Sustaining health: the role of BRAC's community health volunteers in Bangladesh, Afghanistan and Uganda

Full text

Sustaining Health: The Role of BRAC’s
Community Health Volunteers in
Bangladesh, Afghanistan and Uganda
Laura Reichenbach
Shafiun Nahin Shimul
September 2011
Research Monograph Series No. 49
Research and Evaluation Division, BRAC, 75 Mohakhali, Dhaka 1212, Bangladesh
Telephone: 88-02-9881265, 8824180-7 (PABX) Fax: 88-02-8823542
Website: www.brac.net/research

ii
Copyright © 2011 BRAC
September 2011
Cover design
Sajedur Rahman
Printing and publication
Altamas Pasha
Design and Layout
Md. Akram Hossain
Published by:
BRAC
BRAC Centre
75 Mohakhali
Dhaka 1212, Bangladesh
Telephone: (88-02) 9881265, 8824180-87
Fax: (88-02) 8823542
Website: www.brac.net/research
BRAC/RED publishes research reports, scientific papers, monographs, working
papers, research compendium in Bangla (Nirjash), proceedings, manuals, and other
publications on subjects relating to poverty, social development and human rights,
health and nutrition, education, gender, environment, and governance.
Printed by BRAC Printers, 87-88 (old) 41 (new), Block C, Tongi Industrial Area, Gazipur, Bangladesh

iii
TABLE OF CONTENTS
Acronyms and definitions iv
Acknowledgements v
Executive summary vii
Introduction 1
Section 1: Understanding of community health worker models and financial costing
approaches: a literature review 5
Section 2: Description of BRAC and BRAC’s health programme 11
Section 3: Study design, methodology and data analysis 17
Section 4: Results 21
Multi-country study results 21
Study results for Bangladesh 23
Study results for BRAC Afghanistan 45
Study results for BRAC Uganda 55
Section 5: Discussion and conclusion 75
References 82
Appendix 1 – Detailed breakdown of programme costs 84
Appendix 2 – Survey Instrument 86

i
v
ACRONYMS AND DEFINITIONS
ARI Acute Respiratory Infections
BDP BRAC Development Programme
BHP BRAC Health Programme
DOTS Directly Observed Treatment – Short Course
CFPR-TUP Challenging the Frontiers of Poverty Reduction – Targeting the Ultra
Poor
CHP Community Health Promoter (The term used in BRAC Uganda
Health Programme)
CHW Community Health Worker (The term used in BRAC Afghanistan Health
Programme)
EHC Essential Health Care
EPI Expanded Programme on Immunization
MNCH Maternal, Neonatal and Child Health
PHC Primary Health Care
PPP Purchasing Power Parity
SK Shasthya Kormi – (Female, salaried supervisor of the SS)
SS Shasthya Shebika (The term used in BRAC Bangladesh Health
Programme)
VO Village Organization

v
ACKNOWLEDGEMENTS
This study would not have been possible without generous support of the Rockefeller
Foundation and the patience, support and guidance of a number of people in
Bangladesh, Afghanistan, Uganda, and the United States. I want to thank Dr.
Mushtaque Chowdhury who requested my involvement in this study and for his early
guidance refining its scope. I am especially grateful to Mr. Faruque Ahmed, Director
of BRAC Health Programme and Mr. Jalaluddin Ahmed, Associate Director of BRAC
International for their time, inputs, and above all, patience throughout all phases of
the study. I am also very grateful to Late Mr. Aminul Alam and Dr. Imran Matin of
BRAC International for their intellectual and logistical support to make this multi-
country study possible. Dr. Syed Masud Ahmed of BRAC Research and Evaluation
Division (RED) was always available and willing to provide feedback throughout the
study. The RED staff did a wonderful job entering survey data under tight deadline.
Mr. SN Kairy, Group Chief Finance Officer, BRAC and BRAC International and Mr.
Mrinal Kanti Biswas of Accounts Division were consistently helpful. Dr. Md. Ariful
Alam, Programme Coordinator (Training), BHP; Abdus Salam Sarker, Senior
Regional Manager (EHC), BHP; Tapan Kumar Ghosh, Senior Health Coordinator
(EHC), BHP; Md. Rezaul Haque, Sector Specialist (EHC), BHP; and field level staff of
BRAC Health Programme in different sites provided important inputs into the study.
The support of the staff of the country offices in Uganda and Afghanistan was
immeasurable. In the Uganda office I would like to thank Khondoker Ariful Islam,
Country Programme Head, BRAC Uganda, Habib, Programme Coordinator (Health),
Abebual Zerahin, Proloy, and all of the staff of the Research and Evaluation Unit for
their hard work collecting and analyzing data. In Afghanistan, I would like to thank
Md. Fazlul Hoque, Country Programme Head, BRAC Afghanistan; Md. Taufiqur
Rahman, Programme Coordinator (Health), Afghanistan; Sher Shah Amin, Deputy
Programme Manager, Health Programme, Afghanistan; Md. Hanif, Health
Programme, Afghanistan; Nayeem Mujaddidi, Research Associate, Health
Programme, Afghanistan; and field level staff of BRAC Health Programme in different
sites.
This study would not have been possible without the hard work and intellectual
contributions of Shafiun Nahin Shimul, Lecturer, Institute of Health Economics,
University of Dhaka. He assisted in questionnaire development; oversaw the training
and provided field supervision of researchers in Bangladesh and Afghanistan;
collected and analyzed all of the financial cost data on BRAC Health Programme and
the Shasthya Shebika programme in Bangladesh and Afghanistan. He interviewed
programme staff in Afghanistan and carried out important analyses on the data sets.
His flexibility and hard work were critical to the successful completion of this study.

v
i
I thank Dr. Richard Cash for his useful comments and practical suggestions on
several drafts of the report. I am extremely grateful to Susan Davis, CEO, BRAC
USA, who has been immeasurably patient and understanding with unforeseen delays
in the study. I appreciate her support and careful comments on earlier drafts of this
report.
I have tried to address the range of comments and suggestions received on several
earlier drafts of this report. A study of this size and scope undoubtedly omits a range
of data and analyses that could have improved the overall report. We accept full
responsibility for these omissions.

v
ii
EXECUTIVE SUMMARY
This study fills an important gap in current understanding about a critical aspect of
BRAC’s health programmes – the financial sustainability of the community health
volunteers (commonly referred to as Shasthya Shebikas in Bangladesh) that are the
cornerstone of BRAC’s health programmes. Shasthya Shebikas (SS) are a cadre of
female volunteers that are recruited and trained by BRAC to provide a range of
essential healthcare services to their communities. What is unique about BRAC’s
approach is that, while these women can be considered volunteers they do not
receive a salary or monthly stipend, they are provided with financial incentives on the
sale of basic medicines and selected health commodities to their community. This
sets BRAC apart from other health programmes that rely on either entirely salaried or
volunteer cadres of community health workers (CHW) and raises important questions
about the financial and programmatic sustainability and replicability of BRAC’s
approach.
CHWs like the SSs play a crucial role in terms of human resources for health in
Bangladesh and other countries. The 2007 Bangladesh Health Watch reports a
shortage of 800,000 health workers in Bangladesh. Relying on formal institutions to
train health workers requires significant time and financial investment, therefore, the
SSs can be a critical and cost-effective input into the provision of essential health
services in Bangladesh. BRAC was among the first organization to set up a
community health volunteer (CHV) programme in Bangladesh in the 1970s. Its
original CHV programme recruited and trained male paramedics to treat minor illness
for which they received a small fee for referrals. Lessons from BRAC’s early CHV
experience included issues related to remuneration, supervision and accountability.
BRAC addressed these issues by recruiting and training cadres of female health
volunteers. Since then, BRAC has adapted and revised the programme in response
to programmatic and community needs and has effectively scaled up the
programme from 1,080 SSs in 1990 to 80,000 SSs operating in 64 districts in
Bangladesh today. The SSs are an impressive force in terms of their numbers,
geographic coverage, and quick mobilization. The 80,000 SSs provide home visits to
18 million households every month (BRAC 2009).
The BRAC Health Programme operating model clearly relies on the successful
recruitment, training, and retention of female CHVs. The rationale for BRAC’s
approach is that community-based financial incentives of a volunteer community
health workforce can achieve wide programme coverage and respond to community
essential healthcare needs while providing income opportunities to its female
volunteers.

v
iii
This study addressed two main research questions:
1) Is the Shasthya Shebika approach of the BRAC Health Programme a financially
viable model for the SSs themselves as well as for BRAC?; and
2) Is this approach generalizable to other countries?
In addition to review of the published and unpublished literature and relevant
programme documents on CHWs internationally and in Bangladesh, the study
collected primary data related to specific aspects of the SS experience – e.g.,
monthly income, incentives, procurement of supplies, and competition. This data
was collected through a structured questionnaire administered to a sample of 270
SSs in Bangladesh, 210 CHWs in Afghanistan, and 158 Community Health
Promoters (CHP) in Uganda. Survey data were analyzed using summary statistics
and cross tabulations to provide descriptive information to inform questions of
sustainability and replicability of the programme. While this study was conducted in
three countries, it is not intended as a direct comparison between the three health
programmes. The underlying programme context, age, scale and scope are so
different it makes any direct comparison impossible.
Major programme-relevant findings
Review of BRAC health programme’s revenues and expenditures between 2007
and 2010 suggest that in some areas it is successfully self-financed. (e.g., BRAC
Brace and Limb Centre) and other programmes, such as the Essential Health Care
(EHC) Programme are partially financed by BRAC. However, the majority of revenue
sources remain donor grants (86% in 2007). In Uganda and Afghanistan, both more
recent programmes, the health programmes are entirely donor-financed.
BRAC health programme model relies on the SSs; ensuring their financial
sustainability which will reduce dropout rates will help ensure programmatic
sustainability. The unit cost to BRAC of recruiting and training a shebika in
Bangladesh is US$ 89.48 (PPP-adjusted $245) in the first year and US$ 34.20 (PPP-
adjusted $94) in the second year. In Uganda, the costs are significantly higher and
estimated to be US$ 394 (PPP-adjusted $1204) in the first year and US$ 208 (PPP-
adjusted $636) in the second year. In BRAC Afghanistan programme the costs are
US$ 247 (PPP-adjusted $427) and US$ 84 (PPP-adjusted $145). Clearly there are
large cost differences between the programmes and these reflect the different pricing
and cost structures in different settings. Salary costs, for example, are much higher
in Uganda than Bangladesh. The Bangladesh programme is also much older and
more established and can benefit from economies of scale that the newer
programmes cannot. The wide variation in programme costs per shebika in different
settings also signals the need for flexibility – in terms of programmatic design and
financial adaptations when BRAC begins programmes in new settings.
SSs, CHPs, and CHWs are active. In all countries in the study, BRAC’s CHWs are
quite active. The average number of hours SSs in Bangladesh reported working each

ix
day was 3.6 hours and SSs reported that, on average, they could visit 14
households a day. In Uganda, CHPs reported working 3.2 hours per day visiting an
average of 9 households per day. In Afghanistan, women reported working 3.6 hours
per day and visiting 5 households a day. The fewer number of households visited in
Afghanistan likely reflects the increased distance between homes and the geographic
and cultural challenges associated with travel for women in Afghanistan.
Replenish inventory between refresher trainings. Eighty percent of the SSs in
Bangladesh and 85% in Uganda replenished their product supply between refresher
training which suggests that they are able to move at least some of their inventory. It
could, however, also suggest that SSs do not have income to purchase and hold a
lot of inventory each month. It is financially more convenient for them to buy fewer
supplies more frequently. Interestingly, 10% of the SSs in Bangladesh said that they
purchased medicines or supplies outside of BRAC to use in their work suggesting
that there is community demand for specific products. This may be an area that the
programme wants to examine more closely as it has implications for inventory
management.
There was a wide range of reported monthly income between the three country
study sites. For the SSs in Bangladesh, the reported mean monthly income in the
last month was Tk. 360 (PPP $14.07) and mean income per average month was
slightly higher at Tk. 374 (PP $14.62). Almost all women reported monthly
fluctuations in sales. In Uganda, the average monthly income was UX 38,222 (PPP
$58.40) and in Afghanistan it was Af 143 (PPP $4.94). Because of differences in cost
of living and general economic situations, comparing income between such different
settings is not possible. For example, the relatively lower monthly income reported in
Afghanistan likely reflects that the Ministry of Public Health in Afghanistan does not
allow CHWs to sell medicines, thus reducing their income potential.
Characteristics of high performing SS. Basic statistical analysis of survey results
suggests that high performers in Bangladesh were more likely to have a current loan
with BRAC than low performers. The correlation between length of time working with
BRAC and SS’s monthly income is positive and statistically significant, which
suggests that those SSs who have worked with BRAC for a longer period are more
likely to earn more than those working with BRAC for a shorter time. The fact that
high performers are more likely to have received health-related training outside BRAC
is a potentially interesting result. Rahman and Tasneem (2008) also found an
association between SSs’ income and additional training from outside BRAC. This
may be a useful additional selection criterion for recruiting SSs.
Expanding product mix. In the surveys in Bangladesh and Afghanistan, 30% of SSs
and CHWs said that they wished they could sell additional health commodities,
medicines and/or non-health commodities. In Uganda, 67% of CHPs wished to sell
other commodities and medicines. This is an important programmatic question as it
has implications for procurement, inventory management and pricing. It also raises
questions about how much the programme will respond to community demands for
non-health commodities such as school supplies for children.

x
The self-reported motivations for becoming an SS were primarily related to
financial incentives in all settings except Afghanistan. Non-financial incentives were
reported less frequently than financial incentives in Bangladesh and Uganda
suggesting that women see their role as an SS as income-earning opportunity. Many
SSs reported that their monthly earnings made a financial difference to their
household. Seventy-five percent reported that their SS income makes a big
difference to household income. Seven percent of the SSs in Bangladesh reported of
‘obtaining financial independence’ as an initial motivation for becoming an SS, yet
97% reported that being an SS had given them financial independence. In Uganda,
80% of the CHPs reported that being a CHP has given her financial independence.
In Afghanistan, 15% of the women reported their working as a CHW gave her
financial independence.
Utilization of microfinance is variable. Sixty-eight percent of the SSs in Bangladesh
said that they had ever borrowed a loan from BRAC and 34% had current loan with
BRAC. Thirteen percent had more than one loan with BRAC and 36% had loan with
another NGO. In Uganda it was 86%. In Afghanistan, 27% of the CHWs reported of
having loan from BRAC. This suggests that access to microfinance loan is not
necessarily an incentive for all SSs. This may be an area where the programme
needs to adapt some flexibility. It appears that not all SSs want or need access to
microfinance. Several CHPs in Uganda said that they did not want to take another
loan because they were having trouble repaying it while also managing to purchase
their CHP supplies on a monthly basis. In Afghanistan, the low numbers of CHWs
with current microfinance loans may reflect that in Afghanistan programme VO
membership is not a selection criteria for becoming a CHW.
Current VO membership is not universal and there is variation between study
countries. Forty-one percent of the SSs in Bangladesh reported that they were
current VO members despite this being a SS selection criterion (except in
Afghanistan). Thirty-six percent of the VO members had dropped out. VO
membership rates are much higher in Uganda (91%) and much lower (10%) in
Afghanistan. In Afghanistan, the programme is structured closely with the
government of Afghanistan (MoPH). Therefore, VO membership is not a selection
criterion as it is in other BRAC country programmes. Basic statistical analysis of
survey results suggest that high performers were more likely to currently have a loan
with BRAC than low performers.
These results may suggest the need to re-examine the role of VO membership in
recruiting SSs. We asked the respondents to describe how being a BRAC VO
member affects her work and productivity as an SS. In Bangladesh, 22% said that
people trusted her more because she was a VO member, and 13% said that it made
it easier for her to sell products. Only 2% reported that being a BRAC VO provided
her the necessary microfinance to purchase products to work as a SS and 30%
reported that it had no effect. In Uganda, 31% reported that people trust her more
because of her VO membership and 36% said that it made it easier to sell products
while 32% reported that VO membership had no effect. In Afghanistan, the reported
benefits of VO membership were slightly higher: 52% said that people trust her more

x
i
and 21% said that they needed loan to buy supplies – but the overall numbers of VO
members are lower in Afghanistan. However, when we compared high and low
performing SSs in Bangladesh, 51% of the high performers reported being VO
members as compared to 36% of the low performers. Furthermore, 96% of the high
performers replenished their supplies between refresher training as opposed to 73%
of the low performers.
Financial incentives
In terms of financial incentives the quantitative and qualitative data suggest several
things. Firstly, the SSs are financially motivated to sell medicines and health
commodities and the time use data suggest that they spent significant proportion of
their time in these activities. Secondly, the majority of SS/CHP/CHW would like to
expand the products they sell to include additional health and non-health products.
These new products include antibiotics, malarial treatments, ulcer treatments, and
non-health products like school supplies for children, sugar, and cooking oil. Clearly
there is a potential programmatic trade-off here between increasing the SS’s ability
to respond to community demand and in turn increase her sales and monthly
income, while ensuring the preventative and health education aspects of the
programme are sufficiently addressed.
SS comparative advantages
The SSs in each of the settings reported that they felt competition from pharmacies,
private clinics, and other providers that limit their income. It is important that the
comparative advantage of the SS be maximized to address this competition. Her first
comparative advantage is that she provides household delivery of healthcare and
products. By expanding the product range or mix of products that she provides to
households she could potentially increase her income and compete more effectively
in the local market. It appears from the qualitative and quantitative data that BRAC is
mostly meeting the demand for medicines. However, SSs report that there are a
range of health commodities and non-health products that the community is
demanding. There are two major trade-offs here – the first is the added programme
costs for procuring, storing, and transporting additional products to branch offices. If
some of these goods were produced by BRAC enterprises this could keep costs
down, at least in Bangladesh where production centres are functional throughout the
country. This may be more cumbersome in programmes like Uganda that are newer,
do not have BRAC enterprises established and face more challenging procurement
requirements. The second is the potential that as more products are added to her
inventory, the SS will spend more time selling products and less time on the
preventative aspects of her role as SS. Questions about whether or not SS are
becoming a sales workforce as opposed to a health workforce could be raised.
The SSs’ second comparative advantage is the BRAC name or ‘brand’. Ensuring
that BRAC products are responding to community demands and are seen as high
quality will be critical to SS sales performance. In each country some respondents
said that they had difficulties in selling products because people do not prefer BRAC

x
ii
products (29% in Afghanistan; 46% in Uganda; and 25% in Bangladesh). Performing
market research to better understand the perception of the BRAC brand could
improve the likelihood that SS can sell products and compete with other distribution
outlets. The fact that 10% of the SSs in Bangladesh reported buying products
outside BRAC for sale may be in response to a particular community demand for a
product. In order to capitalize on both of their comparative advantages (household
delivery and BRAC brand), additional training in social marketing could help SSs
maximize their monthly sales and income.
When asked if they ever considered quitting their work as an SS, 13% replied yes. Of
them, 83% had considered quitting their work because of too low payment, 6% said
because the work is difficult; and less than 1% attributed the reasons to better paid
job alternatives, the need to care for their children, and disapproval of husband or in-
laws. It is also interesting to note that SSs do perceive that they have other economic
opportunities available to them and yet they continue to work as SS.
Non-financial incentives
Non-financial incentives, while more difficult to measure, must also be addressed.
The survey results found that ‘increased social recognition’ is only somewhat
important. In Bangladesh, 18% of the SSs said that increased social recognition was
an advantage to working as an SS; in Uganda it was reported to be 11% and in
Afghanistan 16%. In Afghanistan, ‘social recognition’ was mentioned by 40% of the
CHWs as the reason for becoming a CHW. It is possible that the incentive of social
recognition wanes over time and as recognition is achieved. This may suggest that
additional non-financial incentives might be required to continue to recognize the SS.
These may include certificates for special training received, awards recognizing
extremely high sales in a period or of a particular product or activity, or providing a
saree or salwar-kameez in recognition of long-term service.
Another non-financial incentive is to offer additional specialized training for SS. In all
of the settings, many SS requested additional and specialized training to learn
additional health skills such as taking temperature and blood pressure and screening
for diabetes. Such training would increase her skill and potentially make her more
competitive in the community. The trade-off is that training add cascading costs to
the programme – from master trainers to supervisors – and not all SSs will be
interested in receiving additional training. In addition, if BRAC responds to the
request of some SSs for additional and more complicated health activities, it might
need to revisit the literacy requirement for recruiting SS.
An additional practical training in social marketing and inventory management can be
offered. As BRAC expands its product line in programmes like Uganda, those
women who have social marketing skills tend to be higher performers. Others, who
are unable to manage their inventory or cannot pay back their loan(s), will not be able
to perform up to their potential. These women may benefit from some training in
social marketing as well as basic financial and inventory management.

x
iii
Future research agenda
The results of this study suggest a rich research agenda for the future:
Better understanding of inventory management and its barriers. Eighty percent of
the SSs in Bangladesh replenished their product supply between refresher training.
Why are they choosing to replenish supplies between refresher training? Is it because
they do not have income available to purchase and hold sufficient inventory each
month and it is financially more convenient for them to buy fewer supplies more
frequently? Is it because they are responding to a particular household need or
demand? Further research on how SSs manage their inventory could help identify
slow or fast-moving items. Also, 10% of women in Bangladesh said that they
purchased medicines or supplies outside BRAC to use in their SS work. Further
examination of which items they purchase outside BRAC and why they do so is also
important for the programme to understand.
Competition and its effect on SS performance. The SSs in all the countries face
competition from pharmacies and public and private clinics. A careful study of how
such competition impacts her ability to sell medicine and commodities could provide
useful information for the programme. It will also be important to document the
community perspective of the medicines and commodities sold by the SS. For
example, are they perceived to be more expensive than drug shops?
Document and understand the community perspective. How does the community
perceive the role of the SS? How does the community perceive BRAC products? Are
community members skeptical of SS quality and ability? If so, what could be done to
increase community awareness about the SS and her role?
Role of VO membership in SS performance and productivity. How does being a
VO member help or hinder an SS in carrying out her activities?
General lessons learned and recommendations
While this report does not make direct comparison between the three country
programmes, when taken as a whole, this multi-country study provides useful
lessons learned and recommendations.
Generate and foster government support
Because of its long duration, the Bangladesh programme enjoys a history of
developing a strong relationship with the government. The fact that the BHP works in
partnership with the government of Bangladesh on several national programmes is
testament to this. As BRAC moves into new countries, establishing and negotiating
its relationships with the government and especially the ministry of health is extremely
important. Experience also suggests that this is an unpredictable and time
consuming process. BRAC has been very effective in building public-private
partnerships in both Uganda and Afghanistan. In Afghanistan the process was

x
i
v
guided by establishment of BSPH. In Uganda, however, BRAC had to develop these
relationships from the ground up. There are important lessons in terms of process
and resource allocation here that other country programmes can learn from.
Be willing to change the programme and be flexible
Country experience suggests that the SS programme is replicable in other countries
but with necessary adjustments to respond to the local environment. In Uganda, for
example, the number of households SSs are able to visit is being reconsidered
because the geographic distance and terrain is very different than in Bangladesh. In
Afghanistan, it was difficult for the CHWs to visit 120 households in a month. This
was due to both the geographic distance between homes; population clustering
around water sources; the conservative nature of the culture that makes travel for
women difficult; and ethnic differences within catchment areas making it difficult for
some women to visit the homes of other ethnic groups/identities.
Have a clear assessment of SS performance and expectations
It was not always clear how the programmes defined high and low performing SS.
Making performance assessments more transparent may help the SS/CHP/CHW in
setting individual performance goals. There is also a question of how the
performance of SSs/CHPs/CHWs should be assessed. Monthly sale of products
may be too narrow a measure and neglect a host of other activities and benefits
(e.g., health education) that the SS provides her community.
Do not lose sight of non-financial incentives
As the programme aims to be financially sustainable, it should not do so at the
neglect of the range of services that SS provides her community, many of which are
not immediately financially remunerated. While it may be possible to model the
appropriate product mix and profit margin to make the programme financially
sustainable, there needs to be continued and careful tending to the other aspects of
how the SS spends her time. There is a potential tension between maximizing the
sales of medicines and health commodities versus providing basic health education
and preventative care in the community. It may be important in future to think about
ways to incentivize general preventative health interventions. One approach might be
to reward SS for overall improvements in basic health social indicators in her
catchment area. These could be measured using BRAC’s existing MIS system and
measured on an annual basis to identify trends. The SS is not just a medicine seller,
she promotes the overall health and well-being of her community. Finding ways to
reward her for overall health improvements in her community would incentivize her to
focus on preventative activities, such as health education, that are currently not
financially incentivized.
Empower SS to manage inventory appropriately
Ensuring regular and affordable supplies to the SS is critical to her success. In
Bangladesh this system has been developed over years and is efficient and well-

xv
functioning. In Uganda, the programme started by giving each CHP a loan for her
first set of supplies. She was expected to pay this back in 40 installments. For some
women, this has not been a problem. These women may be better off or may simply
be better managers of their money or have higher sales and profits from which to pay
back their loans. Other women, and especially those with more than one loan, may
fall behind on repayments for their CHP supply loan creating a situation where they
are unable to resupply and, therefore, cannot earn the money required to repay the
loan. This vicious cycle makes her unable to function properly as a CHP.
Re-examine importance of microfinance as an incentive
SSs are to be recruited from the VOs and are provided access to microfinance loans
to support not only their work as an SS but to other economic opportunities as well.
The data from this study suggest that not all the SSs take advantage of their access
to microcredit. Nineteen percent of SSs in Bangladesh; 4% of CHWs in Afghanistan;
and 60% of CHPs in Uganda reported of using microfinance loans to support their
work. It is thought that SS may perform better with the support of her VO and the
social capital that comes with VO membership. The fact that so many SSs are not
active VO members raises a critical question of how important SS links with VO are.
When asked how being a VO member affects their role as SS/CHW/CHP 30% of
SSs, 32% of CHPs, and 21% of CHWs said that being a VO member had no effect.
In Bangladesh 22% of the women said that being a VO member increased
community trust in their work and 13% said that it made it easier to sell products.
Further exploration of how important VO membership is for the recruitment and
retention of SS will be important for BRAC to consider.
Describe notions and understanding of ‘volunteerism’
The SSs are a volunteer cadre but are also financially incentivized by the creation of
community demand for drugs, commodities, and basic health services. In Uganda,
CHPs expressed concern that their community did not understand that they were
volunteers. The community thought they were receiving a monthly salary from BRAC
(which they got when they went to the refresher training) and that they were selling
drugs to make additional profit. Several CHPs said that if the community understood
better that they were volunteers it would make their work as CHP easier.
Misunderstanding or misperceptions of what it means to work and serve as a
volunteer did come up in discussions with SSs in all three settings. Furthermore,
understanding of what it means to be a volunteer is culturally bounded and
influenced by local context.
Generalizability of BRAC SS approach
The quantitative, qualitative, and participant observation data from this study suggest
that the BRAC Bangladesh SS approach is generalizable to other settings – or at
least to the contexts of Uganda and Afghanistan. These programmes have, however,
had to explicitly adapt the model to reflect the local context. There are some useful
lessons in these examples for BRAC as it branches into other countries. The

x
vi
adaptations these country programmes have made are in three main areas: 1) health
policy and health systems; 2) socio-cultural environment; and 3) management and
logistics. It may be useful to generate a checklist of major issues in these areas that
other programmes have had to adapt or are considering adapting.
This study helps to document the unique and fundamental role that SSs play in
BRAC’s Health Programme and their community. The results suggest that long-term
financial sustainability is possible but complicated by ensuring that all the
components of the SS role are maintained. The SSs are a mixed motive cadre – not
entirely volunteers, but not salaried either. The data from this study suggest that it is
a combination of financial and non-financial factors that motivate a SS to continue
her service to BRAC and her community.

1
INTRODUCTION
This study was carried out between December 2008 and September 2009 in
response to a grant from the Rockefeller Foundation to BRAC USA (Grant
#2008/018).
This study fills a gap in current understanding about a critical aspect of BRAC’s
health programmes – the financial sustainability of the community health volunteers
(commonly referred to as Shasthya Shebikas in Bangladesh) that are the cornerstone
of BRAC’s health programmes. Shasthya Shebikas (SS) are a cadre of female
volunteers that are recruited and trained by BRAC to provide a range of essential
health care (EHC) services to their communities. What is unique about this approach
is that while these women are considered as volunteers they do not receive salary or
monthly stipend. They are not entirely volunteers because they are provided financial
incentives to sell basic medicines and selected health commodities to their
community. This sets BRAC apart from other health programmes that rely on either
salaried or volunteer cadres of community health workers (CHW) and raises
important questions about financial and programmatic sustainability and replicability
of BRAC’s approach.
While there have been studies and reviews of the BRAC Health Programme (BHP),
there have been few detailed analysis of BRAC’s community health volunteer (CHV)
approach from a business or financial perspective (Rahman and Tasneem 2008). Nor
have there been studies that specifically address the question of the long-term
financial sustainability of the CHV approach from the perspectives of both BHP and
the SS herself.
Financial and programmatic sustainability are important issues for BHP which
provides a wide range of health services and programmes all over Bangladesh.
Some of these programmes like the malaria programme and Water, Sanitation and
Hygiene (WASH) programme rely primarily on donor funds. Other programmes, such
as EHC programme, are partially self-sustaining. BRAC is committed to improving
the financial sustainability of its EHC programme. BRAC’s commitment to
sustainability is a driving force behind this study and it is hoped that these results will
inform the financial sustainability of the SS approach and, therefore, protect BHP
from potential fluctuations or reductions in donor funding.
This study also addresses the issue of the replicability or transferability of BRAC
Bangladesh’s CHV approach to other settings – specifically to Afghanistan and
Uganda. This is especially timely as BRAC International programmes expand in scale
and scope. The question of how well the SS model, developed in Bangladesh
decades ago, transfers into settings with different health systems and regulatory,

2
policy and legal environments is critical. Afghanistan and Uganda are useful
examples in which to examine questions of transferability. They provide regional
diversity and both are settings which may yield valuable lessons as BRAC becomes
increasingly involved in other post-conflict areas such as Sudan, Sierra Leone and
Liberia. This is a unique study for BRAC as it is one of the first multi-country
comparative studies that BRAC has undertaken.
While this study provides useful information for BRAC Bangladesh, BRAC
International programmes, and BRAC USA, it also provides valuable lessons and
insights for the broader public health community. The study results inform larger
questions related to the role of unpaid health volunteers in the provision of primary
healthcare services; the role of private providers and non-state sector entities in the
public health system; and overall questions related to the measurement of financial
and programme sustainability in community health programmes. The results are,
therefore, very relevant for and of potential interest to other NGOs and the
international donor community.
Project description
Study rationale and research questions
This study addresses two main research questions:
1) Is the Shasthya Shebika approach of the BRAC Health Programme (BHP) a
financially viable model for the Shasthya Shebikas themselves as well as for
BRAC?; and
2) Is this approach generalizable to other countries?
To answer these questions, the study addressed four research objectives:
1) To determine BHP’s financial operating model;
2) To describe how financially sustainable and viable the BHP has been for BRAC
as an organization and for the SSs themselves;
3) To describe the key strengths and weaknesses of the community health
volunteer model; and
4) To determine to what extent the model is replicable and sustainable in other
countries (i.e., Afghanistan and Uganda).

3
Study caveats
There are several qualifications to this study:
• Because BHP has a range of activities and programmes that rely on the SS, it
was necessary to limit the study to a manageable size and scope in the
Bangladesh study. This was done in consultative discussion with senior
management in BHP, BRAC RED, and BRAC International programmes and it
was decided to focus this study on those SS providing EHC services in rural
areas of Bangladesh.
• The study does not assess the health impact of BHP as this is beyond the scope
of the study. It is in no way an analysis or evaluation of BHP in any of the study
country sites.
• The units of analyses in this study are BRAC and the CHVs themselves. It does
not take into account the perspectives of community members or the patients of
the SS.
• BRAC’s CHWs are referred to by different terminology in each country study
sites; in Bangladesh they are referred to as SS, in Uganda they are called CHPs,
and in Afghanistan they are called CHVs. The report uses the term SS except
when referring specifically to Afghanistan or Uganda.
• This is a study of BRAC’s CHWs in three countries and similar research
instruments and methods were standardized to the degree it was possible.
However, direct comparisons between the countries are not possible (nor
intended) due to the significant variation between the programmes in terms of
age, size and scale, and country level factors.
Organization of the report
This report is organized into five sections. The first section describes some general
findings from a review of the international literature on CHWs and financial and
economic costing of CHW programmes; section two briefly describes BRAC and
BHP; the third section describes the study design and methodology; the fourth
section presents the study results for Bangladesh, Afghanistan, and Uganda; and the
fifth section discusses findings and conclusions in terms of sustainability and
generalizability.

5
Section
UNDERSTANDING OF COMMUNITY HEALTH WORKER
MODELS AND FINANCIAL COSTING APPROACHES: A
LITERATURE REVIEW
Since 1978 and the Alma Ata Conference on Primary Health Care (PHC) there has
been international interest in the role of community health workers (CHW) in the
provision of basic health services. In response, CHW programmes were established
in many developing countries. In the 1970s and 1980s CHWs were considered the
foundation for providing primary healthcare (Haines et al. 2007) and there was great
enthusiasm for their role. The focus on PHC addressed two underlying goals; to
address shortages of health professionals with a low cost alternative; and, more
fundamentally, to create political change agents in communities (Standing and
Chowdhury 2008). Interest in CHW programmes diminished somewhat in the 1990s
for several reasons including the challenges of scaling up CHW programmes in a
sustainable manner while still maintaining their effectiveness (Haines et al. 2007).
Other problems included institutional constraints such as “poor supervision; lack of
training and equipment; tensions between preventive and curative roles” (Standing
and Chowdhury 2008) and the view that vertical programmes were more successful
(Haines et al. 2007).
International interest in the potential role for CHWs in health programmes has been
rejuvenated in part because of concerns about limitations and constraints in human
resources for health, particularly in rural, hard-to-reach areas (Joint Learning Initiative
2004, Haines et al. 2007) and in light of WHO’s renewed focus on primary healthcare
(World Health Report 2008). Pressure to achieve the Millennium Development Goals
has also led to current interest in an increased role for CHWs (Haines et al. 2007).
A wide range of terms have been used to refer to CHWs in the international literature
and this has led to some confusion about their role and definition. They are
sometimes referred to as ‘community health workers’; ‘village health workers’; ‘lay
health workers’; ‘auxiliary health workers’; ‘community health volunteers’; and ‘health
promoters’ among other terms. Despite the different terminology, the description and
role of CHWs is generally similar. The WHO has defined community health workers
as those workers “who live in the community they serve, are selected by that
community, are accountable to the community they work within, receive a short,
defined training, and are not necessarily attached to any formal institution” (Swider
2002, p12). There are other models or understandings of CHWs. Standing and
1

6
Chowdhury (2008) usefully highlight four different models and roles for CHWs in the
future: 1) generic community health worker; 2) specialist CHW that are trained to
focus on a particular disease or set of diseases; 3) expert patient/advocate
approach; and 4) community mediator.
These workers, usually women, are often referred to as the ‘cornerstone’ of any
primary health system. They are usually provided some limited training but do not
have formal professional or paraprofessional certification. In some settings, they are
considered to be part of the formal health workforce but do not replace the need for
facility-based health services.
Arguments for CHW programmes
The main arguments for CHW programmes have been that they are a more
appropriate and cost-effective way to provide basic health services to under-served
populations compared to clinic-based services. In addition to being viewed as less
costly, CHW programmes have also been promoted as reducing social distance,
cultural and linguistic gaps and fostering self-reliance and local participation (Walker
and Jan 2005, Walt 1988). Additionally, they can provide a link between the formal
health system and the community (Bender and Pitkin 1987).
Assessments of the impact and effectiveness of CHW programmes have been mixed
but are generally positive. Lewin et al.’s (2008) systematic review of 48 studies of lay
health workers (LHW) concluded that inclusion of LHWs in health programmes
resulted in improvements in immunization rates, breastfeeding practices, morbidities
related to child illnesses, and TB treatment. Haines et al. (2007), in their evaluation of
CHW studies, conclude that, CHWs “…can reduce the costs of transport and lost
productivity for recipients associated with seeking health care, both of which can act
as barriers to utilization” (page 2125). CHWs can also improve quality of care. A
study in Bangladesh that compared the quality of care provided by health
professionals versus other care givers found that low-level community based workers
(family welfare visitors and nursing aides) did a better job in providing rational
prescription of antibiotics and advice compared to other caregivers (Haines et al.
2007). An additional benefit of using CHWs is in emergency situations – such as the
earthquake in Pakistan in 2005 where 8,000 CHWs were quickly mobilized to
respond to their local communities in earthquake affected areas.
While there is encouraging evidence that CHWs can increase access to care in a
potentially equitable and cost-effective manner, others have cautioned that the CHW
role can ‘be doomed by overly high expectations, lack of a clear focus, and lack of
documentation’ (Swider 2002, p19). One critical issue related to CHW programmes
is a tendency for high rates of attrition. A comprehensive review of CHW
programmes found attrition rates ranged between 3.2 to 77% with higher rates
usually associated with volunteers. CHWs who are financed by their community have
two times the rate of attrition as those who are government salaried (Bhattacharya et
al. 2001). An additional programme concern includes the tendency to overload the
CHW with additional responsibilities (Walley et al. 2008).

7
Compensation of CHWs
How to remunerate and compensate CHWs is a fundamental question and challenge
for CHW programmes and there are several country examples to draw upon. There
is a range of financial incentives including: offering a small honorarium; paying a small
monthly salary from the state; payments for attendance at training sessions; and
provision of cash incentives for drug sales. The source of payment for CHWs also
varies and may come from the community, NGOs, the government or for-profit
companies.
A review of incentives and disincentives for CHWs suggest that payment of CHWs
can bring unforeseen problems. The money can create distrust on the part of the
community about the CHWs underlying motives and the community may start to see
them as employees of the government. The money tends not to be considered
enough by many of the CHWs creating issues for programme management;
sustainability of monetary payment is challenging and projects often face irregularities
in payment or may have to phase out payment when a project ends; and
comparison of salaries and compensation can generate tensions between CHWs
and other health workers (Bhattacharrya et al. 2001).
Many programmes use incentives other than financial remuneration. These include
in-kind payments of food, housing, childcare or material items such as bags and
umbrellas as well as preferential treatment for CHWs and their families for credit or
literacy programmes and at health facilities. “Alonso and Hurtarte (2000) have found,
however, that incentives given too often or in too many forms are unsuccessful and
demotivating in the long term” (Bhattacharya et al. p18). Regardless of the type of
compensation that is provided, prior experience suggests that despite their short-
term positive effects, they may pose problems or issues in terms of long-term
sustainability. Bhattacharrya et al.’s (2001) review of CHW programmes finds that
“incentives were implemented ad hoc rather than as part of a systematic
programme…new incentives are often proposed in reaction to a crisis of low morale
rather than as part of an overall programme effort to maintain high morale.”
Furthermore, they recommend that “[p]rogrammes should consider systematic effort
to plan for multiple incentives over time to build CHW’s continuing sense of
satisfaction and fulfillment” (p36).
Pakistan, for example, pays its Lady Health Workers (LHW) a monthly salary and
considers them to be part of the government health system. The LHW programme
recruits and provides basic training to young married women who provide door-to-
door health services in their communities: these LHWs are rural women who serve
200 households in their community. They are given a 15-month training and they
provide maternal and child health services, general health promotion and education,
and referral to local health facilities. They collect routine health statistics in their
catchment area and must register all women and children in their area. While
Pakistan has roughly 90,000 LHWs, there is variation in regional LHW coverage.
About one-third of rural communities in Sindh, Punjab, and NWFP provinces have an
LHW whereas in Balochistan province only 10% of rural areas have an LHW (Mir and

8
Rashida 2007). The LHW programme has been in existence since 1994 and it has
been externally evaluated three times (at the time of this writing it was undergoing its
fourth evaluation). A thorough evaluation of the programme in 2002 found that the
LHW programme did have positive impacts in terms of health and fertility outcomes
including: increased childhood vaccination rates; increased uptake of antenatal
services; better provision of iron tablets to pregnant women; increased levels of child
growth monitoring; and lower rates of childhood diarrhoea (OPM 2002). The 2002
OPM review of the Pakistan LHW programme also highlights some challenges that
are relevant to other CHW programmes. These include: maintaining stock and
supplies (95% of LHWs in the survey had been out of ORS for more than three
months); supervision of LHW performance – particularly in rural areas; and receiving
salary on time (only 1/3 of LHWs had been paid in the last month and 1/3 had not
been paid for four or more months).
Iran, on the other hand, utilized an entirely volunteer cadre of women to successfully
provide family planning education and services. Other countries utilize a mix of
financial and non-financial incentives for CHWs. In Honduras, AIN-C monitors are
community-based volunteers motivated mostly by non-financial or non-material
incentives. They do however receive one time and annual incentives including free
ministry of health care; a carrying bag; a diploma; and a party at end of the year.
These workers were found to be cost-effective; scenario analyses found that if
monitors were paid the equivalent daily wage of agricultural day labourers the total
costs during the 6-year phase-in period would increase by 20% and long-term
annual recurrent costs would increase by 35%. Furthermore, the MoH provided
services are nine times more expensive than AIN-C services (Fielder et al. 2008).
A literature review supports the importance of non-financial incentives in managing
CHW programmes. Some suggest that “[n]on-financial approaches to improving
performance such as use of visual identification (badges, T-shirts, etc.), acquisition of
skills, and flexible hours, may have less potential to distort care than fee-for-service
payments or those associated with drug sales.” Concerns have been raised that
“fee-for-service payments or payments associated with drug sales may encourage
inappropriate treatment at the expense of prevention and overuse of medications”
(Haines et al. 2007, p2127). As a result, many recommend that policymakers
consider using a mix of financial and non-financial incentives tailored to local
circumstances.
Dieleman et al. (2003) describe the main factors that influence job motivation at the
commune and district health centres in rural areas of North VietNam. They found that
community recognition and appreciation were critical motivating factors in the
programme. They concluded that “salaries and working conditions are important to
retain staff, but alone are insufficient to lead to better staff performance; recognition
and feeling of achievement are more likely to influence staff motivation and therefore
their performance” (page 9 of 10). Nonmonetary incentives such as adequate training
and supervision and community support are also necessary for any successful CHW
programme.

9
Whether and how CHWs are remunerated also has important implications for
programmatic sustainability in terms of how programmes motivate CHWs and ensure
CHW accountability and performance. “Arguments for community support of CHWs
are that it fosters community participation in the health of their community and gives
the community a feeling of responsibility over meeting their own health needs” (page
519). Haines et al. (2007) review of financial incentives and remuneration for CHWs
suggests there are few examples of CHW programmes that are sustained solely by
community financing. “Even non-governmental organizations tend to find ways of
financially rewarding their community workers…Even when the workload is light and
can be fulfilled on a part-time basis, the costs entailed by lost economic
opportunities may be too high” (page 2127). There are fewer examples that describe
volunteers who receive financial incentives such as the BRAC Shebika approach.
Costing and cost-effectiveness of CHWs
Despite general agreement that CHWs play a potentially important role in the
provision of basic health services, there is surprisingly little data on their cost-
effectiveness. One reason for this is the challenge of examining CHWs in traditional
economic analyses. Non-financial values such as volunteerism and sense of duty or
social pressure are not easily reflected in conventional forms of economic analysis
(Walker and Jan 2005). The data on cost-effectiveness that does exist suggests that
CHWs can provide services in a cost-effective manner (e.g., primary healthcare
provision in Kenya). CHWs were found to improve immunization coverage because
houses were chosen with better precision and vaccination days were selected to suit
parents, thus improving uptake. Several studies suggest it is more cost-effective to
use CHWs than government programmes in the provision of TB care. A BRAC study
in Bangladesh found that the cost of BRAC CHWs was US$64 per patient cured
versus US$96 for the government programme (Hadi 2003).
The economic costing approach of CHW programmes raises two main issues:
whether the cost savings from using a CHW approach is simply shifting costs from
the health sector to the community; and whether the cost of volunteer labour is
calculated as a ‘shadow’ price that reflects prevailing wage rates. Traditional
approaches to economic evaluation usually treat volunteer and paid labour
interchangeably. “The problem with this approach is that notions of “volunteerism”
may be tied in with other institutional characteristics such as social capital and trust”
(page 227). An additional challenge is measuring time forgone in order to provide
informal care. The economic evaluations of CHWs that have been carried out tend to
focus on very narrow outcomes (e.g., vaccinations administered or patients treated)
which may neglect key community-based elements of the programme (Haines et al.
2007).

11
Section
DESCRIPTION OF BRAC AND BRAC’S HEALTH PROGRAMME
Background on BRAC
BRAC began in 1972 as a small relief and rehabilitation project in response to the
post-war challenges facing Bangladesh. Since then it has expanded its efforts
toward integrated sustainable development and poverty alleviation and is currently
one of the largest NGOs in the world. In Bangladesh, BRAC serves 110 million
people across 70,000 villages and 2,000 urban slums in each of the 64 districts of
Bangladesh. It operates using 117,000 staff and 80,000 community health
volunteers (CHV) in 4,500 offices (BRAC 2009). BRAC, through its international
programmes, has been expanding its efforts in some of the neediest countries in the
world, with a particular focus on post-conflict settings. In 2002 it started in
Afghanistan, its first programme outside Bangladesh. Since then, it has developed
programmes in Sri Lanka, Tanzania, Pakistan, Liberia, Uganda, Southern Sudan, and
Sierra Leone.
BRAC‘s overall approach and philosophy is one of integrated development to
achieve sustainable poverty reduction through education, health, water and
sanitation, and microfinance interventions. In addition to its multifaceted development
approach, BRAC emphasizes the organizational development of the poor and
capacity building at a national scale. BRAC has several core programmes: 1)
Economic Development Programme (EDP); 2) BRAC Education Programme (BEP); 3)
BRAC Health Programme (BHP); 4) Social Development, Human Rights and Legal
Services; and 5) Disaster Management, Environment, and Climate Change.
The BRAC development programme improves the livelihood of the poor and has
organized 7.2 million poor people into 254,673 VOs to foster economic opportunities
with microfinance services. VO-based credit schemes offer communities a range of
loan products including general loans (for any profitable income generating activity)
and programme loans (e.g., poultry or fisheries where BRAC provides technical
assistance and training in addition to the loan). BRAC’s microfinance is self-
sustaining and generates a surplus for the organization. BRAC’s microfinance
programme uses a ‘credit ladder’ approach that addresses the extreme poor,
moderately poor, and vulnerable non-poor. BRAC emphasizes poverty reduction
through targeting the ultra poor who are often neglected by other programmes and
development interventions. In this regard, in 2002 BRAC initiated the Challenging the
Frontiers of Poverty Reduction – Targeting the Ultra Poor (CFPR-TUP) programme
2

12
which targets the poorest 15% of the population, individuals too poor to qualify for
traditional microcredit services.
In addition to its broad range of social development programmes, BRAC has set up
a wide range of commercial enterprises that help to subsidize and support its
development and poverty alleviation programmes. Financial self-sustainability is a
strong underlying goal of BRAC. As a result, BRAC has become less donor
dependent over time. In 2008 BRAC’s annual expenditure was US$ 535 million of
which 73% was self-financed (BRAC 2009).
BRAC Health Programme (BHP)
BHP is a core programme and one of the original programmes of BRAC. In 1980s it
successfully administered the Oral Therapy Extension Programme (OTEP) throughout
the country to reduce deaths and disability related to diarrhoeal diseases. In 1990s it
expanded its focus to provide more comprehensive service-based programmes like
the Women’s Health and Development Programme. Currently BHP has a holistic set
of programmes with massive coverage based on its platform programme of EHC
which started in 1985. EHC grew out of the community-based distribution of ORS
and family planning supplies. By 1996 it included water and sanitation, immunization,
health and nutrition education, family planning, and basic curative services.
EHC remains the core programme of the BHP and serves as the foundation for
introduction of other health programmes such as WASH, tuberculosis (TB) control,
and maternal, neonatal and child Health (MNCH) programmes. Currently, EHC
includes seven components: health and nutrition education; water and sanitation;
family planning; immunization; pregnancy-related care; basic curative services; and
TB control. In some areas the EHC programme includes additional activities such as
MNCH and malaria treatment. The EHC programme works in 54 districts and
reaches over 92 million people; 86 million people are reached through its TB Control
Programme alone (BRAC 2009). The EHC programme collaborates with the
government in the implementation of national programmes such as vitamin A
supplementation and family planning. EHC is also provided for the ultra poor through
the TUP programme.
BRAC Health Programme approach
BHP is built on a village extension concept and a unique feature of BRAC’s health
programmes is its reliance on a cadre of female health volunteers, known as
Shasthya Shebikas (SS). BRAC recognized early on the importance of health
volunteers as a way to overcome critical human resource limitations in the health
sector – particular in rural and under-served areas. Shasthya Shebikas are provided
basic training to enable them to provide door-to-door health education, treatment of
basic health problems, collect health information, sell medicines and health
commodities, and make referrals to health centres as necessary.

13
BRAC was among the first organization to set up a CHV programme in Bangladesh
in the 1970s with four curative health clinics staffed by doctors and using locally
recruited men as paramedics. These men were high school graduates who were
trained to treat minor illnesses and make referrals for a small fee. In an attempt to
make it self-sustaining, a health insurance system was tried. The programme had
30% cost recovery from the insurance premium but was stopped after evaluations
found that the programme was not reaching the poor (Standing and Chowdhury
2008). Problems with BRAC’s early CHV programme included issues of
remuneration, supervision, and lack of accountability. As a result, BRAC revised the
programme to recruit and train a new cadre of female health volunteers. Females
were thought to be more effective as CHVs because they lived in the community and
the clients were mainly women.
In the mid-1970s in BRAC’s Sulla programme in Sylhet district, BRAC recruited and
trained local women to promote family planning services and provide health and
hygiene education (this was done even before the government programme)
(Standing and Chowdhury 2008). The Shebika programme was piloted in 1977 as
part of the Manikganj Integrated Development Project and since then, “… has grown
into one of the largest national-scale community health volunteer programmes in the
world” (BHP Annual Report 2007, p3). While BRAC had been training community
health volunteers since the late 1970s it did not scale them up until the 1990s. The
growth in scale of these community health volunteers was impressive: in 1990 there
were 1,080 SS in BHP and today there are 80,000 SS functioning in 64 districts all
over Bangladesh. The SS started with selling oral contraceptives and then BRAC
included other items based on community needs and experiences such as WHO
essential drugs, and soap or ash for hand washing. There are currently 21 basic
items that every SS offers for sale (see Table 6 for list of these items).
BRAC provides several underlying arguments for its EHC approach using the SSs.
One is an economic argument. The alternative to using SS in the provision of EHC
services is to recruit and hire graduates to market the same medicines and health
commodities. This would not only be much more expensive but recruiting them to
work in rural and underserved areas would likely be a challenge. The second
argument is increased programme coverage. Because the SS live in the communities
they serve it is easier to increase the geographic accessibility to essential health
services. The SS play a crucial role in terms of human resources for health in
Bangladesh. The 2007 Bangladesh Health Watch reports a shortage of 800,000
health workers in Bangladesh. Relying on formal institutions to train health workers
requires significant time and financial investment. BRAC believes that an appropriate
alternative given the time and resource limitations is to train CHWs within the
communities in which they live (RED staff, personal interview). The SSs are an
impressive force in terms of their numbers and geographic coverage and one that
can be mobilized quickly. The 80,000 SSs provide home visits to 18 million
households every month (BRAC 2009). As a result, millions of people can be reached
with health messages all over the country in a matter of hours if necessary (RED staff,
personal interview). A third argument is that it generates community participation and

14
ownership in the overall health of the community and in conjunction with BRAC’s
other development interventions.
The basis for the EHC programme is, therefore, cost recovery and sustainability,
while achieving the underlying goal of improving health equity and overall community
development. Using the SSs in the delivery of EHC is seen as an important
mechanism for filling gaps in remote and underserved areas. In addition, the SS
become agents for community development. In terms of their impact on health
outcomes there is sufficient evidence to suggest that SS have improved health
outcomes particularly for TB. One study found that TB prevalence was half the rate in
BRAC areas as compared to those areas where BRAC was not present (Chowdhury
et al. cited in Standing and Chowdhury 2008).
Initially the SSs were supposed to work completely voluntarily, but attrition was high
because they had to earn some income. In response to this attrition, BRAC decided
to provide the SSs with financial incentives. However, BRAC decided the financial
incentives should come from the community rather than from BRAC directly in order
to make it sustainable (Interview with programme staff, RED). Several programmes in
BHP are already or very close to being financially self-sustaining. For example, the
BRAC Limb and Brace Centres cover the bulk of their costs through cross-subsidies
and scholarships. BRAC approaches the introduction of new health programmes
with the issue of long-term financial sustainability in mind.
Who are Shasthya Shebikas?
In theory, the SSs are recruited from among VO members and are microfinance
borrowers. VO group members are asked to select a woman based on the following
criteria; 25-36 years old, youngest child is older than two years, not living near a
health facility or a bazaar, and willing to work 2-3 hours per day. BRAC then
assesses the nominees. Their catchment area should be reachable in a 15-20 minute
walk and she should not live near a health facility to avoid competition in the
provision of medicines. Table 1 summarizes the selection criteria. In non-EHC
programmes the SSs are recruited and trained based on the specific needs of the
programme and, therefore, do not always exactly match the profile of the EHC SS.
Table 1. Selection criteria for EHC Shasthya Shebikas
She is a current member of a BRAC village organization
She is married and her youngest child is not less than two years of age
She is 25 years or older
She is willing to provide voluntary services
She is socially acceptable to the village
Her family agrees to her involvement as an SS
She does not live near health facility or bazaar to avoid competition
SSs are expected to visit 250-300 households per month or approximately 10-30
homes per day. They provide basic health promotion and education; treatment of
common ailments; and sell key health-related items some of which are produced by

15
BRAC enterprises and some of which are procured by BRAC from the market. It is
the sale of these items in addition to referral fees that generates a monthly income for
these non-salaried volunteer women. Table 2 narrates their duties and
responsibilities.
Table 2. Shasthya Shebikas’ responsibilities and duties
Health promotion
and education
Treatment of common
ailments
Other medical duties Sale of drugs and
health commodities
Hygiene Fever Early diagnosis and
treatment of malaria
Paracetamol
Nutrition Common cold Provision of essential
newborn care
Vitamins
Family planning Anemia Identifying TB
suspects
Anti-histamines
Pregnancy-related
care
Peptic ulcer Referral for sputum
examination
Oral rehydration saline
Childhood
immunization
Diarrhoea Ensuring DOTS for TB
patients
Antacids
Safe water and
sanitation
Amoebic dysentary Referral to government
facility for temporary or
permanent
contraceptive methods
Anti-helminthics
Mobilization for
national
immunization days
Goiter Identify pregnant
woman and refer her
to SK
Health commodities
Scabies ARI prevention and
treatment services
Iodized salt
Helminthiasis Soap
Ring worm Sanitary napkins
Pneumonia Condoms
Angular stomatitis Contraceptive pills
Safe delivery kits
Reading glasses
Sources: BRAC Annual Health Reports
Shasthya Shebikas’ training and supervision
After her selection, the SS is given an 18-day basic training held at the BRAC
regional office. She also participates in monthly refresher training and is provided
specific orientation training on issues like TB, acute respiratory infection (ARI), and
malaria as needed. The SSs are provided direct and continuous supervision and their

16
performances are assessed by their immediate supervisors, the Shasthya Kormis
(SK). There are 7,000 SKs and each SK supervises 10-12 SSs. Unlike the SSs, SKs
are paid a monthly salary. To qualify to be a SK she must be married, acceptable to
the community and have passed class 10 in school. In addition to their role of
supervising the SSs, SKs conduct monthly health forums, provide antenatal care and
post-natal care, and carry out immunization programmes. The SKs are supervised by
programme organizers (PO) who in turn are supervised by the Area Programme
Manager. Medical officers provide overall technical supervision.
The SK provide timely feedback to the SS at the monthly refresher training when they
solve problems and review individual performance. SSs must report on their activities
either orally or written (for those SSs who cannot write, they sometimes have their
children write for them). There is also an independent monitoring department that
measures inputs/outputs and quarterly performance. The health programme
produces performance data and based on that the MIS is built. These data can be
used to look at health status and identify gaps in knowledge.

17
Section
STUDY DESIGN, METHODOLOGY AND DATA ANALYSIS
The study employed four methods of data collection:
1. Literature review of existing programme documents and published/
unpublished international and national literature related to community health
workers. An attempt was made to collect documents relating to BRAC’s SS
programme. A series of interviews with BRAC staff from across the organization
were carried out and included requests for any materials related to the Shebika
programme. A review of the published literature on CHWs and costing of CHW
programmes was also carried out.
2. In-depth interviews with a range of programme staff in BRAC, Bangladesh
head and field offices and in BRAC, Afghanistan and BRAC, Uganda offices. The
selection of these respondents was based on programme responsibility and also
snowball sampling methods. Appendix II provides a list of individuals who were
interviewed.
3. Secondary analysis of health programme cost data related to the SS
programme. This included analyzing the overall programme budgets for the BHP
including revenue, funding and expenditure detail for the last 3 years. The
sources for this data were BHP and the Finance and Accounts Division in
BRAC’s head office. Data were also collected from field offices whenever
possible. BRAC’s revenue and expenditures were broken down to identify
specific investments and expenditures related to the SS programme. Mr. Shimul
Shafiun, a health economist and the co-author of this study, was responsible for
collecting and analyzing these data.
4. Primary data collection and analysis of surveys of SSs in Bangladesh,
Afghanistan, and Uganda. In addition to designing and fielding a survey
questionnaire in each of the three countries, programme data were collected and
reviewed, and programme staff interviewed to understand the pricing,
procurement, inventory, ordering, distribution, and payment models. Survey data
were analyzed using summary statistics and cross-tabulations to provide
descriptive information on CHWs.
3

18
Sampling frame and design
Country case selection
Because this is a multi-country study the sampling design was adapted based on
differences between countries. In Bangladesh where there is national coverage of
SSs and the programme has been operating for several decades, a 30-cluster
sample frame was used. In Afghanistan where security and geographic accessibility
are issues, a purposive sampling frame was used. And in Uganda, a newer
programme, a convenience sample of all the CHPs who, at the time of the survey,
had been working since the launch of the health programme was used. The
sampling frame for each country is described in more detail below.
Sampling frame for Bangladesh study
We employed the 30-cluster survey method (Milligan et al. 2004) because it is well-
known, standardized and relatively quick and easy to implement. We started by
listing all the relevant upazilas in each of the six divisions of Bangladesh. After
discussion with BRAC Bangladesh staff it was decided to focus the study only in
rural areas where SSs are providing the EHC programme. As a result we excluded
upazilas in urban areas and upazilas where specific donor-funded programmes like
MNCH and malaria were being implemented. We excluded about 100 upazilas from
our listing for a total of 376 upazilas. Once we determined the upazilas to include in
each division we calculated the distribution of study upazilas proportionate to the size
of the division (Table 3).
Table 3. Distribution of study upazilas proportionate to the size of the
divisions
Division Total no. of upazilas % of total No. of sample UHCs out of 30
Dhaka 105 28 8
Chittagong 54 14 4
Rajshahi 101 27 8
Khulna 58 15 5
Barisal 40 11 3
Sylhet 18 5 2
We randomly sampled the appropriate number of upazilas from each division for a
total sample of 30 clusters. We requested the upazila Manager from each of these
30 clusters to rank all of the SSs in the upazila based on their performance measured
in terms of last monthly sales plus service charges for each SS. Performance was
classified into the following three categories:
High performing SSs were classified as those with sales and service charges of Tk.
501 or more in the last month;
Medium performing SSs were classified as those with sales and service charges of
Tk. 301-500 in the last month; and
Low performing SSs were classified as earning Tk. 300 or less in the last month.

19
From these lists generated by the upazila manager, each interviewer systematically
sampled three SSs from each performance category using a random start. This
selection was done in the presence of the upazila manager. As a result, the study
had a total sample size of 270 with 90 SSs in each of the performance categories –
high, medium, and low.
A two-day training of interviewers was carried out in March 2009 at BRAC Centre.
Seventeen interviewers were organized into 5 teams of 3 with 2 women and 1 male
on each team. The remaining 2 trained interviewers provided supervision and quality
checks in the field. Data collection took longer than anticipated because of the travel
times required to maintain the random nature of the sample. In some cases,
individual SS lived in villages that could only be reached by foot. As a result, the
interviewers remained in the field for more than 15 days.
To ensure quality control, spot checks were carried out in the field. One study team
had to be replaced when it was found that there were questions raised regarding the
quality of the data. This further delayed data collection but replacing the study team
in this district ensured the overall quality of the data.
The trained interviewers administered the quantitative survey. Survey data were
entered into SPSS by BRAC RED staff. In addition to the quantitative survey, in-
depth interviews with several low and high performing SSs were carried out to further
understand the barriers and opportunities to their financial sustainability.
Sampling frame for Afghanistan study
Based on discussions with BRAC International and BRAC, Afghanistan staff, three
study sites were selected taking into account security, geographic accessibility, and
budget considerations. It was decided to include two Dari and one Pashto speaking
areas in order to examine possible regional differences. The study sites selected
were Balkh, Nangarhar and Parwan. Seventy CHWs were randomly sampled in each
of the sites.
Parwan is a one-hour drive from Kabul, and Nangarhar and Balkh are two-and-half
and nine-hour drive from Kabul, respectively. From each province 70 respondents
were randomly selected. The population of the study was 320 CHWs in Nangarhar,
305 CHWs in Parwan, and 744 CHWs Balkh. In Nangarhar 23 from Khiwa, 24 from
Surkhrood and 23 from Behsood districts were selected for interview and all the
respondents were female. In Parwan, 24 from Jabal Siraj, 23 from Charikar, and 23
from Bagram were selected and all of them were females. In Balkh, the respondents
were selected from Dehdadi and Balkh. In Balkh BRAC has both male and female
CHWs whereas in Nangarhar and in Parwan, BRAC has only female CHWs. Among
the 70 respondents in Balkh, 10-12 were male CHWs.
Survey data were entered in BRAC Afghanistan’s head office in Kabul. Shimul
Shafiun traveled to Afghanistan in May to perform quality checks, provide support
during the data entry process, and interview programme staff.

20
Sampling frame for Uganda study
Because Uganda is a newer programme established in 2006, we selected the entire
cohort of CHPs who had been originally trained (N=180). This allowed for at least 12
months of CHP experience to inform the survey. The survey instrument was adapted
to the Ugandan context after pre-testing. Selection and training of interviewers were
organized and supervised by the Research and Evaluation Unit (REU) at BRAC
Uganda office in Kampala. Survey data were entered and cleaned by REU staff in
Kampala and they performed initial data analysis as well.
In addition to the quantitative survey, in-depth interviews were carried out with high
performing and low performing CHPs in Kampala and Iganga districts. Focus group
discussion with four high performing CHPs and in-depth interviews with two low
performing CHPs were also carried out. These interviews were carried out in the local
language and translated into English at the time of the interview. These were tape
recorded and transcribed for analysis.

21
Section
RESULTS
For the sake of organizational clarity the report presents a summary of some of the
key findings from the multi-country analysis followed by detailed description of the
results of each country study separately and in the following order (Bangladesh;
Afghanistan; and Uganda). For each country we present: 1) financial analysis of the
health programme; 2) socio-demographic characteristics of the SSs; 3) performance-
related characteristics of the SSs; and 4) description of barriers and challenges.
Section 5 provides overall discussion and presents some recommendations based
on the earlier analyses.
Multi-country study results
Financial analysis
To determine financial sustainability we calculated estimates of the programme costs
to develop an SS in the first year and the costs of maintaining that same SS in the
second year. Table 4 presents those estimates for each of the three study sites. It
presents results in US$ and in PPP adjusted US dollars. Clearly the PPP adjusted
costs are highest for Uganda ($1,204 in the first year) and lowest for Bangladesh
($245 in the first year). All of the costs dropped significantly in the second year when
refresher training and supervision are the main financial inputs of the programme.
Table 4. Comparative per unit cost to BRAC for developing a SS/CHP/CHW
per year 1
Bangladesh Afghanistan Uganda
$US or
Taka
I$US (PPP
adjusted)
$US or
Afghanis
I$US (PPP
adjusted)
$US or
Shillings
I$US (PPP
adjusted)
Total cost in the first
Year $89 $245 $247 $427 $394 $1,204
Total cost in the
second year $34 $94 $84 $145 $208 $636
SS income received
in average month
374 Taka
(Mean)
$US 14.61 143 AFG
(Mean)
$US 4.95 UX 38,222
(Mean)
$US 58.40
Table 5 presents some selected comparative results from the three study sites. It is
striking to note despite the many similarities between the three sites (age; number of
1 See Appendix 2 for detailed breakdown of the analysis.
4

22
hours worked each day; similar health training outside BRAC), there are many areas
of difference between the sites. One, for example, is the percentage of current VO
members; this is highest in Uganda and lowest in Afghanistan.
Table 5. Selected comparative characteristics between three study sites
Bangladesh
N=270
Uganda
N=158
Afghanistan
N=210
Age 39 years 36 years 36 years
Years of schooling 5 10 5.6
Number of family members 5 7 8.5
Currently a VO member 41% 91% 10%
Monthly household income
Monthly income is always greater than
expenditure
Monthly income is always less than expenditure
Monthly income is equal to expenditure
14%
24%
30%
38%
3%
6%
6%
29%
28%
Importance of SS earnings for family
H/H could not continue without SS income
SS income makes a big difference
SS income makes no difference
4%
75%
5%
18%
69%
7%
15%
11%
33%
How SS income is used
Spend on children
Give to husband
Pay school fees
Buy food
Save
Pay back loans
24%
10%
6%
17%
13%
2%
40%
1%
12%
20%
7%
2%
65%
73%
19%
3%
.5%
1%
Hours work as SS per day 3.6 3.2 3.6
Replenish SS supply between refresher trainings 80% 85% 6%
Bought medicines or health commodities outside
BRAC
10% 1% 3%
Loan used to buy SS supplies
If yes, amount spent from loan on supplies
19% (of 183
women)
1530 Taka
(Mean)
65% (of 134)
UX 170,000
(Mean)
4% (of 134)
AFG 3000
(Mean)
Number of visits required to sell medicines
1 visit
2 visits
3-4 visits
4+ visits
26%
50%
21%
.7%
2%
9%
30%
60%
9%
38%
34%
19%
Number of visits required to sell health
commodities
1 visit
2 visits
3-4 visits
4+ visits
44%
40%
12%
1%
1%
9%
18%
72%
3%
49%
32%
15%
Monthly income in last month (mean) 360 Taka
PPP$ 14.07
UX27,680
PPP$ 42.29
145 AFG
PPP$ 5.01
Income received in average month (mean) 374 Taka
PPP$14.62
UX 38,222
PPP$58.40
143 AFG
PPP$4.94
Table 5 continued…..

23
………..Continued Table 5
Monthly fluctuations in sales 97% 67% 11%
SS Loan status
Ever borrowed a loan from BRAC
Current loan with BRAC
More than one loan with BRAC
Loan with another NGO
68%
34%
13%
36%
99%
88%
36%
0.8%
13%
4%
2%
n/a
Why became an SS
Source of income for household
Financial independence
Social recognition
To learn something new
To help my community
86%
7%
3%
3%
1%
22%
3%
47%
27%
34%
4%
40%
12%
2%
Being an SS has given her financial independence 97% 80% 15%
How has BRAC VO membership affected work as
SS
People trust me more
Easier to sell products
Need loan to buy products
Has no effect
22%
13%
2%
30%
31%
36%
1%
31%
N/A
Wish to sell other health commodities and
medicines 30% 67% 30%
Health related training outside BRAC 12% 16% 9%
Minimum monthly salary (without commissions)
required
Taka 1339
(Mean)
PPP$51.80
Taka 1000
(Median)
PPP$36.68
UX 173,150
(Mean)
PPP$264.57
UX 100,000
(Median)
PPP$152.80
3643 AFG
(Mean)
PPP$126
3500 AFG
(Median)
PPP$121.15
In Uganda the CHPs tend to be better educated and a higher percentage of them
are better off than the CHWs in Afghanistan and Bangladesh. In all three countries
the majority of the SS/CHP/CHW income is spent on their children except in
Afghanistan where 73% of women reported giving their income to their husbands. In
all the three countries, the majority of women required at least two household visits
to sell medicines and health commodities, but in Uganda, 60% and 72% of women
require making four or more visits to sell medicines and commodities respectively.
This may reflect that some of the CHP transactions are done on credit and thus,
women have to make several visits in order to receive payment. Or it could be due to
the newness of the programme.
The remainder of this section describes the results of each country study in detail.
Study results for Bangladesh
Financial analysis of BRAC Health Programme
BHP data and BRAC Accounts programme data for the past three years were
reviewed to understand the revenue mix of the programme. This process was more
difficult than expected because of the complex nature of the accounts data and the

24
fact that there are many sources of revenue and expenditure. Data collection was
further hampered by a change in project staff responsible for the accounts data.
BRAC has several funding sources (including grants, donations, and self-financing)
which have been used to fund the health programme in the last several years. BHP
can be divided into two broad areas: 1) the Essential Health Care (EHC) programme
which is partially funded through the EDP of BRAC and the other BHP programmes
which are funded by individual donors (for example, Urban Maternal and Neonatal
Health Project (known as MANOSHI) which is funded by the Bill and Melinda Gates
Foundation; malaria and TB control programmes supported by the Global Fund on
AIDS, TB and malaria, and WASH programme by RNE); and 2) other smaller
programmes and pilot programmes funded by individual donors.
Table 6 shows BHP expenditure for 2005, 2006, and 2007. The results show the
difficulties of examining programme expenditure over time with dynamic
programmes. During this time, some programmes were discontinued while new
programmes were initiated. Funding sources for the programmes are shown
whenever possible. Those programmes that are entirely self-funded by BRAC are
highlighted.
Table 6. BRAC health programme expenditure (in Taka) for 2005, 2006 and
2007
Programme detail 2005 2006 2007
Essential Health
Care
208,174,353 373,775,146 BRAC;
DFID; CIDA;
NOVIB
238,304,497 RAC; DFID;
CIDA; NOVIB;
Ausaid
National Nutrition
Programme
424,261,561 230,244,204 Not listed in annual report for
this year.
Poultry for Nutrition 25,818,817 Not listed in annual report
for this year.
Not listed in annual report for
this year.
Nutrition Gardening
Project
28,809,295 Not listed in annual report
for this year.
Not listed in annual report for
this year.
Tuberculosis
Community-based
TB control
7,207,323
11,020,126
Not listed in annual report for
this year.
TB control 312,883,146 471,651,398 GFATM 514,154,078 GFATM
Strengthening
DOTS
6,554,827 Not listed in annual report for
this year.
Public Private
Partnership
496,210 Not listed in annual report
for this year.
Not listed in annual report for
this year.
Early Childhood
Development
5,902,433 514,710 Not listed in annual report for
this year.
HIV AIDS
Prevention
11,710,736
32,066,500 World Bank 10,107,883 UNICEF
Community based
Arsenic Mitigation
14,462,185 326,786 UNICEF Not listed in annual report for
this year.
Microhealth
Insurance
2,546,214 1,024,981 BRAC 369,257 BRAC
Table 6 continued…..

25
……Continued Table 6
Shushasthya(BRAC
Health Centre)
44,594,592 48,071,007 BRAC 47,695,593 BRAC
BRAC Limb and
Brace Fitting Centre
1,989,202 5,075,575 BRAC
ICRC-SFD
5,663,875 BRAC
ICRC-SFD
Distribution of water
purifying tablets
565,419 1,891,960 Not listed in annual report for
this year.
Inter personal
communication
477,000 Not listed in annual report
for this year.
Not listed in annual report for
this year.
Management
training to NNP
manager
4,916,942 Not listed in annual report
for this year.
Not listed in annual report for
this year.
Women domestic
violence health
policy
655,773 86,669 Not listed in annual report for
this year.
Water, Sanitation
and Hygiene
86,088,339 Royal
Netherlands
Embassy
486,195,713 Royal
Netherlands
Embassy
Strengthening
community based
health care
Programme in
Chittagong Hill
Tracts
Not listed in
annual report for
this year.
6,917,972 Not listed in annual report for
this year.
Improving sexual
and reproductive
health
Not listed in
annual report for
this year.
5,878,207 IDS, Sussex 3,283,985 IDS, Sussex
Scojo-BRAC
Reading Glasses
Pilot Project
Not listed in
annual report for
this year.
773,478 BRAC; Scojo
Foundation
1,385,539 BRAC; Scojo
Foundation
Malaria Control Not listed in
annual report for
this year.
91,090,436 GFATM
Community-based
Maternal, Neonatal,
Child Health – Rural
Not listed in
annual report for
this year.
22,174,33 BRAC
Community-based
Maternal, Neonatal,
Child Health –
Urban
Not listed in
annual report for
this year.
77,894,834 Gates
Foundation
Total 1,123,813,686 1,275,410,058 1,496,320,023
Sources: BRAC Health Programme Annual Reports for 2005; 2006 and 2007; BRAC Annual Reports
2006 and 2007.
Figure 1 shows the expenditures of BHP in 2007. The greatest expenditure went to
the water and sanitation programme (WASH), followed by tuberculosis. The EHC
programme was third in terms of expenditures.

26
Figure 1. Expenditure of BRAC Health Programme 2007 (in Taka)
We tried to examine all sources of revenue for these three years. This was a difficult
exercise as the sources were not easily available from accounts data for reasons
described earlier. We were more confident that data for 2007 were accurate and
more complete. Figure 2 presents the percentage of source of revenue for the health
programme for 2007. Our analysis suggests that approximately 84% of BHP funds
were from donor grants in 2007.

27
Figure 2. Sources of funds for BRAC Health Programme, 2007
Methodology for calculation of programme cost per SS/CHW/CHS
The SS is clearly the foundation for BHP as she is involved in almost all health
activities of BRAC. Because of their pervasiveness, breaking down the costs related
to recruitment, training, and retention of Shebikas is a challenging undertaking. We
reviewed programme cost data and interviewed field and head office staff to estimate
the cost to BRAC of developing an SS per year. A similar analysis was previously
carried out (Ahmed J 2007) and we based our analysis on this initial costing.
The key methodological challenges were: a) determining up to which level of
employee the costing calculation should include; and b) identifying and including
direct or indirect costs. To avoid complexity only direct costs from BRAC side were
included in this costing exercise. The social costs were not included. Employees
directly involved in supervision and assistance of SSs were included. The proportion
of their time spent directly on SS is included in the costing analysis. For example,
interviews with programme staff in the field found that SK spent almost 60% of her
time on SS supervision or SS-related activities and the remaining 40% is spent on
activities not directly related to SS. So if her salary is X amount per month and if she
is responsible for Y number of SSs, then per SS salary cost of SK per year is:
(X/Y)*60%*12months. Salary costs for other human resource were calculated using a
similar approach. To find out costs other than salary, 10% overhead cost was
assumed. Then all other costs other than staff cost such as training cost, material

28
cost, etc were included. The first year is more costly as basic training is required
while in the subsequent years less costly refresher trainings are conducted. As it is a
difficult task to measure the cost of community health volunteers, 5% uncertainty
was also assumed to address sensitivity analysis. All costs were converted into US
dollars using both official exchange rate/nominal exchange rate and PPP adjustment
of that exchange rate.
It is important to note that the programme cost per CHVs will vary over time because
the scale of the programme is different in three countries. Bangladesh is a much
older programme and is operating on a large scale while for newer programmes in
Afghanistan and Uganda this is not the case. It is more likely that cost per SS in
Bangladesh will be less than that of Afghanistan and Uganda for CHW/CHP because
of economies of scale (when large number is produced the average cost declines).
This may suggest that the costs in Afghanistan and Uganda will decline over time as
the programmes become more established and increase in size. The cost calculated
here is neither incremental cost nor marginal cost but is simply the average cost per
CHW in each of the countries.
Table 7 presents the results of the costing analysis. Based on our estimate, it costs
BRAC roughly US (PPP) $245 to train and supervise a Shebika in her first year. The
total cost in the second year drops to $94 with the main cost components being
monthly refresher trainings and salaries of the supervisory staff. These cost estimates
include 10% overhead on the salaries of all supervisory staff including SKs, POs,
upazila managers and SHC. We incorporated estimates of the proportions of time
that these staff reported spending on the SS programme based on interviews with
field staff. Finally, these cost estimates are after deductions for monthly sales income.
The detailed spreadsheet on how these calculations were determined can be found
in Appendix 1.
Table 7. Cost to BRAC, Bangladesh of developing a Shebika per year 2
Taka $US IUSD (implied PPP adjusted)
Total Cost in the First Year 6,263.78 $89.48 $244.82
Total Cost in the Second Year 2,393.78 $34.20 $93.56
Total Cost in 1st year (Min)* 5,950.59 $85.01 $232.58
Total Cost in 1st year (Max)* 6,576.96 $93.96 $257.06
Total Cost in 2nd year (Min)* 2,274.09 $32.49 $88.88
Total Cost in 2nd year (Max)* 2,513.46 $35.91 $98.24
Dropout rates of Shasthya Shebika
An important determining factor for programme financial sustainability is the dropout
rate of SSs. As the unit programme costs are significantly higher in the first year (PPP
$245) as opposed to the second year (PPP $94), minimizing the need to replace SS
is a strategic way to achieve financial sustainability.
2 See Appendix 1 for detailed breakdown of the analysis.

29
This study included active SSs only. However, BHP provided data for 2006, 2007,
and 2008 for SS dropouts in three divisions of the country (Table 8).
The total dropout rate for these three divisions including 4 districts of Dhaka division
is 11.62%. The dropout rates for Dhaka division are higher (15.5% in 2008) than for
rural divisions (11.6%).
Given that this study is focusing on the rural EHC SS, it is most appropriate to
consider an average dropout rate of 12% for EHC SS in 2008. This dropout rate is
relatively low compared to other programmes internationally. It is also lower than
earlier studies of BRAC SS dropout rates which found a dropout rate of 22% but
with regional variation (Khan et al. 1998).
Table 8. Shasthya Shebika dropout, 2006 – 2008
Years
2006 2007 2008
Divisions # of SS Drop
out
% of
Drop
# of SS Drop
out
% of
Drop
# of SS Drop
out
% of
Drop
Raj – 1 6765 715 10.57 7777 828 10.65 8720 876 10.05
Dhaka (4
districts) 3296 525 15.93 3417 420 12.29 3849 596 15.48
Raj – 2 8904 897 10.07 10459 1035 9.90 14014 1620 11.56
Khulna 8135 872 10.72 7906 1328 16.80 7895 915 11.59
Total 27100 3009 11.10 29559 3611 12.22 34478 4007 11.62
Total SS
(EHC) 68029 7551 11 70000 8554 12 80159 9314 12
Source: BRAC Health Programme
Incentive and income structure for Shasthya Shebikas
Given the costs to BRAC associated with recruiting and training a Shebika, ensuring
their continued contribution and active participation in the programme is critical in
terms of financial and programmatic sustainability. As these women are volunteers
who are expected to give up several hours of each day to provide services in their
community, understanding the incentive structure and how women respond to those
incentives is critical for ensuring sustainability.
There are two main financial incentives for the SS. One comes from the access to an
additional microfinance loan from BRAC as part of her SS role and the second is the
small commission she makes selling medicines and health products. In addition to
these sales she also earns a nominal amount from service charges for antenatal care
(ANC) and referral fees to other health facilities. At the time of this study, for ANC
referrals the SS received Tk. 2 for each VO member referred and Tk. 3 for each non-
VO member she referred. The SKs conduct the ANC and the SSs get the referral fee
for these services during their monthly refresher trainings.

30
She also receives some financial incentive when someone installs a slab latrine or
tubewell. While this is a relatively rare occurrence, she receives 10% of the cost. So,
the financial incentive model for the SS is as follows:
Financial incentives for SS = Sales of medicines + sales of health products + referral
fees + service charges
For those SSs working in areas where the TB DOTS programme is in operation there
is an additional financial incentive. In this case, if a SS identifies a suspected TB case
and that person is smear-positive, the SS initiates directly observed treatment
(DOTS) regimen for that patient. If she successfully follows-up the entire course of TB
treatment, she receives Tk. 150 per TB patient. Initially BRAC structured the TB
DOTS treatment so that the incentive came from the patient who paid a Tk. 300
deposit as a guarantee of treatment compliance but 20% of patients were unable to
pay. Since the programme shifted to SS incentive, there is a 93% adherence rate of
TB DOTS (RED Staff, personal interview).
Pricing and procurement for the SS programme
Shasthya Shebikas rely on the sale of a range of basic medicines and health
products to contribute to their monthly income and sustain their work financially.
Clearly without a regular and reasonably priced supply of products they cannot
function effectively. Therefore, the pricing, production, and procurement of these
supplies are critical to the sustainability of the programme.
The SS receives all of her supplies at the BRAC area branch office. BRAC must
supply 2,400 outlets throughout the country to ensure that SS have the necessary
supplies. There are three routes of product supply to BRAC area branch offices: 1)
central supply from BRAC head office; 2) from local sources; and 3) from BRAC
production centres located throughout Bangladesh that produce ORS, sanitary
napkins and iodized salt.
For supply of medicines and drugs, BRAC currently uses three pharmaceutical
companies. (In 2007, BRAC used five pharmaceutical companies but they limited it
to three in 2008 because some of the drug labeling was difficult for SS to read and
interpret properly.) While there is no single formula for acquiring the supply of drugs,
the general procedure is that BHP provides a list of desired drugs (they use the WHO
Prescribed List for non-qualified doctors (CHW) to BRAC procurement and requests
the procurement process be initiated. BRAC Procurement department advertises a
call for tenders in newspapers. The Procurement department negotiates with the
bidders and in the end the pharmaceutical company provides drugs to BRAC at the
institutional price. The SS sells the drug at the manufacturing and retail (MRP) price
and BRAC gets the institutional price reimbursed plus the small markup the SS
keeps as her profit. (Table 9 provides these prices for medicines).
Several health commodities such as sanitary napkins, delivery kits and iodized salt
are produced at BRAC Production Centres. The prices for these goods are set by
the BRAC Procurement Unit. These are supplied to the branch office directly from

31
the regional production centres. Other products that are not produced by BRAC are
procured by BRAC from the market at very low rates.
The SSs order and refill their supplies at the monthly refresher training which are
organized by the Programme Organizers (PO). If they need supplies at other times
they can go each Thursday to the branch office to replenish their supplies. The PO
writes down the amount required by each SS and then the PO prepares an order
plan which she sends to her respective upazila (Sub-district) manager. For
medicines, the upazila manager sends this request to the medical representatives of
the pharmaceutical companies selected by BRAC head office. During the last week
of each month the PO submits a requisition to the Head Office for supplies. The
supplies arrive at the branch office on the first day of the month.
Table 9 provides the details of the medicines and health commodities that are sold
by the SS in 2008. It also shows the institutional purchase price that BRAC pays, the
amount they sell it for to the SS, and the amount of profit per unit that the SS makes.
Those items highlighted are produced by BRAC.
Table 9. Medicines and health commodities sold by SS (in Taka) (January–
December 2008)
Medicine Form/
Strength
Purchase by
BRAC
Sold to SS BRAC’s
margin
Incentive for
SS
1. Paracetamol Tablet 0.41 0.43 0.02 0.13
2. Paracetamol Syrup 10.13 10.50 .38 1.64
3. Histacin Tablet 0.16 0.16 0.01 0.05
4. Histacin Syrup 10.65 11.00 .35 2.50
5. Iron Tablet 0.15 0.15 0.01 0.03
6. Iron Syrup 20.72 21.50 0.78 4.50
7. Metronidazol Tablet 0.76 0.80 0.05 0.21
8. Metronidazol Suspension 17.71 18.00 0.29 5.25
9. Vitamin C Tablet 1.12 1.15 0.04 0.15
10. Vitamin B complex Tablet 0.32 0.33 0.01 0.12
11. Vitamin B complex Capsule 0.49 0.50 0.02 0.07
12. Vitamin B complex Syrup (100ml) 16.81 17.50 0.70 3.30
13. Vitamin B complex Syrup (200ml) 27.80 28.50 0.70 9.00
14. Riboflavin Tablet 0.16 0.17 0.01 0.06
15. Antacid Tablet 0.44 0.45 0.03 0.09
16. Antacid Suspension 25.85 26.00 0.16 6.00
17. Whitfield Ointment Ointment 11.00 12.00 1.00 1.00
18. Benzylbenzoate Emulsion 15.75 16.00 0.25 2.20
19. Albendazole Tablet 1.94 2.25 0.26 1.15
20. Mebendazole Tablet 0.52 0.55 0.03 0.22
21. Mebendazole Suspension 11.50 12.00 0.50 2.60
22. Cotrimoxazole Suspension 16.00 16.50 0.51 4.50
Table 9 continued…..

32
……..continued Table 9
Health Commodities
Pregnancy test strip Single Pouch 4.98 6.00 1.02 14.00
Femicon Pill 9.63 10.00 0.38 2.00 Oral contraceptive pill
Nordette-28 pill 25.30 26.00 0.70 3.86
Hero/piece 1.53 1.60 .07 0.40
Panther/3
pc/pack
7.50 8.00 0.05 2.00
Condom
Sensation/3
pc/pak
7.83 8.00 0.17 2.00
ORS Sachet Single sachet 2.19 2.40 0.21 0.60
Aromatic
Beauty
12.25 12.50 0.25 3.50 Soap
Aromatic Gold 12.25 12.50 0.25 3.50
Sanitary Napkin Nirapad, 10 in
one pack
25.00 26.00 1.00 2.50
Delivery Kit Kallayani, Single
Plastic Sachet
17.00 18.00 1.00 2.00
Iodized Salt Transparent 1
KG plastic pack
10.25 11.00 0.75 2.50
Source: BRAC Health Programme
There are currently 21 items that the Shebika can supply her community. BHP is
considering some additional medicines and health products such as tooth powder;
½ kilogram bags of salt; and laundry detergent powder for pregnant women to
reduce their physical burden.
How Shasthya Shebikas spend their time
While the recruitment, management and incentive structures for the SS are clearly
defined by BHP, there is less detailed understanding of how SSs spend their time
and what factors influence the amount of financial income they receive as SS. The
quantitative survey designed and administered for this study provides a wealth of
data about the basic characteristics of SS and some context about the factors that
contribute to their financial success. The full questionnaire can be found in Appendix
2.
Table 10 presents some basic socio-demographic characteristics of the study
sample. As described in Section 3, this is a random sample representative of SS
providing EHC in rural areas of Bangladesh.
Table 10. Socio-demographic characteristics of SS in study sample (N=270)
Indicator Average
Age 39 years
Number of years of schooling 5
Number of family members 5
Currently a VO member 41%

33
Interestingly, 41% of the women reported to be current VO members despite this
being considered a selection criterion for being an SS. Of those that had been VO
members, 36% had dropped out of the VO. These results are similar to Rahman and
Tasneem’s (2008) study of SSs in Nilphamari which found that 40% of SSs were VO
members before becoming an SS.
The survey asked several questions to ascertain the economic situation of the SS
household. Almost one-quarter of the women reported their household monthly
income to be always less than their expenditure. Only 14% of women reported that
their monthly household income is always greater than expenditure. While only 4%
said that their households could not continue without their SS income, 75% did
report that their SS income made a big difference to household income and 97%
reported that being an SS had given them financial independence.
In terms of how they used their earnings as SS, 24% reported spending their income
on children; 17% bought food for the household; and 13% of women reported that it
was saved (Table 11).
Table 11. Household financial status and monthly income
Monthly household income
Monthly income is always greater than expenditure
Monthly income is always less than expenditure
Monthly income is equal to expenditure
14%
24%
30%
Importance of SS earnings for family
H/H could not continue without SS income
SS income makes a big difference
SS income makes no difference
4%
75%
5%
How SS income is used
Spend on children
Give to husband
Pay school fees
Buy food
Save
Pay back loans
24%
10%
6%
17%
13%
2%
The results related to SSs’ performance suggest that the majority of SSs are quite
active. Women reported that, on average, they worked for 3.6 hours and could visit
14 households a day. However, 95% of the women said that community members
come to their home to buy medicines and it is unclear whether or not this was
counted as a household visit in their response. Eighty percent of the respondents
replenished their product supply between refresher trainings suggesting that they are
able to move at least some of their inventory. It could, however, also suggest that
SSs do not have income available to purchase and hold a lot of inventory each
month but that it is financially more convenient for them to buy fewer supplies more
frequently. Interestingly, 10% of the women said they purchased medicines or
supplies outside BRAC to use in their SS work.

34
Based on their reports of the number of visits required to sell medicines and health
commodities, it appears that SSs are required to make more than one visit –
particularly in the case of selling medicines where 50% of the SSs reported that two
household visits are required (Table 12). This may reflect the SS responding to an
individual demand from a community member or her identification of a particular
illness in a household. Or it could reflect that she is not keeping a large inventory of
medicines and that in order to respond to specific demand for medicines she has to
resupply and then return to the household.
On average, selling health commodities required fewer visits – 44% of the SSs said it
only required one visit. This could reflect that the community easily recognizes health
commodities such as iodized salt or sanitary napkins and/or that demand for these
commodities are more regular and predictable.
Table 12. Performance-related characteristics
Years working as an SS 5.8 years (average)
Reported number of h/h visits possible per day 14 visits(average)
Hours work as SS per day 3.6 hours (average)
(5% reporting working
5 hours per day)
Replenish SS supply between refresher trainings 80%
Bought medicines or health commodities outside BRAC 10%
Loan used to buy SS supplies
If yes, amount spent from loan on supplies
19% (of 183 women)
1530 Taka – Mean
Women come to SS house to buy health commodities and
medicines
95%
Number of visits required to sell medicines
1 visit
2 visits
3-4 visits
4+ visits
26%
50%
21%
.7%
Number of visits required to sell health commodities
1 visit
2 visits
3-4 visits
4+ visits
44%
40%
12%
1%
Financial performance and monthly income of SS
BRAC categorizes SS performance in three ways: 1) very active – those that earn Tk.
300-500 per month; 2) moderately active – those that earn Tk. 150-300 per month;
and 3) low performing – these SS are superficially involved and show up for monthly
refreshers but have low sales.
For the SS in the survey, the mean monthly income in the last month was Tk. 360
(PPP$14.07) and the median was Tk. 250 (PPP$9.75). The reported mean income
per average month was Tk. 374 (PPP$14.62). Almost all women reported monthly
fluctuations in sales. Sixty-eight percent of the respondents said that they had ever

35
borrowed a loan from BRAC. This suggests that access to a microfinance loan may
not be a strong financial incentive for all SS. Thirty-four percent had a current loan
with BRAC; 13% had more than one loan with BRAC; and 36% had a loan with
another NGO3 (Table 13).
Table 13. Monthly income, financial performance and incentives of SSs
Monthly income in last month 360 Taka – Mean (PPP$14.07)
250 Taka – Median (PPP$9.75)
Income received in average month 374 Taka – Mean (PPP$14.62)
250 Taka – Median (PPP$9.75)
Monthly fluctuations in sales 97%
SS Loan status
Ever borrowed a loan from BRAC
Current loan with BRAC
More than one loan with BRAC
Loan with another NGO
68%
34%
13%
36%
Why became an SS
Source of income for household
Financial independence
Social recognition
To learn something new
To help my community
86%
7%
3%
3%
1%
Being an SS has given her financial independence 97%
How has BRAC VO membership affected work as SS
People trust me more
Easier to sell products
Need loan to buy products
Has no effect
Not a VO member
22%
13%
2%
30%
33%
Eighty-six percent of the respondents said that they became an SS to contribute as
a source of income to their household. Only 3% reported ‘social recognition’ as their
motivation for becoming an SS and 1% reported ‘helping her community’ as her
initial motivation for becoming an SS. Interestingly, 7% of the SSs said that obtaining
financial independence was an initial motivation for becoming an SS yet 97% of them
reported that working as an SS has given them financial independence. This may
suggest that the expectations of SS change and grow as she spends time working
as an SS and that what initially motivated her to join as an SS are no longer sufficient.
We also asked the SS how being a BRAC VO member affects her work and
productivity as an SS. Forty-one percent were current VO members while 36% of the
sample had dropped out as VO members. Twenty-two percent said that people
trusted them more and 13% said that it made it easier for them to sell products. Only
2% reported that being a BRAC VO provided her the necessary microfinance to
purchase products to work as an SS.
3 The study did not record which NGO.

36
Time allocation of SS
We asked SSs to report the frequency and average time spent in the last month on
activities related to financial incentives. Unfortunately, this means we do not have any
data on the amount of time SSs spend on health education and promotion. Because
this was a study on financial aspects of the SS programme we did not include it in
the questionnaire. However, this is an important area that should be addressed in
future. The majority of SSs reported that they spent their time selling medicines and
health commodities. Only 2% said that they did not sell medicines. For each of the
main activities expected of the SSs we asked whether or not she provided it in the
last month, the average time to perform the activity, and the average number of
times each month she provides the activity. This provides a rough picture of how
SSs are spending their time. The results are summarized in Table 14.
The biggest portion of the SS time spent in attending refresher training (275 minutes
per month) followed by selling medicines (216 minutes per month) and selling
commodities (120 minutes per month). Seventy-seven percent had performed
pregnancy identification in the last month. In terms of activities for which there is a
financial incentive for the SSs, fewer (14%) reported providing referrals to BRAC
Health Centres (although this likely reflects the fact that BRAC Health Centres are not
everywhere); and 59% were attending delivery and newborn care.
Table 14. SS reported activities related to financial incentives and how she
spent her time in the last month
Activities related to
financial incentives
Provided in the last
month
Average time
to perform
service
Average
number of
times
provided
Average total
number of
minutes per
month
Pregnancy identification Yes – 77%
No – 23%
20 minutes 2 40
Attending delivery and
providing newborn care
Yes – 59%
No – 41%
27 minutes 3 81
Referral to Shushatya
(BRAC health referral
centre)
Yes – 14%
No – 26%
No BRAC Shushasthya
– 60%
17 minutes 2 34
Referral to government
clinic or hospital
Yes – 41%
No – 59%
31 minutes 2 62
Treat TB (DOTS) Yes – 40%
No – 56%
No DOTS Programme –
4%
14 minutes 2 28
Attending refresher
training
Yes – 92%
No – 8%
275 minutes 1 275
Selling medicines Yes – 98%
No – 2%
12 minutes 18 216
Selling health
commodities
Yes – 97%
No – 3%
10 minutes 12 120

37
Previous analysis suggests that SSs spent 60% of their time selling medicines and
40% of their time selling health commodities (Ahmed J, personal communication).
BRAC would like to increase the proportion spent on selling health commodities but
they recognize that this requires social marketing skills.
Competition and other constraints on SS performance
In the survey the SS was asked whether she felt the presence of other service
providers in her area (such as pharmacies or shops that sell medicine, village doctor,
TBA, NGOs, public and private facility etc.) affect her income. Seventy-four percent
of the respondents said that they have pharmacies or shops that sell medicine in
their area and about 80% of these respondents said that this limited their income.
Sixty-seven percent reported that they felt competition from private and public clinics
in their areas while 76% and 33% respondents said that the village doctors and
traditional healers limited their income respectively. Competition from the traditional
healers was lower at 33% (Table 15). The survey data, therefore, suggest that SSs
face the greatest competition from village doctors followed by pharmacies or drug
shops.
In interviews with SSs, it was stated that in some cases village doctors recommend
lower quality and lower priced drugs that are easily available in local drug shops.
Often these are lower priced than the drugs that SSs are selling which sometimes
makes it difficult for SSs to sell their products. In other cases the SSs reported that
people are less aware about the brand they were selling and therefore were more
skeptical of their quality. SS drugs are sometimes more expensive than drug shops
and, therefore, some villagers are skeptical of SS quality and ability and the high
price they charge. The SSs reported, however, that once they were able to sell their
higher quality product it becomes easier for them to do so in the future.

38
Table 15. Competition and other potential constraints to performance
Other health providers
Existence in SS area
If yes, these limit SS
income
Pharmacy or shop that sells medicines
NGOs
Government clinic or hospital
Private clinic
Village doctor
Traditional healer
TBA
74%
10%
34%
data not interpreted
74%
53%
49%
80%
65%
67%
76%
33%
16%
Have any trouble selling BRAC suggested
medicines or health commodities
25%
Frequency Valid Percent If trouble selling products, why?
People buy products from shops
Product is too expensive/cheaper elsewhere
Less preference for BRAC products
People do not trust SS
30
18
17
2
44%
27%
25%
3%
Wish to sell other health commodities and
medicines
30% - Yes
70%- No
There is some anecdotal evidence that SSs face less competition from government
facilities in part due to the perceived and actual limitations of government facilities.
For example, while government facilities generally provide good quality drugs, they
often do not have sufficient supply to meet the local demand. In addition,
government facilities often suffer from problems in drug distribution – sometimes
drugs, even those of high quality, are given in loose form without any foil pack which
can reduce or destroy the efficacy of the drug and can make prescribing it for
patients difficult. BRAC selects drugs with clear labeling and packaging. Finally, there
are often hidden costs at the hospital and they do not provide door-to-door services.
In the survey, 30% of the SSs said that they wished they could sell additional health
commodities or medicines (Table 16).
Table 16. Medicines and health commodities SS reported they wish to sell
Name of medicine/health commodity Responses (%) Percent of cases
Renitidine (for stomach upset) 14 28
Femicon(Eye drops) 8 17
Cinkara (Herbal tonic for range of symptoms) 5 10
Napa (Paracetomol) 5 10
Diclofenac (Anti-inflammatory/pain reliever) 5 10
It is interesting to note that among 10% of the SSs who reported buying supplies
from outside BRAC purchased several of these medicines (Table 17).

39
Table 17. Medicines/commodities purchased outside of BRAC for sale as SS
Name of medicine/health commodity Responses (%) Percent of cases
Renitidine (for stomach upset) 17 37
Napa (Paracetomol) 5 11
Civit 3 7
Histacin (Anti-histamine) 3 7
Femicon (Eye drops) 3 7
While a detailed analysis of products was beyond the scope of this report, we asked
SSs to report the medicines and health commodities that they sold most and least
frequently. The results are shown in Tables 18 and 19. Because women could select
up to three responses and due to individual differences, some items, such as soap,
appear as most frequently and least frequently sold. Paracetamol, vitamins and
antacids are the most frequently sold medicines (and were not mentioned as least
frequently sold medicine). Soap, salt, and saline are the most frequently sold health
commodities (although soap and salt also were mentioned by some SSs as a least
frequently sold commodity).
Table 18. Most frequently sold medicines and health commodities
Medicine Responses (%) Percent of cases
Paracetamol 16 45
Vitamin-B-Syrup 10 28
Antacid Plus 9 24
Histacin (anti-histamine) 8 24
Iron Tablets/Syrup 8 22
Health commodities
Soap 38 83
Salt 28 70
Saline (ORS) 22 55
Delivery kit 7 17
Sanitary napkin 6 15
Table 19. Least frequently sold medicines and health commodities
Medicine Responses (%) Percent of cases
Dermin Balm (Skin ointment to relieve pain and itching) 18 39
Ascabiol(Lotion for treatment of scabies and body lice) 13 28
Histacin (anti-histamine) 9 20
Riboson (Vitamin B2) 8 18
Mebendazole (Treatment for pinworms) 7 16
Health commodities
Sanitary napkin 38 73
Delivery kit 32 61
Condom 8 15
Salt 5 9
Soap 4 8

40
We asked women to describe the advantages and disadvantages associated with
working as an SS. The main advantages reported were that she could work from her
home; more prestigious in their community; better hours; and the expectation of
better earnings in future. The main disadvantage that was expressed was less
earnings in future. Forty-three percent of the respondents said there was no
disadvantage to being as an SS. When asked how being an SS affects carrying out
their household duties, the majority (66%) said that being an SS did not affect their
household responsibilities. Thirty-three percent reported of working harder at
household because of their role as SS (Table 20).
Table 20. Advantages and disadvantages of working as an SS4
N Percent Percent of
cases
What are the advantages of working as an SS
Better earning in future
Better work environment
Can work from home5
Better hours
More prestigious among the community
Family approves
Less distance to travel
No advantage
104
29
150
111
104
33
31
7
18.1%
5.1%
26.1%
19.3%
18.1%
5.7%
5.4%
1.2%
39.1%
10.9%
56.4%
41.7%
39.1%
12.4%
11.7%
2.6%
N Percent Percent of
cases
What are the disadvantages of working as an SS
Less earnings in future
Not good work environment
Must work outside of home
Hours not convenient
Less prestigious among the community
Family disapproves
Greater distance to travel
No disadvantage
134
2
12
2
3
4
8
122
47%
.7%
4.2%
.7%
1.0%
1.4%
2.8%
43%
51%
.8%
5%
.8%
1%
2%
3%
46%
How does being an SS affect household duties
Does not affect h/h duties or responsibilities Have
to work harder in h/h duties
Less time for children
Less time for cooking
66%
30%
3%
.4%
Economic opportunity cost and dropouts
Clearly a major factor in determining the financial sustainability of the SS programme
is minimizing the number of dropouts. As the financial cost data presented earlier
show the initial training costs for each SS are substantial. Replacing each dropout
becomes very expensive for the programme. BRAC considers an SS to be active if
she participates in two consecutive refreshers and visits 15 households per month.
4 More than one response possible.
5 The survey response ‘can work from home’ is slightly misleading as the SS job requires moving outside
her home to make household visits. It is likely that the response “work from home” reflects a broader
response of flexible working hours and conditions.

41
Khan et al. (1998) found total EHC SS dropout to be 22% but with significant
regional variation (31% in Mymensingh compared to 44% in Fulbaria). This in-depth
qualitative study identified the following as some of the reasons for SS dropout: lack
of time because of the need to care for children and household chores; lack of profit
for the amount of effort; family disapproval; and not enough time to visit BRAC office.
While not a part of this study, there is anecdotal evidence that SS dropout rates are
higher in urban areas, because they have greater economic opportunities (garment
sector or work as domestic help) available to them (S. Taskeen, personal interview).
Our study sample did not include SS dropouts, so our sample may be biased
towards those SS who are less likely to dropout. However, when asked if they ever
considered quitting their work as SS, 13% replied yes. Of these women, 83% said
that they had considered quitting their work because the payment is too low; 6%
said because the work is difficult; and less than 1% attributed the reasons to better
paid job alternatives, the need to care for their children, and disapproval of husband
or in-laws. The SSs were asked whether or not a fixed monthly salary would
encourage them to be more active as SS; 92% felt that a monthly salary would
motivate them to work harder. The minimum monthly salary they expected was Tk.
1,339 (PPP-adjusted $51.80) (mean response) (without commissions from sale of
medicines or drugs).
Economic opportunity cost of SS
Clearly the decision whether to continue as an SS is closely influenced by the
availability of other economic opportunities and how she might use her time
otherwise. This study attempted to address the economic opportunity cost of the SS
by asking if the SS had done or was currently doing other work while serving as an
SS, whether this work was regular or seasonal, and what was the mean earning in
an average month? We also asked about her perceptions of other work available to
her. Three-fourth of the women (74%) reported that they had or were concurrently
raising poultry while working as SS. On average, this provided women an additional
Tk. 298 per month and was regular rather than seasonal work. Handicraft work was
the next most frequently reported work in addition to being an SS with 26% reporting
this mostly regular work (Table 21).
Other economic opportunities that pay more than the average monthly income for
the SS tend to be regular work that does not have the flexibility in terms of hours and
location that the SS work does. For example, women reported the average earnings
of factory work to be Tk. 1,300, but 2% of the SSs in the survey reported having ever
done or were currently doing that work. Factory work generally requires long and
regular hours, which would constrain her from carrying out her duties as an SS.
We also asked the respondents to identify what other jobs would be available to
them if they were not an SS and to report the monthly average income possible from
that job. Poultry raising had the highest average monthly income possible followed by
factory worker and tailor (Table 22).

42
Table 21. Reported work that SS has done or currently does while also
working as an SS
Type of work Has done or currently
doing
Regular or Seasonal Mean earnings in an
average month
Factory worker Yes – 2%
No – 98%
Regular – 83%
Seasonal – 17%
1300 Taka
(PPP$ 50.29)
Handicraft Yes – 26%
No -74%
Regular -80%
Seasonal -20%
616 Taka
(PPP$ 24.07)
Poultry raising Yes – 74%
No -26%
Regular -98%
Seasonal -2%
298 Taka
(PPP$ 11.65)
Agriculture Yes – 13%
No -87%
Regular -46%
Seasonal -54%
833 Taka
(PPP$ 32.55)
Small business/hawking Yes – 9%
No -91%
Regular -80%
Seasonal -20%
1088 Taka
(PPP$ 42.52)
Tailor Yes – 13%
No -87%
Regular -91%
Seasonal -9%
611 Taka
(PPP$ 23.88)
Domestic worker Yes – 5%
No -95%
Regular -79%
Seasonal -21%
585 Taka
(PPP$ 22.86)
Midwifery/TBA Yes – 14%
No -86%
Regular -92%
Seasonal -8%
276 Taka
(PPP$ 10.79)
Table 22. Availability of other jobs and monthly income possible
Other jobs available Average monthly income
possible (In taka)
Factory worker Tk. 246 (PPP$ 9.62)
Handicraft 166 (PPP$ 6.50)
Poultry raising 459 (PPP$ 17.94)
Agriculture 109 (PPP$ 4.26)
Small business/hawking 199 (PPP$ 7.78)
Tailor 230 (PPP$ 8.91)
Domestic worker 51 (PPP$ 1.99)
Midwifery 51 (PPP$ 1.99)
It is interesting to note that, except for poultry raising, the perception of the average
monthly income possible from alternative jobs is relatively low and certainly in the
range of what she can or is already making as an SS. It is especially interesting that
the perceived salary for a factory worker is so low (Tk. 246) as compared to the
reported average monthly earnings of those SS who work as factory workers (Tk.
1,300). This may suggest that there are misperceptions or imperfect information
about the financial potential of other economic alternatives available to these women.
It is also interesting to note that the SSs do perceive that other economic
opportunities are available to them. And despite the availability of these jobs, they
continue to work as SS.

43
Factors that account for SS income performance
Identifying potential factors that explain why some SS earn more monthly income
may yield useful lessons for BHP and its sustainability. The survey data were
examined to identify some of the key factors that might explain why some SS
perform better in terms of monthly income.
There were 47 SSs in the sample who fit the criteria of a high performing SS as
earning more than Tk. 501 in the last month. Of them 51% were current VO
members, 43% currently had a loan from BRAC, and they work an average of 3.4
hours each day.
Table 23 shows other descriptive characteristics of these high performing SSs.
Table 23. Selected descriptive statistics for high performing SS (n= 47) and
low performing SS6 (n=149)
Indicator Average for High
Performing SS
(n=47)
Average for low
performing SS
(n=149)
Age 41 years 39 years
Number of years of schooling 6 5
Number of family members 5 4.7
Currently a VO member 51% 36%
Monthly household income
Monthly income is always greater than
expenditure
Monthly income is always less than expenditure
Monthly income is equal to expenditure
21%
26%
23%
12%
22%
32%
Currently have BRAC loan 59% 48%
Loan used to buy drugs and SS Supplies 27% 23%
Replenish supplies between refresher trainings 96% 73%
Buy medicines or health commodities outside
BRAC
17% 7%
Average hours worked per day 3.4 4
Any health related training outside BRAC 19% 10%
Minimum monthly salary (without commissions)
required
Mean – 1757 Taka
(PPP$ 60.67)
Median - 1500 Taka
(PPP$ 50.62)
Mean – 1244 Taka
(PPP$ 48.62)
Median – 1000 Taka
(PPP$ 39.08)
The study findings suggest that high performers were more likely to currently have a
loan with BRAC than low performers. Having more than one loan did not affect
performance however. The correlation between length of time working with BRAC
and SSs’ monthly income is positive and statistically significant. This suggests that
the SSs who worked with BRAC for a longer period are more likely to earn more than
those who are with BRAC for short time. Having children aged <2 years is not a
6 Low performers defined as earning less than 300 Taka in last month (n=149).

44
predictor of SS performance. There was no statistical association between
educational status (having primary education) and monthly earning.
The fact that high performers are more likely to have received health-related training
outside BRAC is a potentially interesting result. Rahman and Tasneem (2008) also
found an association between SS income and additional training from outside BRAC.
This may be a useful additional selection criterion for recruiting SS.
An important factor that might explain the performance of an SS, beyond the number
of hours she works each day, is how she spends her time as an SS. We tried to
compare this for low and high performing SSs in the two charts below. These
represent average number of minutes spent per month on each SS activity. While
these should be viewed cautiously because they are averages, it is striking to note
the differences in the reported number of minutes spent on selling medicines and
commodities between the two groups (Fig. 3 and 4). While it is not surprising that the
high performers would be selling more medicines and commodities, the additional
amount of time they spend on these activities is quite significant – 171 minutes more
per month for selling medicines and 110 minutes more per month for selling health
commodities. In this analysis high performers spent more time on average each
month in these particular SS activities than did low performers – 1,158 versus 751
minutes. It is also important to note that the questionnaire did not ask SS to report
the amount of time spent on health education and health promotion as that is
currently not financially incentivized in the SS programme.
Figure 3. High performing SS – average minutes per month on selected SS
activities

45
Figure 4. Low performing SS – Average minutes per month on selected SS
activities
Study results for BRAC Afghanistan
BRAC Afghanistan, the first BRAC programme outside Bangladesh, began
operations in 2002. The health conditions and status in Afghanistan are extremely
poor due to decades of war, challenging physical terrain that limits access to
services, and generally poor knowledge and awareness of health conditions and
health-seeking behaviour. Social and development indicators are generally poor: life
expectancy at birth is 43.6 years; adult literacy is 28%; maternal mortality is
extremely high at 1,600 deaths per 100,000 as is infant mortality (165). Afghanistan
ranked 181st out of 182 countries in the Human Development Index, a composite
measure of human well-being and development.
BRAC operates in all 34 provinces in the country and includes 2,100 Health Posts,
66 Basic Health Centres and has over 3,600 CHWs. The programme has both male
and female CHWs and the catchment area is smaller than in Bangladesh. The health
programme treated over 130,000 patients in 2007 (BRAC Afghanistan Annual Report
2007).
BRAC Afghanistan initiated and introduced the concept of Community Health
Volunteers (CHV) in 2002. During that period, there was no unique policy from the
Ministry of Public Health (MoPH) with regard to CHV activities in the field. Several

46
NGOs were working in the health sector according to their own policies and
strategies. Some of the NGOs trained community people and called them CHWs and
paid them varying amounts of salaries between US$2 to US$50 per month.
The CHV, which BRAC Afghanistan initiated in 2002, were all female, working
voluntarily, and not paid any salary. BRAC Afghanistan recruited female CHWs to
supervise on average 10 CHVs. The CHWs were paid about 750 Afghani per month
(equivalent to US$15).
In 2004, the Afghan MoPH introduced a new strategy called the Basic Package of
Health Care Services (BPHS) throughout the country. The objective of the BPHS is to
improve quality and access to health services, particularly for women and children.
Following the introduction of BPHS, the MoPH established a mechanism for
contracting out the delivery of health care services throughout Afghanistan. This
changed how local and international NGOs provided health services. The local and
international NGOs were contracted to implement the BPHS in 34 provinces of
Afghanistan. This was intended to help the MoPH coordinate efforts in the health
sector and minimize duplication. Under the implementation of BPHS, the NGOs are
responsible to establish Health Posts beside the Basic Health Centres (BHC),
Comprehensive Health Centres (CHC) and district hospital (DH). The health post is, in
fact, the house of the CHW. As per this policy, the NGOs are responsible for training
two CHWs (one male and one female) for every 100-150 households. In BPHS the
role of the CHW is clearly identified.
The CHW programme of BRAC Afghanistan
At the time of this study, BRAC Afghanistan health programme was operating in 7
provinces out of 34 provinces in the country. BRAC Afghanistan provides healthcare
services through 7 districts hospitals, 20 CHC, 67 BHC, 44 Sub Health Centres and
2,055 Health Posts. A total of 3,483 CHWs are working in the catchments areas of
the above mentioned health facilities in 7 provinces. Unlike in other BRAC
programmes, both male and female CHWs are working under BRAC health
programme. About 61% of all CHWs are female. However, BRAC still has only
female CHWs in two provinces (Nangarhar and Parwan). Since introduction of BPHS,
BRAC Afghanistan must follow the MoPH policy for its CHWs in Afghanistan.
Province-wise information about the CHWs is provided in Table 24.
Incentives for CHWs
Because of its close and institutionalized partnership with the Afghanistan
government, the CHW programme is different from the SS programme in
Bangladesh and Uganda. In Afghanistan, where BRAC is being contracted to provide
services by the government, BRAC must follow the MoPH policy in terms of the
services provided. The government regulations do not allow CHWs to sell drugs so
the financial incentive structure for the Afghanistan programme is slightly different
than Bangladesh or Uganda.

47
Table 24. Description of BRAC Afghanistan CHW coverage
Province Project/
Donor
# of
Health
Post
# of
Female
CHWs
# of
Male
CHWs
Total
CHWs
Notes
Badghis PPA/World
Bank
331 331 331 662 There are 7 districts in
Badghis and whole
province is covered
Balkh
BRAC
PPA/World
Bank
372 406 338 744 There are 15 districts in
Balkh province, all the
districts are covered
Balkh BDN PPA/World
Bank
656 553 509 1062 BDN is implementing
partner of BRAC in
Balkh province
Nimroz 150 140 150 290 There are 5 districts, all
the province is covered
Nangarhar CBHP/
Oxfam
Novib
241 362 58 420 There are 22 districts in
Nangarhar, BRAC
works only in five
districts
Parwan CBHP/
Oxfam
Novib
305 305 0 305 There are 10 districts in
Parwan, BRAC works
in 3
Total 2055 2097 1386 3483 61% are female
Source: BRAC, Afghanistan Country Office
The financial incentives for CHWs in BRAC, Afghanistan are:
CHWs get 100 AF (PPP$ 3.46) for attending refresher training in Balkh whereas
every attendee gets 60 AF (PPP$ 2.07) in Ningrarhar and in Parwan;
• CHWs get 50 AF (PPP$ 1.73) for detection of a TB patient and 150 AF (PPP$
5.19) for completion of TB treatment;
• CHWs get 20 AF (PPP$ 0.69) per delivery;
• Sale of iodized salt and delivery kit; and
• CHWs get 5 AF (PPP$ 0.19) for referring women for injectable family planning
services.
Fourteen items of drugs are given to CHWs during the refresher training. These are
provided and paid for by the donor.
CHW training
Trainings for the CHWs are different than those for SS in Bangladesh. CHWs are
provided more training which is provided in three phases rather than a single two-
week session. In the first phase they have three weeks of training; in the second
phase they have another three weeks of training after spending two months in the
community; and in the third phase they have another three weeks after two months
in the community. The monthly refresher training are three hours in length.

48
Because of these fundamental differences associated with the BPHS programme,
the cost structure of developing a CHW per year is also quite different. We tried to
use the same costing approach as we did for Bangladesh and Uganda. The cost in
the first year is about $PPP 247 per CHW and for the second year is about $PPP 83
which includes the cost of refresher training and the staff cost. Results are
summarized in Table 25 and detailed analysis is found in Appendix 2.
Table 25. Cost to BRAC, Afghanistan of developing a CHW per year
Afghani (AFN) US$ IUS$ (PPP-Adjusted)
Total Cost in First Year 12,248 246.96 $427.38
Total Cost in Second Year 4,178 83.56 $144.61
Total Cost in First Year (min) 11,113.20 234.61 $384.65
Total Cost in First Year (max) 13,582.80 259.31 $470.12
Total Cost in Second Year (min) 3,760.20 234.61 $130.15
Total Cost in Second Year (max) 4,595.80 87.74 $159.07
The basic socio-demographic characteristics of the CHWs in our sample highlight
some of the contextual differences in Afghanistan as compared to Bangladesh and
Uganda. The CHWs in our study have less education (mean of 5.6 years); larger
family sizes (mean of 8.5); and higher numbers of children (mean of 4.4 children).
Only 10% of women are VO members but this likely reflects the programmatic
change in VO membership not being a selection criterion in response to government
regulation (Table 26).
When looking at the performance characteristics of CHWs in the sample it is clear
that they are active as CHWs. The mean number of years worked as a CHW is 3.7
and the mean number of hours worked per day is 3.6 hours and reported number of
possible household visits per day is 5.3. Ninety-five percent report that women come
to their home to buy health commodities and medicines. It is not clear whether these
are reported as household visits by the CHWs. The majority of women report
needing to make two or more visits to sell both medicines and health commodities
(Table 27). Only 10% of CHWs were VO members and this is likely influenced by the
MoPH removal of VO membership as a requirement for being a CHW.
Table 26. Socio-demographic characteristics of CHWs in study sample
(N=210)
Indicator Average
Age 36 years
Number of years of schooling 5.6 years (Mean)
Number of family members 8.5 (Mean)
Number of children 4.4(Mean)
Number of children under 2 years of age 14 %
Currently a VO member 10 %

49
Table 27. CHW performance-related characteristics
Years working as a CHW 3.7 years (mean)
Reported number of h/h visits possible per day 5.3 (mean)
4 (Median)
Hours work as CHW per day 3.6 hours (mean)
3 hours (median)
Number of household visits expected from BRAC 36 (mean)
30 (median)
Replenish CHW supply between refresher trainings 6 %
Bought medicines or health commodities outside BRAC 3 %
Loan used to buy CHW supplies
If yes, amount spent from loan on supplies
4% (of 134 women)
AFG 3000 – Mean
(PPP$ 104)
Are you a VO member? 10%
Women come to CHW house to buy health commodities and
medicines
95 %
Number of visits required to sell medicines
1 visit
2 visits
3-4 visits
4+ visits
9%
38%
34%
19%
Number of visits required to sell health commodities
1 visit
2 visits
3-4 visits
4+ visits
3.3 %
49 %
32 %
15 %
Financial performance and monthly income of CHW
The reported average monthly income was PPP$ 4.95. The fact that the MoPH does
not allow the CHWs to sell medicines obviously limits the monthly income for these
CHWs. Only 11% of the women reported monthly fluctuations in income. Relatively
fewer CHWs had ever borrowed a loan from BRAC (13%) and 4% of the women
currently had a BRAC loan. Access to BRAC loan was not a big incentive for
becoming a CHW (2% reported this as reason for becoming a CHW). The most
common reason for becoming an SS was ‘social recognition’ (40%) and ‘source of
income for household’ (30%). In the culturally conservative setting of Afghanistan, the
opportunity for increased social recognition is likely to be a strong incentive. The
majority of CHWs (79%) also reported that being a CHW had increased her status
within her household significantly (Table 28).

50
Table 28. CHW monthly income, financial performance and incentives
Monthly income in last month 145 AF – Mean
(PPP$ 5.01)
110 AF – Median
(PPP$ 3.81)
Income received in average month 143 AF – Mean
(PPP$ 4.95)
100 AF – Median
(PPP$ 3.46)
Monthly fluctuations in sales 11 %
SS Loan status
Ever borrowed a loan from BRAC
Current loan with BRAC
More than one loan with BRAC
Loan with another NGO
13%
4%
2%
Why became an SS
Source of income for household
Financial independence
Social recognition
To learn something new
To help my community
To get BRAC loan
34%
4%
40%
12%
8%
2%
Being an SS has given her financial independence 15%
How has BRAC VO membership affected work as SS
People trust me more
Easier to sell products
Need loan to buy products
Has no effect
Not a VO member
Data not available.
How CHW status has changed in household
Importance increased significantly
Importance increased somewhat
No change at all
Importance decreased somewhat
Importance decreased significantly
79%
12%
3%
3%
3%
The average reported time to perform CHW activities is greater in Afghanistan than in
the other study countries. This likely reflects the distance between houses, the more
difficult terrain in some parts of Afghanistan, and the socio-cultural constraints on
female mobility in many parts of the country. Referral to government and NGO clinics
takes the most time (51 minutes on average) while treating TB/DOTS takes the least
(41 minutes) (Table 29).
In terms of the activities performed almost all CHWs had provided family planning
counseling and services in the last month (96%) and a large number had identified
pregnancy and attended a delivery in the last month (88% and 87%, respectively).
Ninety-six percent had referred patients to a BRAC Health Centre in the last month.
They spent the most time providing medicines in the last month (650 minutes).

51
Table 29. CHW reported activities and how she spent her time in the last
month
Activity Provided in
the last month?
Average time
to perform
service
Average
number of
times
provided
T
otal average
number of
minutes
Pregnancy identification Yes – 88%
No – 11%
47 minutes 4 188
Attending delivery and
providing newborn care
Yes – 87%
No – 10%
46 minutes 4 184
Referral to BRAC Health
Centre
Yes – 96%
No – 3%
43 minutes 5 215
Referral to Government,
NGO clinic or hospitals
Yes – 66%
No – 19%
51 minutes 2 102
Treat TB (DOTS) Yes – 73%
No – 12%
41 minutes 8 328
Provide Family Planning
Services and Counseling
Yes – 96%
No – 1%
48 minutes 7 336
Attending refresher training Yes – 85%
No – 6%
Don’t know–7%
182 minutes 1 182
Providing medicines Yes – 80%
No – 17%
50 minutes 13 650
Providing health
commodities
Yes – 63%
No – 31%
45 minutes 9 405
The reported advantages of working as a CHW are better earnings in future (55%),
increased social prestige in the community (16%) and better work environment
(19%). It is interesting that so many women reported better earnings in future given
that 38% reported financial reasons for becoming a CHW. The most frequently
reported disadvantage of working as a CHW was ‘less earnings in the future’ (50%).
It is possible that women initially were motivated to become CHWs for non-financial
reasons but as they serve as CHWs they are more motivated to earn income (Table
30).
Forty-two percent reported no disadvantages of working as a CHW and the majority
of women reported that being a CHW does not affect her household duties.
Relatively fewer women reported other work while also serving as a CHW. The most
commonly reported concurrent work activity was ‘handicraft’ which 13% of CHWs
reported having done or currently doing. For most women (82%) this was seasonal
work as opposed to regular opportunity. Twelve percent reported working in poultry
raising and 8% in agricultural activities. The range of monthly average income from
these activities was roughly PPP$ 50 to PPP$ 109. Obviously this is much higher
monthly income than the average CHW income (PPP$ 4.95) (Table 31).

52
Table 30. Advantages and disadvantages of working as a CHW
N Percent
What are the advantages of working as a CHW
Better earning in future
Better work environment
Can work from home
Better hours
More prestigious among the community
Family approves
Less distance to travel
No advantage
116
39
5
5
34
4
2
2
55%
19%
2%
2%
16%
2%
1%
1%
N Percent
What are the disadvantages of working as a CHW
Less earnings in future
Not good work environment
Must work outside of home
Greater distance to travel
Hours not convenient
Less prestigious in community
Family disapproves
No disadvantage
104
3
1
2
3
4
3
87
50%
1%
.5%
1%
1%
2%
1%
42%
N Percent
How does being a CHW affect household duties
Does not affect h/h duties or responsibilities Have to
work harder in h/h duties
Less time for children
Less time for cooking
Less time for cleaning
Miss out on family events
87
42
28
19
9
20
42%
20%
14%
9%
4%
10%
Competition
The CHWs were asked whether the presence of some service providers (such as
pharmacies or shops that sell medicine, village doctor, TBA, NGOs, public and
private facility etc.) affected their income. Fifteen percent responded that they had
pharmacies or shops that sold medicine and 15% said that there were village
doctors in their community. About 50% respondent said that having the presence of
pharmacy limited their income. Forty-five percent said they felt competition from a
private and public clinic in their area. Forty-five percent and 44% of the respondents
said that the village doctors and TBAs respectively limited their income. For the
traditional healer this rate is 33%. It appears that CHWs faced the most competition
from pharmacies or drug shops followed by public and private health facilities and
village doctors and TBAs.
In terms of other limitations on their performance, 2% reported of having trouble in
selling BRAC medicines or commodities. Only 30% wished they could sell or provide
other health commodities or medicines (Table 32).

53
Table 31. Reported work that woman has done or currently does while also
working as CHW
Type of work Has done or
currently doing
Regular or
Seasonal
Mean earnings in an
average month
Factory worker Yes – .5%
No – 99.5%
Regular – 100%
Seasonal – 0
450 AFG
(PPP$ 15.76)
Handicraft Yes – 13%
No -87%
Regular -18%
Seasonal -82%
1527 AFG (mean)
(PPP$ 52.85)
1700 AFG (median)
(PPP$ 58.84)
Poultry raising Yes – 12%
No -26%
Regular -32%
Seasonal -68%
1306 AFG (mean)
(PPP$ 45.20)
1000 AFG (median)
(PPP$ 34.61)
Agriculture Yes – 8%
No -92%
Regular -76%
Seasonal -24%
3123 AFG (mean)
(PPP$ 108.09)
2000 AFG (median)
(PPP$ 69.22)
Small
business/hawking
Yes – 3%
No -97%
Regular -0%
Seasonal -100%
3140 AFG (mean)
(PPP$ 108.68)
2000 AFG (median)
(PPP$ 69.22)
Tailor Yes – 5%
No -94%
Regular -27%
Seasonal -64%
1490 AFG (mean)
(PPP$ 57.57)
1250 AFG (median)
(PPP$ 43.26)
Domestic worker Yes – 0%
No -100%
No data No data
Midwifery/TBA Yes – 3%
No -97%
Regular -17%
Seasonal -83%
2000 AFG (mean)
(PPP$ 69.22)
2000 AFG (median)
(PPP$ 69.22)
The three most frequently sold medicines were paracetomol, cotrimixazol (antibiotic)
and ORS. The three least frequently sold medicines were oral pills, condoms and
anemia tablets.
The three most frequently sold health commodities were iodized salt, soap and
toothbrushes. The three least frequently sold health commodities were condoms,
toothpaste and soap.

54
Table 32. Competition and other potential constraints to performance
Existence in CHW
area
If yes, these limit
CHW income
Other health providers
Pharmacy or shop that sells medicines
NGOs
Government clinic or hospital
Private clinic
Village doctor
Traditional healer
TBA
29%
7%
18%
16%
32%
Data not available
35%
50%
11%
45%
47%
45%
33%
44%
Have any trouble selling BRAC suggested
medicines or health commodities
2%
Frequency Valid Percent
If trouble selling products, why?
People do not prefer BRAC products
People buy products from shops
Product is too expensive/cheaper elsewhere
People do not trust CHW
2
2
2
1
29%
29%
29%
14%
Wish to sell other health commodities and
medicines
30% Yes
70% No
If yes, which ones (selected)
Rubbing alcohol
Amoxicilin syrup
Ampicillin
Antibiotics
Eye drops
Pampers for children
Plasters
There are several important differences between the BRAC CHW programme in
Afghanistan, Bangladesh and Uganda.
1. Health system differences
The context in which the health system functions is different in Afghanistan. Most of
the population is deprived of very basic facilities, particularly in the rural areas. Most
professional medical doctors prefer to migrate to other countries and those who do
stay, prefer to be in Kabul and are reluctant to go outside Kabul. Afghanistan is a
particularly conservative and follow traditional culture for women. Women do not
leave their home or compound very often and have few opportunities to interact with
others. It is thought that they can be effectively utilized for the betterment of the
community, and serving as a CHW is one way to do that.

55
2. Non-financial incentives versus financial incentives
While the context of Afghanistan suggests that non-financial incentives (e.g., social
recognition, helping ones community) will be driving aspects of sustaining the
programme, it is also recognized that if CHWs cannot earn in the long run it will
jeopardize programme sustainability. However, programme staff feels that the
programme can be considered sustainable in terms of its approach because the
communities have accepted the concept, although there is some variation by
province. BRAC, Afghanistan’s future plan is to ensure generation of income for the
CHWs so that they will continue to work as CHWs. Possible considerations are to
sell some health commodities and to strengthen their involvement with other
programmes thus increasing the incentives for service charges and referrals.
However, because this is a partnership with the government, these changes will have
to be carefully negotiated with that in mind.
3. Public-private partnership
BRAC, Afghanistan can be considered an example of a successful public-private
partnership. BRAC works very closely with the government and the government is
supportive of BRAC’s work and has asked BRAC to participate on many projects.
The careful collaboration and adherence to government regulation has inevitably
changed some of the fundamental aspects of the SS programme. For example,
originally the BRAC, Afghanistan included VO membership as a selection for being a
CHW but this was removed because of government regulations.
Study results for BRAC Uganda
CHP programme - financial and economic analysis
Description of BRAC Uganda
While the GDP per capita in Uganda is $1,500 and its growth rate is 5-7%, 85% of
the Ugandan population lives on less than $1 per day. Furthermore, many of those at
the bottom of the economic scale remain without basic services, including
healthcare. As part of its mission to help the poor and reach those most affected in
post-conflict settings, BRAC began its operations in Uganda in 2006 with 10 branch
offices. Since then it has opened 64 branch offices in 23 districts and created over
3,000 VOs with 76,000 group members and has become one of the largest NGOs in
the country. As of the time of this study it was providing loans to 100,000 women.
BRAC Uganda has two area offices in Kampala and Iganga districts. These two
offices cover seven districts and each area office covers 5-8 branches. Each branch
organizes 60-90 village lending groups. All BRAC borrowers are within 5 km of a
branch office.
BRAC Uganda health programme receives funding from two sources: 1) Living
Goods (LG) and 2) the Master Card Foundation (MCF). LG began supporting BRAC

56
Uganda in 2007. In September 2008 Master Card Foundation provided a two year
$19.6 million dollar grant. At the time of this study the health programme was
working in 80 branches.
BRAC Uganda health programme
BRAC began its health programme in Uganda in July 2007. The BRAC Uganda
health programme is very much based on the model of BHP in Bangladesh. It relies
on female volunteers called Community Health Promoters (CHP) who are recruited
from BRAC VOs. The EHC programme in Uganda has 10 components: health and
nutrition education; safe water, sanitation and hygiene; family planning; immunization
and vitamin A; basic curative care; pneumonia; pregnancy related care; malaria
control; tuberculosis control; and HIV/AIDS.
Uganda faces a range of challenges in its health sector that BHP felt it could directly
respond. A major problem facing the health sector in Uganda is a severe shortage of
health workers. Only 49% of the population lives within 5 km of a health facility.
Reliable and safe drug supply is another problem facing the Ugandan health system.
Irregular supply of drugs and problems related to drug quality, price and stockouts
are common and mismanagement and corruption in the health sector means that
drugs are often sold for several hundred percent of their cost. There is a heavy
reliance on an unregulated private sector which results in overprescribing and missed
diagnosis. Public health sector facilities are overburdened and poorly supplied and
underfunded. The landscape for drug sales in Uganda is fragmented with no chains
or franchise networks. In response to this scenario, BRAC Uganda felt they could cut
out one or more middlemen in the distribution of drugs and health commodities.
Community Health Promoters (CHP)
BRAC Uganda supports networks of CHP to target the provision of basic and
primary health care particularly for women and young children. An additional goal is
to create a sustainable livelihood for the health workers, CHPs, and themselves.
Based on the same BRAC approach in Bangladesh, the programme relies on female
volunteers recruited from BRAC VOs in their communities to provide basic health
education, referrals and the sale of medicines to address basic health problems.
Initially, CHPs were assigned to cover 200-240 households every month and
expected to visit 10-15 households each day. It became apparent that in some areas
this was too large a catchment area and CHPs, because of physical distance
between households and travel times, were not able to visit all of these households.
Since then the coverage expectations are being reconsidered by the programme
staff.
As in Bangladesh, CHPs are selected from among the VOs. They are usually
nominated by their peers in their VO group or they can volunteer themselves. The
training of the CHP is 12 days at the branch office. Before CHPs can sell they must
complete a heath survey of the households in their area. This allows them to build

57
trust and also to better understand the health needs of their community. CHPs have
a probationary period of 30-60 days during which they only do household visits and
health education. After their training and this initial field work they are given an
inventory loan and can begin selling health commodities.
Each branch office has two Community Health Assistants (CHA) which, like Shasto
Kormis in Bangladesh, supervise the CHP and provide additional health services like
ANC. CHAs work 5 days a week, and on Saturdays they are in branch office. Each
CHA supervises 10 CHPs. They visit each CHP at least twice a month. On Saturdays
they are involved in refresher training and dispensing drugs and supplies at the
branch office. The CHAs spend much of their time moving around visiting CHPs at
their houses. The CHA tracks progress of each CHP in a ‘CHP Movement Register’
which is kept in the home of each CHP. They keep a copy and give a copy to Project
Officer.
Project Officers (PO) are responsible for one area which includes 5 or 6 branches.
The POs move around to visit the CHAs and CHPs and supervise their progress.
BRAC provides them with a motorcycle but the distance between some branches is
as great as 40-50 km which is difficult by motorcycle so they often travel by bus.
There are currently four Regional Health Coordinators (RHC), all male Bangladeshis,
who provide overall supervision of the health programme.
Specific costs for CHP programme
As with the analysis for the SS programme in Bangladesh, we collected and
analyzed data from programme staff to determine an average cost of recruiting,
training and managing a CHP per year. We employed a similar methodology as in
Bangladesh and attempted to standardize as much as possible to compare between
the programmes. Programme cost data for Uganda is clearly much higher than that
in Bangladesh. The estimates of US$1,204 (PPP adjusted) in the first year and
US$636 (PPP adjusted) in the second year reflect the higher costs of salaries and
basic supplies in Uganda and also the newness of the programme. Table 33
summarizes the total costs. Appendix 2 provides detailed analysis.
Table 33. Cost to BRAC, Uganda of developing a CHP per year7
Ugandan shillings $US I$US (implied PPP
adjusted)
Total cost in the first year UGX 788,254 $394.13 $1,204.45
Total cost in the second year UGX 416,555 $208.28 $636.49
Total cost in 1st year (min)* UGX 709,429 $354.71 $1,084.01
Total cost in 1st year (max)* UGX 876,080 $433.54 $1,324.90
Total cost in 2nd year (min)* UGX 374,899 $187.45 $572.85
Total cost in 2nd year (max)* UGX 458,210 $229.10 $700.14
7 See Appendix 2 for detailed breakdown of the analysis.

58
Incentive and income structure for CHP
The CHPs are entirely incentivized through the sale of medicines and health
commodities. The programme is considering adding pregnancy identification and
ANC check-up which would provide additional financial incentives for service
charges.
Procurement of inventory for the CHPs requires more time and resources than it
does in Bangladesh. Procurement in Uganda must involve the head office, unlike in
Bangladesh where it is decentralized. The logistics and supply component of the
programme is very complicated and BRAC Uganda is considering hiring two staff to
manage procurement and supply logistics. In Bangladesh, BRAC is able to contract
with pharmaceutical companies once a year to supply all branches. While Uganda
has pharmaceutical companies, they do not have all the essential medicines in stock.
As a result, the programme is considering importing medicines from Bangladesh.
Other CHP products such as diapers and sanitary napkins are imported from Egypt.
The CHPs have an inventory of 30 products including in its product mix a range of
traditional health commodities (e.g., ORS) but also basic consumer items (e.g., soap
and sanitary napkins) to bolster CHP sales and increase the likelihood that they will
achieve financial sustainability.
BRAC purchases the products (or in some cases receives them for free) at the
lowest possible price. The CHP buys the products from BRAC at the wholesale price
and then sells them to her community at a markup thereby making a profit. The profit
margin on health products varies based on the product but in general the margin for
BRAC is 10% and the profit margin for CHPs is 15% (Table 34).
Each CHP starts with a complete set of products. She is provided this initial inventory
based on a revolving loan fund. They are expected to pay back this loan on a
monthly basis. They can qualify again for revolving fund if necessary.
Efficiently and effectively supplying the CHPs is a critical function to ensure
sustainability. Each branch office has a storeroom in which they try to maintain at
least two months of stock. Initially the procurement of supplies was not based on
need and, therefore, the storerooms became very crowded. Even maintaining just
two months of stock means that there is sometimes not enough space as the BRAC
branch offices are not large and all BRAC programmes (health, education and
microfinance) are operating in the same branch office. Some of these items, like
sanitary napkins and pampers, require a lot of storage space.
For renewing supplies, branch offices make a requisition to the Kampala office and
usually receive supplies in a week or less. Transport of supplies is by hired truck
between Kampala and the branches or sometimes manufacturers and distributors
supply the branches directly. Previously CHPs could come anytime to the branch
office to resupply their medicines but recently they were told to come only on
Saturdays. There was concern that because of the time and cost of coming during
the week it reduced the time CHPs had to make their household visits.

59
Table 34. BRAC health programme: branch product and price list
Req Product description Brand Retail pack Price
CHP sells for
Pain, cold, and cough
R Pain reliever (500 mg) Panadol 4 tabs 100
R Cold capsule - 2 tab blister
pack
Cold cap 1 Tab 100
Cough mixture - 100ml Delesed Bottle 2,200
Family planning and reproductive health
R Oral contraceptive Pillplan 1 cycle 300
R Condoms 3 pack Protector 3 pack 250
R Condoms 3 pack Lifeguard 3 pack 250
R Mama kit PSI 1 kit 10,000
Health – Other -
R ORS Medipharm 1 sachet 200
R De-worming Albendazole
(size) 200 or 400 mg
Bendex 1 pill 200
Eye drops - Gentamycrin Tetra 1 bottle 500
R NEW zinc tabs tbf Tbd -
Malaria prevention and treatment
R Anti malarial Kamsidar 3 tabs 300
R Bed net - Long lasting 5x6 Permanent
(LLN)
1 net 12,000
Sanitation and hygiene
R Sanitary pads Perfect 1pad 1,200
R Sanitary pads Allways 1pad 1,500
Disposable gloves Glovemax 1 pair 100
Toothpaste – 70 ml Delident 1 tube 700
R Hand soap - 25 g Samona 25 g Bar 900
R Water purification Waterguard 1 Tab 40
Vitamins and supplements
R Vitamin A 0 2 tabs 200
R Vitamin B 0 6 tabs 50
R Iron tablets 0 6 tabs 75
Multi-vitamin tablet (Vitamin
A, B1+2, D)
0 6 tabs 100
Multi-vitamin syrup – 100 ml
(Vit A, B, B12, C, D,
riboflavin)
Renavit 1 bottle 1,300
R Iodized salt - 1/2 kg Safi 0 450
Wound Care
Antiseptic ointment – 20 g Burnem
cream
1 tube 1,100
Antiseptic liquid – 100 ml Savlon 1 100 ml
bottle
2,150
Miscellaneous
Facial Jelly - 50gms Samona 50 g bottle 1,000
Diapers - Size 2 (3-6 kg) Pampers 1 Diaper 3,500
Diapers - Size 3 (4-9 kg) Pampers 1 Diaper -
Cotton 50 g 0 1 50 g pkg 500
Laundry Soap Bar- 600 g Mukwano 1 XX gm
Bar
1,250
Source: BRAC Uganda office, May 10, 2009

60
Factors that influence CHP performance
Identifying and understanding the potential factors that influence performance of
CHPs in terms of their monthly sales and profits is critical to thinking about whether
or not the CHP programme can and will become financially sustainable. In order to
understand this both quantitative and qualitative research was carried out.
Table 35 shows the selected socio-demographic characteristics of the CHPs in the
sample. The majority of CHPs are current VO members (91%) and the majority is
above the age of 35 years. Nineteen percent of the CHPs are either divorced or
widowed. The mean number of family members (6.6) is quite high (Table 35). Sixty-
nine percent of the respondents reported that the income they earned as an SS
made a big difference to the household (Table 36).
Table 35. Socio-demographic characteristics of CHPs in study sample
(N=155)
Indicator Average
Age
20-24 years
25-29 years
30-34 years
35-39 years
40 and above
3.9%
13.5%
23.2%
30.3%
29%
Marital Status
Married or living together
Divorced/separated
Widowed
Never married
81%
12%
7%
0.6%
Number of years of schooling 10.10 (Mean)
Number of family members 6.6 (Mean)
Number of children 4.35 (Mean)
Number of children under 2 years of age 19%
Currently a VO member 91%
Table 36. Household financial status and monthly income
Monthly household income
Monthly income is always greater than expenditure
Monthly income is always less than expenditure
Monthly income is equal to expenditure
38%
3%
6%
Importance of CHP earnings for family
H/H could not continue without SS income
SS income makes a big difference
SS income makes no difference
18%
69%
7%
How CHP income is used
Spend on children
Give to husband
Pay school fees
Buy food
Save
Pay back loans
40%
1%
12%
20%
7%
20%

61
Examining performance-related characteristics suggests that most CHPs are active.
They reported that on average they could visit nine households a day and work just
over 3 hours each day. Eighty-five percent replenish their supplies between refresher
training and 65% have used a BRAC loan to buy CHP supplies. The number of visits
that are required to sell medicines is quite high – 60% of the CHPs reported to make
four or more visits to sell medicine. For health commodities, such as bed nets, even
a larger proportion reported to make four or more visits to sell health commodities
(Table 37). This could reflect the generally higher price for commodities or that the
CHP does not always carry these commodities with her when she makes her
household visits. Also, commodities such as sanitary napkins and pampers are quite
popular and some CHP said that these were not always in stock at the branch office.
Table 37. CHP performance-related characteristics
Years working as a CHP
Less than one year
1-2 years
More than 2 years
23%
51%
26%
Reported number of h/h visits possible per day 9.2 (average)
Hours work as CHP per day 3.2 hours (average)
Replenish CHP supply between refresher trainings 85%
Bought medicines or health commodities outside BRAC 0.6%
Loan used to buy CHP supplies
If yes, amount spent from loan on supplies
65% (of 134 women)
UGX 169,204 – Mean
UGX 180,000 – Median
Women come to CHP house to buy health commodities
and medicines
98%
Number of visits required to sell medicines
1 visit
2 visits
3-4 visits
4+ visits
2%
9%
30%
60%
Number of visits required to sell health commodities
1 visit
2 visits
3-4 visits
4+ visits
1%
9%
18%
72%
Average monthly income and financial performance for CHPs
We asked CHPs what their income was in the last month from working as a CHP.
The mean was UGX27,680 (PPP-adjusted $42.29) and the median was UGX24,000
(PPP-adjusted $36.67). This was slightly lower than the monthly income they
reported in an average month (UGX38,222 or PPP-adjusted $58.40) (Table 38).
Sixty-seven percent of the respondents reported that there was monthly fluctuation
in sales. In our qualitative interviews we found that sales tend to increase around the
time when children are going to school. At these times, there is an increased
demand for sanitary napkins, soap, and other basic health and medicines that
children take with them to school.

62
The majority of the CHPs has ever had or currently had a loan with BRAC (88%).
Thirty-six percent of women had more than one loan with BRAC. In the qualitative
interviews with high and low performing CHPs, the discussion of the ability to repay
the CHP loan for supplies in addition to repaying another BRAC loan came up
frequently. It was clear that some CHPs simply could not keep up with payments for
both loans. The CHAs complained that because the Credit Officer (CO) was trained
to collect repayments for the microfinance loan, the CO always reached the CHP first
and successfully got her microfinance loan repayments. Once the CHA arrived, the
CHP often said she had no money left to repay her CHP supply loan.
“The problem is that I have to meet my weekly installment but also need
money to resupply. If the community requests a medicine from me but I
don’t have it and don’t have the money to buy it then the community will
lose trust in me.” CHP In-depth interview
Table 38. CHP monthly income, financial performance and incentives
Monthly income in last month UGX27,680 – Mean
(PPP-adjusted $42.29)
UGX24,000–
(PPP-adjusted $36.67) Median
Income received in average month UGX 38,222 – Mean
(PPP-adjusted $58.40)
UGX30,000–
(PPP-adjusted $45.00) Median
Monthly fluctuations in sales 67%
CHP Loan status
Ever borrowed a loan from BRAC
Current loan with BRAC
More than one loan with BRAC
Loan with another NGO
99%
88%
36%
0.8%
Why became a CHP
Source of income for household
Financial independence and social recognition
To learn something new
To help my community
22%
3%
47%
27%
Being a CHP has given her financial independence 80%
Ever considered quitting work as CHP 20% - Yes
85% - No
If yes, why considered quitting (N=26)
Payment is too low
CHP role is difficult
96%
4%
How has BRAC VO membership affected work as CHP
People trust me more
Easier to sell products
Need loan to buy products
Has no effect
31%
36%
1%
32%
It is interesting to note that 22% of the women said that they became a CHP to
provide a source of income for the household. Forty-seven percent said that they did

63
it “to learn something new” and 27% reported it was in order “to help my
community”. However, 80% reported that being a CHP had given them financial
independence. This may suggest that her expectations about financial returns to
being a CHP were much lower before she joined and that she did not necessarily
appreciate the extent to which she could earn money as a CHP.
For many women the desire to help their community is likely linked with the health
problems they have seen in their communities, particularly the devastation of
HIV/AIDS. For many women, being a CHP is providing them a chance to be a part of
the health workforce in their communities.
“Since I was born I wanted to be a nurse but I failed. But I have that interest.
I had it inside my heart. So, that is why I raised my hand up and they picked
me.”
Of the 20% of women (n=26) who reported that they had ever considered quitting
their work as CHPs, 96% said it was because the payment was too low. So it
appears that for a sub-set of CHPs, their financial expectations or goals are not
being met.
In terms of the CHP activities that are being carried out, almost all of the CHPs
reported providing pregnancy identification and attending delivery and providing
newborn care in the last month. Almost all had sold medicines and health
commodities in the last month. Ninety-four percent reported attending refresher
training in the last month (Table 39).
It appears that in addition to attending the refresher training, CHPs are spending the
bulk of their time selling medicines and health commodities. They reported an
average of 46 medicine sales and 38 health commodity sales in the last month. It is
striking that 72% of CHPs must make four or more visits to sell health commodities.
This likely includes visits to obtain payment for the commodity from women who buy
products on credit.
The three most frequently reported medicines that are sold are Bendex, Deworming
tablets and Panadol. The least frequently sold medicines are Kevlon, Pill Plan (oral
contraceptives) and vitamin syrup.
In the in-depth interviews with CHPs it became clear that the size and price of some
of the CHP inventory were not competitive with what the market was selling.
Question: Are there any products you aren’t selling or that people don’t want?
Response: “Iron tablets are not moving – have to explain a lot about them….; Kevlon
is not moving because people can’t buy the whole tin…just do a cotton swab
(individual treatment) but they don’t pay for it. Samona (jelly) is expensive and of small
size. So, the community is complaining about the size and price is too expensive.
The community wants a bigger size.”

64
Table 39. CHP reported activities and how she spent her time in the last
month
Activity Provided in
the last
month?
Average time
to perform
service
Average
number of
times provided
Average total
number of minutes
per month
Pregnancy identification Yes – 98%
No – 2%
25 minutes 7 175
Attending delivery and
providing newborn care
Yes – 97%
No – 3%
28 minutes 7 196
Referral to Government,
NGO clinic or hospitals
Yes – 88%
No – 12%
21 minutes 11 231
Attending refresher
training8
Yes – 94%
No – 6%
231 minutes 28 461
Selling medicines Yes – 99%
No – 1%
23 minutes 46 1058
Selling health
commodities
Yes – 97%
No – 3%
21 minutes 38 798
In terms of how CHPs coped with slow moving products, some adapted their pricing
structure to sell products.
Question: Do you ever lower the price just to sell it?
Response: “Yes…like iron tablets…regular price is 10 tabs for 100 shillings now I am
selling 15 tabs for 100 shillings. People like the ORS but they say that these are given
for free in the hospital… Now I have to lower prices.”
Question: Do you ever give something for free if buying something else?
Response: “No.”
Question: Do you sell things for less profit to get rid of things?
Response: “When things are approaching the expiry date I will sell it at a lower price
rather than throwing it out.”
CHP Turnover
The BRAC Uganda programme is very new so there are less data to draw upon than
in the Bangladesh programme. While the CHPs have been recruited since 2007, they
have only been fully functioning and selling medicines since 2008. Of the 180 original
CHPs that were trained, 60 have dropped out.
8 Refresher trainings are held once a month; it is likely that some SSs included visits to the branch office to
resupply in their response to this question which is why there are more than one time listed in the last
month.

65
The programme identifies a CHP as a dropout if they are not interested in selling
products and are not attending the refresher training. The programme originally
thought women who have shops outside the home would be successful CHPs but
that was not the case. There was too much competition in terms of their time. As a
result BRAC has changed the selection criteria to those women who have shops in
their house or community but not in the market. Another perception about the high
dropout rate is that the CHP’s expectations about the income they could earn may
have been too high.
While it is difficult to determine definitely the threshold at which a CHP drops out of
the programme, the survey did ask CHPs if they would work for a salary and if so
what that minimum salary would be. All the respondents said that a fixed monthly
salary would motivate them to be more active as CHPs. Of those who responded as
to what that minimum salary would be (n=97), 76% said that it would be in a range of
150,000 to 250,000 shillings per month (PPP-adjusted $229 - $382).
When asked what the advantages of working as a CHP, about 1/3 replied: better
earnings in future, and can work from home. Interestingly, when asked about the
disadvantages of working as a CHP, 57% said that less earning in future while 31%
reported that there are no disadvantages to being a CHP. Ninety-one percent
reported that being a CHP has no effect on carrying out her household duties and
responsibilities with only a few women reporting it leaves them less time for caring for
their children and cooking and cleaning (Table 40).
Question: What do you like the most about being a CHP?
Response: “Community looks at me as the person who brings ideas and knowledge
to them. They come to me with issues and questions and I am able to give ideas to
them. Been looked at by community so good. … People come to me and I help
them and they go out happy and preach to others that I can help the community.”

66
Table 40. Advantages and disadvantages of working as a CHP
N Percent
What are the advantages of working as a CHP
Better earning in future
Better work environment
Can work from home
Better hours
More prestigious among the community
Family approves
52
18
50
17
17
3
33.1%
11.5%
31.8%
10.8%
10.8%
1.9%
N Percent
What are the disadvantages of working as a CHP
Less earnings in future
Not good work environment
Must work outside of home
Greater distance to travel
No disadvantage
87
3
1
15
47
57%
2%
.7%
9.8%
31%
N Percent
How does being a CHP affect household duties
Does not affect h/h duties or responsibilities
Have to work harder in h/h duties
Less time for children
Less time for cooking
Less time for cleaning
Miss out on family events
141
5
3
3
2
1
91%
3%
2%
2%
1%
0.6%
Would fixed monthly salary encourage you to be more active as
CHP?
100% (N= 155)
Frequency Percent
What is minimum monthly salary that you would require?
100,000 – 249,999 (PPP-adjusted $152-382)
250,000-500,000(PPP-adjusted $382-764)
Above 500,000 (PPP-adjusted $764 and above)
74
22
1
76%
23%
1%
Economic opportunity costs for CHPS
Understanding the economic opportunity cost structure in Uganda, and identifying
potential differences with Bangladesh are critical for designing the CHP programme
that is financially sustainable. In Bangladesh, women are more likely to be involved in
income generating activities within their home and then the males in the family will
transport it or sell it in the market. In Uganda, women are more likely to go out
themselves and run the business and/or market their products directly. Societal
differences between the treatment and expectations of women also contribute to
this. For example, there are many more female-headed households in Uganda
compared to Bangladesh.
In this economic and societal backdrop, BRAC Uganda health programme is tackling
a potentially difficult question. Can the CHP earn a sustainable income as a CHP? Or
is it simply complimenting other work she is already doing and therefore providing
value added to create a sustainable situation. The concept of sustainable income

67
and CHP incentives may need to be thought of differently in the Ugandan context
because the economic opportunity cost for her may be different. We asked the
respondents to identify what other jobs are available to them if they were not a CHP.
The most frequently reported alternative jobs are small business/hawking, raising
poultry, and agriculture (Table 41). Many CHPs are already engaged in additional
jobs such as these and in qualitative interviews with them, they report that the
biggest advantage of being a CHP is that it is easily integrated into the work or
business they are already doing. If they are a seamstress at home or sell second
hand clothes out of their house, they already have a clientele that is coming to their
home. Interviews with BRAC Uganda programme staff mention that while Ugandan
women are very mobile, the programme has found getting them to visit 10-15
households per day difficult. This may be due to their need to stay fixed at home or in
a particular spot in the village to carry out their other work.
Table 41. Availability of other jobs as reported by CHPs
Other jobs available Percent reporting availability
Handicraft 11%
Poultry raising 27%
Agriculture 20%
Small business/hawking 29%
Tailor 5%
Domestic worker 0.8%
Midwifery 8%
Non-financial incentives may be different in Uganda as well. In Bangladesh women
are more likely to stay at home and work while Ugandan women are out running
businesses outside home…this sets up a situation where Ugandan women are
“sacrificing income” versus Bangladeshi women who are “sacrificing
time”….therefore, their opportunity cost calculation is different.
Competition and potential constraints on performance
Both the quantitative and qualitative data suggest that CHPs face competition in their
catchment areas. The most frequently mentioned source of competition were
pharmacies or shops that sell medicines, private clinics, government clinics or
hospitals, and traditional healers. The competition that they felt limit their income as
CHPs are the pharmacies and the government and private clinics - 88% of the
women reported that the presence of a government clinic limits their income. Some
clinics are providing certain items that may be donated by international agencies of
NGOs for free (condoms for example). This creates difficulties for CHPs who are
trying to sell the same or similar products. Forty-six percent of those women who
said they had trouble selling BRAC CHP products said that it was because people
did not prefer BRAC products.
When asked if they wished to sell additional health products, 67% replied that selling
additional products or medicines would be preferred. The list of these products was

68
quite long and a few of the more frequently mentioned responses are shown in Table
42. There are also many non-health items (lotion, cooking oil, rice, etc) that were
mentioned and 69% of the women responded that they would like to sell additional
non-health items as a CHP.
Table 42. Competition and other potential constraints to performance
Existence in CHP area If yes, these limit
CHP income
Other health providers
Pharmacy or shop that sells medicines
NGOs
Government clinic or hospital
Private clinic
Village health team
Traditional healer
TBA
74%
29%
62%
74%
19%
44%
29%
78%
48%
88%
65%
48%
9%
11%
Have any trouble selling BRAC suggested
medicines or health commodities
36%
Frequency Valid Percent
If trouble selling products, why?
People do not prefer BRAC products
People buy products from shops
Product is too expensive/cheaper elsewhere
People do not trust SS
26
18
10
2
46%
32%
18%
4%
Wish to sell other health commodities and
medicines
67% Yes
33% No
If yes, which ones (selected)
Antibiotics
Antimalarials
Better contraceptives
Fansidar
Ulcer medicine
Are there non-health-related products you wish
you could sell or women ask about?
69% - Yes
31% - No
If yes, which ones (selected) Cooking oil
Sugar
Tea
Toilet paper
School books for children
Lotions
Rice
When asked to describe the biggest challenges they face, one CHP replied:
“Some people ignore you in the community – especially new people in the
community. In the beginning people don’t listen to you. And during the rainy
season it is very hard to travel and visit -- in rainy season it is tough.”
In response to questions about being a ‘volunteer’ and how the community views
their work as volunteers, many replied that there were misunderstandings in the

69
community that the CHPs were volunteers. Some community members believe that
BRAC is giving the CHPs the medicines and commodities for free and she is selling
them to make personal profit.
Table 43. Selected descriptive statistics for high performing CHP (n= 55) and
low performing CHP (n=67)
Indicator Average for High
Performing CHP
(n=55)
Average for low
performing CHP9
(n=67)
Age 35 years 36 years
Number of years of schooling 10 10
Current marital status
Married/living together
Divorced/separated
Widowed
82%
11%
6%
84%
9%
7%
Number of family members 7 6
Currently a VO member 91% 91%
Monthly household income
Monthly income is always greater than expenditure
Monthly income is always less than expenditure
Monthly income is equal to expenditure
24%
4%
9%
43%
4%
3%
Currently have BRAC loan 87% 85%
Currently have more than one BRAC loan 17% 41%
Loan used to buy drugs and SS Supplies 58% 62%
Replenish supplies between refresher trainings 91% 80%
Buy medicines or health commodities outside BRAC 0% 2%
Average hours worked per day 3.6 2.8
Any health related training outside BRAC 13% 24%
Ever considered quitting as CHP 15% 18%
Minimum monthly salary (without commissions)
required
UGX 100,000-
249,000 (49%)
UGX 250,000-
500,000 (35%)
UGX 100,000-
249,000 (93%)
When comparing basic descriptive statistics of low and high performing CHPs it is
striking how many similarities there are between the two groups (Table 43). One
major difference is whether or not the CHP has more than one BRAC loan. In this
analysis, having more than one BRAC loan suggests that the CHP is more likely to
be a low performer. This could reflect that the CHP is behind on paying back her
loans and, therefore, does not have the financial resources to buy new supplies. This
sets up a vicious cycle, for without resupplying she cannot sell items and cannot earn
any income. In our qualitative interviews with low performing CHPs in the field this
was a definite problem. Two of the women we talked to had not resupplied for the
past 5–7 months because they were having trouble paying back their loans.
9 Defined as earning less than 24,000 on average per month

70
Another interesting difference is that the high performers are more likely to resupply
between refresher training than low performing CHPs (91% versus 80% respectively).
This may reflect that high performers have access to the financial capital required to
resupply whereas the low performers might not.
Perhaps not surprisingly, high performers reported of working on average a greater
number of hours each day than low performing CHPs (3.6 hours versus 2.8 hours). If
low performers do not have medicines or commodities to sell, this will automatically
limit the amount of time and the activities they are able to participate in as CHPs. In
the in-depth interviews it became clear that high performers used different marketing
techniques. Some purposefully did not return to areas for two weeks in order to build
up demand; others always took different routes to the market to introduce herself to
a larger community of clients; while others used her successful cases and cures to
promote her business through word of mouth.
Even high performers had suggestions and ideas on how they could perform better
and earn more.
Question: How do you think you are doing as a CHP? Can you earn more?
Response: “Yes, if I get a big stock. I am intending to open a shop next year. I could
sell more if I had a big store, and the timing of buying. Now [in the] changed system
we have to buy only on Saturday. Before I could buy every morning. New system will
be difficult because on Friday I go for prayers (away and overnight) and getting back
on Saturday is difficult. For me now I don’t know how I’m going to do it. Now I have
to wait until Saturday and I am out of soap and pads. Want to purchase more
Bendex I only have one packet and I know it will get purchased today.”
Many CHPs expressed a desire to learn more and have more training. “[We]…want a
two-week training to understand more about checking blood pressure,
temperature…” A focus group discussion with high performers suggested that BRAC
build a clinic with a trained doctor and then each CHP would take shifts each week
to assist the doctor as a way to earn more income.
Key successes and challenges
While the BRAC Uganda programme is heavily modeled on the Bangladesh
programme, there are several important differences and adaptations that the BRAC
Uganda programme has made since it began operating its health programme in
2007.
1. Products they sell are different
The health situation is different in Uganda than in Bangladesh and as a result,
variation in not only the products offered is required but the training materials must
also be adapted to reflect this. Malaria treatment is necessary everywhere in Uganda,
unlike in Bangladesh. In Bangladesh SSs are doing ARI treatment while in Uganda

71
they currently cannot provide medicines for ARI treatment. Pit latrines may be more
popular than slab latrines in Uganda and at the moment, the BRAC programme is
not providing latrines – just health education on sanitation. However, it suggests the
need to do careful background research on local needs and preferences in Uganda
before introducing new products.
2. Health policy effects
A challenge faced by the BRAC health programme has been the health policy history
of Uganda. The Ugandan government tried to implement volunteer village health
teams (VHT) and drug distributors as part of its earlier programme. However, there
were very high dropout rates and the programme, which was poorly funded, did not
do well. Because of this generally negative experience, the Ugandan government and
communities were skeptical that another volunteer community based approach
could work. There is still a long memory of this as a failed health experience. In
response to this, BRAC has worked very hard to collaborate closely with the Ministry
of Health and gain the trust of the government. They now have a full-time position of
a liaison with the ministry of health at district and national level.
Another health policy difference is that in Bangladesh SS can do TB identification
and treatment. Uganda does not have GFTAM money and, therefore, CHPs do not
provide TB treatment.
3. Difficult to recruit and hire trained medical officers
The Uganda health programme currently has a Medical Officer from Bangladesh
because of difficulties the programme has had successfully recruiting and retaining a
Ugandan qualified medical officer. There is a dearth of trained medical personnel in
Uganda and among those that do exist, it is a challenge to incentivize them to move
to rural areas where BRAC is operating. There is also a lot of competition with other
NGOs, both national and international, which require medical doctors for their
programmes and pay higher salaries than BRAC traditionally pays its Medical
Officers. BRAC may have to reconsider its salary structure to ensure that it can
recruit and retain Ugandan medical officers.
4. Procurement is more challenging and time consuming
There are no pharmaceutical companies in Uganda that are producing drugs so all
drugs for the programme must be imported. This results in frequent price changes
which is frustrating to both the CHPs and their clients. It is also requires a huge time
commitment on the part of BRAC Uganda Health programme as prices for drugs
need to be renegotiated – sometimes as often as every 6-8 weeks. In Bangladesh
there is a strong network and connectivity in Dhaka with pharmaceutical companies
and depots that make distribution of products easier. In Uganda this is simply not the
case. Furthermore, programme production of materials is more centralized in
Uganda than it is in Bangladesh due to lack of infrastructure and materials. As a
result producing items such as CHP training manuals are more time consuming and
often more expensive in the Ugandan context.

72
5. Economic opportunity cost and motivations of CHPs are different
There is a perception among BRAC staff that women in Uganda are more
commercial-minded and very much motivated by financial incentives as opposed to
non-financial incentives. The programme suggested the need to do more
motivational work with CHPs in terms of identifying appropriate non-financial
incentives.
6. BRAC is becoming better known in Uganda
At the time of this study, BRAC was the largest NGO in Uganda. However, not all
Ugandans understand what BRAC is or what they do. Many have a mixed or limited
understanding of BRAC - “it is microfinance” while others say “it is health” and still
others ask, “what is BRAC?” When the Programme Manager first recruited 60 staff,
he advertised in the newspaper and received only 12 applications. Now 10-20
women show up each day to submit their application even when no job has been
advertised. So, clearly the word of mouth of BRAC and what it is doing is creating
demand for jobs.
The lack of common understanding about BRAC and what it does has negatively
impacted their performance, according to several CHPs. Several CHPs said that
community education and sensitization about BRAC and what they do would
improve their performance and sales. Clarifying to the community that CHPs are
volunteers and are not paid salaries and given the drugs they sell for free may help
alleviate some of the mistrust and suspicion that some CHPs experience in their
communities.
7. Strong relationship with the government of Uganda
BRAC currently enjoys a strong relationship with the Ugandan government and their
ministry of health. However, when BRAC’s programme manager first met the state
minister for health he was asked how many ambulances BRAC was bringing to
Uganda and was reminded that the previous health volunteer programme failed in
Uganda. Despite this, BRAC got written permission to operate in 10 branches to
start and then in 85 branches. BRAC signed an MOU and over time the relationship
between BRAC and the government has become strong. While there is no financial
support from the government, they are providing moral and other support. The state
minister for health wrote a letter to districts to ensure their cooperation with project.
8. Issues related to sustainability
Programme staff in BRAC Uganda expressed concern about future and long-term
donor support and they are actively thinking about the issue of programme
sustainability. They expect it will take seven to eight years to achieve sustainability.
The basis for this calculation is to project average household expenditure on CHP
products each month. If each household spends $1 each month on CHP items, then
each CHP would be turning over $200 per month by the year 7 or 8 of the project.

73
Currently 15% of profit goes to the CHP, so for $200 this would mean a profit of
$35. BRAC has a 10% margin which would give them $20 per CHP. Multiplied by
1,500 CHPs (their target in 8 years) = $30,000 per month. The calculation of
operating expenses for the health programme (including training cost for CHP) is
$75,000 per month. This is the current philosophy and thinking of BRAC Uganda
regarding financial sustainability of the health programme.

75
Section
DISCUSSION AND CONCLUSION
Study limitations
This study is potentially affected by several limitations and it is important to consider
the results presented in this report. Firstly, in some cases, data were not available or
were incomplete, and despite our best efforts we were unable to collect it. For
example, data related to revenues and expenditures for BHP over several years were
not complete or were collected in different forms in different years. This has not
allowed us to reliably examine trends over time.
Secondly, some questions in the interview survey such as those about monthly
income might be sensitive, and therefore, lead to misreporting and possibly over
reporting. It is difficult to cross-check this reliably.
The study did not consider the patient or community perspective. For example, we
cannot comment on how patients perceive the quality of the SSs – clearly an
important influence on SSs monthly sales and performance.
Finally, we only included currently active Shebikas in the study. Because we did not
include those Shebikas who had dropped out, this study sample may be biased
towards women with more positive experiences. While we did try to ensure inclusion
of high and low performing SS, our respondent selection and study results may still
underestimate the constraints faced by SSs.
Despite these limitations, the findings from this study yield useful data to inform the
question of sustainability and generalisability of BRAC’s SS model. These are
discussed below.
Sustainability
The BHP operating model clearly relies on the successful recruitment, training, and
retention of female CHWs in each of the countries included in this study. The
rationale for BRAC’s approach is that community-based financial incentives of a
volunteer community health workforce can achieve wide programme coverage and
respond to community EHC needs while providing income opportunities to its female
volunteers.
5

76
The costs to producing a SS or CHP or CHW are very real. This study suggests that
in the first year it costs BRAC Bangladesh US$ 89 to recruit, train, and supervise a
SS. In Afghanistan and Uganda the costs are significantly higher (US$247 and
US$374 respectively). While the costs in the second year drop, if BRAC has trouble
retaining SS and experiences high dropout rates, the overall costs to the programme
increase significantly.
Trying to minimize dropouts, therefore, is key to programmatic and financial
sustainability. There are several ways to approach this and as the literature suggests,
both financial and non-financial incentives may be required.
In terms of financial incentives the quantitative and qualitative data suggest several
things. Firstly, SSs are financially motivated to sell medicines and health commodities
and the time use data suggest they spend significant proportions of their time
engaged in these activities. Secondly, the majority of SS/CHP/CHW would like to
expand the products they sell to include more health and non-health products.
These products include medicines like antibiotics, malarial treatment, ulcer treatment,
and non-health products like school supplies for children, sugar, and cooking oil.
Clearly there is a potential programmatic trade-off here between increasing the SS
ability to respond to community demand and in turn increase her sales and monthly
income, while still ensuring that the preventative and health education aspects of the
programme are being sufficiently addressed.
Increasing financial performance and addressing competition
The SSs in each of the settings reported that they felt competition from pharmacies,
private clinics, and other providers limited their income. It is important that the
comparative advantage that the SS brings be maximized to address this
competition. The first comparative advantage is that the SS provides household
delivery of care and products. By expanding the product range or mix of products
that she provides to households she could potentially increase her income and
compete effectively in the local market. In each of the settings the Shebikas
expressed an interest in expanding their product line to meet community demand. It
appears from the qualitative and quantitative data that BRAC is mostly meeting the
demand for medicines. However, there are a range of health commodities and non-
health products that SS said the community was demanding. There are two major
trade-offs here – the first is in terms of additional programme costs for procuring,
storing, and transporting additional products to branch offices. If some of the goods
were produced by BRAC enterprises this might keep costs down, at least in
Bangladesh. The second is the potential for the SSs to spend more time selling
products and less on the preventative aspects of her role as SS. Questions about
whether or not SS are becoming a sales workforce might be raised.
The second comparative advantage that the SS has is the BRAC name or ‘brand’.
Ensuring that BRAC products are responding to community demands and are seen
as high quality will be critical to SS sales performance. In each country some women

77
said that they had difficulty selling products because people do not prefer BRAC
products (29% in Afghanistan, 46% in Uganda and 25% in Bangladesh). Performing
market research to better understand the perception of the BRAC brand for
products could improve the likelihood that SSs can sell products and compete with
other distribution outlets. The fact that 10% of SSs in Bangladesh reported buying
products outside BRAC for sale as SS may be in response to a particular community
demand for a product. In order to capitalize on both of these comparative
advantages (household delivery and BRAC brand), additional training in social
marketing could help SSs maximize their monthly sales and income.
Non-financial incentives, while more difficult to measure, should not be ignored.
Currently non-financial incentives are couched in difficult to measure concepts such
as increased social recognition or prestige within the home or community. The
survey results suggest that increased social recognition is important: in Bangladesh,
18% of SSs said that increased social recognition was an advantage to work as an
SS; in Uganda it was reported to be 11%; and in Afghanistan it was 16%. It is
possible that the incentive for social recognition wanes over time and as social
recognition improves or is perceived to be achieved. This could suggest that
additional non-financial incentives might be required to continue to satisfy or validate
social recognition of the SS. There are tangible non-financial incentives that the
programme could consider as additive to the overall incentive package. These may
include certificates for special training received, certificates recognizing extremely
high sales in a period or of a particular product or activity, providing a saree or
salwar-kameez in recognition of long-term service. Clearly the costs of these would
need to be carefully assessed, however, these tangible non-financial incentives might
validate for the SS the appreciation that both BRAC and her community has for her
volunteer services.
Another non-financial incentive is to offer additional specialized training for SS. In all
of the settings, many SSs requested additional and specialized training to learn
additional health skills such as taking blood pressure or taking temperature. Such
trainings would certainly increase her skill set and potentially make her more
competitive in the community. The trade-off here is that training add cascading costs
to the programme – from master trainers to supervisors – and not all SSs will be
interested in receiving additional training. In addition, if BRAC responds to the
request of some SSs for additional and more complicated health activities, it might
need to revisit the level of literacy required for SSs.
An additional and practical training may be considered in social marketing and
inventory management. As BRAC expands its product line in places like Uganda,
those women who have social marketing skills tend to be higher performers. Others,
who are unable to manage their inventory or cannot pay back their loan(s), may not
be able to perform up to their potential. These women may benefit from some
training in social marketing as well as basic financial management.
Underlying this is the challenge of incentivizing SSs to perform a greater number of
tasks while being careful not to overload them. Not all SSs necessarily want to do

78
more. It is important to note the high rates of satisfaction associated with being an
SS: 43%, 31%, and 42% of SSs in Bangladesh, CHPs in Uganda and CHWs in
Afghanistan said there was no disadvantage to being a CHW in BRAC programme.
Selection/recruitment of SS
The motivation for becoming an SS varies between countries. For many SSs, the
initial motivation is a financial one. Several programme staff raised the question as to
whether or not it is beneficial to recruit SS with more social marketing skills. Women
who have received other health training or have worked for other NGOs seem to
perform better in both Bangladesh and Uganda. This may serve as an informal
additional assessment tool when recruiting SS.
Lessons learned and recommendations
• Generate and foster government support
Because of its long duration, the Bangladesh programme enjoys a history of
developing an effective working relationship with the government. The fact that
the BHP works in partnership with the government of Bangladesh on several
national programmes is testament to this. As BRAC moves into new countries,
establishing and negotiating its relationships with the government and especially
the ministry of health is extremely important. Experience also suggests that this
is very time consuming and can be unpredictable in how long it takes. BRAC has
been very effective in building public-private partnerships in both Uganda and
Afghanistan. In Afghanistan the process was guided by the establishment of the
BSPH. In Uganda, however, BRAC had to develop these relationships from the
ground up. There are important lessons here that other country programmes can
learn from.
• Be willing to change the programme and be flexible
Country experience suggests that the programme must make adjustments to
respond to the local environment. In Uganda, for example, the number of
households is being reconsidered because the geographic distance and terrain
is very different than in Bangladesh. In Afghanistan, it was difficult for CHWs to
visit 120 households in a month. This was due to both the geographic distance
between homes, populations clustering around water sources, the conservative
nature of the culture that makes travel for women difficult, and ethic differences
within catchment areas making it difficult for some women to visit the homes of
other ethnic groups/identities.
Clearly, not all SSs/CHPs/CHWs are VO members. The programme will have to
continue to be flexible about this. Not all women require a microfinance loan.

79
• Have a clear assessment of SSs’ performance and expectations
In some settings, the programme clearly defined high and low performing SSs in
terms of monthly sales and monthly income. In other places, performance
measurement was less clear. Making performance assessments more
transparent may help the SSs/CHPs/CHWs set individual performance goals.
Clearly assessing performance and using this as a form of motivation for the SSs
is another mechanism to recognize goal achievement with non-financial means.
Another issue that this analysis raises is how the performance of
SSs/CHPs/CHWs should be assessed. If it is solely on the sale of products and
monthly income it may be too narrow a measure and neglect a host of other
activities and benefits (e.g., health education and counseling) that the SSs
provides her community.
• Don’t lose sight of non-financial incentives
As the programme aims to be financially sustainable, it should not do so at the
neglect of the range of important services that SS provide in her community
many of which are not immediately financially remunerated. While it may be
possible to financially model the appropriate product mix and profit margin to
make the programme financially sustainable, their needs to be continued and
careful tending to the other aspects of how SS spend her time. There is a
potential tension between maximizing the sales of medicines and health
commodities versus providing basic health education and preventative care in
the community. Currently, responding to disease specific issues, such as TB
management, is better incentivized than general preventative health. It may be
important in future to think about ways to incentivize general preventative health
interventions. The SS is not just a medicine seller; she plays a vital role in
promoting the overall health and well-being of her community.
• Empower SSs to manage inventory appropriately
Ensuring regular and affordable supplies to the SS is critical to her successful
functioning. In Bangladesh this system has been developed over years and is
well-functioning. The SS is given revolving capital and not a loan. In Uganda, the
programme started by giving each CHP a loan for her first set of supplies. She
was expected to pay this back monthly in 40 installments. For some women, this
has not been a problem. These women may be better off or may simply be
better managers of their money or have higher sales. For some women with
more than one loan, they fall behind on repayments for their CHP supply loan
and get into a situation where they are unable to resupply and therefore cannot
earn the money they require to repay the loan.

80
• Re-examine role of microfinance as an incentive
SSs are to be recruited from VOs and are provided access to microfinance loans
to support not only their work as an SS but to other economic opportunities as
well. The data from this study suggest that not all SSs take advantage of their
access to microcredit. Nineteen percent of the SSs in Bangladesh; 4% of CHWs
in Afghanistan; and 60% of CHPs in Uganda reported using microfinance loans
to support their work. It is thought that one value of developing the SS out of the
microfinance model is that it empowers SS to perform better with a VO
supporting her and the social capital that comes with that. The fact that so many
SSs are not active VO members raises a critical question of how important SS
links with VO are. When asked how being a VO member affects their role as
SS/CHW/CHP: 30% of SSs, 32% of CHPs, and 21% of CHWs said that being a
VO member had no effect. In Bangladesh, 22% of women said that being a VO
member increased community trust in her work and 13% said that it made it
easier to sell products. Further exploration of how important VO membership is
for the recruitment and retention of SS will be important for BRAC to consider.
• Describe notions and understanding of ‘volunteerism’
The SSs are a volunteer cadre but are also financially incentivized by the creation
of community demand for drugs, commodities, and basic health services. In
Uganda, CHPs expressed concern that their community did not understand that
they were volunteers. The community thought they were receiving monthly salary
from BRAC (which they got when they went to the refresher training) and that
they were selling drugs to make additional profit. Several CHPs said that if the
community understood better that they were volunteers it would make their work
as CHPs easier. Understanding of what it means to be a volunteer are culturally
bounded and influenced by local context. For example, in Uganda where
HIV/AIDS has destroyed homes and communities, there is for many CHP a
strong incentive to volunteer as a CHP “to give back to my community” or “to
help my neighbours”. The CHP role may resonate more in Uganda as one of
community service as opposed to a ‘health volunteer’. Misunderstanding or
misperceptions of what it means to work and serve as a volunteer did come up
in discussions with SSs in all three settings.
Generalizability of BRAC SS approach
The quantitative, qualitative, and participant observation data from this study suggest
that the BRAC Bangladesh SS approach is generalizable to other settings – or at
least to the context of Uganda and Afghanistan. These programmes have, however,
had to explicitly adapt the model in order to reflect the local context. There are some
useful lessons in these examples for BRAC as it branches into other countries. The
adaptations these country programmes have made are in three main areas: 1) health
policy and health systems context; 2) socio-cultural environment; and 3)
management and logistics context. It may be useful to generate a checklist of major
issues in these areas that other programmes have had to adapt or are currently
considering adapting.

81
This study has just scratched the surface of critical questions about the sustainability
and generalizability of BHP’s SS approach. The results suggest that long-term
financial sustainability is possible but complicated by ensuring that all the
components of the SS role are maintained. The SSs are a mixed motive cadre – not
entirely volunteers, but not salaried either. The data from this study suggest it is a
combination of factors that motivate a SS to continue her work – both financial and
non-financial.
This study helps document the unique and fundamental role that SSs play in BHP
and her community. The SSs are not simply ‘medicine sellers’ and as one BRAC
Uganda staff member aptly put it…“selling products is not the nature of the
programme. Rather it is to change the health behaviour of customers.” It is through
her role as a change agent in health behaviour that the BRAC CHW will improve the
essential health of her community.

82
REFERENCES
Ahmed J (2007). Community Health Volunteers (Shasthya Shebika) at BRAC: post Alma-Ata
experience. September 16, 2007. (Powerpoint presentation)
Bender DE and Pitkin K (1987). Bridging the gap: the village health worker as the cornerstone
of the primary health care model. Soc Sci Med 24(6):515-28.
Bhattacharrya K, Winch P, LeBan K, Tien M (2001). Community health worker incentives and
disincentives: how they affect motivation, retention, and sustainability. Arlington: United States
Agency for International Development.
BRAC (2009). BRAC Annual Report 2008. Dhaka: BRAC. 133p.
BRAC (2007). BRAC Afghanistan Annual Report 2007.
BRAC (2007). BRAC Health Programme Annual Reports 2007.
Dieleman M, P Viet Cuong, L Vu Anh, Martineau T (2003). Identifying factors for job motivation
of rural health workers in North Viet Nam. Human Resources for Health. 1:10.
Fielder JL, Villalobos CA, De Mattos AC (2008). An activity-based cost analysis of the
Hondorus community-based integrated child care (AIN-C) programme. Health Policy Plan
23;408-27.
Hadi A (2003). Management of acute respiratory infections by community health volunteers:
experience of Bangladesh Rural Advancement Committee (BRAC). Bull WHO 81,3:183-9.
Haines A, Sanders, Lehmann U, Rowe AK, et al. (2007). Achieving child survival goals:
potential contribution of community health workers. Lancet 369:2121-31.
Joint Learning Initiative (2004). Human resources for health: overcoming the crisis. Cambridge:
Harvard University Press.
Khan SH, Chowdhury AMR, Karim F, Barua MK (1998). Training and retraining shasthyo
shebika: reasons for turnover of community health workers in Bangladesh. Health Care
Supervisor 17(1):37-47.
Lewin SA, Babigumira SM, Bosch-Capblanch X, Aja G, et al. (2008). Lay health workers in
primary and community health care: a systematic review of trials. Cochrane EPOC Review.
Milligan P, Njie A, Bennett S (2004). Comparison of two cluster sampling methods for health
surveys in developing countries. Int J Epidemiol 33,3:469-76.
Mir AM and Rashida G (2007). Challenging the gender paradigm in rural Pakistan: a case
study of empowered lady health workers. In: Reichenbach, L (Editor). Exploring the gender
dimensions of the global health workforce. Cambridge: Global Equity Initiative, Harvard
University.
OPM (2002) Key findings for lady health worker evaluation. Oxford Policy Management, April
2002.

83
Rahman M and Tasneem S (2008). Determinants of income of the Shasthya shebikas:
evidence from a pilot MNCH initiative in the Nilphamari district of Bangladesh. Dhaka: BRAC.
42p. (RED working paper No. 6)
Standing H and Chowdhury AMR (2008). Producing effective knowledge agents in a pluralistic
environment: what future for community health workers? Soc Sci Med 66:2096-107.
Swider SM (2002). Outcome effectiveness of community health workers: an integrative
literature review. Public Health Nursing 19(1):11-20.
Walker DG and Jan S (2005). How do we determine whether community health workers are
cost-effective? Some core methodological issues. J Commun Health 30(3):221-9.
Walley JJ, Lawn A, Tinker M, de Francisco M, Chopra Z, Bhutta I, Rudan R, Black (2008).
Primary health care: making Alma Ata a reality. Lancet 372(9642):1001-7.
Walt G (1988). CHWs: Are national programmes in crisis?” Health Policy Plan 3(1):1-31.
World Health Report (2008). Primary healthcare: now more than ever. Geneva: World Health
Organization.

84
APPENDIX
Appendix 1. Detailed breakdown for programme cost calculations
Official exchange rates:
1 US$ = BDT 70
1 US$ = AFN 50
1 US$ = UGX 2000
Source: Official Exchange rate of 2008 (approximation): www.onda.com
Implied PPP (Purchasing Power Parity Exchange rate)
1 US$ = BDT 25.585
1 US$ = AFN 28.892
1 US$ = UGX 654.451
Source: World Economic Outlook (WEO), IMF, 2008
Table A.1. Cost per SS in Bangladesh
Item Spent per SS
per year (BDT)
Spent per
SS per year
(US $)
% of total
cost of
(1styr)
% of total
cost of
(2nd yr)
Salary SK 906.60 12.95 13.23% 30.38%
Salary PO 1024.56 14.64 14.95% 34.34%
Salary UM 270.00 3.86 3.94% 9.05%
Salary SHC 75.00 1.07 1.09% 2.51%
Total salary cost 2276.16 32.52 33.21% 76.28%
10% overhead cost 227.62 3.25 3.32% 7.63%
Total cost per SS per year(exc. Tra cost) 2503.78 35.77 36.53% 83.91%
Basic training cost 3910.00 55.86 57.05%
Refresher training cost(40Tk/RT) 440.00 6.29 6.42%
First year cost of training 4350.00 62.14 63.47%
Second year cost of training 480.00 6.86 16.09%
Total cost in the first year 6853.78 97.91 100.00% 100.00%
Total cost in the second Year 2983.78 42.63
Total cost in 1st year (Min)* 6168.40 88.12
Total cost in 1st year (Max)* 7539.15 107.70
Total cost in 2nd year (Min)* 2685.40 88.12
Total cost in 2nd year (Max)* 3282.15 46.89
*5% Uncertainty range assumed
BRAC's earning from SS through sales/yr
Total cost after deducting sales income
Total cost in the first year 6263.78 89.48
Total cost in the second year 2393.78 34.20
Total cost in 1st year (min)* 5950.59 85.01
Total cost in 1st year (max)* 6576.96 93.96
Total cost in 2nd year (min)* 2274.09 32.49
Total cost in 2nd year (max)* 2513.46 35.91
note: 10 SS/SK, 40 SS/PO, 80 SS/UM, 480 SS/SHC assumed

85
Table A.2. Cost per CHP in Uganda
Item Spent per CHP
per year (UGX)
Spent per CHP
per year (US $)
% of Total Cost
of (1styr)
% of Total
Cost of
(2nd yr)
Salary CHA 278400.00 139.20 35.32% 65.61%
Salary PO 66120.00 33.06 8.39% 15.58%
Salary regional coordinator 14250.00 7.13 1.81% 3.36%
Salary master trainer 12825.00 6.41 1.63% 3.02%
Total salary cost 371595.00 185.80 47.14% 87.57%
10% overhead cost 37159.50 18.58 4.71% 8.76%
Total cost per SS per year(exc. Tra
cost) 408754.50 204.38 51.86% 96.32%
Basic training cost(first year only) 379500.00 189.75 48.14%
Refresher training cost (from 2nd
year) 15600.00
First year cost of training 379500.00 189.75 48.14%
Second year cost of training 15600.00 7.80 3.68%
Total cost in the first year 788254.50 394.13 100.00% 100.00%
Total cost in the second year 424354.50 212.18
Total cost in 1st year (min)* 748841.78 374.42
Total cost in 1st year (max)* 827667.23 413.83
Total cost in 2nd year (min)* 403136.78 201.57
Total cost in 2nd year (max)* 445572.23 222.79
5% Uncertainty range assumed
note: 10 CHP/CHA, 100 CHP/PO, 200 CHP/RHC, 400 CHP/ MT assumed
Table A.3: Cost per community health workers in Afghanistan
Item
Spent per
CHW per year
(AFs)
Spent per
CHW per year
(US $)
%of Total Cost
of (1styr)
%of Total Cost
of (2nd yr)
Salary CHW Supervisor-CHS 900 18 7.29% 21.54%
Salary PO 780 15.6 6.32% 18.67%
Salary medical officer 300 6 2.43% 7.18%
Salary master trainer 400 8 3.24% 9.57%
Total salary cost 2380 47.5 19.27% 56.97%
10% overhead cost 238 4.75 1.93% 5.70%
Total cost per CHW per year(exc.
Tra. cost) 2,618.00 52.36 21.20% 62.66%
Basic training cost (first year only) 9730 194.6 78.80%
First year cost of training 9730 194.6 78.80%
Second year cost of training 1560 31.2 37.34%
Total cost in the first year 12,348.00 246.96 100.00% 100.00%
Total cost in the second year 4,178.00 83.56
Total cost in 1st year (min)* 11,113.20 234.61
Total cost in 1st year (max)* 13,582.80 259.31
Total cost in 2nd year (min)* 3,760.20 234.61
Total cost in 2nd year (max)* 4,595.80 87.74
*5% Uncertainty range assumed
Note: One CHW Supervisor is Responsible for 10 CHW, One PO is Responsible for 10 CHW Supervisor,
One MO is Responsible for One PO
And there is one CHW Master Trainer at Provincial level

86
Appendix 2. Survey instrument
Financial and economic analysis of BRAC’s
Community Health Volunteers
BANGLADESH: Survey of Shasthya Shebikas
TO BE READ TO EACH STUDY PARTICIPANT:
Hello my name is _____________ and I work with BRAC. BRAC is carrying out a study of
Shasthya Shebikas in your area to find out more about how Shasthya Shebikas spend their
time, their monthly earnings, the products they sell, and other aspects of their job that affect
them financially. The information from this study will help BRAC’s Health Programme to better
understand how the Shasthya Shebika programme can run as smoothly as possible .
We would like to ask you a series of questions to get your inputs into this study. There are
about 65 questions and the entire questionnaire should take about 35 minutes of your time.
Your answers to these questions will be kept entirely confidential and your responses will not
affect your job in any negative way. Your participation in the study is completely voluntary and
you can stop answering questions at any time during the survey if you choose to. If you
choose not to participate or to stop the survey it will not negatively affect your work as an SS.]
Questions and filters Responses and codes
A1. Do you have any questions you would like
to discuss at this point?
Yes----------------------------------------------------------1
No--------------------------------------------2 ÆGo to A3
A2. [IF YES] what are the questions? [Specify]
__|__
__|__
__|__
A3.
May I start the interview?
[If No, then complete A4-A9 and end
interview]
Yes--------------------------------------------- ------------1
No----------------------------------------------------------2
I have read the informed consent and study
description to the respondent and she has
indicated her understanding.
Interviewer’s Signature here
_________________________________________
A4. Name of Interviewer
|___________________________________|
A5. Date of Interview ___|_____|___
Date Month
A6. Result of Interview Completed ------------------------------------ 01
Partially completed --------------------------- 02
Respondent is absent----------03 Æ End interview
Refusal---------------------------04 Æ End interview
Other_______________________77 Æ End interview
A7. Reason for no or incomplete interview [__________________________________________]
A8. Time interview started (write in 24 hours) __|__|__|__
A9. Time interview ended (write in 24 hours) __|__|__|__
A10. Questionnaire checked by in field (Name) |_____________________________________|
A11. Questionnaire checked by in office (Name) |_____________________________________|
A12. Questionnaire entered by (Name) |_____________________________________|
Appendix 2 continued………

87
Continued appendix 2………
Identification of area and respondent
A13. What is your name? |________________________________|
A14. What is the name of your Village/Mohalla? |________________________________|
A15. What is your Branch Code number? |________________________________|
DK/Can’t recall____________________3
A16. What is your VO Number? |________________________________|
DK/Can’t recall____________________3
A17. What is your Member Number? |________________________________|
DK/Can’t recall____________________3
A18. What is your age? |_________| years
A19. How many years of schooling have you completed? |_________| years
A20. Do you have any children under the age of two
years? [________]
Household financial and economic information
B1. How many members
live in your household?
(a) Total family members__|__
B2. On average, what is
your household
income/expenditure
status each month?
Income is always greater than expenditure________________1
Income is sometimes greater than expenditure____________2
Income is equal to what we spend______________________3
Income is sometimes less than expenditure______________4
Income is always less than expenditure__________________5
Not sure/Don’t know_________________________________6
B3. What is your average
monthly expenditure?
[_____________] Taka______________________________1
Don’t Know______________________________________3
Main earner
Average monthly earning
B4. Who is the main earner
in your family and what
is his/her average
monthly earning?
1) Taka [_____________]______1
DK_______________________3
B5. Are you currently a VO
member of BRAC?
Yes, current VO member (skip to B7 )______________________1
No, not a current VO member______________________________2
B6. Have you ever been a
VO member of BRAC?
Yes, I used to be a VO member_____________________________1
No, never been a VO member (skip to B14)___________________2
B7. Were you a VO member
when you first became
a SS?
Yes_____________________________________________________1
No_____________________________________________________2
B8. Did you ever borrow a
loan from BRAC?
Yes_____________________________________________________1
No (skip to B14)_________________________________________2
B9. Do you currently have a
loan with BRAC?
Yes_____________________________________________________1
No (skip to B11)________________________________________2
B10. Do you have more than
one loan with BRAC?
Yes_____________________________________________________1
No _____________________________________________________2
B11. Do/did you use any of
the loan(s) to buy
medicines or other
supplies for your work
as a SS?
Yes_____________________________________________________1
No (skip to B14)__________________________________________2
B12. Approximately what
amount of the loan
is/was used to buy
supplies for your role
as SS?
[_______] Taka__________________________________________1
Can’t recall/DK__________________________________________5
Loan 1 Loan 2 Loan 3 B13. How much is/are the
current BRAC loan(s)
now?
[______] Taka______1
Can’t recall/DK____3
[______] Taka_____1
DK_____3
[___] Taka___1
DK_____3
Appendix 2 continued..............

88
Continued appendix 2………
B14. Do you have any loans
with NGOs other than
BRAC?
Yes_________________________________________________1
No___________________________________________________2
Employment details
C1. How long have you
been working as an
SS?
[_____] Years_________________________________________1
[_____] Months_______________________________________2
C2. How many households
are you able to visit in
one day?
[_____] Number of households___________________________1
Can’t recall/DK________________________________________3
C3. How many households
are you responsible for
each month?
[_____]Number of households___________________________1
Can’t recall/DK________________________________________3
C4. How many hours on
average do you work
each day?
[_____]Number of hours________________________________1
Can’t recall/DK________________________________________3
Do you provide
the following
service in the
last month?:
Provide this service
in the last month
Average time (or
number of visits)
to perform
service/transacti
on
Average
number of
people served
in last month
Is last month
performance
about average
for each
month?
Pregnancy
identification
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Attending
delivery and
providing
newborn care
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Referral to
Shushasthya
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Referral to
government
clinic/hospital
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Attending
refresher
training
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Treating TB
(DOTS)
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______] Yes_______1
No_______2
DK_______3
Selling
medicines
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______]
Numbers of
people sold
medicines to
Yes_______1
No_______2
DK_______3
C5.
Selling health
commodities
Yes____________1
No____________2
Cannot
recall/DK______3
[______] minutes
[_____] visits
[_______]
Numbers of
people sold
heath
commodities
to
Yes_______1
No_______2
DK_______3
Appendix 2 continued………

89
Continued appendix 2………
Others (specify) ____________________________
____________________________
____________________________
[_______] Times per month
[_______] Times per month
[_______] Times per month
C6. What are the three
most frequently sold
medicines (in order of
frequency)? (Specify)
1)_______________________
2)_______________________
3)_______________________
C7. On average, how
many household visits
are required to sell
medicines?
1 visit_______________________________________________1
2 visits_______________________________________________2
3-4 visits______________________________________________3
4+ visits______________________________________________4
Don’t know___________________________________________5
C8. Do women ever come
to your house to buy
medicines?
Yes_________________________________________________1
No_________________________________________________2
C9. What are the least
frequently sold
medicines? (Specify)
1)_______________________
2)_______________________
3)_______________________
C10. Why do you think
these medicines are
not popular or difficult
for you to sell?
________________________________________________________
_______________________________________________________
_______________________________________________________
C11. What are the most
frequently sold health
commodities?{Specify
in order of frequency}
1)_______________________
2)_______________________
3) _______________________
C12. On average, how
many household visits
are required to sell
health commodities?
1 visit_______________________________________________1
2 visits_______________________________________________2
3-4 visits______________________________________________3
4+ visits______________________________________________4
Don’t know___________________________________________5
C13. Do women ever come
to your house to buy
health commodities?
Yes_________________________________________________1
No_________________________________________________2
C14. What are the least
frequently sold health
commodities?
{Specify}
1)_______________________
2)_______________________
3)_______________________
C15. Why do you think
these health
commodities do not
sell well? (open-
ended response)
___________________________________________________________
___________________________________________________________
___________________________________________________________
C16. Do you ever replenish
your supplies of
medicines or
commodities between
refresher trainings?
Yes________________________1
No________________________2
C17. Do you ever buy
medicine or health
commodities from
outside of BRAC for
selling purposes?
Yes_______________________________1
No (skip to C19)___________________2
Appendix 2 continued………

90
Continued appendix 2………
C18. If yes, which ones do
you buy outside of
BRAC for sale?
{Specify}
[_____________________________]
[_____________________________]
[_____________________________]
C19. Do you face any
problems in the
community selling
BRAC suggested
medicines or health
commodities?
Yes___________________________________________1
No (skip to C21)___________________________________2
C20. Why do/did you have
problems in the
community with
selling BRAC
suggested medicines
or health
commodities?
{circle all that apply}
People do not prefer BRAC products_________________________1
People buy products from shops_____________________________2
Products are too expensive/cheaper elsewhere________________3
People do not trust me_____________________________________4
Other reasons (specify)_____________________________________5
_____________________________________________________
C21. How does being a
BRAC VO member
affect your work as an
SS? (specify)
People trust me more_______________________________________1
Easier to sell products_______________________________________2
I need the loan to buy supplies________________________________3
Being a BRAC VO member has no effect_______________________4
Other (specify)______________________________________________5
C22. Are there other
medicines or health
products you wish
you could sell as an
SS or other products
that your community
asks about?
Yes____________________________________________________1
No (skip to C24)_________________________________________2
C23. If so, what are these?
{Specify}
1)[___________________________________________]
2)[__________________________________________]
3)[___________________________________________]
4)[___________________________________________]
C24. Are there products
that are not health-
related that women
would be interested
and willing to buy
from you? {If yes,
specify which
products}
Yes___________________________________________________1
Specify[_____________________________________
____________________________________________]
No____________________________________________________2
Don’t know____________________________________________3
C25. Do you face any
problems making
referrals to
Shushasthya?
Yes___________________________________________________1
No (skip to C27)________________________________________2
Don’t know____________________________________________3
C26. What type of
problems do you have
making referrals to
government clinics?
Staff do not treat referralswell______________________________1
Facilities are not always available___________________________2
Clinic does not have adequate supplies______________________3
Too far away_____________________________________________4
Other (specify)____________________________________________5
C27. Do you face any
problems making
referrals to
government clinics?
Yes___________________________________________________1
No (skip to C29)________________________________________2
Don’t know____________________________________________3
Appendix 2 continued………

91
Continued appendix 2………
C28. What type of
problems do you have
making referrals to
government clinics?
Staff do not treat referralswell______________________________1
Facilities are not always available___________________________2
Clinic does not have adequate supplies______________________3
Too far away_____________________________________________4
Other (specify)____________________________________________5
Shasthya Shebika motivation and incentives
C29. How important are
earnings from your
SS activities for your
household?
Household could not continue without SS income______________1
SS income makes a big difference to household_________________2
Would not make a big difference if no SS income_______________3
Would not make any difference if no SS income________________4
Don’t know_______________________________________________5
C30. Does being an SS
give you financial
independence?
Yes______________________________________________________1
No______________________________________________________2
C31. How do you use the
money you make as
an SS? (circle all that
apply)
Spend on children_________________________________________1
Give to husband___________________________________________2
Pay for school fees_________________________________________3
Buy food__________________________________________________4
Save _____________________________________________________5
Pay back loans____________________________________________6
Other (specify)_____________________________________________7
C32. Have you ever or do
you currently work for
other NGOs/health
programmes/clinics
while working as an
SS?
Yes_____________________________________________________1
Specify[_________________________________]
No_____________________________________________________2
C33. Have you received
any health-related
training outside of
BRAC SS training?
Yes____________________________________________________1
[If yes, specify which __________________________________]
No____________________________________________________2
C34. Does being an SS
affect your ability to
carry out your
household
responsibilities?
Have to work harder in household duties___________________1
Less time for children____________________________________2
Less time for cooking_____________________________________3
Less time for cleaning____________________________________4
Miss out on family events________________________________5
Children’s education suffers______________________________6
No, doesn’t affect household responsibilities_________________7
C35. Why did you start
working as a SS?
{circle all that apply}
Source of income for household___________________________1
Financial independence__________________________________2
Social recognition________________________________________3
To learn something new__________________________________4
To help my community___________________________________5
To get BRAC loan________________________________________6
Other (specify)___________________________________________7
C36. Has working as an
SS changed your
status in your
household?
Importance increased significantly___________________________1
Importance increased somewhat_____________________________2
No change at all___________________________________________3
Importance decreased somewhat____________________________4
Importance decreased significantly__________________________5
C37. What was your
income from working
as a SS last month?
[_____________] Taka in last month
Appendix 2 continued………

92
Continued appendix 2………
C38. What is the total income
you receive in an average
month for your SS work?
[_____________] Taka per month
C39. Is there fluctuation in your
monthly sales?
Yes__________________________________________________1
No__________________________________________________2
Don’t Know__________________________________________3
For each of the following
providers, ask:
Do any [ ] provide services
in your area?
If yes, how many of
them work in your
area?
Do you
feel these
providers
limit your
income as
a SS
Pharmacy or shop that
sells medicine
Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
NGO Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
Government clinic or
hospital
Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
Private clinic Yes________1
No________2
DK________3
[_____] Number___1
DK______________
2
Yes____1
No_____2
DK_____3
Village doctor Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
Traditional healer Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
TBA (dai) Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
C40.
Other (specify) Yes________1
No________2
DK________3
[_____] Number___1
DK_____________2
Yes____1
No_____2
DK_____3
O
pportunity cost
Please report any work you have done or currently do while also working as an SS:
JOB or WORK Is this income regular or
seasonal?
If seasonal how many
months/weeks in last
year?
How much do
you earn from
this work in an
average
month?
Factory worker Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Handicraft Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Poultry raising Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
C41.
Agriculture Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Appendix 2 continued………

93
Continued appendix 2………
Small business/
hawking
Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Tailor Regular________________1
Seasonal________________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Domestic
worker
Regular________________1
Seasonal_______________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Midwifery Regular________________1
Seasonal_______________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Others
(specify)_______
_______
Regular________________1
Seasonal_______________2
[______] weeks____1
[______]months___2
Can’t recall_______3
[_______] Taka
Other jobs available Monthly average income possible
Factory worker____________1 [__________] Tk/month______1
DK________________________3
Handicraft________________2 [__________] Tk/month______1
DK________________________3
Poultry raising_____________3 [__________] Tk/month______1
DK________________________3
Agriculture________________4 [__________] Tk/month______1
DK________________________3
Small business/hawking_____5 [__________] Tk/month______1
DK________________________3
Tailor_____________________6 [__________] Tk/month______1
DK________________________3
Domestic worker___________7 [__________] Tk/month______1
DK________________________3
Midwifery_________________8 [__________] Tk/month______1
DK________________________3
C42. What other jobs
are available to
you if you were
not an SS?
Specify and list
all mentioned
Others (specify)____________9 [__________] Tk/month______1
DK________________________3
C43. What are the
advantages of
working as an
SS as
compared to
these other
jobs?
Better earnings in future____________________________________1
Better work environment___________________________________2
Can work from home_______________________________________3
Better hours_______________________________________________4
More prestigious among the community______________________5
Family approves___________________________________________6
Less distance to travel______________________________________7
No advantage____________________________________________8
Other (specify)___________________________________________9
C44. What are the
disadvantages
of working as an
SS as
compared to
these other
jobs?
Less earnings____________________________________________1
Not good work environment_______________________________2
Must work outside home___________________________________3
Hours not convenient______________________________________4
Less prestigious among the community______________________5
Family disapproves_______________________________________6
Greater distance to travel__________________________________7
No disadvantage_________________________________________8
Other (specify)___________________________________________9
C45. Have you ever
considered
quitting your
work as an SS?
Yes_____________________________________________________1
No______________________________________________________2
Appendix 2 continued………

94
Continued appendix 2………
C46. If yes, why? Payment is too low_______________________________________1
SS role is difficult_________________________________________2
I can’t keep up with H/H responsibilities/pressures____________3
Husband/in-laws disapproves_____________________________4
Community disapproves__________________________________5
I have a better paid job alternative__________________________6
Other (specify)___________________________________________7
C47. Would a fixed
monthly salary
encourage you
to be more
active as an
SS?
Yes_______________________________________________________1
No (End Interview)_________________________________________2
C48. If yes, what is
the minimum
monthly salary
(with no
commission)
that you would
require?
[___________] Taka per month_______________________________1
Don’t know________________________________________________3