Quality (RADWQ)
(hard copy and electronic)
• 2005 MCH PSI demographics questionnaire
• PUR Packet branded and printed in Amharic by PSI
Population Services International Research Division (2004) “PSI Behavior Change
Framework “Bubbles”: Proposed Revision.” Washington, DC.
PSI Research & Metrics. (2007). “Ethiopia (2006): Maternal and Child Health TRaC
Study among Caregivers of Children Fourteen Years and Younger in Addis Ababa and
SNNPR.” First Round. Addis Ababa.
PSI (2007) “Sales of Waterguard 2006 & 2007.” Powerpoint Presentation.
PSI (2006) “Diarrhea DALYs Ethiopia 2006.” Unpublished report.
133
PSI (2007) “DALYs prevented using SWS 2007.” Unpublished report.
Central Statistical Agency [Ethiopia] and ORC Macro. (2006) “Ethiopia Demographic
and Health Survey 2005.” Addis Ababa, Ethiopia and Calverton, Maryland, USA:
Central Statistical Agency and ORC Macro.
Crapper, D. (2007). “Q3 presentation.” Population Services International Ethiopia,
Unpublished manuscript.
Tadesse, D. et al. (2006) “Rapid Assessment of Drinking-Water Quality in the Federal
Republic of Ethiopia: Country Report.” Addis Ababa.
PSI (2007) “D3: Doing Development Differently: Annual Report 2006-2007.” Addis
Ababa.
134
Menassie Kifle – EtMedix, Program Manager
Kassahun Birru- Medentech, Africa Representative located in Addis
January 10th 2008
Addis Ababa, Ethiopia
Implementation Background
• Starting in 2008, EtMedix, an established pharmaceutical distributor in Addis
Ababa will repackage Aquatabs into boxes with the name EtMedix and sell to
pharmacies, retailers, kiosks as a wholesaler as well as to NGOs for
emergency use on a commercial basis, insuring full cost recovery for
marketing, management and product
• Both businessmen interviewed had a limited understanding of how the product
worked, to the point that they were conjecturing its advantages over other
chlorine products and could not recommend appropriate usage procedures
outside that on the existing packaging
Training
• None other than what is written on Aquatabs packaging (explained in
Aquatabs Effective Use write-up)
Effective and Sustained Use
• EtMedix intends to market Aquatabs as a lifestyle product to promote disease
prevention as well as to get government approval and co-promotion before
breaking into the emergency market, in order to help with sustained use
Monitoring and Evaluation Activities
• None planned at outset of product launch
• No formal process to capture customer information
• Reliance on supply chain stakeholders for post sales services including
customer complaints and customer service etc.
• Intended indicator: “Average number of customer complaints per customer.”
Although without the stated intention of monitoring customers, this metric
might actually be the number of complaints per sleeve of Aquatabs sold, or
some variation therein.
• Users are not going to be monitored
Materials Collected
• Preliminary market survey results
135
Gladys Inzofu <[email protected]>
Oxfam
January 10th 2008
Addis Ababa, Ethiopia
Interviewee’s Role and Organization
• Consultant brought in to evaluate Oxfam’s response to acute watery diarrhea
(AWD, aka cholera) emergency outbreak in the Southern Nations,
Nationalities, and People’s Region (SNNPR) of Ethiopia, December, 2007.
Implementation Background
• Oxfam distributed Waterguard and PUR to communities hit by cholera in late
2007 for three months.
• Similar to UNICEF’s responses, Inzofu found that the community’s
dependence and trust in the government for provision of water and
government of Ethiopia not promoting POU as a permanent solution to celan
water greatly hampered sustained use of the product following emergency
distribution
• She claims that emergency implementation hampers the private sector
implementation by changing user perception to that of only being needed in
emergency. Similarly, free distribution makes users reluctant to pay for the
products following the emergency.
Effective and Sustained Use
• By the end of 3 months of handing out PUR/Waterguard and the winding
down of the emergency phase, Oxfam switched to source treatment because
they found hoarding/disuse of the products in the home
• Unable to monitor household use effectively, so switched policy to that which
they could monitor more easily (microbial water quality testing), yet did not
create a safe solution either way in the home due mainly to recontamination.
• Although emergency was never declared by government, initial acceptance of
PUR and Watergaurd was high due to visible health improvement and
trainings. This was reversed once the tangible health impacts (death due to
AWD) subsided and community was no longer in “emergency” situation, with
Oxfam moving to development phase. People rejected that they were unsafe
once the cholera died down and no effective exit strategy for the products was
present. Thus, use was discontinued despite Oxfam still handing them out.
• AWD = problem, POU = solution; daily life ≠ problem, POU ≠ solution; It
may take a generation of education and awareness to change belief and
behavior systems.
Monitoring and Evaluation Activities
• Claims that evaluating usage post emergency is not feasible due to budgeting
and timeline constraints
136
Jesse Jones Agbanyo – Product / General Manager
Ebenezer Aidoo – Sales and Marketing Executive
Precision dx (Sole distributor for Medentech in Ghana)
January 15th 2008
Accra, Ghana
Interviewee’s Role and Organization
• Both Jesse and Ebenezer joined Precision in 2007 to work on the launch of
Aquatabs, when sole licensing of Aquatabs was granted by Medentech to
Precision in Ghana
• Precision is now Medentech’s partner in Ghana and will take over all
importation and distribution of Aquatabs, as well as handling secondary
distributors such as New Energy in Tamale. Their official launch is scheduled
for February, 2008.
Implementation Background
• Precision’s first business was the distribution of mosquito nets, which started
in 2006
• Medentech’s partnership with the Ghanaian government and AED has helped
significantly to build awareness
• This partnership has supplemented Precision’s marketing and training costs
• Government endorsements help significantly
• Precision also partners with Guiness Ghana for emergency relief
• Precision has special prices for NGOs
• Similar to at EtMedix in Addis, the businessmen at Precision had little
technical knowledge of the Aquatab product(e.g., all parties had a limited
understanding of free available chlorine and the differences between NaDDP
HOCl), but had investigated the success of Aquatabs in other African markets
(e.g., Kenya) and were astute businessmen.
Training
• Aquatabs/Medentech has partnered with the USAID Academy for Educational
Development (AED) to help develop educational materials
o This partnership was initiated by Kevin O’Callaghan because it was
essential for Aquatabs success
• Launching in Cape Coast first in early 2008, piggybacking on the training of
600 community health volunteers by the government and AED. Precision
plans to specifically train these volunteers on use of the Aquatabs at the same
time
Effective and Sustained Use
• Plans to work with AED to monitor usage, yet has no specifics at this time.
137
Monitoring and Evaluation Activities
• Similar to other consumable social/commercial marketing programs, Ebenezer
did a willingness to pay (WTP) study at outset of program, and concluded that
5 pesuis/67mg Aquatab was reasonable. Thus, he set the price per tab at 4
pesuis (~ $0.04). It was unclear how they intend to monitor the WTP once the
product hits the streets.
• Household monitoring will not likely be part of their program. The author did
not hear them say that monitoring is part of the budget
Materials Collected
• N-193 Hardcopy of Initial Market Survey in June 2007
• Sleeve of Aquatabs, labeled with instructions and Precision dx’s name
(directions written out in Usage section of Aquatabs Effective Use)
138
Abaazan Peter Adagwine, Shak Ibrahim, Peter Alhassan
Pure Home Water
January 2008
Tamale and Upper East Region, Ghana
Interviewee’s Role and Organization
• Sales representatives and multi-faceted employees of Pure Home Water
(PHW), a social enterprise and legally registered non-profit organization
based in Tamale, Ghana founded in 2005 by Susan Murcott, with local
partners
Implementation Background
Three implementation models:
• Salespeople go directly into targeted communities and provide a
demonstration and training of the ceramic water purifier locally branded as the
Kosim filter. After a community liaison collects money from community
members, PHW delivers filters to households with appropriate training in
house, and make a $1 US on each filter sold
• Emergency distribution of filters in Upper East (UE) Region to flood-affected
victims (FAVs) under UNICEF funding.
• Retail sales through shops in district capitols of Tamale (Northern Region)
and Bolgatanga (Upper East), with new retail operations intended also in Wa
(Upper West).
Training
• Distributes a poster relating health to clean water to proper and consistent use
of the Kosim to most users
• Group training to FAVs in UE by salesman, using posters and engaging group
participation
• Community demonstrations and training as part of first visit to new
communities
Effective and Sustained Use
• Kate Clopeck’s survey of PHW users for a total of 221 surveys in 28 villages
in January 2008 specifically targeted “sustained use”
• See more specifics in Appendix C: Household Monitoring Reports
Monitoring and Evaluation Activities
• Many of Murcott’s student projects through MIT have helped to boost
monitoring and reporting capabilities of PHW staff, whose responsibilities
focus on sales and administration
• Salespeople do a good job keeping in touch with users in various communities
(from what I saw), but demand for new Kosim filters cannot be met at key
times (i.e., following the arvest, due to manufacturing constraints and a large
priority order from UNICEF in January 2008), limiting the ability of PHW to
139
•
reach communities in Northern Region in the first half of 2008 and making
salespeople not meet their projected deliveries to communities
community volunteers work with PHW salesmen to gather money, hold
community meetings, install and keep in touch with users of the filter in their
community after installation
Field Visit Notes
• Author accompanied Kate Clopeck in one day of her surveying, with
“Monitoring Observations and Water Quality Results” written up in Appendix
C: Household Monitoring Reports
• Author went with Shak and Abaazzan to the Upper East Region for three days
to distribute filters for the UNICEF flood relief contract; did some monitoring
while there, and reported on this under “Monitoring Observations” in
Appendix C
Materials Collected
Murcott, S. (2005). “Behavioral and financial targets behavioral and financial targets in
implementing, scaling up, in implementing, scaling up, monitoring and evaluating
monitoring and evaluating household water treatment household water treatment and safe
storage technologies and safe storage technologies”. Annual Meeting of International
Network to Promote HWTS, Quito, Ecuador.
Murcott, S. (2006). "Implementation, Critical Factors and Challenges to Scale-Up of
Household Drinking Water Treatment and Safe Storage Systems." Background Paper on
HWTS for the Electronic Conference May 12-22, 2006, Hosted by USAID / Hygiene
Improvement Project (HIP).
Murcott, S. (2007) “Guinea Worm Cloth Filter: Household Water Treatment and Safe
Storage Product and Implementation Fact Sheet.” http://stellar.mit.edu/S/project/hwtsnetwork/materials.html#topic3/
Pure Home Water-Ghana, (2008) "Ceramic Pot ("Kosim") Filter Training Manual."
http://stellar.mit.edu/S/project/hwts-network/materials.html#topic6
140
Mumuni K. Osman
International Aid
January 24th 2008
Tamale, Ghana
Jim Niquette
GWEP, Carter Center
January 25th 2008
Tamale, Ghana
Carl Allen
Peace Corps
January 2008
Tamale, Ghana
Interviewee’s Role and Organization
• Mumuni is the Program Manager (Country Water Initiative) for International
Aid in Ghana as well as the leader of Watersites, his Ghanaian consultancy to
NGOs in the water and sanitation sector
• Includes info from an informal lunch meeting with Jim Niquette, Resident
Technical Advisor of the Guinea Worm Eradication Program (GWEP) based
in Tamale, on behalf of the Carter Center, January 25th 2008
• Includes info from conversations with Carl Allen, Peace Corps Volunteer in
Tamale who helped with installing the filters and liaison to the affected
communities
Implementation Background
• International Aid partnered/hired Mumuni to install a large number (2250
were delivered to Accra in a container in 2007) of the HydrAid cylindrical
plastic biosand filters (BSFs) that are manufactured in the US; in 2006,
Mumuni went to Aquinas College in Michigan for a week long training held
by Dr. David Manz, the designer of the original BSF
• Mumuni started implementation in Kpanvo, near Tamale, Ghana with 100
BSFs in partnership with the Carter Center and the voluntary assistance of
Carl Allen
• International Aid’s intent was to give these original 2000+ BSFs away for free
to partner organizations while the partner agency was left to handle the
implementation in their active communities
• Niquette says that this was not understood by the Carter Center when joining
on, and the policies of the Carter Center do not allow them to
collaborate/partner on this basis, so he can not get funding for implementation
or monitoring and Carter Center’s involvement is finished
Training
• Adventist Development and Relief Association (ADRA), a faith-based
organization operating in Northern Region Ghana, was to do a training of
biosand construction in February, 2008 prior to implementing 500 filters after
collecting baseline survey in 5 communities in which they currently operate.
ADRA will bear implementation costs minus hardware, but will eventually be
on their own for monitoring. Mumuni hopes to use the 500 filter users for an
International-Aid funded health impact study under the direction of Dr. Mark
Sobsey of the University of North Carolina.
• Hands out CAWST BSF usage poster, included in Appendix F: BSF Usage
Instructions
141
Effective and Sustained Use
• Post implementation in Kpanvo, Mumuni is currently collecting money from
BSF users to show investment in BSF as well as to provide users with jerry
cans in order to encourage safe storage. These actions, although necessary to
effective use of the BSF and Safe Storage, were not put into the action plan or
funding upfront and seem like an afterthought.
Monitoring and Evaluation Activities
• Mumuni claims to have “given” the Carter Center 100 filters, and helped
install them and test them with the understanding that they would do
monitoring of the systems once installed
• Carter Center’s GWEP community volunteer for Kpanvo was enlisted to help
monitor the BSF use while conducting household visits in the community,
although Mumuni claimed that his assistance was not necessary and that
Mumuni himself would be able to establish community-based assistance for
his program on his own, if needed.
• Carl Allen and Jim Niquette provided physical assistance for training and
installing the filters and have not been able to do much follow up. KCarl
completed his Peace Corps assignment and left Ghana as of April 2008.
• Mumuni did microbial testing the day after installation, and got inconsistent
results from the filters (this information was not provided to the author,
although the testing was premature as the schmutzdecke would not have
developed). He has no more money to spend on water quality testing. He
was interested to see the microbial results of Izumi Kikkawa and Sophie
Walewijk.
• Mumuni self-monitored the program following implementation and witnessed
“appreciation” and neighbors using the filters too.
Field Visit Notes
• The MIT-Pure Home Water-Peace Corps team, including Susan Murcott,
Peter Alhassan, Sophie Walewijk, Izumi Kikkawa, Mike Dreyfuss, the GWEP
volunteer and the author conducted surveys in 7 Kpanvo households and took
water quality samples in 30 households (see Kikkawa, p.98-101). The surveys
were especially useful in comparing user-acceptability of the BSF with the
Pure Home Water Kosim ceramic pot filter that was already in use in many of
the households, apparently unbeknown to the BSF implementers at the time of
installation.
• Monitoring observations, water quality data w, and pictures for 7 of the
households using both Kosim and BSF has been written up in Appendix C:
Household Monitoring Reporting
Materials Collected
• Flowrate analysis of “Sibi” BSFs in August and September 2007
• Household questionnaire on BSF usage with answers from visits (Internal
report)
142
Atsu Titiati
Enterprise Works
January 29th 2008
Cantonments, Accra, Ghana
Interviewee’s Role and Organization
• Titiati is the general manager of Enterprise Works, a Washington D.C.-based
non-profit supporting the enterprise/market approach to development with
local offices in many poor and middle-income countries.
Implementation Background
• In 2006, Enterprise Works received funding for an initial 5000 ceramic pot
filters from Diageo Foundation PLC UK; Guinness Ghana followed up with
4000 more filters for flood affected victims; 1000 additional filters were
donated from an anonymous donor at Guinness UK (parent company to
Diageo)
• CWP is branded the Adokuro filter, a Twi word meaning the clean, naturally
filtered water that comes from under trees in the forest
• Enterprise Works is a customer of Ceramica Tamakloe, buying and
distributing exactly the same product (minus the taps) as Pure Home Water in
Tamale
• At first Enterprise Works was selling with full cost recovery to the funder, but
when sales were too slow, Diageo asked Titiati to sell at 50% subsidy of the
original selling price of $5 US.
Choosing the communities
• Titiati does not target communities with very turbid water sources for the use
of the ceramic filter, as he does not want the customers to be dissatisfied with
insufficient flow rates
• Once a community has been identified (generally a peri-urban area on the
outskirts of Accra), a community meeting is organized with the help of the
assembly man or chief
• At the meeting, Enterprise Works introduces the filter and tells of its
importance to health. They then appoint a retailer within the community
(usually someone with a shop, often a trusted community figure chosen by
chief or the chief himself, but someone who agrees to do household trainings
to end users and health promotion (as trained by Enterprise Works), and
provided with training material to hand out to users (same as Potters for Peace
Materials in Appendix F: Ceramic Pot Filter Usage Instructions)
• Sammy, Enterprise Works’ field liaison for the community visited by the
author chose to distribute filters in the community because it was close to the
road and to Accra. He originally arranged meetings through the chief to
streamline things
143
Training
• Titiati claims that the CWP requires a lot of health education, through both the
retailer and promoter; need to teach health impact for people to want to afford
15 Cedis (~$15 US) for the filter (a profitable cost, given high cost from
Tamakloe manufacturer)
• (PFP) The Potters for Peace training materials distributed by Enterprise Works
claim 3 a year life span for the filter because colloidal silver wears out (see
Appendix F: Ceramic Pot Filter Usage Instructions)
• Supplied a brush for cleaning in beginning of program, but could only
recommend correct brush to be used when funds fell short later in program
Effective and Sustained Use
• Enterprise Works undertook testing the flowrate of each batch of filters, as
Tamakloe was not doing this (although required to by production protocol).
Titiati was displeased with the slow production and inconsistent quality.
Monitoring and Evaluation Activities
• Monitoring is a deliverable for Diageo (funder) from Enterprise Works
• Retailers take 3$ of the 5$ selling cost, and are now buying straight from CT
(some confusion noticed in the field about who was supposed to place order,
Enterprise Works or retailer)
• Retailer keeps a record of users for monitoring purposes
• Titiati said that “the generation for the pot filter is passing now,” and,
unhappy with the performance of the filter (both treatment and sales-based),
he plans not to sell any more filters, and is looking for new technologies to
promote, such as the Tulip ceramic vacuum filter(?); Diageo contract finished
in November, 2007, and there is no more money in the budget for M&E, but
retailers are expected to keep in touch with consumers in their communities,
and to report broken pots or need for new pots such that Enterprise Works can
order more filter from Tamakloe
Field Visit Notes
• Field visits to two villages in a peri-urban area outside of Accra, monitoring 6
households in total, were conducted on January 29, with Monitoring
Observations written up in Appendix F: Household Monitoring Reports
Materials Collected
• Hardcopy of CWP promotional fliers (Rivera format) and training materials
144
Appendix C: Household Monitoring Reports
Organization
Monitoring Notes
Page
Ethiopia:
Kale Heywet Church
8 biosand filter users
145
Ghana:
Pure Home Water
Pure Home Water/Unicef
International Aid/Carter/PHW
Enterprise Works
Kate Clopeck Survey of Kosim CWP users
Distribution of CWP to flood affected areas
3 joint users of biosand and ceramic filters
6 ceramic filter users
157
165
167
181
These reports give an overview of the household monitoring visits conducted in January,
2008. Refer to Appendix B: Field Interviews for background information on the
implementing agencies and their specific projects.
145
Kale Heywet Church biosand filter users
Filtino and Gondogorba communities, near Debre Zeyt town, Oromia region, Ethiopia
January 5th and 12th 2008
Field Site Overview
• The community of Filtino received many of their biosand filters from Kale
Heywet Church (KHC) and Samaritan’s Purse’s original pilot scale
implementation in 1999, with filters working well since then and consistent
community involvement of the technicians at the nearby factory/field office for
the program if any problems are reported, as has been rare over the past 9 years of
use; Many of the users had blue plastic “safe storage” containers with lids that
had holes in them in order to catch the treated water, as handed out by KHC with
their scaled up implementation since 2003
• The sight is located a few hours east of Addis Ababa, in the Ethiopian highlands,
within the Oromia region of Ethiopia that surrounds Addis Ababa. A largely
denuded countryside, the rivers that serve as water sources for residents of
Gondogorba, the first community visited, neighboring Filtino, flow with very
turbid water (up to 1000 TU measured in-house). This area is home to smallscale agriculturalists who generally do not own their own land, as can be
witnessed from the large flower farm adjacent to the community, from whose
irrigation ditch the women of Filtino fetch their water.
• The author went to sight with different KHC employed technicians on each of the
two Saturdays, reaching the villages with a ride in the KHC field vehicle.
Because Saturday is market day, many houses with BSFs were empty, and often
children answered the questions in lieu of their mothers, the main caretakers for
the BSF who were away at the market.
• All waters were of pH 8.5-9 (basic volcanic soils) and high in turbidity
Note the holes dug in the riverbank for pre-filtering the turbid water. Use of such prefiltering by HH1 resulted in collected water of significantly lower turbidity and pathogen
load as well as correspondingly long filter run times in between cleaning.
146
This irrigation ditch serves as the sole source for Filtino Village=250NTU, 20°C, flowing
Results
Observational Monitoring Results for Kale Heywet Church Biosand Users
HH
Pretreat
HH1
Riverbed
filtation
HH2
None
HH3
HH4
HH5
HH6
HH7
None
None
None
None
Days
since
Cleaning
120, not
needed!
60, unable
w/child
Cleaning
Method
Storage
clean?
Storage
cover?
Handling
Stir w/
finger**
?
Clean
Yes
Clean cup
No
No
7
Stir w/
finger**
Remove
sand
Stir w/
finger**
Remove
sand
Remove
sand
Clean
Yes
Dirty
hands in
storage
Clean cup
Clean
No
Clean
7
7
14
7
Dist to
Source
(m)
>100
Proper
Use? *
>100
Use
(Liter
/day)
60120
?
100
100
Yes
Good
<100
50
No
No
Good
<50
>50
Yes
Clean
No
w/hand
100?
100
No
Very
bad
No
Yes
No
Wash w/
>100
50?
No
Dirty
None
water, use
hand
*Proper Use is defined by passing the main components use as labeled in the Observational Monitoring
section of the Biosand Filter Effective Use write-up.
**
Stir w/ finger refers to a wet harrowing cleaning method involving stirring the top few centimeters of sand
with the finger down to the second knuckle and then scooping out dirty water, as promoted by KHC and
Samaritan’s Purse.
147
Water Quality Methods: In order to look at treatment efficiency and likelihood of
diarrheal disease to the user, water was tested for bacterial quality. The 3M Petrifilm
allows for counts of greater than 1 colony forming unit (CFU) per ml (which is
equivalent to ≥100 CFU per 100 ml of both E.coli and total heterotrophic coliforms) by
the simple addition of 1 ml of sample to the disposable film plate under sterile conditions
followed by incubation. All samples were taken in Whirlpak bags with sodium
thiosulphate, kept on ice for <6 hours before addition to the Petrifilms and incubated at
36±2 degrees Celsius for 24 hours, as called for in the Petrifilm protocol. The author
carried a portable laboratory setup including a phase-change incubator designed by Amy
Smith of MIT. This incubator required reheating after 12 hours of night time ambient
temperatures in order to achieve the stated sustained temperature range.
Water Quality Results
Unfiltered
Treated
Storage
E.coli/
T.coli/
E.coli/
T.coli/
E.coli/
T.coli/
Turb
100ml
100ml
100ml
100ml
100ml
100ml
TU
HH1 1/5/08
500
2000
<100
100
<5*
HH2 1/5/08
2000
14000
<100
25000
<5*
HH3 1/5/08
500
14000
<100
<100
<100
1400
<5*
HH4 1/12/08
12
<100
20000
<100
<100
700
3600
HH5 1/12/08
12
5000
18000
100
6200
600
10400
HH3 1/12/08
30
<100
18000
<100
<100
<100
2400
HH6 1/12/08
6
200
5000
14000
<100
700
1000
1900
-HH7 1/12/08
7
1000
29000
<100
1900
100
2600
*visually clear; assumed turbidity of <5NTU was not measured due to minimal amount of sample available
** Source for HH3-HH8 was an open, flowing irrigation ditch of Turbidity >500TU possessing 4000
E.coli/100ml and 22000 T.coli/100ml.
HH
Date
Flow Rate
L/hr
Turb
TU
<5*
1000
500
Treatment and Effective Use through Water Quality Monitoring for Kale Heywet Church
Biosand Filter Users
HH
Removal via Treatment
T.coli
T.coli
Absolute
%
Log
Risk*
Actual Removal via Storage
E.coli
T.coli
T.coli Absolute
%
%
log
Risk*
Contam
via
Storage
HH1
Low/int
>80
95
1.3
Low/int
No
HH2
Low/int
>95
-80
-0.3
Low/int
?
HH3
99.3
2.2
Low/int
>80
90
1.0
Low/int
No
>80
HH4
99.5
2.3
Low/int
recontam
80
0.7
High
Yes
98
HH5
65
0.5
High
88
40
0.2
High
Yes
HH3
99.5
2.3
Low/int
85
0.9
Low/int
No
98
HH6
95
1.3
Low/int
80
85
0.9
High
Yes
90
HH7
93
1.2
Low/int
90
90
1.0
High
Yes
* Risk levels based on WHO E.coli risk categories (WHO, 1997). Presence of E.coli on the 3M Petrifilm
indicates high risk water with >100 E.coli/100ml.
** ? Question marks indicate that the level of detection for E.coli of the Petrifilm method is above that
needed to discern low risk from microbial contamination and thus Effective Use was not judged for these
results.
E.coli
%
Effe
ctive
Use?
?**
No
?**
No
No
?**
No
No
148
• Despite claims by technicians and the program manager Tsegaye of retraining all
users to maintain the filter by stirring the top sand layer down to the second knuckle,
some users were still removing sand to clean the filter. Retraining was needed due to
losses of sand incurred by cleaning method which involves the removal of sand during
the pilot implementation of 1999.
Household # 1
1/5/08
Gondogorba community, Oromia district, Ethiopia
Note the elevated and dedicated safe storage
unit, separate small-mouthed jerrycan for
fetching water, tile floor, and visible
presentation of KHC’s maintenance poster
although they are missing a suitable top for the
storage (top was in place when I walked in, but
has a hole for the dripping of the filtered water
and was not very clean). This first household
was a good example of positive Effective Use in
terms of Observational Monitoring.
Name and status of person interviewed
Mother of a family of four who is the primary
water fetcher and caretaker of BSF, which she
has had for over 5 years
Household visit notes
• 30 minutes carrying time from river source
• Does she like it? “It takes whatever dirt we bring” and is adamant about lack
of diarrhea in her family of four
• Corrugated zinc plated iron roofing (CGI) and dirt/tile floor
• Her pit latrine had “no flies” and was thus clean, as part of KHCs intervention
was to provide safe concrete bases to improve the pit latrines
• Mother was uncomfortable, especially at the beginning
Monitoring Observation
• See notes on picture above
• Storage clean, but not properly covered
• Claims that she uses BSF treated water for all uses (except washing, which
can be done with riverbed-filtered water) during the wet season, when the
turbidity is higher; preferable for cooking because njera (fermented
unleavened bread) gets better holes in it with BSF water
149
•
•
•
•
•
•
Mother claims consistent use: will not drink non-filtered water; family will
either bring their own in water bottles, or drink before and after going to
market/work; she worries about sickness with non-filtered water
Filters 2-4 X 30L jerrycans per day; neighbors use the filtered water too
Claims that is takes 30 minutes for 20L to filter; she likes the slow flow rate
(makes the water clear and coldÆadvantage of concrete biosand)
Filter has not been cleaned in last 4-5 months; filter tends to block up during
heavy rains (more turbid water, using the filtered water for all uses)
She pre-filters at the source by digging into the riverbed and pulling water
from there, bringing the water down from 1000NTU to nearly clear!
Brought clean, dry cups when the technician was prompted to ask for a glass
of water
Water Quality Monitoring
• Clearish influent from effective riverbed filtration at source; <5NTU effluent
• Good flowrate (not measured)
• Effectively removed E.coli, within the limits of detection of 3M Petrifilm
(<100 E.coli per 100ml); best stored-water quality surveyed, as effective use
protocol were carefully followed by this user
Effective Use Assessment
• Most effective use witnessed of all the KHC filters, especially because of
prefiltering through the riverbank
150
Household # 2
1/5/08
Gondogorba community, Oromia district, Ethiopia
Name and status of person interviewed
• Mother, sole caretaker for the BSF
Household visit notes
• She admits that she has not been able to upkeep her BSF or clean her storage
unit as she is a young mother with two small children, one newly born. She
thus relies on her nieces/neighbor’s children to get water, and it is too much to
ask of them to filter it through the riverbank, as she has been trained. Æ BSF
presented too much maintenance for a mother with a newborn and a turbid
water source to keep up with; she knows it
• markedly poorer living conditions than HH1, her neighbor
• Chicken was inside, with access to the open storage unit
• Corrugated iron (CGI) roof, tile floor
• I did not run the interview very long as she was recently pregnant and I felt
that I was invading her privacy, with her husband away.
Monitoring Observation
• Storage unit very dirty with the top off and a dirty cup fallen inside
• woman put her hands (dirty from handling the baby) into the storage to pull
out the cup, and fetched water that way
• I did not witness flowrate, as there was not enough water around to filter
• Raw water in storage: 1000+TU
• Not cleaned in a long while
Water Quality Monitoring
• Despite all signs of neglect and improper water handling, the water in storage
was of low/moderate risk (<100 E.coli per 100ml), with effective treatment of
turbidity (<5NTU)
Effective Use Assessment
• Failed the monitoring due to a lack of ability to properly maintain the system,
yet still managed to create a moderate risk water with low turbidity
151
Household # 3
1/5/08; revisited and retested 1/12/08 for confirmation
Filtino community, Oromia district, Ethiopia
Notes from Picture: Shares positive hygiene and
observational monitoring characteristics as HH1, as
well as conditions suitable to minimal contamination.
Note the elevated and dedicated safe storage unit,
separate small mouthed clean jerrycan for fetching
water, tile floor, and visible presentation of KHC’s
maintenance poster and sticker although they are
missing a suitable top for the storage. The user was
rightfully proud of her BSF, as was her neighbor, who
was very helpful on the second interview as Ashetee
was at the market.
Name and status of person interviewed
Ashetee, the mother of the household and main caretaker of the BSF, as picture
above.
Household visit notes
• Has had the filter 8 years, likes it; she and her husband were happy users of
the BSF and happy and proud to share info on it
• Appreciates that it was free, but would pay any amount to buy one, even
1000birr (110$) when prompted
• No animals in house; CGI roof, dirt floor
• 7 people drink from the filter
• They prefer BSF to chlorination (better taste and cool), and does not desire a
borehole because she has the BSF
Monitoring Observation
• Uncovered storage
• No pre-treatment from irrigation-ditch source
• Fetches 4X25 liters every day from source 100m away
• With 500TU influent from the source, she cleans the filter weekly
• On second testing, flowing at 30 L/hour with just an inch of water above the
diffuserÆfast; user does not know when last time it was cleaned
Water Quality Monitoring
• Total coliform (TC) recontamination in storage as compared to directly
treated water, but absolute level of risk low/moderate (<100 E.coli per 100ml)
• Same treatment and storage characteristics during second monitoring round,
but influent was only moderate risk during second week
Effective Use Assessment
152
•
effective on both fronts, although could have used a top on the storage in
order to eliminate the minor recontamination noticed in storage
Household # 4
1/12/08
Filtino community, Oromia district, Ethiopia
Name and status of person interviewed
• Fanu Gareshu, mother of the household
and primary caretaker of the BSF
Household visit notes
• Household never has diarrhea
• CGI roof, tiled floor
• Chickens inside
•
•
•
•
Monitoring Observation
• Very happy with filter, shows her
appreciation to the technician as he is
part of KHC
• Treats 50L/day
• Only drinks from BSF, shows bottle
used for traveling and carrying water
Lid on BSF, and diffuser in place, but diffuser has sedimented sludge on it
Missing lid for storage and dips cup in it to fetch us a glass of water but is
careful not to get her hand wet
Cleans the filter every week by removing the top layer of sand to a bucket,
uses filtered water to stir and rinse the sand, pours off water, replaces sand,
and uses the water right away
Storage unit raised off the ground
Water Quality Monitoring
• With filter reservoir filled, flowrate=12L/hour
• Effective reduction in turbidity (<5NTU) as well as E.coli as far as can be
known (<100/100ml)
• Large recontamination in the storage unitÆneed to ask: how often do you
clean the storage?
Effective Use Assessment
• While observed use characteristics do not set her apart from HH3 in any
significant way, she treats her water well with the BSF but suffers massive
recontamination from unsafe storage practicesÆhow can usage characteristics
be more refined?
153
Household # 5
1/12/08
Filtino community, Oromia district, Ethiopia
Name and status of person interviewed
• Zenagu Gutama, mother of the
household and primary caretaker for
BSF
Household visit notes
• BSF located in dark empty room behind
door, and has been in use for 7 years
• CGI roof with pigeons, tile floor
• Family of 4, no diarrhea this week
Monitoring Observation
• BSF treated water used for drinking,
cooking, and washing of bodies, but not
for washing clothes
• Storage is raised off of the ground but
•
has no cover
Cleans every week, by stirring the sand surface and scooping off the water, as
retrained
Water Quality Monitoring
• 12L/hr, but not sure how full the filter was at time of measurement
• Treated water had at least 100 E.coli/100ml, and stored water had 600
E.coli/100ml= not properly treated or stored
Effective Use Assessment
• Ineffective treatment and storage
• Need to cover those storage units and handle properly
• Do not know why the filter did not workÆshould have asked when was last
time cleaned, but forgot
154
Household # 6
1/12/08
Filtino community, Oromia district, Ethiopia
Name and status of person interviewed
• Pre-adolescent daughter who fetches
water, fills and knows how to operate
filter, although her mother generally
does the maintenance
• She is very candid and frank and proud,
without the pretenses and worries
exhibited by the adult interviewees, and
can speak a bit of English (educated,
unlike her sister)
Household visit notes
• BSF serves 8 people in the household
• CGI roof, dirt floor
• In use for 7 years
Monitoring Observation
• Storage for unfiltered water is out of the sun
• Storage is raised off of the ground but has no cover, and is almost empty
despite actively filtering upon our arrival
• Claims 80 L/day is filtered, yet this would mean constant filtering.
• Not cleaned for at least 2 weeks prior, but cleans every 2-3 weeks or 15 days
• When her mother or father cleanse it, they remove sand for washing, using
treated water, like in HH4
• When asked for a glass of water, she brings a seemingly clean ladle that gets
dirtied by being upside down on the wet lid of the filter; washes a glass with
treated water, and then rubs her hand in it before pouring water
Water Quality Monitoring
• When filled to two inches from the top, flowrate=6 L/hr
• 200TU influent, too clean to measure treated water (probably <10NTU)
• Effective treatment (<100 E.coli per 100ml), yet very bad recontamination
through storage and handling (1000 E.coli per 100ml)
Effective Use Assessment
• Monitoring observations and water quality assessment both show effective
treatment from the BSF, and both show ineffective handling and storage.
However, without asking for a glass of water, I would not have noticed the
bad handling, as there was little water in an otherwise visibly clean storage
unit.
155
Household # 7
1/12/08
Filtino community, Oromia district, Ethiopia
Name and status of person interviewed
• Bekelu Kebeda, the mother of the
household and caretaker of the BSF
Household visit notes
• Markedly poorer and less healthy household
than the neighbors I just had visited
• 9 people in family; kwariorsher belly on 3
year old, showing protein deficiency and/or
worms, but mother claims no diarrhea
Monitoring Observation
• Improper placement of filter (see picture):
located in a goat pen, accessible to animals,
holes in thatched roof allowed direct sunlight of
the filter housing, mud/hay floor with sheep
poop surrounding filter
• Hygiene: mother slapped young son away
from drinking directly from the tap (she knew it
was incorrect, but had not taught the children correctly; conversely, the kid’s best bet was
to get it straight from the filter, as the storage unit was very dirty.
• This filter was 9 years old, from the pilot distribution when the blue plastic
safe storage containers were not in use, and as such, a jerrycan that was
identical in wear and dirtiness to the one used to fetch water (claimed they
were of separate uses, but I was not convinced), but still dirty, without lid, and
open access to animals
• Claim consistent use, that they are “never too far not to drink it”
• Cleaning: removes sand to clean once a week
• Hygiene: when fetching a glass of water, cleaned the cup with unfiltered water
and hand before pouring filtered water into it
Water Quality Monitoring
• 7L/min flowrate (water level unknown)
• Turbidity: 210TUÆ10TU: fails the effectiveness test (<5NTU after treatment)
• Inside water temp = 20°C; cool despite direct sunlight
• Seemingly effective microbial treatment (<100 E.coli per 100ml), and only
100 E.coli/100ml in storage, which fails the test (high risk), but is better than
many of the other filters sampled)
Effective Use Assessment
• Ineffective observed usage results in adequate treatment without significant
recontamination (recontaminates to the level expected by Levy, 2002 in Safe
Storage section.
156
Survey conducted by Kate Clopeck concerning sustained use of the Kosim ceramic pot
filter in the rural villages west of Tamale, Northern Region, Ghana.
Accompanied by author for two days, with PHW salesman Peter Alhassan translating
Pure Home Water
January 17th and 18th 2008
Synopsis:
The largest survey of monitored usage done in the field in January and July, 2008 was
Kate Clopeck’s survey of 221 CWP users associated with Pure Home Water in Northern
Region, Ghana. Much was learned from this survey in terms of appropriate survey
questions, included in the Best Practices for field monitoring of the Results Chapter and
Common threads in household monitoring of the Discussion Chapter. Although not
conclusive for judging Effective Use, Petrifilm analyses of the 56 filters investigated by
Kate Clopeck resulted in only one showing of E.coli, resulting in a high risk level (as
defined by the WHO) for one filter and low/intermediate risk (<100 E.coli per 100 ml)
for the other 55 samples. The average total coliform count of treated water from storage
units was 1000 T.coli / 100ml. The results of this survey will be formally presented by
Clopeck in her thesis and available at http://web.mit.edu/watsan after June 2009.
Field Survey Overview
• Survey instrument developed by Kate Clopeck is included at the end of these
notes
• Water Quality: Of 56 household samples of treated water taken from the taps
of the Kosim ceramic pot filter, storage units, only one had reportable (200
E.coli/100ml) counts of E.coli using the Petrifilm method. Thus moderate or
low risk was associated with >98% of users. An average of 1000 Total
coliforms per 100 ml (range between less than 100 and over 10000 per 100
ml) was found for treated waters. Significant sources of E.coli and total
coliforms were found from stored untreated waters, indicating reductions
through treatment.
• Throughout this survey, prefiltration with the Guinea Worm Eradication
Project (GWEP) cloth filter was investigated and almost 100% effective use
(proper use, storage, condition, and cleaning as well as knowledge of the
Carter Center community volunteer) was witnessed among people also using
the Kosim filter. Although these cloth filters were inspected by the survey
conducters, the question of whether prefiltration occurs on a regular basis was
led on by Peter Alhassan and may not have led to truthful answers. Either the
Carter Center or Pure Home Water community volunteer for the village led us
to the necessary Kosim users for surveying as they knew everyone in the
village. Carter Center volunteers claimed to do weekly monitoring of the
households and inspection of their cloth filters.
157
1/18/2008: (House 901-907) In Kpilo village located next to
Peter Alhassan’s own village, visited on my second day with
Kate and Peter, people were not using the filter but were
rather using the storage unit for piped water, as piped water
happened to be flowing that week.
All of the household samples showed <100CFU/100ml for
both E.coli and TC, including water directly from the pipe
source, water directly stored without filtration, as well as
water stored after filtering. I lament that I did not test these
water sources for residual chlorination, one likely cause of
the high microbial quality witnessed. All of the storage units
were very clean, nonetheless. So, no post contamination
through storage and/or filtration was noted, and effective use
was noted among these households from both monitoring observation and water quality
monitoring based on safe storage and hygiene practices. In the homes 901-907, often the
filters were not located in proper places, as never letting them touch the ground was
explicitly emphasized in PHW trainings (see picture above, left).
In the same community of Kpilo a few houses were found
not to have proper knowledge of how to use the filter,
including one man who had inherited the filter from his
brother and had no knowledge of its intended use as well
as one woman who left the filter on the bottom of the
storage unit which was full of (piped) water (see picture
on left). Although the water was clean and the filter could
have provided some minor cooling from this use, improper
training and knowledge of this user was suspected.
Pictured to the left is an educated man from the
community with the flowing piped water who collaborated
with Peter Alhassan to sell/gain support/distribute filters
within his community. He had excellent use of his filter
(clean cup on top of clean filter, half full of water, stably
situated, and displaying the PHW training poster above the
filter), especially for the fact that despite the filter being
located in the head of household’s room, he encouraged
his children to use it directly (rare among men, the author
observed, who may control of their CWP by locking it
their room and using it only themselves).
Figure 20 GWSC liaison for Kpilo
158
In the house of the community liaison (pictured above) who collected user fees for the
piped water distributed Kpilo by the Ghana Water and Sewerage Company (GWSC), I
saw the bill for the community’s piped water on this man’s chair, and asked to look at it.
The community was charged 83GHC for 126m3 flowing over 20 days. That’s about
1.50USD/m3, or .15¢ US cents per liter. Although the man said he is employed to collect
1GHC from each of 98 users in the community every month, the backlogged bill for the
community was 1,200GHC (~1200USD), at least a ten month backlog without interest.
While people complain of intermittent flow of the prized and trusted pipe-born water in
their community, the GWSC is apparently providing it with a large subsidy and without
expecting return payment from the community.
Use by the man who collects user fees:
• This man used the Kosim unit for storage when the pipe was running and for filtration
when the surface and other unimproved sources had to be used (I did not witness the
dam/stream for this community and cannot comment on when this was used).
• He fills the entire storage unit twice a week
• He dad to fix his tap with a polyethylene bag (pictured on the previous page)
• Likes the look of the filter; sees it as new, clean, expensive, useful
• Does not let his child touch the filter
• Loves the tap!
• He was one of the few people I met who answered that his 3 year old has had diarrhea
during the last week
Water Quality Results
Household
Number
302
303
304
305
306
307
308
309
401
402
403
404
405
407
409
410
411
412
Date Collected
1/10/2008
1/10/2008
1/10/2008
1/10/2008
1/10/2008
1/10/2008
1/10/2008
1/10/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
1/11/2008
Household
samples
E.coli/100ml
<100
<100
<100
<100
<100
<100
200
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
Household
samples
T.coli/100ml
200
200
1400
10000
500
5800
5600
<100
1400
400
10000
100
300
200
100
100
<100
500
Method
Petrifilm
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
159
501
502
504
506
507
509
510
511
512
601
603
605
606
607
608
609
611
612
613
702
706
709
712
801
802
804
805
807
808
809
810
901
902
903
904
905
906
907
Average
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/14/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/15/2008
1/16/2008
1/16/2008
1/16/2008
1/16/2008
1/17/2008
1/17/2008
1/17/2008
1/17/2008
1/17/2008
1/17/2008
1/17/2008
1/17/2008
1/18/2008
1/18/2008
1/18/2008
1/18/2008
1/18/2008
1/18/2008
1/18/2008
(Arithmetic)
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
100
100
<100
<100
200
800
2700
7700
3900
<100
1200
500
<100
<100
<100
<100
100
<100
<100
<100
900
<100
<100
100
<100
100
<100
100
200
<100
<100
<100
<100
<100
<100
<100
<100
<100
1000
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
PF
160
Addendum: Kate Clopeck’s Sustained Use Survey (January, 2008)
Ghana Household Survey: Sustained Use of the Kosim Filter
Hello, my name is Kate Clopeck, and I am student from MIT in the United States.
We are conducting a household survey about the KOSIM filter you purchased from Pure
Home Water. We would like to talk with a woman of the household for about 30
minutes. Participation is voluntary; you may decline to answer any or all of the
questions, and you may end the questionnaire early if you wish. All information will be
kept confidential. Do you understand? Will you be willing to participate?
Yes
No
(If no, thank and close)
Identification code: ___ ___ ___ and GPS Setting _________________________
Date of interview: ___/___/___
Interviewer: ______________________________
Name of person interviewed:____________ __________________
Last Name
First Name(s)
Age and gender of respondent: _____________________________
Household status: _______________________________________
Filter Use
1.
Can you show me the water you use for drinking?
OBSERVE:
a.
b.
c.
d.
e.
How high is the filter from the ground?
Is the ceramic filter installed in the unit?
Do they use water from the bottom of the Kosim unit?
Is the filter covered with a lid?
Is there water in the bottom unit?
2. From where do you collect water?
3. Is the water dirty from that source?
4. How did you first hear about this kind of filter?
161
5. Is your filter working?
6. When did you purchase your filter? (check this with PHW records)
7. Did you receive any training papers when you bought your filter?
a. If yes, can you please show me these materials?
8. From whom did you purchase the filter?
9. Did the sales person come to your house and show you how to use filter?
10. Can you act out for me how you use the filter?
OBSERVE
a.
b.
c.
d.
Clean the filter first?
Filter with cloth filter first?
Use Alum?
Let water settle?
11. How many people use the filter every day?
12. How many adults? How many children?
13. Who collects the water to be filtered?
14. Do you ever drink unfiltered water?
a. If yes, why?
15. Can you show me the water that you use for cooking?
a. Where does this water come from?
b. Do you filter this water?
OBSERVE:
c. Does the water appear turbid?
d. Showed cloth filter? (if applicable)
e. Is the water being stored in a covered container?
16. Can you show me the water that you use for cleaning the dishes?
a. Do you filter this water?
162
OBSERVE:
b. Does the water appear turbid?
c. Showed cloth filter? (if applicable)
d. Is the water being stored in a covered container?
17. Can you show me the water that you use for washing your hands?
a. Do you filter this water
OBSERVE:
b. Does the water appear turbid?
c. Showed cloth filter? (if applicable)
d. Is the water being stored in a covered container?
18. How often do you filter water (days/week)?
19. Is it hard work?
a. If yes, why?
20. Do you ever buy water?(DO NOT ASK IN RURAL)
a. If yes, from whom?
b. Can you show me some of the water you have bought?
Filter Maintenance
21. When was the last time you cleaned the filter and the storage unit?
22. Did the sales person come to your house and show you how to clean the filter?
a. Did this person provide you with a brush to clean the filter?
OBSERVE
a. Saw brush?
23. Can you act out for me how you clean the filter?
OBSERVE:
a. Did they only touch the top lip of the filter?
b. Do they place the filter on a cloth that has been washed in chlorinated or
bleached water?
c. Did the place the filter on the lid of the unit?
d. Did the place the filter in a clean basin?
163
e.
f.
g.
h.
i.
j.
k.
l.
m.
Do they fill the filter halfway with filtered water?
Do they use the provided brush?
Do they only brush the inside of the unit?
Did they clean the storage unit?
Did they use soap and filtered water to clean the storage unit?
Did they use filtered water to clean the storage unit? Cloth filter?
Did they use pipe water to clean the storage unit?
Did they disinfect the storage unit after cleaning?
Did they disinfect the spigot?
Perception
26. Do you like the taste of filtered water?
27. What does it taste like?
28. Is the filter easy to use?
29. What do you like about the look of the filter?
30. Have you had any problems with the filter breaking?
a. If yes, can you show me what the problem is/was?
31. Before you got the filter, did you treat the water at all?
a. If so, how?
b. Can you show me?
c. Did that work?
32. When was the last time someone in your house had diarrhea?
a. how old was this person?
Thank you!
164
Pure Home Water/UNICEF
Distribution of ceramic pot filters to flood affected communities
January 21-23, 2008
Pwalugu, Arigu, and Baluugu communities, Upper East Region, Ghana
Sight Overview
• In their flood relief efforts in the Upper East Region following large scale
flooding during September 2007, UNICEF purchased 5,000 CWPs from Pure
Home Water to distribute to these communities of displaced people who had
recently moved back into their homes and were rebuilding their lives.
• The author went with PHW staff Shak and Peter for three days to the Upper
East to accompany them in their delivery of the filters, distributing them and
conducting trainings to the women of the communities, and then to monitor
users who had received the filters during the previous weeks. We distributed
around 400 filters (as many as could fit in the truck, and then some from
storage in Bolga Tanga) to the women of three communities.
(1) Shak gathering signatures of Kosim recipients; (2) Long lines gathered around the
assembled filters; (3) Women participating in the group training on how to assemble, use,
and maintain the filter.
The procedure of filter distribution was inefficient. The community liaison aided the
PHW staff in gathering the women of the community together, who waited while filter
parts were organized and signatures of recipients were taken. After recipients were thus
identified, a group presentation was made by the PHW staff on how to assemble, operate
and maintain the Kosim. Certain women did all this and did not receive a filter
(UNICEF’s was to give one to each household, but the truck held well less than the
number of women that showed up), resulting in arguing and a bit of confusion over who
would get the last few filters.
165
On the left, Shak with a young woman who had excellent effective use characteristics:
very clean setup, stably situated, tap not leaking, half full, actively filtering, good
maintenance techniques, and a clean cup associated with use of the filter. This same
woman was very helpful in assisting Shak to reeducate and reassemble the filter of her
neighbor (shown demonstrating correct tap installation and storage cleaning on the right).
This neighbor had many user faults including lifting the filter with water in it, placing the
filter on the floor, situating the filter on a non-stable, non-flat base, and generally poor
hygienic habits including washing the system and her drinking cup with unfiltered water,
a common user habit witnessed throughout all of the author’s monitoring visits.
Household visit notes
After two days of distributing filters and conducting trainings, the team conducted
monitoring of 5 households to which filters had been distributed the week before. The
monitoring program was supposed to cover every household a few weeks after
distribution, but it was clear that the PHW staff charged with distribution were not the
best-suited to carry out the follow-up monitoring. A separate monitoring campaign by
independent agents was established in June 2008.
Monitoring Observation
All around good use among the 5 HH’s visited during follow up monitoring by the PHW
staff and the author in January 2008, despite a few leaky taps and unstable bases
(corrected by Shak).
Water Quality Monitoring
Waters were visually clear after treatment, with no microbial water quality measurements
taken. Unknown source.
Effective Use Assessment
Very good from an observational standpoint, despite a few leaky taps and one old
woman’s ignorance of proper use (adequately retrained). It was impressive that people
went from not knowing anything about this system to adopting it very well in the few
hours that Shak had spent distributing these filters the previous week. Group training
was effective.
166
International Aid and Pure Home Water households with both biosand and Kosim
ceramic pot filters; Carter Center and Carl Allen of Peace Corps helped with installation
January 20th 2008
Kpanvo community, Tamale, Ghana
1/20/2008 Comparative Survey among joint Kosim and Biosand Users
Kpavno Community
A team comprised of Peter Alhassan, Matt Stevenson, and Susan Murcott conducted a
written survey among three households of the Kpavno community who had purchased the
Kosim filter midway through 2007 and then received a free Hydraid Biosand filter from
International Aid late December, 2007. Peter Alhassan conducted the survey in the local
language of Dagbani.
Water Quality Monitoring:
Sophie Walewijk of Stanford University conducted membrane filtration microbial tests of
many of the Kpanvo biosand filters in households that were visited and informally
interviewed by this author. Using a membrane filtration method from the 11th Edition
(Standard Methods, 1960), Walewijk conducted testing of 100ml samples using the
Millipore portable membrane filtration unit with a 47 μm filter paper and mColi-Blue24
broth incubating for 24 hours at 35° ± 1° C. By this method, counts of E.coli and Total
Coliforms in 100ml of sample can be determined, yielding the resolution necessary to
investigate low risk conditions that are generally created through the use of HWTS
technologies.
Water Quality Results for International Aid Biosand Filter Users
House
Sample
Date
Flowrate Turbid
E.coli/
T.coli/
Method
Effective
hold
L/hr
TU
100ml
100ml
**
Use
HH1
BSF Inlet
1/21/08
<1
100000
MF
HH1
BSF Outlet
1/21/08
<1
350
MF
Yes
HH1
BSF Inlet
1/18/08
28
<100
4100
PF
HH1
BSF Outlet
1/18/08
32
.3
<100
<100
PF
?*
HH1
BSF Storage
1/18/08
2.3
<100
2100
PF
?*
HH2
Raw/BSF In
1/20/08
<1
21000
MF
HH2
BSF Outlet
1/20/08
<1
600
MF
Yes
HH2
CWP Storage
1/20/08
<1
11
MF
Yes
HH2
BSF Inlet
1/19/08
15
<100
2500
PF
HH2
BSF Outlet
1/19/08
8.6
2.7
<100
4600
PF
?*
HH2
BSF Storage
1/19/08
2.4
<100
1300
PF
?*
HH3
BSF Inlet
1/22/08
40
<100
36000
PF
HH3
BSF Outlet
1/22/08
12
(6)
<100
100
PF
?*
HH3
BSF Storage
1/22/08
(15)
<100
900
PF
?*
* Question marks indicate that the level of detection for E.coli of the Petrifilm method is above that needed
to discern low risk from microbial contamination and thus Effective Use was not judged for these results.
** Petrifilm =PF; Membrane filtered = MF
167
Results Overview:
HH1 achieved Effective Use of BSF from monitoring observation. Microbial testing
shows that filter treated water has <1 E.coli/100ml, conforming to WHO guidelines and
within our definition of effective treatment. The storage unit was clean to the eye and did
not store water for a long period of time. Unfortunately, due to limited time and testing
capability, the storage unit of HH1 was not tested and we cannot conclude that the
storage practices were effective. Disuse of Kosim was described for HH1, such that no
microbial testing or direct monitoring of the Kosim was possible.
Household 2 HWTS water management
At HH2, effective microbial treatment was measured in both the BSF and the CWP. BSF
storage practices are not ideal, however, consisting of an open, rusted iron can. He was
well informed of usage procedures for both systems. The women of the house preferred
the biosand for its quick pouring and access, while the husband enjoyed the taste of his
Kosim CWP.
Effectiveness of treatment is seemingly insured in the BSF of HH3, yet probable
recontamination occurs in the storage unit because it is rusty and uncovered and
accessible (microbial testing hints at this with regrowth of total heterotrophic coliforms,
although the Petrifilm method by itself lacks the resolution to show low levels of risk
from E.coli). She practices secondary safe storage in her CWP storage unit when primary
storage overflows.
Full Interviews:
First Respondent:
Rematu Musah is a 29 year woman who had given birth the previous day. She
was very gracious to have given us an interview. She is the wife of the head of the
household and lives in a brick house with cement floors, a corrugated iron roof and a
limited rainwater harvesting capability. She collects the water herself from the Kpanvo
dugout throughout the year.
168
When Peter asked her for water, she brought a tin full of Guinea worm clothfiltered dugout water from a large ceramic pot in the courtyard. This water had a
turbidity of 60 Turbidity Units.
After learning about the Kosim filter from a demonstration in town by the Pure
Home Water (PHW) volunteer Nachina, Rematu purchased her Kosim filter for 60,000
(~US $6) cedis through 3 installments of 20,000 cedis (~US $2), 7 months prior to our
visit. At the time of the visit, the Kosim was dismantled, with the storage unit in a
separate room and the clay pot filter being used for storage. We asked her to put the filter
together in our surveying room for comparison purposes.
She claims to have used the filter for the last two months of the dry season before
utilizing unfiltered rainwater when the rains came. After the rainwater was depleted, she
was in the third trimester of her pregnancy and relied on others to fetch water. Because
of this added burden, she was not able to put her Kosim filter into service for the few
months after the rainy season. She solely cloth-filtered (good condition) her water until
she received a Biosand filter in late December. The Biosand filter was actively in use
when we entered, with a low visible flow rate.
When asked about maintenance of her Kosim filter, she responded that when
cleaning is needed (3 times per week), she places the filter on a clean surface and uses the
provided brush to clean out the ceramic with Kosim-filtered water, and then uses clothfiltered water, soap and sponge to clean the plastic storage unit. The Kosim filtered
water’s taste was described as “pure water” by her.
Currently, Rematu uses the cloth filter followed by Biosand to treat her water.
She is the only person who operates and maintains the Biosand filter. When Peter asked
her about how she had heard about the Biosand filter, she originally responded that it was
through Pure Home Water. Whether this was actually her perception of the Biosand
intervention or not is debatable, as we seemed to clarify later that she received it from a
white man, probably Carl of the Peace Corps (hence the possible confusion as Matt and
Susan are also both white). Something may have been lost in translation here, but it is
obvious that she did not relate Carter Center or International Aid with donating her
Biosand filter. Regardless, she received the filter a month before our visit, and with it
came a laminated pictorial cleaning instruction from CAWST (see Appendix F: Biosand
Filter Usage Instructions and picture below). Peter claimed that training for the filter
happened at the house, but this too may have been a mis-interpretation, as evidenced in a
later interview. The filter was placed out of direct sunlight in her bedroom, was actively
filtering and spotlessly clean, with a small uncovered white wash basin (also very clean)
for storage. Six people (two adults) use the filter for drinking every day. Water is
constantly added to the filter, and she claims to clean it every two days. Whether this
high cleaning rate is based on need due to the high turbidity and consequent clogging, or
based on a recommendation to clean the filter every three days as instructed by Carl is
unclear, but Rematu said that the water becomes dirty after a few days. Because this high
rate of cleaning is common to many of the Biosand households in Kpanvo, accurate
microbiological testing will tell us if this is a sound cleaning regimen, or whether is it
continually disturbing the schmutzdecke (see data tables that follow for each of the three
households, as well as compiled data in the Field Results chapter). To that end, the
cleaning style described by Rematu is very gentle and most likely not very obstructive of
the biological layer. To clean, first she rinses the filter, and then attempts to make the
169
top of the sand smooth by rubbing it softly with the open face of her hand. After another
“rinse,” it is ready to filter again. She also was very rare in that she claimed to clean the
storage unit with soap and sponge using filtered water (a visual inspection of the storage
container confirmed its cleanliness, but hopefully we can also test it microbiologically).
These cleaning techniques contributed to an overall impression of effective use of the
biosand through observational monitoring characteristics.
Rematu describes the taste of the biosand water as good, like “pure water,”
similar to her claim about the Kosim. The Biosand filter is easy to use and has caused her
no problems. She likes the Biosand because it is beautiful as well as its clean water. She
likes the “model” (a seeming buzz word to Peter, unclear of the intended translation) of
the Kosim, saying that it is transportable, and produces clean and cool water. She could
not say a bad thing about them, she explained, because she liked them both, and they
were too important not to like. She seemed too uncomfortable during the interview to
make opinionated claims. This could be attributable to many things, including a
reluctance to show strong opinions in front of her two male and/or white interviewers. It
is possible that she simply did not have strong preferences between the two filters, and
her answers to questions on which produced better water, health, taste, and aesthetics
were positive to each filter. She did however claim that the Biosand filtered faster and
that the spigot of the Kosim was too slow, as compared to the open storage unit of the
Biosand. She reported no recent diarrhea in the household, although these results are
suspect based on her noticed discomfort in answering this question to strangers.
When asked if she would buy one for a friend, Rematu replied that she would buy
a Kosim for someone else, as to ensure that the storage container was safe for them. As
for herself, she knows how to keep the storage container clean, and would only buy the
Biosand for herself. Willingness to pay for either filter was not inquired.
Effective Use judgment:
• Observationally, the user showed very effective practices with the BSF. Microbial
testing confirmed this by showing that E.coli from the filter is <1/100ml, conforming to
WHO guidelines. The storage unit was clean to the eye and did not store water for a long
period of time, yet no microbial water test was done from storage and we cannot
conclude that the storage practices were effective.
• Disuse but seemingly ineffective use of Kosim, but did not witness first hand and no
microbial testing.
Rematu’s Water Quality Results
Data
source
Sophie
Sophie
Izumi
Izumi
Izumi
Sample
Date
BSF Inlet
BSF Outlet
BSF Inlet
BSF Outlet
BSF Storage
1/21/08
1/21/08
1/18/08
1/18/08
1/18/08
Flowrat
L/hr
Turbid
TU
.3(?)
32
28
.3
2.3
E.coli/
100ml
0
0
<100
<100
<100
T.coli/
100ml
100000
350
4100
<100
2100
Notes
Membrane filtered = MF
MF
*has not cleaned ever
Æsame data source?
Petrifilm =PF
Second Respondent
At the second household we came to, both the Kosim and the Biosand are used in
parallel to provide drinking water. The “landlord,” or head of the household was home,
named Suliemana Ibraham. His wife, who maintains the filters, was out fetching
170
firewood, and we were not able to garner adequate information on cleaning practices
from either Suliemana or his adolescent daughter as they were not the custodians of the
two filters. His wife and children retrieve water from the same Kpanvo dugout as
Rematu throughout the year, unless their dugout is dried up, when they have to travel to
the next community to collect water from their dugout.
The two filters sat side by side out of the sun in a food store room of concrete
floor and thatched roof that seemed not to have access by animals, both with plenty of
filtered water in storage (the Kosim storage was half full and both were actively filtering).
The Kosim unit was raised three inches off of the floor. No vessel was nearby to drink
from, and when Peter asked for a drink of water, the young daughter found two very dirty
cans with which she sampled from the two storage units, dipping directly into the small
open-top steel drum under the Biosand, though there was a ladle-cup already inside.
This man bought his Kosim filter 8 months prior to the interview for 60,000 cedis
(~US 6$. He heard about it during a community demonstration by Nachina, the PHW
community liaison, and later purchased it from him. Suliemana received no training
materials about the filter other than this PHW-led community demonstration. The liaison
also made one trip to his house during installation to fix a leaking gasket seal (washer) on
the tap. Although his household used the cloth filter before buying the Kosim, they had
cases of Guineaworm in their family. Suliemana told us that God answered his prayers
with Pure Home Water, and that it has solved the Guineaworm problem. The cleaning
brush for the Kosim was present, and Suliemana claims that the Kosim is cleaned every
three days (but not actually by him, such that this information may be unreliable). The
filter was very clean inside. The daughter also claimed that the Kosim storage was
cleaned with soap and sponge using cloth-filtered water.
Suliemana likes his Kosim very much, and compares the taste to that of piped
water. He claims that it is easy to use, and said rather inconsistently that he has had no
problems with it, although he admitted to a leak in the tap earlier.
He was introduced to the Biosand filter by the teacher, Joseph, who is the
community volunteer for the Carter Center Guinea Worm Eradication Project. One
month ago, a group of people (notably including mostly white people) came to his house
with the filter and showed him how to use it, and provided the CAWST poster and
appropriate sponge that International Aid recommended to use for decanting dirty water
during cleaning, which resembles that used to pour off oil, a common practice in
Northern Ghana. Four people, including two adults drink from the Biosand filter. He has
no complaints about the Biosand breaking.
For Suliemana, the Biosand changes the scent of the water to that resembling
algal growth, and prefers the cool and earthy water of the Kosim, in which the natural
scent is not altered. He will only drink tea made from the Kosim. Later in the interview,
however, he says that the taste of Biosand is like piped water, with similar taste to Kosim.
The women in his family, however, prefer the Biosand’s taste as they perceive it
to add some type of chemical treatment, and they like using it for cooking as well, as the
flowrate is plentiful. In some cases, the flowrate is too high, and Suliemana complained
that you cannot leave the biosand alone for it will overflow. He shares the sentiment with
Rematu that while the Biosand water is cleaner, it is more susceptible to contamination
after treatment. He wishes to fit a spigot to the Biosand to avoid contamination, and said
ultimately that a Biosand with a tap would clean the water much better than the Kosim.
171
A spigot would ruin the perceived benefit to the women, as they appreciate the ability of
the Biosand to produce a lot of water to an open storage vessel so that they can fetch
water quickly so as not to spoil when making TZ (tea-zed), a complaint that Rematu also
shared. His wife prefers to make soup from the Biosand water as well.
In terms of perceived health impact, Suliemana did not comment on any notable
diarrhea yet said that his wife’s stomach pains have lessened since drinking from the
Biosand.
Suliemana prefers the Kosim, and the women in the family prefer the Biosand.
Part of this preference may be a wish to see a return on his recent investment in the
Kosim (author’s conjecture). In response to our comparative questions (# 47-58 on
survey), he responded that the Biosand cleans the water better and had a very interesting
explanation. He took a wooden bowl to show us the clarity of the stored water from both
filters, describing the Biosand water as “light,” or clear like kerosene, and then the water
from the Kosim as “thick,” or dirty with some particulates. They used to use the Kosim
more, but it is now easier to use the Biosand because of the fast flow rate. The children
like it. The Biosand design looks better to him. He would recommend the Biosand over
the Kosim, and after a good long thought, gave his estimated price of 100,000 cedis (~US
$10) for the Biosand, based on his buying the Kosim at 60,000 cedis (~US $6). He thus
values the Biosand a reasonable amount more than the Kosim. He expects the Biosand to
fetch a higher price, and sees it as a long-term investment.
(1) Suliemana’s Kosim, BSF, metal BSF storage, and fetching jerrycan. BSF
storage practices are not ideal (uncovered rusty drum). Note the BSF training material
shown. (2) Suliemana and his daughter during interview
Suliemana’s Water Quality Results:
Sample
Date
Raw/BSF In
BSF Outlet
1/20/08
1/20/08
Flowrat
L/hr
Turbid
TU
E.coli/
100ml
0
0
T.coli/
100ml
21000
600
Notes
MF
MF
172
CWP Storage
BSF Inlet
BSF Outlet
BSF Storage
1/20/08
1/19/08
1/19/08
1/19/08
8.6
15
2.7
2.4
0
<100
<100
<100
11
2500
4600
1300
MF
PF
PF
PF
Suliemana and his wife showed effective microbial treatment for both of their BSF and
CWP.
Respondent 3
Mata Baba is the woman in charge of water procurement in her house. She lives in brick
wall house with cement floors and corrugated iron roofing, and has a newborn child.
After hearing about the Kosim from the community liaison Nachina, she bought one for
60,000 cedis (~US $6) of her own money 7 months ago. She used it happily until
receiving a Biosand “as a gift from whites” one month ago. She now uses the Kosim
storage unit for occasional overflowing of the metal drum in which she stores the Biosand
treated water, for she likes the tap on the Kosim. At the time of the interview, the
ceramic filter was sitting on the bottom of the storage unit, moist and with condensation
inside.
Mata Baba and her husband are the only ones who operate the BSF, but seven
people in total drink from it. They add water every day to the BSF, and she indicated that
she tends to clean it every two days (!) The cleaning method is that described by the
previous two households, namely using the palm of the hand to flatten the top layer and
using the sponge provided to extract the dirty water. Peter told us that Mata Bata cleans
the open storage unit that receives the Biosand filtered water three times a week with
sponge and soap and cloth-filtered water, but this is a bit hard to believe. The taste is that
of “piped water.”
When asked to fetch a glass of water, the woman’s daughter went to the ceramic
storage pot in the yard to fetch water to clean a glass that she then dipped into the
Biosand storage container and gave to Peter for to drink.
Mata Baba appreciates that there is always water available with the Biosand,
which she did not say for the Kosim. She also likes being able to fetch it quickly, without
having to wait for the tap to pour. As for the Kosim, it looks good, has a nice tap and
always has water on hand (a little inconsistent). For the preference questions of taste,
clean water, flowrate, ease of use, and health impact, Biosand was rated better by this
woman. Both filters look equally good, however. She has noticed a reduction in the
number of skin boils since using these products, and would recommend the Biosand to
her relatives. She said the maximum that she could pay for the Biosand was 100,000
cedis (~US $10), but would pay up to 200,000 cedis (~US $20) if she had the money
available. She had the strongest inclination in favor of the BSF of the households
interviewed in Kpanvo.
Mata Baba’s Water Quality Results:
Sample
Date
BSF Inlet
BSF Outlet
BSF Storage
1/22/08
1/22/08
1/22/08
Flowrat
L/hr
12
Turbid
TU
40
(6)
(15)
E.coli/
100ml
<100
<100
<100
T.coli/
100ml
36000
100
900
Notes
Uses everyday, PF
Cleans when she can, PF
PF
173
Effectiveness of treatment is seemingly insured, yet probable recontamination in the
storage unit for BSF because it is rusty and uncovered and accessible (microbial testing
hints at this with regrowth of total coliforms). Secondary safe storage in CWP storage
unit, although would recommend chlorine treatment.
Water Quality: Although the design flow rate of the
HydrAid BSF is 47 L/hr, the flow
rates were not measured at maximum head. The
average flow rate was 17 L/hr, much slower than the
design flowrate but in good operation range
(Kikkawa, 2008).
Note the picture of Mata Baba’s Kosim and BSF:
Kosim storage unit was empty, and located on the
floor, inaccessible to use and with risk of
contamination to the tap. Unsuitable rusty metal.
Jan. 23, 2008. Peter Alhassan, Susan Murcott and Joseph, the Kpanvo school teacher,
conducted the same survey described above, translated into Dagbani orally by Peter
Alhassan. On this day, four households of the Kpavno community who had purchased the
Kosim filter midway through 2007 and then received a Biosand filter as part of an
International Aid donation in late December, 2007 were surveyed. Water samples were
subsequently collected and analyzed.
Respondent 4 – Dawni (grandfather) and Ayishetu (grandmother)
Dawni and Ayishetu were the elders of this household. Peter Alhassan referred to Dawni
as “senior sister.” (Miscommunication?). This household also collects their water from
Kpanvo Dam. The daughter of the household, who is responsible for the operation and
maintenance of the filters, was not at home at the time of this survey. Therefore, the
questions were answered by Dawni first, and Ayishetu second.
At the time of this survey, only the BSF was being used. They had used the Kosim and it
had been kept in the daughters (and her husband’s?) bedroom. Their 2-year old, whom
we met, also slept in that room. The Kosim had been on an unstable stool directly beside
the bed. The child had knocked the Kosim over in the night and had broken the pot. The
Kosim also leaked, because the hole was too large for the tap. And, there was a black
washer – only one, not two as should be the case. The washer did not fit properly for the
size of the hole and diameter of the tap. In addition, the hole was rough – it was not
properly filed down, which also may have contributed to the leakage. This was a mute
174
point though, because they were not using the Kosim after it broke. After that, they
received their Biosand filter. They use the Biosand filter and it is kept in a common room.
This family also said it had been given 10 Lifestraw filters (one for every family
member) – they brought out several of the small size black ones to show us. These were
given out in November 2007, according to Dawni, and every person in Kvanpo has one.
He said they use it when they go to their farm fields and get thirsty. In Dec. 2007, three
people in Kvanpo had guinea worm. Now, they are cured and there are no new cases,
according to Dawni.
This household received their BSF in December from Carl/Osman. They received
training materials. We observed that they used their water collection can (the 40 liter
metal type) as their receiving water container. When Susan questioned them about this,
Ayishetu, who was doing the demonstration of how they use the BSF, said that they used
jerry cans as receiving containers. There were two jerry cans beside the BSF. Ayishetu
commented that they were both leaky and she sent someone to get a non-leaky jerry can.
It was brought, and the interesting thing was that when we observed decanted BSF
filtered water going into the jerry can, one had to watch it, because of the narrow
opening. You couldn’t just let it flow, as you could with a wide, open-mouthed receiving
container.
There were about 10 people in this household using the BSF, including 4 children. They
clean the filter 3 times per week. They received instructions in their home on how to
clean the filter. They like the taste of the filtered water, it is not hard to use the filter and
they have not had any problems with it. They like the “good water, not spoiled.” There is
nothing they don’t like about the filter.
Regarding the Kosim (which they no longer use because it is broken), they liked the taste
of Kosim filtered water. It “never tastes salty, just like pipe-borne water.” The Kosim was
not difficult to use. Before they acquired the Kosim, did they treat the water at all? Yes,
they treated the dam water at the source with Abate, about every three months.
Q: How long did they have the Kosim?
A: Long time.
Q: Who was responsible for it?
A: Rabi, their daughter.
Q: How did they hear about the Kosim?
A: Nachina, the PHW liaison
Q: How much did they pay.
A: They paid GHC 2, which was partial payment.
Q: Did they receive any training papers with the Kosim
A: No.
Q: Did the sales person come to their house?
A: No. They learned about it at the PHW demonstration.
Q: How often do they pour water into the ceramic pot?
A: They add whenever the water gets low.
175
Q: How often did you clean the Kosim?
A: Every day
Q: Did the sales person come to your house and show you how to clean the filter?
A: No, they learned at the PHW demonstration
Q: Was a brush provided?
A: Yes.
Comparing the BSF and Kosim, they liked both filters. Both cleaned the water well. The
BSF was faster, in terms of flowrate. The BSF was easier to use, but you have to watch it.
The Kosim you could leave and go back later – you didn’t need to watch it. In contrast,
you have to stay close to the BSF when you pour water into it (as we observed when they
tried to filter directly into a jerry can). The BSF flows like piped water. If they were
going to buy one of these two filters for their family, they would buy the BSF. How much
would they want to pay? They would want to pay GHS 6 (they knew that this was the
cost of the Kosim). They stated that they had no diarrhea in their family.
Respondant 5 – Ibrahim Abdul Rahaman
Ibrahim Abdul Rahaman is the village chairman and this was the second visit our MIT
team had paid to his home – we had stopped by there briefly the previous week with Carl,
Sophie, Izumi and others. However, we didn’t know this household had a Kosim when
we paid our first visit. In addition to being the village chairman, Mr. Rahaman is a
butcher by trade, which was explained to us as meaning that he is a bit wealthier. He has
three wives: 1st = Lansah, 2nd = Asibi and we didn’t learn the name of the third. Neither
were available, so he became our survey respondent.
Their water source is Kpanvo Dam, which they say dries up in April or May. Ibrahim’s
wife and children collect the water.
Ibrahim had purchased the Kosim and kept it in his room for his private use. His also had
roughness around the tap hole from not having been properly filed down.
Regarding who is responsible for cleaning the BSF, it was his 1st and 2nd wives. Ibrahim
first heard about the BSF in the community, then his household became the site of the
first installation, together with the installation at the chief’s palace. The BSF is working
well, and it has been doing so for the 2 months since its installation.
Observation of their storage post BSF filtration showed that there was likely
contamination – they used random jerry cans which did not appear clean. Ibrahim also
said that although they have lifestraw filters, they prefer to take jerry cans of BSF water
with them to the fields – they found it easier.
The number of people using the BSF was 15, with 4 adults (1 husband, 3 wives) + about
11 children. The BSF was last cleaned the day before yesterday and it was cleaned about
every 4 days. The verbal, but not acted out, description of how to clean the BSF seemed
176
accurate. Ibrahim said that he and his family liked the taste of BSF water, that it tasted
like pipe0borne water. It was not difficult to use and they had no problems with the BSF.
Regarding the experience with the Kosim, he likes the taste of the Kosim as well. “It
tastes like BSF water.” The only difficulty he has had with the Kosim is leakage – exact
same problem as Respondant #4 – roughness around hole, the hole is too big for the tap,
and again, there was only one washer (black) which was not the right size. They have not
had any problem with the Kosim breaking – on the other hand, it is only in Ibrahim’s
room, not shared with the women and children.
Comparing the BSF and Kosim, he prefers the Kosim, his wives prefer the biosand.
Which water tastes better – the same. Kosim cleans the water better, the BSF has a faster
flow rate. The BSF is easier to use. Both filters are good for health. The Kosim is a better
model because it has a tap. Ibrahim suggests adding a tap to the BSF. He paid GHS 6 for
the Kosim, and he would pay GHS 10 - 20 for the BSF. Would he pay up to GHS 20 for
the Kosim? No, only up to GHS 10. Regarding diarrhea in the family – yes, they have
seen a change in the rates of diarrhea since they started using the filters.
Respondant #6: Bhinayili = father, Mde. Absuli = mother, Idurisu Adbuli =
unmarried son, age 16.
In this household, Idruisu, the 16 year old son had purchased the Kosim, and he kept it in
his room. The BSF was used generally by the household, and it was kept in a common
room. It was maintained by his mother, Mde. Abduli, and she answered the BSF
questions.
The household received the BSF one month ago from Carl/Osman/IA. They came to the
house to do the installation. In terms of acting out how to clean the BSF, Mde. Abduli
demonstrated correct cleaning procedures. She used a cup, not a sponge, to decant the
BSF during cleaning. She explained that there were about 12 people who used the filter
every day, including 7 children and 5 adults. They like the taste of BSF water – it tastes
like “pure water.” The filter is not hard to use.
The last time the BSF was cleaned was yesterday and she uses BSF filtered water to clean
the filter. Mde. Abduli said that if she doesn’t clean it every two days, the water will
come out dirty, looking like dam water.
We were shown into Idurisu’s room, and the Kosim pot was on the floor and the storage
container, which was on a small stool with the Kosim pot under it, was nearly full to the
brim with water. It was unclear how he could have filtered so much water through the
Kosim pot, so it was unclear whether this was all filtered water or only partly filtered
water.
When asked if he liked the taste of filtered water the answer was yes, that it tasted like
piped water. The Kosim filter was not difficult to use. Before he obtained the filter, he
did not treat his water.
177
The comparison between the two filters included the father, mother and son. What they
liked about the BSF was that the water was “light” and very clean, cleaner than the
Kosim. The water from the BSF tastes good and there was nothing they did not like about
the BSF. Mde. Abduli likes the “mold shape” of the Kosim and there is nothing they
don’t like about the Kosim either. All three liked the BSF better – the father said he liked
it because it removed bacteria and guinea worm and that it was fast. Both the father and
mother thought the BSF water tasted better than the Kosim. It was faster, they use it more
often, it was easier and better for their health. It also looked better. If they were going to
buy a filter for their family, they would pay GHS 6 for the Kosim and they would pay
GHS 10 for the BSF, but they would not pay GHS 15 for a BSF – that was too much.
They had not seen any change in the family’s diarrhea as a result of using the filters.
Respondant #7: Amim Fuseini
Amim Fuseini is a health extension worker at the Kpanvo Health Clinic. He is the
nephew of the chief of Kakpagayili, where we had been the previous day. When we met
him at Kakpagayili, he had explained that when he was with his uncle, he used the Kosim
regularly, but that when he was at home in Kpanvo, he used the BSF, as his household,
like all in Kpanvo, had received a free one.
Amin is age 28, married with one child. His family gets its water from Kpanvo Dam, and
his wife collects it. Because we visited him at the Health Clinic, we did not see his BSF,
however, he was able to compare them for us.
Asked if he likes the taste of Kosim water, the answer was yes. It tastes “like chemical.”
Sometimes the same, sometimes different. (?) The only problem with the Kosim is the
slow rate of flow, but if you have several Kosim, no problem.
Asked if he treated the water before he had the Kosim, the answer was yes, he used the
guinea worm cloth and also, used alum when the dam was turbid. Alum was purchased in
Tamale, and it was not so easy to use, and it was expensive.
Amin’s wife had received their BSF from Joseph, the school teacher, who had come to
his house during the installation. There were 5 people who used the BSF every day,
including 3 children and 2 adults. The BSF was filled once per day. Cleaning took place
once the flow rate comes to a stop – that is the indication that it is time to clean it. That is
the latest cleaning instructions they have received. Previousl7y, they were told to clean
the filter every 3 days. This is what was told to them by Carl/Osman.
Do they like the taste of BSF water – yes, it tastes “like chemical.” “It seems like they put
some chemical into it.” The filter is not difficult to use and they have no problems with it.
Comparing the BSF and Kosim, the BSF is easy to fetch water – there is no delay.
However, it is a lot of work to regularly clean the BSF – every several days.
178
What they like about the Kosim water is that it is cool. What they don’t like about the
Kosim is that if you don’t wash the pot, after several days, the filter will not flow. Also, if
you are not careful, the filter can break.
Which water tastes better? Both
Which cleans the water better? Both
Which one filters faster? Both
Which do you use more often? BSF.
Which is easier to use? Kosim
Which water is better for your health? Don’t know.
Which filter looks better? Both.
What are they willing to pay for either filter? They would pay GHS 2 for either.
Because everyone in the community who wanted a Kosim had to pay for it, only 7 or 8
people got it, but Osman/Carl/IA have brought clean water to everyone in the community
via the BSF distribution. According to Amin, most youth have no jobs and most adults
don’t have three square meals per day in this community, so although people may want a
filter, they cannot afford to pay for it.
*********************************************************
According to the school teacher, Joseph, the only problem with the Kosim is the
breakage. Apart from that, there is no problem. With the BSF, if one is not patient, it is a
lot of work to wash it. As soon as you use it, you need to clean it. This requires a lot of
water, and water is in limited supply and is hard to come by. Whereas the cloth filter only
removes guinea worm, the Kosim and BSF take out all bacteria, same as with piped
water. With alum, a chemical, if you use too much, you will get stomach pains, and it
may not removal bacteria.
Household 8
Nachine Ziblila, Male, ~50 years old, maybe less
Nchine’s wives and children collect water from Kpanvo dam throughout the year. When
asked for a glass of water, he cleaned out a cup (using unfiltered water) and brought
Kosim-filtered water to us. It had clay particles in it, making it seem as if the storage unit
had not been cleaned recently. The Kosim was not situated correctly, only a few inches
off the ground on a wobbly piece of Styrofoam. Storage was almost empty. Nachine is
the community liason for PHW, and thus received his filter from Shak as a donation
along with that of the chief’s filter. He perceives the filter rate as adequate to his needs.
His GWEP cloth filter had holes in it. He cleaned the Kosim two days ago correctly,
except that he claims to use unfiltered water to clean the storage unit of the Kosim.
Influent turbidity level from clay-pot storage is 40 TU.
179
Nachine and his wives prefer to use BSF water if that unit is full, and will use Kosim
water if no BSF water is available. They received the BSF 1 month ago, and still have
the CAWST training materials posted on the wall above the unit. Cleans the BSF 4 times
a week, saying that they had forgotten to clean it that morning, which emphasizes that
more cleaning was thought to be better cleaning. Could not determine if they clean the
BSF correctly through translational issues and time restraints. Has no permanent storage,
and filters directly for use. Wife prefers taste of BSF, but Nachine does not show a
preference and uses the Kosim once the BSF water is finished.
a
name
Abdulai-Iduriso
Ibrahim-AbdulRahaman
Fuseinikipem
b
E.Coli
Total
CFU/100 ml
0
CFU/100 ml
18000
0
5000
4000
116000
c
E.Coli
CFU/100
ml
0
Total
CFU/100
ml
7
0
980
0
1400
E.Coli
CFU/100
ml
0
Total
CFU/100
ml
100
0
1100
180
Guided by Sammy, a field attendant for Enterprise Works who collaborates directly with
the monitored communities
Enterprise Works
January 29th 2008
Ahentia Community, Afutu Ewutu Senya District, Central Region
A peri-urban area one hour west of Accra, in villages a few miles north of the main road.
Synopsis:
Of 6 filter users monitored, 5 of them passed by demonstrating excellent use of the CWP,
showing 83% of households practicing Effective Use for this small subset of users. The
one remaining household claimed consistent use yet their filter had not been filled in
three days, and showed poor hygiene in her handling technique. No water quality
measurements were made.
Field Site Overview for first village, with the Chief as Community Retailer
• No microbial water quality measurements were taken at this site
• HH1-HH5 reside in a village which was sold 130 CWP filters produced by
Ceramica Tamakloe (CT), Accra, identical to the ones sold in Tamale by Pure
Home Water. This was the first community in which Enterprise Works (EW)
sold filters. The first 30 filters sold at US $12 on installments, but due to low
selling rates, the last 100 were subsidized at US $5. For each filter sold, the
chief (the appointed retailer in this village) receives US $2. He stopped
selling filters in July, and although 4 filters have broken, he has not contacted
CT or EW to buy replacements, as he has not received any money for the
replacements from users (he probably does not receive a commission on the
replacements). When a user has a problem, he comes to the chief with a
question, and then the chief visits the house (according to the chief). One of
the chief’s 3 filters was currently broken. He would like to sell more. He has
seen many of the taps spoil. A borehole is the main source for the village, as
“the harmattan [dry season dusty winds] has finished the river source.”
• Perceived Health Benefit: Chief sees reduction in guinea worm using the
filter, and has stopped using the Carter Center cloth that he and most other
villagers possess in lieu of the ceramic pot’s effects. He has not seen diarrhea
in a long time, and his friend attributes his lack of eye problems to the filter.
• Cleaning: The chief teaches this cleaning regiment to users, for a
commission: Every three weeks, wash the storage unit with soap and sponge,
use a brush to clean the pot. (Note the teaching that cleaning is done on a time
basis, and probably assumed the clean borehole water for a source. EW
encourages a nylon fishnet sponge for cleaning but does not provide one as
part of their sales).
181
Household # 1
The Chief and retailer for the community, with
one of his filters.
Name and status of person interviewed
• Chief’s house, interviewed his wife
briefly
Household visit notes
• She says the water has a “fine taste
on the tongue”
Monitoring Observation
• Filter is set up in clean entryway, on
stable table off of the ground.
• Ceramic pot is empty but damp,
storage half fullÆgood technique
• Separate drinking glass used just for
filter
• Tap is dirty
• Cleans twice in three weeks
Effective Use Assessment
• Other than dirty tap, very well used.
Household # 2
Name and status of person interviewed
• Chief’s neighbor
Monitoring Observation
• Same as previous case, except non-dirty tap.
Effective Use Assessment
• Very good, based on monitoring observation alone
Household # 3
Name and status of person interviewed
• Gladys, daughter of owner of the CWP
Household visit notes
• Her father bought the CWP from the chief
• 5 people drink from it
• Lots of flies in house
182
Monitoring Observation
• No designated drinking cup
• Little water in pot, has not filled for three days
• Has had the filter for about a year now, and says that it has stopped diarrhea
for a long time
• Filter is dark with dirt, yet is cleaned every four days
• Uses sponge and no soap to clean the pot
• Sponge for cleaning and training materials are kept in father’s room, and not
available for me to see
• When asked for a glass of water, she rinses the cup with filtered water yet
washes with her unclean hands before pouring a cupÆineffective hygiene
• No clay dust in the sampled water=clean storage unit(?)
Effective Use Assessment
• Can not be using is it for all drinking purposes if filled three days agoÆ
suspect inconsistent use
• Having no designated drinking cup and not washing hands before handling
inside of cup negates the potential benefits of CWPÆineffectively used
Household # 4
Name and status of person interviewed
• Gifty, the mother of home and
caretaker of CWP
Household visit notes
• CWP stays in the parent’s bedroom,
with a thatch roof and cement floor
• Gifty learned of the CWP from a
display given by EW staff member
Sammy in the village
Monitoring Observation
• Although pot is empty, she claims that
it has been consistently filled
throughout the past 4 days
• CWP sits on specially fabricated table
•
•
•
•
(see picture above)
A dedicated drinking cup sits next to the CWP
User displays the usage poster on the side of the CWP
She grabbed a nylon sponge from the kitchen to demonstrate proper cleaning
of the pot
CWP “reduces small-small sickness that has been worrying them”; no
diarrhea in a very long time
183
Effective Use Assessment
• Very well informed, and attentive to her CWP
Household # 5
Name and status of person interviewed
• Joyce, the mother of home and
caretaker of CWP
Household visit notes
• clean cement floor, CGI roof
• paid US $12 (full price,
unsubsidized)
•
•
•
Monitoring Observation
• Using the CWP for 1 year so far
• CWP sits on specially fabricated
table (see picture above)
• a dedicated drinking cup sits next to
the CWP, although it does not look
very clean
• Joyce learned how to use and clean
the CWP from a demonstration
training given by Sammy in the village
She keeps it full, with water above the bottom of the pot in storage
Cleaned it three days ago with a nylon sponge
Looks like there is a crack in it (I did not inspect due to there being water in
the pot, but it still holds water while filtering), but she is not worried about
cracks in the filter until it breaksÆmisperception of how the unit works??
Effective Use Assessment
• Monitoring observation checked out all round
Site Overview for 2nd village, with Emmanuel Amponasah as Community Retailer
• This site was less than a mile away from the previous one, a bit more towards
the main road but just as rural.
• As the retailer for the community, Emmanuel sold 40 filters. He finds that
without subsidies, at US $12 as currently charged by the producer Tamakloe,
the filters will not sell.
• Training: Promotes the use of ceramic water purifier (CWP) by wearing
“good water” T-shirt and talking up the health benefits. He notices that
diseases such as malaria, bilharzias, diarrhea have come down since using the
filter.
184
•
•
•
•
Monitoring by community retailer: Two weeks after installation, he went
around to the houses and checked up on them. Found one bad filter, which
was replaced. Last time he checked the filters was with Sammy (my liaison
for the day) in November, and found no problems except one leaky tap, which
they fixed.
Sources of water in the community: rain water harvesting, 2 boreholes, and 1
hand dug well with a pump. Emmanuel does not pretreat the water, never did.
He claims consistent use for his family of five, and says that he buys sachet
water when out of the home. He likes the taste of the clay.
His children (ages unknown) fetch water from the storage unit, but he has had
no breakage of the tap.
Household # 6
Name and status of person interviewed
• Hannah, mother of household and caretaker of CWP
Household visit notes
• Reduced diarrhea rate, can’t remember last incident
• Would buy again for 7-8GHC if she had the ability to pay, but she was under
financial strain at the time
Monitoring Observation
• Half-full storage
• Clean tap
• Sitting on blocks
• Two separate cups for drinking
• Pot almost full
• Lid dirty
• Filtered water is clean
• Filter in use for a year
• Cleaning: cleans when dirt in potÆdone last week, with good flow; uses
nylon sponge, esp. for filter; puts pot in clean basin while washing
• Learned how to clean it by household visit by retailer because she was not at
community meeting, though her husband was
• Refills pot when it drains to empty
• Was prefiltering with a white cloth filter (not the one handed out by GWEP) at
one time, but not anymore.
Effective Use Assessment
• Achieved the observational monitoring effective use criteria
185
186
Appendix D: Portable Laboratory Testing Addendums
DelAgua Turbidity Tube Instructions (from www.delagua.org)
3M Petrifilm Instructions (from Bob Metcalf)
IDEXX Colilert Instructions (from Bob Metcalf)
187
Turbidity Analysis using the DelAgua Turbidity Tube
Note: The turbidity tubes cover the range 5 to 2,000 TU
1. Remove the two turbidity tubes from their clips in the lid of the test kit case.
Carefully push the upper tube (open at both ends) squarely into the lower tube. Look
through the open end of the tube at the black circle printed on the base of the tube. Ensure
that there is good illumination available. Normal daylight is adequate for this purpose.
2. Fill the turbidity tube with the water sample to the very top. Allow a few moments for
the water to settle (as there may be formation of bubbles) before you read the tube. Hold
the tube between the thumb and index finger at the joint of the two tubes. Have your arm
fully extended. Do not strain to see the black circle as this can sometimes cause biased
results. Continue to pour small amounts of the water slowly out of the tube checking each
time to see if you can see the black circle. As soon as you can clearly see the black circle
then read the level on the outer markings on the tube.
3. Alternatively, pour the water sample into the tube from the sample cup until the black
circle just disappears when viewed from the top of the tube. Avoid creating bubbles, as
these may cause false readings.
4. The turbidity tubes are graduated in a logarithmic scale with the most critical values.
The result is the value of the line nearest the water level. This permits a reasonable
estimation of the turbidity of the water sample. As the scale is Logarithmic it will be
difficult to get an accurate reading when the water level is between scales. It is always
better to take the reading which appears below the water level.
Taken from:
OXFAM – DELAGUA (2000) “Portable Water Testing Kit Users Manual.” (abridged
web download version) Revised and updated 4th edition. www.delagua.org
188
189
190
Appendix E: Effective Use Monitoring Checklists
Sodium Hypochlorite Solution
Aquatabs
SODIS
Cloth Filter
Ceramic Pot Filter
Biosand Filter
PUR
191
192
Sodium Hypochlorite Solution Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist ( Yes/No/ NA)
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
manufacturer’s specifications, without prompting from the monitor:
1.1. Add a single dose to clear water of the correct
volume.
1.2. Double dose for water that is visibly dirty and/or
from an unimproved source.
1.3. Allow visibly dirty (turbid) water to settle and/or
filter through a clean folded cloth before double
dosing.
1.4. Shake thoroughly after chlorine addition.
1.5. Let sit for 30 minutes prior to drinking.
6. Separate containers for fetching water and
Storage
disinfection/storage of water are used.
7. The dosing volume as specified on the hypochlorite
product is easily measurable in the safe storage
container used for treatment and storage.
8. The design of the safe storage unit (for treatment) has
a tap or a small sealable opening for pouring.
9. Safe storage container is clean, and has no leaks.
10. Safe storage container is out of the sun.
11. Safe storage container is indoors.
12. Safe storage container is raised off the floor and
stably situated.
13. Safe storage container is out of reach of animals and
small children.
14. Lids are kept on tight, and only opened for addition
or pouring of treated water.
Maintenance 15. Regularly scheduled cleaning of the storage unit.
16. Soap or disinfectant used to clean storage unit can be
produced by user.
193
17. User knows expiration date as specified by
manufacturer or distributor on bottle.
18. Unexpired sodium hypochlorite solution sufficient
for at least ten treatments is in stock and easily
accessible if consistent use is claimed.
19. Water bottles for use during travel or school are clean
and producible to the interviewer if consistent use is
claimed outside the home.
20. A dedicated clean cup is used with the safe storage
unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
Replacement
Period
Physical
Inspection
Water Quality Monitoring
Free available chlorine presence is shown if
Chlorine Residual
treatment is claimed.
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial Testing
Notes:
Sample from
Storage of
Treated
Water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
( Yes/No/ NA)
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Test for presence of chlorine residual in stored water while at the household if chlorine
treatment is claimed.
Procedure
2. Take a sample of treated water from the storage unit for microbial analysis. Use a Sodium
Thiosulphate sampling bag if transporting sample to laboratory. Keep the sample out of the
sun and start microbial test within 6 hours.
194
Aquatabs Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist ( Yes/No/ NA)
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
manufacturer’s specifications, without prompting from the monitor:
1.1. Add 1 tablet per 20 liters of clear water
1.2. Add 2 tablets for 20 liters of visibly turbid water
1.3. Before double dosing, filter the water through a
clean folded cloth.
1.4. Let sit for 30 minutes prior to drinking.
5. Pretreatment is practiced for turbid waters.
6. Two separate 20 liter containers for fetching and
Storage
disinfection/storage are used.
7. Design of safe storage unit (for treatment) has a tap
or a small sealable opening for pouring.
8. Safe storage container is clean, and has no leaks.
9. Safe storage container is out of the sun.
10. Safe storage container is indoors.
11. Safe storage container is raised off the floor and
stably situated.
12. Safe storage container is out of reach of animals and
small children.
13. Lids are kept on tight, and only opened for addition
or pouring of treated water.
Maintenance 14. Regularly scheduled cleaning of the storage unit.
15. Soap or disinfectant used to clean storage unit can be
produced by user.
Replacement 16. User knows that product expires 5 years after date of
manufacture, as printed on Aquatab sleeve.
Period
17. Water bottles for use during travel or school are clean
Physical
and producible to the interviewer if consistent use is
Inspection
claimed outside the home.
18. At least one sleeve of ten non-expired tablets is in
stock and easily accessible.
195
19. A dedicated clean cup is associated with the safe
storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
Water Quality Monitoring
Turbidity is <80 NTU.
Turbidity
Free available chlorine presence is shown if
Chlorine Residual
treatment is claimed.
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial Testing
Notes:
Sample from
Storage of
Treated
Water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
( Yes/No/ NA)
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Test for presence of chlorine residual in stored water while at the household if chlorine
treatment is claimed.
Procedure
2. Take a sample of treated water from the storage unit for microbial analysis. Use a Sodium
Thiosulphate sampling bag if transporting sample to laboratory. Keep the sample out of the
sun and start microbial test within 6 hours.
196
SODIS Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist ( Yes/No/ NA)
1. User demonstrates knowledge of treatment and dosing as intended by
Treatment
manufacturer’s specifications, without prompting from the monitor:
1.1. Fill clean bottles with water and close lid tightly.
1.2. Place the bottles on a corrugated iron sheet or on
the roof, and in a place with continuous direct
sunlight throughout the day
1.3. Leave in direct sun from morning to dusk.
1.4. If ≥50% overcast, leave out for 2 days
5. Use of clean and clear PET bottles that are ≤5 liters
in volume and not heavily scratched
Safe Storage 6. Storage bottles with treated water are stored safely
and out of reach of small children.
7. Lids are kept on tight, and only opened for addition
or pouring of treated water.
8. Secondary safe storage is not witnessed.
Maintenance 9. Bottles are clean.
10. Soap used to clean bottles can be produced.
Replacement 11. Bottles are not scratched or opaque.
Period
12. Bottles do not leak.
13. Treated water is available.
Physical
Inspection
14. If weather conditions permit, water is currently being
treated.
15. A dedicated clean cup is associated with the safe
storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
197
Water Quality Monitoring
If when one’s hand is placed under a filled bottle
Turbidity
laying horizontally and the fingers are still
visible, then the turbidity requirement is satisfied
Microbial testing shows <10 E.coli CFU/100 ml
Microbial Testing
Notes:
Sample from
Storage of
Treated
Water
24 hr Colilert (Yes/No)
Yellow? Flouresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
( Yes/No/ NA)
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If tube is clear following incubation, <10 of both Total Coliform and E.coli/100ml are present;
If the water is yellow, >10 Total Coliform/100ml are present;
If tube is yellow and fluoresces, >10 of both Total Coliform and E.coli/100ml are present.
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Take a sample of the treated water from the bottles in storage for microbial analysis. Keep
the sample out of the sun and start microbial test within 6 hours.
Procedure
198
Cloth Filter Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
No water sampling or testing is needed to measure Effective Use of the Cloth Filter.
Monitoring Observations
Checklist
1. Fastens manufactured cloth tightly to water storage
Treatment
vessel before adding water to maintain separation of
filtered water.
2. Always use manufactured cloth filters with the same
side up
3. Fold sari cloth at least 4 times and wrap tightly
around rim of storage vessel inlet before adding.
4. Filter all water immediately while at the source or
upon returning home from the source.
5. Use filtered water for all domestic water uses
6. Maintain separation of filtered water from nonStorage
filtered water.
Maintenance 7. Rinse off filter after each use, with a final rinse of
cloth filtered water.
8. Leave cloth in the sun for decontamination.
9. Clean cloth filter with soap regularly (if instructed).
10. Monitor witnesses that cloth filter is clean.
11. Soap or detergent used to clean cloth filter can be
produced by user (if applicable).
Replacement 12. Monitor witnesses that cloth filter has no tears or
holes.
Period
13. User stores cloth filter in a safe and accessible place
Physical
Inspection
14. User knows where to get a new cloth filter when
needed (if bought or distributed).
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
( Yes/No/ NA)
199
200
Ceramic Pot Filter Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist
1. Water is added to the CWP every day.
Treatment
2. Ceramic pot is frequently topped off in order to
achieve faster flow rate.
3. Ceramic pot is not overfilled. Water level is not
above 3cm below the lip of the pot.
4. Storage unit is not filled above the bottom of the
ceramic pot.
5. Lid for the CWP is kept in place except when being
filled.
6. Turbid waters undergo settling for at least one hour
before ceramic filtration.
7. CWP is raised above the ground to near table height
8. CWP its level on a stable base.
9. CWP located out of direct sunlight.
10. CWP out of reach of young children and animals.
11. Pot remains in place throughout use as directed,
Storage
maintaining a closed storage unit.
12. Storage unit is clean inside and out (if accessible).
13. Secondary safe storage is not used without chlorine
disinfection.
Maintenance 14. Does the user have a good scheduling mechanism for
cleanings?
15. User correctly demonstrates scrubbing the inside of
the pot with a hygienic brush and rinse with filtered
or boiled, cooled water.
16. User never uses soap or disinfectant with the ceramic
pot itself.
17. Ceramic pot, storage unit and tap are clean with no
visible leaks or cracks.
( Yes/No/ NA)
201
18. There is water in the storage unit.
19. Ceramic pot is partially full or at least damp.
20. Water bottles for use during travel or school are clean
and producible to the interviewer if consistent use is
claimed.
21. User demonstrates hygienic method when asked to
add or fetch water to the CWP.
22. Instructional material is displayed with the CWP, if
provided during purchase or installation.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
Physical
Inspection
Water Quality Monitoring
Treated water is expected to be clear (<5NTU)
Turbidity
unless influent is >100NTU.
Free available chlorine presence in secondary
Chlorine Residual
safe storage if chlorine treatment is claimed.
Microbial testing shows <10 E.coli CFU/100 ml
Microbial Testing
of treated water from storage unit(s).
Notes:
Sample from
Storage of
Treated
Water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
( Yes/No/ NA)
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. If treated water is visibly dirty, check the turbidity if sufficient volume exists.
Procedure 2. Check that the flow rate sounds like one drip a second or so.
3. Take a sample of treated water from the CWP storage unit for microbial analysis. Keep the
sample out of the sun and start microbial test within 6 hours.
4. If a secondary safe storage container is used, take a sample for microbial analysis. If
chlorine treatement is claimed by user, test for presence of chlorine residual while at the
household and use a Sodium Thiosulphate sampling bag for transporting sample to
laboratory.
202
Biosand Filter Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
(Yes/No/ NA)
Monitoring Observations
Checklist
1. Water is added daily to the filter.
Treatment
2. Uses separate containers to fetch/pour dirty water and
store filtered water.
3. Adds water slowly with the diffuser plate in place.
4. Pretreatment is claimed for turbid waters (>100NTU).
5. The spout is unobstructed and clean.
6. Smooth and level sand bed at water depth of 4-6 cm.
7. BSF is sitting flat on firm ground.
8. The lid to the filter is in place and clean.
9. System is out of direct sunlight.
10. System is out of reach of animals.
11. Filter has no visible leaks or cracks.
12. Filter flowrate is ~0.6 L/min.
13. Dedicated safe storage unit is used.
Storage
14. Design of safe storage unit incorporates a tap or a small
sealable opening for pouring.
15. The safe storage container has a lid that is kept on tight
except for adding or pouring treated water.
16. Safe storage container is located with the BSF indoors,
out of the sun, off of the floor, in a stable position and
out of reach of animals and small children.
17. Safe storage unit is visibly clean.
Maintenance 18. User uses and demonstrates “swirl and dump” cleaning method:
18.1. Adds ~4 liters of water to the top of the filter
18.2. Scoops out dirty water with small container,
levels sand and replaces diffuser plate.
18.3. Fills with water and repeats the process if flow
rate is still slow.
21. Filter cleaning schedule is determined by significant
reduction in flowrate.
22. BSF cleaned less than once a week.
203
23. User cleans the spout and storage unit with treated
water and soap or chlorine solution each week.
24. Soap or disinfectant used to clean storage unit can be
produced by user.
25. Water bottles for use during travel or school are clean
Physical
and producible to the interviewer if consistent use is
Inspection
claimed outside the home.
26. User demonstrates hygienic method when asked to add
water to filter and fetch a glass of water.
27. A dedicated clean drinking cup is associated with the
safe storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU).
Turbidity
Chlorine Residual Free available chlorine presence in safe storage if chlorine
Microbial Testing
( Yes/No/ NA)
treatment is claimed
Microbial testing shows <10 E.coli CFU/100 ml in water from
both running spout and storage unit.
Notes:
Sample from
running spout
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
# E.coli/
100ml
Risk
Level
Sample from
storage of
treated water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Take a sample of treated water from the storage unit for microbial analysis (if available). If
chlorine treatment is claimed in stored water, test for presence of chlorine residual while at
Procedure
the household and use a Sodium Thiosulphate sampling bag for transporting sample to
laboratory. Keep the sample out of the sun and start microbial tests within 6 hours.
2. Fill the BSF to a consistent level (not to the top).
3. Check the turbidity of the filtering water if it is visible and sufficient volume exists.
4. While taking a sample for microbial analysis from the pouring BSF spout, take a flow rate
measurement by counting seconds until 100ml is full in the Whirlpak bag.
204
PUR Effective Use Monitoring Checklist
Monitor Name:
Community:
Interviewee Name:
Household/Code:
Date and Time:
GPS Coordinates:
____________________
_____________________
Notes:
Instructions: For each observation, fill in Yes, No, or NA for observations that do not apply. Add up the
total #Yes, divide by the total # of observations made, and multiply by 100 for % Observational Effective Use.
Monitoring Observations
Checklist ( Yes/No/ NA)
User demonstrates knowledge of treatment and dosing as intended by
Treatment
Proctor and Gamble, without prompting from the monitor:
1. Add: Cut open one packet and add contents to ten
liters of water
2. Mix: Stir aggressively for 5 minutes and let sit for 5
minutes; if non-flocculated after the wait, stir again
until floc falls out.
3. Filter: Poor water into clean storage container
through a clean and dry cotton cloth without holes.
4. Drink: Wait 20 minutes to drink. Do not consume if
yellow.
5. Complete consumption of the ten liters of treated
water should occur within 24 hours.
6. Two separate, dedicated 10 liter containers for
Storage
fetching/flocculation and disinfection/storage are
used.
7. The volume for treatment as specified on the
hypochlorite product is easily measurable in the safe
storage container.
8. Design of safe storage unit has a tap or a small
sealable opening for pouring.
9. Safe storage container is clean, and has no leaks.
10. Safe storage container is out of the sun.
11. Safe storage container is indoors.
12. Safe storage container is raised off the floor and
stably situated.
13. Safe storage container is out of reach of animals and
small children.
14. Lids are kept on tight, and only opened for addition
or pouring of treated water.
205
15. Rinse off the cloth filter after each use, with a final
rinse of cloth filtered water.
16. Leave cloth in the sun for decontamination.
17. Regular cleaning of cloth filter with soap.
18. Regular cleaning of the treatment and storage
containers with soap or disinfectant.
19. Soap or disinfectant used to clean storage unit and
cloth filter can be produced by user.
Replacement 20. User knows that product expires 3 years after date of
manufacture, as is printed on sachet.
Period
21. Water bottles for use during travel or school are clean
Physical
and producible to the interviewer if consistent use is
Inspection
claimed outside the home.
22. The household contains a supply of unexpired sachets
for consistent use.
23. A dedicated clean cup is associated with the safe
storage unit.
Percentage of observations passed
= #Yes / (#Yes + #No) X 100%
Notes:
Maintenance
Water Quality Monitoring
Treated water is clear (Turbidity of <5 NTU)
Turbidity
Free available chlorine presence is shown if
Chlorine Residual
treatment is claimed.
Microbial testing shows <10 E.coli CFU/100 ml.
Microbial Testing
Notes:
Sample from
Storage of
Treated
Water
24 hr Colilert (Yes/No)
Yellow? Fluoresces?
24 hr Petrifilm (Count)
# Blue w/gas
# w/gas
( Yes/No/ NA)
# E.coli/
100ml
Risk
Level
Incubate Colilert and Petrifilm at body temperature (35°C) for 24 hours (or until results appear), then check:
Colilert: If the water is clear:
<10 Total Coliform/100ml and <10 E.coli/100ml
If the water is yellow:
>10 Total Coliform/100ml
If the water is yellow and fluoresces: >10 Total Coliform/100ml and >10 E.coli/100ml
Petrifilm: # of colonies w/gas X 100= # of Total Coliform/100ml; # of Blue w/gas X 100= # of E.coli/100ml;
No Blue colonies with gas= <100 E.coli/100ml; No colonies with gas = <100 TotalColiform/100ml.
Risk Level: Low is <10 E.coli /100ml; Intermediate is 10-100 E.coli /100ml; High is >100 E.coli /100ml.
Sampling 1. Test for presence of chlorine residual in stored water while at household if chlorine
treatment is claimed.
Procedure
2. Take a sample of treated water from the storage unit for microbial analysis. Use a Sodium
Thiosulphate sampling bag if transporting sample to laboratory. Keep the sample out of the
sun and start microbial test within 6 hours.
206
Appendix F: Usage Instructions per Technology
Sodium Hypochlorite Solution
Aquatabs
SODIS
Cloth Filter
Ceramic Pot Filter
Biosand Filter
PUR
207
Sodium Hypochlorite Solution Usage Instructions
PSI, Kenya. Printed in Kiswahili, this label is well suited to Kenya’s highly literate
population.
(POUZN, 2007)
PSI, Madagascar. This label is well suited for both literate and non-literate users.
208
(POUZN, 2007)
PSI, Ethiopia. 150ml WahuAgar (Waterguard) product, Printed in Amharic without
pictorial representations (Left).
CAWST Disinfection Black English 27 maintenance poster as handed out to users (right)
209
PSI Guinea Educational Materials (Lantagne, 2008)
210
Aquatabs Usage Instructions
From Aquatabs Sleeve of Precision dx Ltd., Accra, Ghana. All information is printed
for every two tabs on the ten-tab sleeve:
NaDCC 67mg
Use one tab to treat 20 litres of clear water in a jerrycan. If the water is dirty, filter it first
with cloth then treat with two tabs. Close your jerrycan and wait for 30 minutes before
use. Do not swallow the tablet. Medentech, Ireland. Distributed by Precision dx Ltd.
On the reverse side, bin number and expiration date are listed.
Additional information from Medentech Website:
How do I Use Aquatabs?
• The tablet is added to the appropriate volume of water. Wait at least 30 minutes
before using the water. The tablets do not need to be crushed; they will selfdissolve to give clear solutions.
• No stirring or shaking is necessary for the smaller, strip-packed tablets.
• For larger volumes of water (200 litres and above) the water should be mixed, for
example by re-circulation, to ensure an even distribution of the chlorine.
• Where the water is very turbid, for example greater than approximately 80 NTU,
then it should be filtered to reduce the turbidity before adding the Aquatabs
211
from Medentech’s AquatabTechnical Report 06:
212
From Muriuki, G. (2007):
213
SODIS Usage Instructions
From SODIS Tech Note 13
Exposing Procedure
• Fill the bottles completely with raw water
• Screw the plug tightly
• Expose the bottles in the morning hours to sunlight on a place which is irradiated the
full day
• Place the bottles in horizontal position on a firm support, preferably on a corrugated
iron sheet/roof or on a tile roof
• Collect the bottles in the late afternoon and bring them to a safe place for cooling
• Consume the treated water directly from the bottle using a clean glass or a cup, store it
possibly overnight for additional cooling
Additional Prescriptions
• Use clean water free of settleable solids and of a low turbidity (maximum turbidity 30
NTU). Separate coarse and settleable solids by storing the raw water for one day and
reduce turbidity possibly by flocculation/sedimentation using alum sulfate or crushed
Moringa oleifera seeds or by filtration.
• Use aerated water. Standing water with a low dissolved oxygen concentration should be
aerated by shaking the containers or before filling the containers.
• Expose the water for one day. Should the sky be covered with clouds, expose the water
for two consecutive days before consuming it.
• Collect rain water from a clean area (e.g. from a corrugated or tile roof) during rainy
days to cover your drinking water demand.
From SODIS Manual (Meierhofer, 2002)
Application Procedure
Preparation
1. Check if the climate and weather conditions are suitable for SODIS.
2. Collect plastic PET-bottles of 1-2 litre volume. At least 2 bottles for each member of
the family should be exposed to the sun while the other 2 bottles are ready for
consumption. Each family member therefore requires 4 plastic bottles for SODIS.
3. Check the water tightness of the bottles, including the condition of the screw cap.
4. Choose a suitable underground for exposing the bottle, for example a CGI (corrugated
iron) sheet.
5. Check if the water is clear enough for SODIS (turbidity <30 NTU). Water with a
higher turbidity needs to be pretreated before SODIS can be applied.
6. At least two members of the family should be trained in the SODIS application.
7. A specific person should be responsible for exposing the SODIS bottles to the sun.
8. Replace old and scratched bottles.
From KWAHO (2005) Promotional Poster
How to use SODIS
1. Wash the bottle well before the first time you use it.
214
2. Use the cleanest water you can get. If your water is dirty leave it in your bucket
for some time to settle it down. Use a clean cup to fill your SODIS-bottle and
leave the residue at the bottom.
3. Fill the bottle ¾ full with water.
4. Shake the bottle for 20 seconds.
5. Fill up your SODIS-bottle completely with water and close it. Only a small air
bubble should be seen after turning around the bottle.
6. Lay down your SODIS-bottle in the sun, e.g. on your roof.
7. Leave your SODIS-bottle for at least 6 hours from morning till evening in the sun.
If it is cloudy, expose your SODIS-bottle at least 2 days to the sun.
8. The water is now ready for drinking.
Keep your SODIS-bottle clean. Replace your bottle when it got too many scratches and is
not clear any more.
Sun Water GRI framework:
To use the sun to purify water, put water in clean and clear or slightly blue plastic or
glass bottles with tops. Remove bottles. Use 1 or 2 liter bottles.
• Leave one inch of air at the top of each bottle. Shake for 15 seconds.
• The water cannot be too cloudy. Large printed letters should be visible through
the bottle. If necessary, filter water through clean sand or several layers of cloth
before putting it in bottles.
Place the bottles outdoors in the sun for 5 hours or for a full day in cloudy weather. Put
them in a clean place away from animals and not in the shade. It is best if the bottles are
inclined so they receive the most sunlight and placed on a black surface in order to warm
them.
This process produces clean drinking water that is safe from bacteria and viruses.
However, clean water is not medicine. It will not protect you if you also drink unclean
water.
This process does NOT remove chemicals, pesticides, worms, or cysts.
215
Ceramic Pot Filter Usage Instructions
Instructions for Use of Ceramic filter (PFP, 2007) from (Swanton, 2008)
216
Pure Home Water Usage Poster (PHW, 2008)
217
The Instructional Sheet or Sticker pictorially shows users how to assemble, set-up and
operate and maintain their Kosim filter. It also shows a STOP line, indicating to users the
acceptable level of water in the safe storage container. The sticker is placed on the
storage unit such that the Stop line is below the bottom of the ceramic pot, preventing
back flow into the filter from overfilling. Finally the sticker shows “Do’s and Don’ts”
which are some of the common mistakes made by users in their operation and
maintenance of the Kosim filter and the manner by which to correct each error. The Tap
Installation Sheet shows the correct position of the washers and nut when installing the
tap. Additionally, each filter sold by Pure Home Water includes one Aquatab, a chlorine
tablet made by Medentech. This Aquatab is to be used in the first cleaning of the plastic
parts of the filter, as explained in the instructional sticker (PHW, 2008).
B
218
Biosand Filter Usage Instructions
Samaritan’s Purse “Biosand Water Filter: User Instructions” (2001)
(As used by Kale Hewyet Church, Ethiopia)
1. ONLY pour water in the filter with the diffuser basin in place - failing to do this will
damage the filter.
2. ALWAYS use two buckets: one to pour in dirty water and one to collect filtered water.
If only one bucket is used, the dirty bucket will contaminate the filtered water.
3. NEVER attach anything to the tap, such as a longer pipe, a hose or a valve.
4. ALWAYS use filtered water for as many tasks as possible: drinking, cooking, cleaning
food, cleaning clothes, washing children, and feeding animals. Using the filter for all
your water needs will contribute to better health.
5. NEVER put bleach in the water before pouring it into the filter and NEVER pour
bleach directly into the filter - this will damage the filter.
6. ALWAYS pour the water into your filter SLOWLY.
7. NEVER move the filter once it has been installed - unless it is an emergency.
Moving the filter will cause water to come out more slowly. If moved, the filter must be
placed in a level position before using.
8. ALWAYS keep the lid on the filter when not in use.
9. DO NOT touch the tap of the filter unless cleaning it - keep animals and children
away.
10. MAINTENANCE:
a) CLEAN tap once each week with a diluted bleach solution or soap.
b) When the flow of water out of the filter becomes much, much slower than its
original flow—this will be a slight trickle, almost dripping rather than flowing in a
stream—it is time to maintain the sand. At this point there will be a visible, thick layer
on top of the sand either brown or green in color. To maintain the filter, put your hand or
a spoon in the filter and down into top 2-3cm of sand. Stir in a circle until the water
becomes dark and then scoop this water out with a cup. Continue to stir and scoop water
out until all the water is gone above the sand. Be careful not to scoop out sand. Add
water to the filter and repeat this process until the water is clear. Then level the sand,
replace the diffuser basin and pour water back into the filter. Do not take sand out of the
filter. Finally, always check the level of water above the sand once you are finished. it
should be 5cm or the second finger joint.
(Samaritan’s Purse, 2001)
Samaritan’s Purse recommended maintenance procedures
(As used by Kale Hewyet Church, Ethiopia. Taken from Appendix 6 of Earwaker, 2006)
Used during Phase 1, pilot implementation
• Remove the filter cover and diffuser plate
• Lower the water level in the filter by scooping out water from the top of the filter
with a small cup.
• Remove approximately 2.5-5cm of sand which should be discarded or washed and
reused.
219
•
Add water to the filter until it begins to drain. Sand should always be added to
water.
• Add fresh or washed sand such that the sand surface is 5cm below the water level.
• Level the surface of the sand.
• Replace the diffuser plate and lid.
• The diffuser plate should not touch the surface of the standing water.
(Dejachew, 2002)
Used during Phase 2, scale-up
• Remove the diffuser plate and lid.
• Put your hand or a spoon in the filter and down into top 2-3cm of sand. Stir in a
circle until the water becomes dark and then scoop this water out with a cup.
Continue to stir and scoop water out until all the water is gone above the sand. Be
careful not to scoop out the sand. Add water to the filter and repeat this process
until the water is clear.
• Level the sand, replace the diffuser basin and pour water back into the filter. Do
not take sand out of the filter. Finally, always check the level of water above the
sand once you are finished. It should be 5cm or the second finger joint.
• Pour water into the filter until it begins to drain.
• The diffuser plate should not touch the surface of the water.
(Samaritan's Purse, 2001)
220
Posters distributed by the Kale Hewyet Church for their work in Ethiopia.
(Taken from Appendix 3 of Earwaker, 2006).
A plastic safe storage container was included in the update of this poster (not shown) as
handed out among their second intervention to ~10,000 households.
221
CAWST Latin America English 21 poster, similar to that distributed to users by the
International Aid implementation in partnership with Watersites Int., Tamale, Ghana.
From:
CAWST “Installation Operation & Maintenance Manual: Biosand Water Filter.”
Version 2007-01
DAILY USE
Educate all of the users, including children, on how and why the filter works and on the
correct operation and maintenance. Children are frequently the main users of the filter.
• Slowly pour raw (untreated) water into the filter daily (at least 20 litres, twice per day)
• Using the same source of water every day will improve the filter effectiveness
• Use the best source of water (least contaminated) available – the better the raw water is,
the better the treated water will be
• Pre-filter or settle raw water if not relatively clear – less than 50 NTU
Tip: A simple test to measure the turbidity is to fill a 2 litre clear plastic soft drink bottle
with raw water. Place the bottle on top of large print such as the CAWST logo on this
manual. If you can see the logo, the water probably has a turbidity of less than 50 NTU.
• The diffuser must always be in place when pouring water into the filter – never pour
water directly onto the sand layer
• The lid should always be kept on the filter
• Use a designated bucket for fetching raw water
• Use a designated safe storage container to hold the treated water which has:
222
○ a small opening to prevent recontamination due to dipping with cups or hands
○ a tap or spigot
• Place the receiving container as close to the spout as possible (i.e. place it on a block) to
reduce dripping noise and prevent recontamination
Note: The dripping noise can be irritating. The closer you place the container to the
spout, the less dripping noise there is. A container with a small opening also reduces
dripping noise.
• Water must always be allowed to flow freely from the filter – never plug the spout or
connect a hose to it
Note: Plugging the spout could increase the water level in the filter, which could kill the
biolayer due to lack of oxygen. Putting a hose or other device on the spout can siphon
or drain the water in the filter, dropping the water level below the sand layer.
• No food should be stored inside the filter
Note: Some users want to store their food on the diffuser plate because it is a cool
location. The water in the top of the filter is contaminated, so it will contaminate the
food. Also, the food attracts insects to the filter.
• The treated water should be chlorinated after it passes through the filter to ensure the
highest quality of water and to prevent recontamination (1-5 drops/litre or up to 1
teaspoon/gallon)
How to Use and Take Care of The Biosand Filter
How to Use:
1. Use the filter daily - this will maintain the water level 5 cm above the sand (measured
during the pause period) and keep the bio- layer alive.
2. Ensure water quality is from the best possible source. Always use the same source if
possible. If water is very dirty allow the water to settle for 24 hours then pour the clear
water through a fine woven-cloth (folded many times).
3. Use two separate containers; one container should be used as a receiving container to
properly store and disinfect water from the filter, a second container should be used as a
source container to collect the water from the water source. Ensure both containers are
kept clean.
4. Typically, add between 1 to 5 drops of bleach for each litre (or up to 1 teaspoon per
gallon) to the empty receiving container - for example, if the container is 20 litres then
add at least 20 drops.
5. Remove the filter lid
6. Slowly pour contents of the source container into the filter, without letting the
sediments enter the filter, and then replace the lid. As the water fills the receiving
container, it mixes and reacts with the chlorine to treat any remaining bacteria.
7. Remove the filter lid.
8. When filtration is complete, cover receiving container.
9. Repeat process at least once a day.
10. Clean the spout daily.
11. Do not store food on the diffuser plate.
12. Keeps animals away from the spout and filtered water.
223
How to Take Care of:
• Location - Protected from the weather (dust & wind), birds, animal, mosquitoes
and insects. Placing the filter indoors is preferred.
• Level - Filter placed on a level spot- even floor, not slanted, no bumps.
• Leaks or Cracks - Drips of water or wet spots under the filter will indicate a leak
in the concrete box.
• Lid - Clean on the outside and inside; no rotting wood parts; tight fitting but not
sealed.
• Diffuser - Clean regularly; sand under diffuser should be level and smooth; rotten
wood should be replaced; diffuser should rest securely on the lip. This should be
approximately 5 cm ( 2”) above water level.
• Sand Level - The surface of the sand should be 5 cm (2”) below the water level.
Contact your technician to add (or remove) sand if this dimension is not correct;
the sand should be smooth and level.
• Spout - Clean daily; eliminate any direct human and animal contact with spout
and filtered water.
• Receiving Container - 5-10 cm (2” - 4”) – a small opening will prevent
contaminants from entering the container that now hold treated water. Sanitize the
container frequently (every second day) by washing it with soap and water or with
a chlorine cleaning solution. Ensure the container has a lid. Do not scoop water
out of receiving container. It is best to pour the water out.
• Flow Rate - Measure the outlet flow rate from the spout when filter reservoir has
just been filled with water; 0.6 litre/minute (100 seconds per liter)is the design
rate for the standard concrete filter; if the flow rate is less than about 0.3
litre/minute (1/3 quart/min), clean the sand in the filter by using the “swirl and
dump” technique.
Swirl and Dump
• Remove the lid to the filter; remove the diffuser
• “Swirl” your hand,(up to the first knuckle), or an appropriate tool, (2 cm deep),
around in the water at least 5 times. You will disturb the surface of the sand but
don’t mix the surface layer below the top 5 cm of sand. The water above the sand
will become dirty.
• Scoop out dirty water with small container (i.e. cup or cut open plastic pop bottle)
Avoid scooping out sand.
• Throw out dirty water outside the house in an appropriate location
• Repeat this until all the water has been removed from the filter
• Replace diffuser
• Add 20 litres or 5 gallons of water- replace lid
• Check flow rate
• Repeat if flow rate is still low
224
PURTM Usage information:
From Aquaya Standard Operating Procedure for the Deployment of Procter & Gamble’s
PURTM PuRifier of Water in Emergency Response Settings. http://www.psi.org/Puremergency-relief/resources/aquayaSOP.pdf Accessed May 22, 2008.
Supplies needed to use PUR:
PUR requires only a few simple tools that most target beneficiaries of disaster assistance
should have at their disposal:
• a scissor or knife to open the sachet,
• a spoon or other implement to stir the water,
• a cloth fabric to filter the treated water, and
• two vessels (i.e. buckets) with volume capacity of 10 liters or more – the first
to be used for the treatment process and the second to be used for storing
the treated water.
How to use PUR:
The treatment procedure is as follows:
1. Open a PUR sachet using a pair of scissors. Add the contents of the
sachet to a vessel containing 10 liters (2.5 gallons) of contaminated water.
One simple way to measure a 10 liter volume is to use a 2-liter bottle five
times. Extreme precision is unnecessary: if there are slightly more or less
than 10 liters, the treatment procedure will still be effective.
2. Stir the powder steadily and vigorously in the water for five minutes.
After adding the powder to the water, the water will become temporarily
colored, and after a minute or two, large particles or “floc” will begin to form,
with the water becoming clear in the process. At the end of five minutes, stop
stirring and let the floc settle to the bottom of the container. If the water is still
colored, it can be mixed again and left to rest for another few minutes.
3. Once the water looks clear, and the floc, or precipitated material, is at
the bottom of the bucket, filter the water through a clean cloth into a clean
storage container. The filter must be a cotton cloth that prevents the floc
particles from passing through.
4. Wait 20 minutes before drinking the water. This is an important step,
because it is during this time that remaining pathogenic bacteria are killed.
The water should be stored in a container with a lid if available to keep it safe
from recontamination.
- Roughly a single sachet per household per day for emergency use is an
appropriate amount for distribution. Two 12-sachet strips will treat 240 liters
of water, and should be sufficient to support a household for three weeks.
- It is important to stir the water vigorously for the floc to form properly. This
visual sign is the signal that the product is working properly. The floc will
form even if PUR is added to clear water.
- The floc from the water treatment process should be disposed of in the
latrine or on the ground away from children and animals.
- Water that is still colored or cloudy after treatment should not be drunk.
225
If floc accidentally gets into the treated water (by accidentally dropping the
filter cloth into the water, for example), then another cloth should be used to
refilter the treated water into a clean container.
- The chlorine in the water gradually disappears, and after 24 hours it will not
be present in sufficient concentration to remove microbes. It is important to
store and dispense drinking water so as to avoid recontaminated.
- Water from the storage container should always be dispensed into another
container, such as a cup or glass for drinking. Unwashed hands and utensils
should never be dipped into the treated water because this is how treated
water becomes re-contaminated. A solid cover on the treated water is preferred.
If a lid is not available, a large plate or a towel may be used.
- A simple test to determine whether the cloth is adequate is to use it to filter
the water. If the “floc” does not pass through the cloth then it is working
correctly. A cotton cloth works best and you should not be able to see
through the cloth. On the other hand, the cloth should not be so thick that it
takes a prohibitively long time to filter the water.
From PUR Usage Instructions in 4 Languages, from the PSI website:
1. Open a sachet using a pair of scissors.
2. Add the contents of the sachet to a clean mixing vessel containing 10 liters of water.
3. Agitate the powder vigorously in the water for 5 minutes. Be sure a vortex is created
when mixing. Then, let the water stand until it clarifies.
4. After adding the powder to the water, the water will become colored. The color
indicates that the product is working. When the process is finished, the water will be
crystal clear.
5. If you see the water is still colored, you can mix again and let it rest for another few
minutes.
6. Once the water looks clear, and the floc is at the bottom of the bucket, filter the water
through a clean cloth filter into a storage container and cover with a lid.
7. The filter must be a cotton cloth that prevents floc from passing through.
8. Wait 20 minutes before drinking the water.
9. Do not drink the water if it is colored or cloudy after treatment. If the floc accidentally
gets into the treated water, use another cloth to filter the floc out of the treated water. The
water is still good to drink.
10. The treated water should be preferably consumed within 24 hours after its
preparation. Water that is left over should be used for cooking, washing, watering
animals, or otherwise discarded.
11. Always dispense the water from storage container into another container, such as a
cup or glass for drinking.
12. Discard the floc from the water treatment process in the latrine, or on the ground
away from children and animals.
Do not ingest the powder; Maintain out of children’s reach.
Contents: Fe2(SO4)3: 352 mg Fe(III); Ca(OCl)2
Also translated into Spanish, French and Arabic
226
Appendix G:
PSI Participatory Hygiene and Sanitation
Training Materials for use with Waterguard
Sodium Hypochlorite Solution
Population Services International, Addis Ababa Office6
When dealing with emergency situations, training of correct dosing is important to
encourage product use and hygiene as well as to prevent improper dosing. Included in
Appendix G are the posters used by PSI Ethiopia in community trainings when supplying
WuhaAgar (Waterguard) in acute watery diarrhea (cholera) outbreaks, as paid for by
various international aid organizations.
The writing on the back of each poster is broken into three segments. The first section
concerns a story that highlights the subject of the picture. The second section lays out
specific questions which, if they are not brought to bear in the ensuing discussion among
the meeting’s participants, should be raised directly by the community educator leading
the discussion. The third section succinctly restates the main points of the poster and
discussion therein.
6
PHAST (Participatory Hygiene And Sanitation Training) posters provided by Henock Gezahegn of PSI
Ethiopia. Translation from Amharic to English done by Bete, a friend and student at UMass Boston and
recorded by Matt Stevenson on March 24, 2008. These pictures and stories are also translated into a
culturally relevant Somali version for PSI’s work in the Somali State (not included here).
227
228
Poster 1: “Basic Health Protection through Washing Hands”
Marta is a young woman who has completed the tenth grade in school, and is on her
annual summer visit to her rural home to spend time with family. As she serves lunch,
she brings water and soap for hand washing. Her mother tells her to leave her alone and
let her eat in peace. To this, Marta replies “Mamma, you must wash hands each time you
eat, cook, or use the bathroom to protect from diarrhea and germs.”
??? Questions:
1. What’s the main point?
2. Who was the main character?
3. What was her education trying to teach them?
4. In the story, what’s the cause of diarrhea?
5. When and where does she recommend washing?
!!! Main points/ideas:
*The main cause of diarrhea is bacteria in improperly cleaned water
*Use soap when washing, or soda ash, or whatever is available
*In order to protect, one must wash before:
eating
cooking
feeding children
using the bathroom
cleaning infant and kids
handling animals
229
Poster 2: “How to clean the water area”
Marta is trying to teach her mother, saying “Don’t leave the pot open because germs can
get in there. They’re invisible and they can cause lots of pain and diarrhea. You must
always keep the top on the pot.”
???
1. In your house, how can germs affect you from your water?
2. Can we see the germs and know if it is clean?
!!!
*You can not see the germs.
*You must always cover the pot.
*Keep a separate drinking glass.
230
Poster 3: How diarrhea seriously affects humans.
“All the families behind the fire.” (n.b., adage implies the danger to all humans)
While people were discussing the baby’s sickness, Marta brought up how serious the
diarrhea is. Adessa (the father) did not want to listen to Marta’s “diarrhea talk,” and was
angry with her. She replied “You may not think that it is serious, but kids with diarrhea
are losing water, and will not resist death.” Her mother’s reply was that she did not know
this, and wanted to know what she can do…? Marta: “Diarrhea comes from water and
food germs, so you must wash and cover pots for water.” Mom: “I can do this in the
future.”
???
1. How does diarrhea affect the body?
2. Have your kids had diarrhea? What was the cause/visible symptoms?
!!!
*Affected water’s germs and bacteria cause diarrhea.
*diarrhea is a very dangerous pain with which we can lose the minerals from the water.
*kids with diarrhea must go to the hospital
*there are many signs: loss of appetite, sunken eyes
231
Poster 4: “Wuha agar”
“After Shopping”
Marta told her father that she went to the store to buy WuhaAgar. Adessa (her father):
“What does Wuhaagar mean?” Marta, who works for Wuhaagar, explains that she brings
chlorine to people to put in the pot to protect from diarrhea and typhoid by killing germs.
Drinking this clean water prevents diarrhea. Adessa: “Maybe this is expensive?” Marta:
“Baba, it is easy and cheap. Wuhaagar especially protects the children under five years
old, which is important. Especially for treating river water, they can be treated.” Adessa:
“I did not know about it, let’s use it.”
???
1. What’s the benefit of Wuha’agar?
2. When we use Wuha’agar, which water can we use?
3. Whose health benefits from using Wuha’agar and why?
!!!
•
•
•
•
•
Wuha’agar is a chemical which cleans water
Wuha’agar kills bacteria and germs, typhoid and diarrhea, etc.
Wuha’agar is very cheap, easy to use.
Especially when taking water from the river, it is very useful.
Wuha’agar is very important for families, especially those with children under
five years of age
232
Poster 5: How to use Wuha’agar
“Family and Wuha’agar”
When Marta brings Wuha’agar to her family, they are originally surprised and do not
know what it is. Marta: “Do you know how to use it?” Mom: “Yes, you have to measure
it, because we already know its benefits.” Marta: “For 20 liters, one cap. Shake it, leave
it for 30 minutes. You can drink it from a cup after this. That is how to use it.” Dad:
“You have to be cautious, because children are not supposed to get chemicals.” Marta is
very pleased because they have learned and understand.
???
1. Discuss about the steps for using Wuha’agar
!!!
•
•
•
•
20L=1 cap
Shake it, leave it for 30 minutes. After this, you can transfer to other containers
Don’t touch containers with dirty hands!
Missing some rules here, as the line is long on the poster
233
Poster 6: Wuha’agar measurements
“Discussion between Martha and her Mother”
Mom asks: “If water is from anywhere, does it matter how much Wuha’agar we add?”
Marta: “One cap= 20L only, for water which is clean (pipe borne, for example). For river
water, must cloth filter, then use 2 capfuls of Wuha’agar.” Momma: “Wuha’agar is good,
it will keep us clean and healthy.”
???
How do measure Wuha’agar? What’s the difference when using tap and river water?
!!!
Tap = 1 cap/20L
Non-tap (“river”)= pre-filter, then 2 caps/20L
234
Fly UP