WORKING PAPER

Community-Based
Health Information
Systems in the
Global Context
A Review of the Literature
MEASURE Evaluation

June 2016

WP-16-161
 WORKING PAPER

Community-based
Health Information
Systems in the
Global Context
a Review of the Literature
MEASURE Evaluation

June 2016
WP-16-161

MEASURE Evaluation
MEASURE Evaluation is funded by the U.S. Agency for International Development
University of North Carolina at Chapel Hill (USAID) under the terms of Cooperative Agreement AID-OAA-L-14-00004 and
400 Meadowmont Village Circle, 3rd Floor implemented by the Carolina Population Center, University of North Carolina at
Chapel Hill, in partnership with ICF International, John Snow, Inc., Management
Chapel Hill, North Carolina 27517 Sciences for Health, Palladium, and Tulane University. The views expressed in
Phone: +1-919-445-9350 • [email protected] this publication do not necessarily reflect the views of USAID or the United States
government. WP-16-161
www.measureevaluation.org
Background
In developing countries, high disease burden from HIV, malaria, tuberculosis, poor nutrition, and
high maternal mortality often intersects with low access to health services (Global Fund, 2014).
Community workers (CWs), including community health workers (CHWs), community health
extension workers (CHEWs) and community social service providers, have met critical service gaps.
Generally, CWs are affiliated with either health facilities or community-based organizations. These
providers live in the communities they serve and provide outreach services to people who may
otherwise not receive them. The work of community worker providers has helped to improve health,
especially maternal, child, and newborn outcomes (Lunsford, 2015; Golding, 2014).

National governments and donors are increasingly emphasizing and supporting the work of CWs to
expand health care coverage (de la Torre, 2014). The important role of CWs has been emphasized in
recent global health policy, such as USAID’s policy for ending preventable child and maternal deaths
(EPCMD) (USAID, 2015) and the President’s Malaria Initiative. Similarly, there is a recognition
that decentralized, community-based approaches to HIV care and support will be required to meet
UNAIDS’ 90-90-90 target (90% of all people living with HIV will know their status, 90% of people
diagnosed with HIV to receive antiretroviral treatment (ART), and 90% of all people receiving ART
will be virally suppressed by 2020) (UNAIDS, 2014). Community-based programs implemented by
CWs are uniquely suited to help achieve the goals laid out in these policies in the most vulnerable
parts of the world. Furthermore, as global health systems move toward universal health coverage,
CWs will become even more important to increase health equity (Golding, 2014).

CWs require key health information to do their jobs, and their managers require information to
monitor their work. Likewise, Governments and donors require information to better plan
community health programs, and to monitor progress against goals. This information, how it is
collected, and how it flows, is how we define a community-based health information system(s)
(CBHIS). These systems should involve data collection, management, and analysis of health and
related services provided to communities outside of facilities (de la Torre, 2014). They should enable
information to be shared among community-based services and between community-based services
and higher-level health facilities. Moreover, to some extent, they should feed into national health
management information system(s) (HMIS). CBHIS have the potential to engage community
members, provide them with an avenue to health services, and hold them accountable, contributing
to the goals of sustainability (Jeremie, 2014; Sabitu, 2004).

Little is known about how CBHIS are being implemented globally. This literature review seeks to
explore the innovations in and different manifestations of CBHIS in developing countries.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 2
Methods
We conducted a literature search in May 2015 using the following databases: PubMed, POPLINE,
and USAID’s Development Experience Clearinghouse. We also searched specific journals, including
the Journal of Health Informatics in Developing Countries, International Journal of Medical
Informatics, and Electronic Journal of Information Systems in Developing Countries, as well as
Google and Google Scholar. We used the following medical subject headings (MeSH) terms:
community network; health status indicators; information systems; and information management.
Title/abstract search terms were “community-based information” and “community health services”
AND “information systems.” Phrase search terms were “community-based information systems” and
“community-based health information systems.” We focused on low- and middle-income countries.
Ultimately, we found 23 eligible references.

Results
We grouped results in the following CBHIS functional areas:

• Case management
o Documenting individual-level needs to support individual care planning
o Enabling bidirectional referrals
o Tracking patients lost to follow-up
• Accountability
o Reporting of inputs and outputs
• Planning, resource allocation, and advocacy
o Assessing population level needs (outcomes)

Facilitating Community-Level Case Management

CBHIS can be used for case management, to document individual needs to inform care plans, to
enable bidirectional referrals, and to track patients lost to follow-up. Examples from the published
literature of each of these are outlined below.

Documenting Individual Needs to Support Care Planning

The Child Status Index (CSI) collects data on orphans and vulnerable children (OVC) on 12
dimensions of well-being (MEASURE Evaluation, 2012). While data are used for program
monitoring and to facilitate planning at higher levels, the individualized approach of the system
makes it especially appropriate for case management, allowing child and family services to be tailored
to the needs identified at each encounter/presentation. The resulting individualized care plan may
also include referral to other agencies and community-based organizations. Sometimes, CSI
information is entirely paper-based, and in some instances, programmers have captured data
electronically to allow for higher-level planning. The CSI is being used in at least 17 different
countries and has been adapted in others to facilitate the same purpose.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 3
Enabling Bidirectional Referrals
Originally designed for the SIDALE (“AIDS Go Away”) project in Haiti, IQReferrals is a tool used
to track clients who have been referred to and from multiple programs and link them across the
continuum of care. IQReferrals provides a communication portal for major hospitals, supporting
satellites and community service clinics to capture referral and counter-referral information, so that
the overall healthcare ecosystem serves clients efficiently and effectively. IQReferrals facilitates
tracking and easy provision of feedback between a site or program that has referred a client and the
one that has received the client. IQReferrals can be used to track clients who are referred among
various program areas—for example, tuberculosis, voluntary counseling and testing, and the
prevention of mother-to-child transmission of HIV —as well as those transferred to other sites
within one or several programs and/or catchment areas. IQReferrals can produce automated reports
both at the facility and central levels as well as across multiple databases, thereby improving data
quality and feedback.

The USAID-funded Linkages across the Continuum of HIV Services for Key Populations Affected
by HIV Project (LINKAGES) is applying a similar approach to facilitate real-time tracking of clinic
referrals among key populations in Papua New Guinea, Laos, and Thailand. Through the mobile
application, built on Dimagi's CommCare platform, outreach workers enter clients’ responses to
HIV risk questions onto a smartphone. Based on the answers given to early questions, the outreach
workers are prompted to ask additional questions. When a client's answers indicate that they should
be referred for HIV testing or other clinical services, the referral is automatically created by the
application and sent to an application on a smartphone at the clinic's receiving desk. The clinician
can then confirm through the app that the client appeared for the referral, thus providing direct and
immediate statistics about follow-up. After a period of time, if a client has not appeared for a referral,
the system will send an SMS message to the client’s mobile phone, providing a reminder of the
appointment and offering further assistance.

Tracking Patients Lost to Follow-Up

One key innovation was keeping a family folder for every family in the catchment area of a health
post as part of CBHIS (MEASURE Evaluation, 2012; Chewitha and Azim, 2013). Along with the
family folders was a simple tickler system, whereby wooden boxes were used to organize health cards
according to the month in which follow-up services were needed for family members. If a health
card had been left in a previous month’s box, health workers could see when a service had not been
provided and then reach out to the family to provide care. Health extension workers also used the
boxes/health cards to plan follow-up with pregnant women, family planning clients, and children for
immunization. For example, in one zone, the family folder system showed that not many children
were getting vaccinated. Health workers were able to identify children who had dropped out or who
had missed an immunization service, initiate contact, and provide the vaccinations.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 4
Now, family folders connected with HMIS and CBHIS enable health workers to build trust with
client households, and thus clients return for critical health services and have more confidence in
service quality. With family folders as a tool for identifying pregnant women who have missed
services, workers are able to reach more women for continuity of services and provide better
antenatal care.

A program in India deployed informal educators to collect data on child immunization dropout
rates, community health needs, and resources (Singh, 1997). Computer-generated immunization
dropout lists were given to CHWs and auxiliary nurse midwives for outreach targeting. Follow-up
data showed increases in “fully immunized children” for the diphtheria, pertussis, and tetanus
vaccination.

Accountability
Perhaps the most familiar functional aspect of a CBHIS is to enable reporting both to the
government and donors. The following examples are illustrative of systems that aggregate and report
community data to higher levels.

The family folder program in Ethiopia not only facilitates service provision at the community level
through its referral and follow-up reminder system but also is integrated in the national HMIS
(Mutale, 2013; Damtew, 2013). HEWs report the data they collect to the nearest health centers
monthly; in turn, district health offices as well as other partners receive reports monthly, quarterly,
and in emergencies. The districts then compile these reports and sends them to zonal health offices,
which forward them to regional and national offices quarterly.

The Connect Project, in Tanzania, links community data collected by community health agents
(CHAs) with the district and national HMIS through its aggregate data from community registers
(Mutale, 2013). CHA supervisors at the facility level are responsible for meeting with CHAs to
create action plans, and for reporting data collected by CHAs to the district level. There, a council
health management team develops comprehensive council health management plans and reports to
the national level.

Mobile technology can facilitate the reporting of community data. The Nigeria Evidence-based
Health System Initiative (NEHSI) used CommCare to connect its CBHIS with the provincial and
national HMIS for planning (Asangansi, 2013). Community field workers were given Android
mobile devices, which they used to register 5,600 people and link that data to the larger information
systems. Benefits of the mHealth system were ease of use as expressed by field workers; savings both
in material and human resources, because fewer man-hours were needed to collect, manage, and
synchronize data; and improved data quality.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 5
Planning, Resource Allocation, and Advocacy
CBHIS are used to identify health needs in communities to inform planning and resource allocation
decisions at the local, sub-national and national levels. These data also enable advocacy for most-in-
need populations. Sahay and Braa (2012) distinguish data collection to support action planning at the
facility and systems level, with data collection and management to support high level planning. For the
purposes of this review, we have wrapped these together. There are diverse examples from the literature
of instances were community data are routinely collected and documented to influence change.

For example, in Nyanza Province, Kenya, CHEWs and CHWs were trained to facilitate community
dialogue and maintain the CBHIS and the village register (Kaseje, 2010). Health status data from
each household were routinely collected and maintained by CHWs through the village register, and
then analyzed by community members and committees and at the health center. Data-driven
participatory action planning by the community and health centers helped to improve services.
Reports are sent to the district-level, where they are processed electronically. The districts saw
statistically significant improvements in immunization coverage, health-facility childbirth, use of
insecticide-treated bednets, and treated drinking water in comparison with control sites. In another
example, under the Kenya Essential Health Package, CHEWs inform community health committees
on key community health indicators, whereupon data collection is planned (Jeremie, 2014). Data
collected are then fed back to the facility to identify health utilization gaps and outbreaks, and
ultimately to improve services.

In the Community-Level Nutrition Information System for Action (COLNISA) project in northern
Nigeria, supported by the United Nations Children’s Fund (UNICEF), volunteers collected
maternal and child health data and analyzed them with the COLNISA committee; then together
they created a workplan to address areas of need (Sabitu, 2004). These data informed several
community interventions: community-based growth monitoring and promotion, outreach antenatal
and immunization services, exclusive breastfeeding promotion, and education about oral rehydration
therapy and complementary feeding. Follow-up data collected by CHWs showed statistically
significant improvements on all indicators except under-five nutritional indices.

In a family-focused program in the Philippines, CHWs visited households and completed family
profiles, assessing child health, family planning practices, and desire to space births (Management
Sciences for Health, 2001). CHWs then used a geographic information system to create a “spot
map” of communities with high child health and family planning needs, which were then targeted
by additional CHW intervention.

The Sustainability through Economic Strengthening, Prevention and Support for OVC and Other
Vulnerable Populations Project (STEPS OVC) is the largest USAID-funded project supporting
OVC in Zambia. STEPS OVC has a fully-functional monitoring and evaluation system that helps to
fulfill reporting requirements, as well as facilitate service provision by subgrantees through
community caregivers. Data regarding services received and referrals made are collected during

Community-Based Health Information Systems in the Global Context: A Review of the Literature 6
routine household visits by community caregivers. Data are then transferred to subgrantee M&E
officers, who enter data in the project-specific Community Prevention Information System database
for analysis at the project level. Data are also reported to the District AIDS Task Forces using
government forms to enable planning. Similarly, the AIDS, Population and Health Integrated
Assistance (APHIAplus) project, in Kenya, as well as Pamoja Tuwalee, in Tanzania, use CHWs to
collect data on child well-being, shelter, water and sanitation conditions, food security, and
household income, and use those data to inform services provided by CHWs based on children’s
needs (MEASURE Evaluation, 2014; MEASURE Evaluation, 2014).

Challenges
As with any system, CBHIS are not without challenges. A common challenge experienced was lack
of technical capacity of CWs (Chewicha, 2013; Jeremie, 2014; MEASURE Evaluation, 2009). An
evaluation of a CBHIS in Kenya highlights the need for intensive training with periodic refresher
courses for CHWs involved in data collection (Jeremie, 2014). Another challenge regarding the
CWs or volunteers is the added workload of data collection and associated activities (Byrne, 2004;
MEASURE Evaluation, 2009; MEASURE Evaluation, 2014; MEASURE Evaluation, 2014;
MEASURE Evaluation, 2014). The added responsibilities may or may not be feasible for CWs,
depending on such factors as whether or not community-level staff earn a salary.

Another challenge in adding to new CW responsibilities is the lack, or early stages, of integration of
CBHIS in formal HMIS. Without complete integration, there are duplicative efforts in data
collection, analysis, and reporting. The family folder system in Ethiopia, for example, faced the
challenge of redundant data elements in registers within and across health posts and districts
(Damtew, 2013). A report on the uThukela District Child Survival Project also cited
nonoverlapping data sets as a major challenge (Byrne, 2004). Because CHW catchment areas are not
the same as that of the Department of Health, reporting to higher levels is difficult. Time spent
collecting and harmonizing redundant or nontransferable data is time that could otherwise be spent
serving the community (Damtew, 2013).

In two separate CBHIS efforts in Ethiopia, the need for a streamlined physical system is highlighted
(Chewicha, 2013; Damtew, 2013). The family folders were bulky and vulnerable to damage from
rain when carried from house to house, so HEWs ended up recording in registers instead and
transferring the data to folders later on (Damtew, 2013). In the Southern Nations, Nationalities and
People’s Region (SNNPR) of Ethiopia, each health post generated a large amount of data, which
became manually unmanageable (Chewicha, 2013).

Other challenges mentioned are: data collection tools often not in a language CWs are most
comfortable using (Chewicha, 2013); weak data quality (MEASURE Evaluation, 2009); high
turnover or vacant district-level positions (Byrne, 2004); and lack of data collection supplies in
remote areas (Chewicha, 2013). Asangansi (2013) addresses challenges unique to those CBHIS using
mobile technology, including the potential for lost or stolen phones and lack of existing policy

Community-Based Health Information Systems in the Global Context: A Review of the Literature 7
around electronic data privacy and security. Some CWs have reported that community members at
times express frustration when they provide information during data collection, complaining that
they often don’t see the improvements in programming or services that are needed (MEASURE
Evaluation, 2014; MEASURE Evaluation, 2014).

Discussion & Conclusion

As this review demonstrates, CBHIS come in various structures and cover diverse health areas, with
different uses of technology and integration in formal HMIS. COLNISA, in northern Nigeria, is
community-focused, with all data collected and used for interventions by CWs and volunteers and
no integration in facility or formal HMIS (Sabitu, 2004), whereas NEHSI, also in Nigeria,
automatically synchronizes data with the national HMIS (Asangansi, 2013). CBHIS can be used for
different goals, including identifying people in need of specific services, facilitating access to
community health services, and reporting to higher HMIS for program planning.

Some common challenges are low technical capacity of community staff (Chewicha, 2013; Jeremie,
2014; MEASURE Evaluation, 2009); burden of new data collection responsibilities on CHWs
(Byrne, 2004; MEASURE Evaluation, 2009); redundant data collection (Damtew, 2013; Byrne,
2004); and cumbersome paper-based data (Chewicha, 2013; Damtew, 2013). Duplicative data may
be reduced if key indicators from CBHIS are harmonized with and integrated in national HMIS.
Similarly, USAID/Kenya has identified ways to streamline reporting systems across implementing
partners, which paves the way for standardized country-level data (MEASURE Evaluation, 2014).
One potential solution to cumbersome physical data is use of electronic data collection systems, but
this is not always possible in lower-resource settings, and priority should be placed on collecting the
right data.

Low technical capacity and the burden of new job responsibilities of CHWs and other CWs are
challenges potentially connected to that of integration in the national HMIS. If CBHIS become
integrated, governments would likely need to consider employing more CWs, as well as providing
training courses and livable wages. Some CWs, such as CHEWs and HEWs, are already employed
by health ministries; others, including volunteers, may not be able to dedicate as much time and
effort to CBHIS, if the work can’t support their livelihoods.

This review is not without limitations. Its primary weakness is that only published articles and gray
literature were included. It is likely that other CBHIS exist that have not yet been discussed in the
literature or presented on a website. Therefore, the practices, strengths, and challenges outlined here
are not comprehensive. Thorough documentation and publication of CBHIS experiences is
encouraged to provide a fuller picture.

While CBHIS are relatively nascent and few comprehensive systems have been well-documented,
this review has shown some promising practices. CBHIS have the potential to shed light on unique
health issues and their causes in communities, as well as aid in improving case management, health

Community-Based Health Information Systems in the Global Context: A Review of the Literature 8
programming, and outcomes. As increased healthcare coverage and health equity become more
important to governments and other donors, the role of community health systems and their
information systems will continue to increase (Golding, 2014). Integration in national HMIS and
putting resources into the training and support of CWs will strengthen CBHIS.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 9
References
Asangansi, I., Macleod, B., Meremikwu, M., Arikpo, I., Roberge, D., Hartsock, B., Mboto, I.
(2013). Improving the routine HMIS in Nigeria through mobile technology for community
data collection. Journal of Health Informatics in Developing Countries, 7(1): 76-87.

Byrne, E. (2004). Participatory design of a community-based child health information system in

South Africa. eGovernment for Development. Available at:
http://www.egov4dev.org/health/case/childhealthis.shtml.

Chewicha, K., Azim, T. (2013). Community health information system for family-centered health
care: Scale-up in Southern Nations, Nationalities and People's Region. Ethiopia Ministry of
Health Quarterly Health Bulletin, 5(1):49-53.

Damtew, Z.A., Moges, A.S. (2013). From multiple register to family folder: the transition of data
collection and reporting tools for health extension workers in Ethiopia. Journal of Health
Informatics in Developing Countries, 7(2): 99-11.

de la Torre, C., Unfried, K. (2014). Monitoring and evaluation at the community level: a strategic
review of MEASURE Evaluation, Phase III accomplishments and contributions. Chapel Hill,
NC: MEASURE Evaluation, University of North Carolina.

DeRenzi, B. S. (2011). A framework for case-based community health information systems. Paper
presented at the iEEE Global Humanitarian Technology Conference in Seattle, WA., Oct. 30-
Nov. 1, 2015.

Global Fund. (2014). Community systems strengthening framework. Geneva: The Global Fund.

Golding, L. (2014). Strengthening community health systems through CHWs and mHealth.
Washington, DC: CORE Group.

Jeremie, N., Kaseie, D., Olayo, R., Akinyi, C. (2014). Utilization of community-based health
information systems in decision making and health action in Nyalenda, Kisumu County,
Kenya. Universal Journal of Medical Science, 2(4): 37-42.

Kaseje, D., Olayo, R., Musita, C., Oinda, C.O., Wafula, C., Muga, R. (2010). Evidence-based
dialogue with communities for district health systems' performance improvement. Global
Public Health, 5(6): 595-610.

Lunsford, S., Fatta, K., Stover, K., Shrestha, R. (2015). Supporting close-to-community providers
through a community health system approach: case examples from Ethiopia and Tanzania.
Human Resources for Health, 13:12.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 10
Management Sciences for Health. (2001). Expanding the delivery of health services through a
community-based monitoring and information system. Washington, DC: Management Sciences
for Health.

Marston, C., Renedo, R., McGowan, C., Portela, A. (2013). Effects of community participation on
improving uptake of skilled care for maternal and newborn health: a systematic review.
PLOS One, (8)2: e55012.

MEASURE Evaluation. (2009). Report of a technical consultation on information systems for

community-based HIV programs. Chapel Hill, NC: MEASURE Evaluation, University
of North Carolina.

MEASURE Evaluation. (2010). Community-level program information reporting for HIV/AIDS

programs: Introduction. Chapel Hill, NC: MEASURE Evaluation, University of
North Carolina.

MEASURE Evaluation. (2012). Community health information system in action in SNNPR. Chapel
Hill, NC: MEASURE Evaluation, University of North Carolina.

MEASURE Evaluation. (2012). Clarification regarding usage of the Child Status Index (CSI). Chapel
Hill, NC: MEASURE Evaluation, University of North Carolina.

MEASURE Evaluation. (2014). Case study series. Community-based information systems: Kenya.
Chapel Hill, NC: MEASURE Evaluation, University of North Carolina.

MEASURE Evaluation. (2014). Case study series. Community-based information systems: Tanzania.
Chapel Hill, NC: MEASURE Evaluation, University of North Carolina.

MEASURE Evaluation. (2014). Case study series. Community-based information systems: Zambia.
Chapel Hill, NC: MEASURE Evaluation, University of North Carolina.

MEASURE Evaluation. (2014). Case study series: Community-based information systems. Overview.
Chapel Hill, NC: MEASURE Evaluation, University of North Carolina.

Medic Mobile. (2015). Our tools [Website]. Retrieved from Medic Mobile: http://medicmobile.org/

Moneti, F. (1992). Community-based information systems. RRA Notes, 31-36.

Munyaneza, F., Hirschhorn, L., Amoroso, C., Nyirazinyoye, L., Birru, E., Mugunga J., Murekatete,
R., Ntagnira, J. (2014). Leveraging community health worker system to map a mountainous
rural district in low resource setting: a low-cost approach to expand use of geographic
information systems for public health. International Journal of Health Geographics, 13:49.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 11
Mutale, W., Chintu, N., Amoroso, C., Awoonor-Williams, K., Phillips, J. Baynes, C., Michel, C.,
Taylor, A., Sherr, K. (2013). Improving health information systems for decision making
across five sub-Saharan African countries: Implementation strategies from the African Health
Initiative. BMC Public Health Services Research, 13(2):S9.

Sabitu, K., Iliyasu, Z., Hassan, S., Mande, A. (2004). Evaluation of a community level nutrition
information system for action in a rural community of Zaria, Northern Nigeria. Annals of
African Medicine, 120-125.

Sahay, S., Braa, J. (2012). Integrated health information architecture - power to the users. India:
Matrix Publishers.

Singh, A. K. (1997). Primary care informatics: Bhorugram, India: revisited. Stud Health Technol
Inform, 43, Part B:884-888

Sudhof, L., Amoroso, C., Barebwanuwe, P., Munyaneza, F., Kaamaga, A., Zambotti, G., Drobac,
P., Hirschhorn, L. (2013). Local use of gepgraphic information systems to improve data
utilisation and health services: mapping caesarean section coverage in rural Rwanda. Tropical
Medicine and International Health; 18(1):18-26.

Tropical Institute of Community Health and Development (TICH) in Africa. (2005). Community
based information system (CBIS) development process. Available at:
http://portal.unesco.org/shs/en/files/9580/11477031801unisol_cbis_development.pdf/uniso
l+cbis+development.pdf.

UNAIDS. (2014). 90-90-90: An ambitious treatment target to help end the AIDS epidemic. Geneva:
Joint United Nations Programme on HIV/AIDS (UNAIDS).

USAID. (2015). USAID Global Health Programs: FY 2016 president's budget request ending
preventable child and maternal deaths. Washington, DC: USAID.

Community-Based Health Information Systems in the Global Context: A Review of the Literature 12
 WORKING PAPER

MEASURE Evaluation
University of North Carolina at Chapel Hill
400 Meadowmont Village Circle, 3rd Floor
Chapel Hill, North Carolina 27517
Phone: +1-919-445-9350 • [email protected]
www.measureevaluation.org

MEASURE Evaluation is funded by the U.S. Agency for International Development (USAID) under
the terms of Cooperative Agreement AID-OAA-L-14-00004 and implemented by the Carolina
Population Center, University of North Carolina at Chapel Hill, in partnership with ICF International,
John Snow, Inc., Management Sciences for Health, Palladium, and Tulane University. The views
expressed in this publication do not necessarily reflect the views of USAID or the United States
government. WP-16-161