C o m Community Health Worker m u n i t Incentives and Disincentives: y

H

e How They Affect Motivation, Retention, a l t h and Sustainability

W o r k e r

I n c e n t i v e s

a n d

D i s i n c e n t i v e s :

H o w

T h e y

A f f e c t

M o t i v a t i o n ,

R e t e n t i o n ,

a n d

S u s t a i n a b i l i t y

Contributors BASIC SUPPORT FOR INSTITUTIONALIZING CHILD SURVIVAL 1600 Wilson Blvd., Suite 300, Arlington, VA 22209 • Tel: 703.312.6800 • Fax: 703.312.6900 Karabi Bhattacharyya Peter Winch E-mail: [email protected] • Website: http://www.basics.org Karen LeBan Marie Tien

Community Health Worker Incentives and Disincentives: How They Affect Motivation, Retention, and Sustainability

Karabi Bhattacharyya Peter Winch Karen LeBan Marie Tien Abstract This paper examines the experience with using various incentives to motivate and retain community health workers (CHWs) serving primarily as volunteers in child health and nutrition programs in developing countries. It makes recommendations for more systematic use of multiple incentives based on an understanding of the functions of different kinds of incentives and emphasizes the importance of the relationship between a CHW and community. Case studies from Afghanistan, El Salvador, Honduras, and Madagascar illustrate effective use of different incentives to retain CHWs and sustain CHW programs.

Recommended Citation Karabi Bhattacharyya, Peter Winch, Karen LeBan, and Marie Tien. Community Health Worker Incentives and Disincentives: How They Affect Motivation, Retention, and Sustainability. Published by the Basic Support for Institutionalizing Child Survival Project (BASICS II) for the United States Agency for International Development. Arlington, Virginia, October 2001.

Credit Cover photo: Johns Hopkins Center for Communications Program (JHU/CCP).

BASICS II BASICS II is a global child survival project funded by the Office of Population, Health, and Nutrition of the Bureau for Global Programs, Field Support, and Research of the U.S. Agency for International Development (USAID). BASICS II is conducted by the Partnership for Child Health Care, Inc., under contract no. HRN-C-00-99-00007-00. Partners are the Academy for Educational Development, John Snow, Inc., and Management Sciences for Health. Subcontractors include Emory University, The Johns Hopkins University, The Manoff Group Inc., the Program for Appropriate Technology in Health, Save the Children Federation, Inc., and TSL.

This document does not represent the views or opinion of USAID. It may be reproduced if credit is properly given.

1600 Wilson Boulevard, Suite 300 • Arlington, Virginia 22209 USA Tel: 703-312-6800 • Fax: 703-312-6900 E-mail address: [email protected] • Website: www.basics.org Table of Contents

Acknowledgments ...... v

Acronyms ...... vii

Executive Summary ...... ix

Section 1 CHWs: The Context ...... 1 Background ...... 1 Attrition in CHW Programs ...... 2 Definitions of Key Words ...... 2 CHWs and Comprehensive Primary Health Care ...... 3 CHWs in the Context of Health Sector Reform ...... 4 CHWs and Community-IMCI ...... 5 Policy Environment and Decentralization ...... 6

Section 2 Methodology ...... 9 Organization of This Review ...... 9 Methodology ...... 9

Section 3 Who are CHWs and What Do They Do? ...... 11 Characteristics of CHWs ...... 11 Duties of CHWs ...... 11

Section 4 Monetary Incentives and Disincentives ...... 15 Money as an Incentive ...... 15 Problems with Using Money as an Incentive ...... 16 In-kind Payments as an Effective “Compromise” ...... 17

Section 5 Nonmonetary Incentives and Disincentives ...... 19 Supervision or Recognition ...... 19 Personal Growth and Development Opportunities ...... 21 Training ...... 21 Peer Support and CHW Networks ...... 24

Section 6 Relationship with the Community ...... 25 Enhancing the Relationship Between CHWs and Communities ...... 25 Selection of CHWs ...... 26 S

Community Recognition of CHW Work ...... 27 T N

Community Organizations that Support CHW Work ...... 29 E T N O C

F O

E L B A T

iii Section 7 Putting It All Together: Multiple Incentives ...... 31 Behavioral Model ...... 31 Examples of Multiple Incentives ...... 31

Section 8 Conclusions and Recommendations ...... 35 Support the CHW’s Relationship with the Community ...... 36 Use Multiple Incentives ...... 36 Match Incentives with Duties ...... 37 Employees or Volunteers? ...... 37 Importance of Monitoring ...... 37 Topics for Research ...... 38

Section 9 References ...... 39

Annexes Annex 1: Examples of the Use of Multiple Incentives ...... 43 Annex 2: Questionnaire for E-mail and Initial Interviews ...... 49 Annex 3: Interview Guide for BASICS Examples ...... 51

Tables Table 1. Alternative Titles for CHWs ...... 2 Table 2. Comparison of CHWs with Professional Health Staff ...... 3 Table 3. Roles of CHWs in the Implementation of the Three Elements of the HH/C IMCI Framework ...... 7 Table 4. Motivation Model ...... 32 Table 5. CHW Incentives and Disincentives Organized by a Systems Approach ...... 33 S T N E T N O C

F O

E L B A T iv Acknowledgments

he authors would like to acknowledge the following people who made comments on the various drafts of the paper: Renata Seidel, Mark Rasmuson, Rene Salgado, Marilyn Rice, T Paul Ickx, Lisa Sherburne, Stephan Solat, Judianne McNulty, William Brieger, Eric Swedberg, Peter Gottert, and Michael Favin. Adwoa Steel, Marcia Griffiths, Tina Sanghvi, and Alfonso Contreras were interviewed for some of the examples. Paul Ickx drafted the example from Afghanistan. The authors also thank Wendy Hammond and Kathleen Shears for editing the paper and Kathy Strauss for layout and design. S T N E M G D E L W O N K C A

v

Acronyms

ADRA Adventist Development and Relief Agency

AHSSP Afghanistan Health Sector Support Project

AIN Atención Integral a la Niñez

ARI acute respiratory infection

BHT bridge to health team

BHW barangay health worker (Philippines) or basic health worker (Afghanistan)

CARE Cooperative for Assistance and Relief Everywhere

CHA community health agent

CHC community health center

CHP community health promoter

CHVW community health volunteer worker

CHW community health worker

CRS Catholic Relief Services

DHO district health office

HA health agent

HAC health action committees

HC health center

HH/C household and community

IMCI Integrated Management of Childhood Illness

MOH Ministry of Health

NGO nongovernmental organization

ORS oral rehydration solution

TBA traditional birth attendant

UNICEF United Nations Children’s Fund

USAID United States Agency for International Development

VHC village health committee

VHW village health worker S M Y N O R C A

vii

Executive Summary

ver the past couple decades, a on experience. Indeed, the very effectiveness number of studies have shown that of CHW work usually depends on retention. Ocommunity health workers (CHWs) This paper examines experience with can help reduce morbidity and mortality in various incentives for CHWs and their impact certain settings. Health programs have on retention of CHWs and the sustainability of recruited and trained these primarily volunteer CHW programs. It reviews the types of workers to carry out a variety of health incentives that are needed to motivate promotion, case management, and service involvement, to retain CHWs once they have delivery activities at the community level. been trained, and to sustain their performance CHWs can serve as a bridge between at acceptable levels. Although there are professional health staff and the community important lessons to learn from other and help communities identify and address community-based development workers, this their own health needs. They can provide paper focuses primarily on community-based information to health system managers that workers who provide some type of health or may otherwise never reach them and can nutrition service. encourage those in the health system to understand and respond to community needs. Organization of the Paper CHWs can help mobilize community The paper is organized into the following resources, act as advocates for the sections. Section 1 discusses the context of community, and build local capacity. CHW programs, including the changes The overall environment of international brought on by health sector reform and the public health has changed dramatically with IMCI approach. Section 2 describes the health sector reform and decentralization. objectives and methodology of the review. Local governments have greater autonomy Section 3 reviews the typical characteristics and authority to develop and finance health of CHWs and provides an overview of the solutions appropriate to their locales. In child wide range of their functions. Section 4 looks survival programs, the Integrated at the various ways that cash can be both an Management of Childhood Illness (IMCI) incentive and a disincentive and reviews in- approach is changing the way sick children kind incentives. Section 5 discusses some of are managed and health facilities are the other program features that can motivate organized. As these dramatic changes in the CHWs to continue their work, including their public health context create new opportunities relationship to other health staff, their sense for programs that include community health of personal growth and accomplishment, workers, this is a critical time to review the training opportunities, and peer support. past experience and draw lessons for the Section 6 examines the critical relationship future. between CHWs and their communities. One of the most critical problems for Section 7 reviews how the use of multiple CHW programs is the high rate of attrition. incentives can contribute to CHW retention, Attrition rates are reported between 3.2 and Section 8 contains conclusions and Y percent and 77 percent. Higher rates are recommendations. R A

generally associated with volunteers. Such M high attrition rates lead to a lack of continuity Conclusions M U S in the relationship between a CHW and Perhaps the most important conclusion of this E V community, increased costs in selecting and review is that there is no tidy package of three I T

training CHWs, and lost opportunities to build incentives that will ensure motivated CHWs U C E X E

ix who will continue to work for years. Rather, a transportation to go to all the households. complex set of factors affects CHW Many CHWs are restricted to preventive and motivation and attrition, and how these factors promotive roles that leave them unable to play out varies considerably from place to respond to community demands for curative place. However, program planners can draw care (and usually medicines). on the extensive experience of the public Monetary incentives can increase health community with CHW programs. A retention. CHWs are poor people trying to summary of the main conclusions of the support their families. But monetary review follows. incentives often bring a host of problems CHWs do not exist in a vacuum. They because the money may not be enough, may are part of and are influenced by the larger not be paid regularly, or may stop altogether. cultural and political environment in which Monetary incentives may also cause they work. The process of health sector problems among different cadres of reform, the adoption of the IMCI strategy, development workers who are paid and not and the progress made by community-based paid. However, there are some success nutrition programs have generated renewed stories of programs paying CHWs. Many interest in the potential contribution of CHWs. programs have used in-kind incentives Health sector reform has changed the effectively. supervisory structure within health systems Non-monetary incentives are critical to and given more autonomy to peripheral the success of any CHW program. CHWs health facilities. It has also decentralized the need to feel that they are a part of the health control of health funds, allowing greater system through supportive supervision and flexibility in spending for various types of appropriate training. Relatively small things, CHW incentives. The IMCI strategy includes such as an identification badge, can provide a a training curriculum on assessing and sense of pride in their work and increased treating mild and moderate childhood status in their communities. Appropriate job illnesses. Such training allows CHWs to play aides such as counseling cards and regular a curative role, which is usually what replenishment of supplies can help ensure communities demand. Policies on CHW that CHWs feel competent to do their jobs. distribution of antimicrobials and Peer support can come in many forms, such antimalarials can have tremendous effects as working regularly with one or two other on their relationship with the community. CHWs, frequent refresher training, or even The motivation and retention of CHWs is CHW associations. influenced by who they are in the community In the end, the effectiveness of a CHW context. The inherent characteristics of comes down to his or her relationship with the CHWs, such as their age, gender, ethnicity, community. Programs must do everything and even economic status, will affect how they can to strengthen and support this they are perceived by community members relationship. First, program planners must and their ability to work effectively. recognize the social complexity of At the micro level, the specific tasks and communities and that communities are not all duties of CHWs affect their motivation and alike. Different communities will need retention. When given too many tasks, CHWs different types of incentives, depending on the Y

R feel overwhelmed with information or may other job opportunities available, prior A

M spend so much time in training that they experience with CHWs, the economic

M rarely practice what they have learned. Often situation of the community, and other factors. U S

the catchment areas they cover are too large Unfortunately, very little experience or E V

I with too many households, making it difficult guidance is available on how best to T

U for a CHW to spend the time or find the differentiate communities. It is important to C E X E x CHW Incentives and Disincentives Organized by a Systems Approach

Incentives Disincentives

Monetary factors ■ Satisfactory remuneration/ ■ Inconsistent remuneration that motivate Material Incentives/Financial ■ Change in tangible incentives individual CHWs Incentives ■ Inequitable distribution of ■ Possibility of future paid incentives among different employment types of community workers

Nonmonetary factors ■ Community recognition and ■ Person not from community that motivate respect of CHW work ■ Inadequate refresher training individual CHWs ■ Acquisition of valued skills ■ Inadequate supervision ■ Personal growth and ■ Excessive demands/time development constraints ■ Accomplishment ■ Lack of respect from health ■ Peer support facility staff ■ CHW associations ■ Identification (badge, shirt) and job aids ■ Status within community ■ Preferential treatment ■ Flexible and minimal hours clear role

Community-level ■ Community involvement in ■ Inappropriate selection of factors that motivate CHW selection CHWs individual CHWs ■ Community organizations ■ Lack of community that support CHW work involvement in CHW ■ Community involvement in selection, training, CHW training and support ■ Community information systems

Factors that motivate ■ Witnessing visible changes ■ Unclear role and communities to ■ Contribution to community expectations (preventive support and sustain empowerment versus curative care) CHWs ■ CHW associations ■ Inappropriate CHW behavior ■ Successful referrals to ■ Needs of the community not health facilities taken into account

Factors that motivate ■ Policies/legislation that ■ Inadequate staff and MOH staff to support support CHWs supplies and sustain CHWs ■ Witnessing visible changes ■ Funding for supervisory activities from government and/or community involve communities in all aspects of the interactions with all community members and Y

CHW program but especially in establishing in how to respond to difficult people or R A

criteria for CHWs and making the final situations. Community-based organizations, M selection. Programs can provide opportunities such as religious groups or youth clubs, can M U S for quick visible results that will promote provide support to CHWs and significantly E V community recognition of CHWs’ work. CHWs lessen their load by taking on health I T

must be trained in appropriate and respectful education activities. U C E X E

xi Many successful programs use multiple systems approach that helps an implementer incentives over time to keep CHWs consider what can be done to support a CHW motivated. A systematic effort that plans for at different levels of the system. multiple incentives over time can build up a CHW’s continuing sense of satisfaction and Vary the incentives based on CHW duties. fulfillment. CHWs continue to play an important role in many international primary health care Recommendations programs. While continuing their preventive Support the CHW’s relationship with the and community mobilization tasks, CHWs are community. increasingly becoming involved in community- The fact that the effectiveness of the work of based case management of prevalent the CHW depends almost entirely on his or childhood diseases. The Community-IMCI her relationship with the community is framework lays out three elements of surprisingly often overlooked. Many programs implementation and describes different types focus on clinical training, supervisory of communities where those elements are checklists, and logistics (all of which are appropriate. The role of CHWs, and extremely important), to the exclusion of consequently their incentives, will vary among activities that support the community the elements. relationship. Effective programs have (explicitly or implicitly) oriented the whole Are CHWs volunteers or employees? program to support and strengthen every In general CHWs are not paid salaries interaction that a CHW has with community because the MOH or donors do not consider members. Many examples of such efforts are salaries to be sustainable. Yet CHWs are often given throughout the paper, from public held accountable and supervised as if they recognition of CHW work by supervisors to job were employees. CHW programs must aides that support the ongoing dialogue recognize that CHWs are volunteers, even if between community members and CHWs. they receive small monetary or nonmonetary Programs must continually ask what can be incentives. They are volunteering their time to done to promote beneficial interactions. serve the community.

Use multiple incentives. Continue to understand your program. In most of the programs reviewed, incentives Many programs do not understand why their were implemented in an ad-hoc manner rather CHWs drop out. Programs would be well than as part of a systematic program. It would served by monitoring some of the most be useful to identify the functions of each of important factors that affect a CHW’s the incentives used to understand which are motivation and desire to stay on the job. the critical functions and how those might vary Program managers must stay abreast of the based on the CHW role and type of community. “competition”: what other jobs and Intrinsic incentives work to promote a sense opportunities are available for the CHWs? that the work is worthwhile, while extrinsic When new tasks or functions are added, incentives include salary and increased status programs should assess how CHWs are within the community and with colleagues. It is managing the increased workload. How do Y

R clear that both intrinsic and extrinsic incentives community members and the CHW interact? A

M should be implemented and monitored. The What demands are community members

M table on page xi shows a list of CHW making on the CHWs? How do their U S

incentives, including key disincentives supervisors and other staff in the health E V

I mentioned in the literature and through system treat CHWs? Do the CHWs have the T

U personal communi-cations, organized into a training and job aides they need to be C E X E xii effective and feel competent in their jobs? realistic attrition rates? What is the most What monetary or nonmonetary incentives efficient way to monitor CHW programs? What would increase their motivation and support are successful ways of reducing attrition and their work? increasing retention of CHWs? What are some financing strategies to pay CHWs in a regular More research is needed. and sustainable manner? What are the critical The question of how to sustain a long-term functions that are achieved by different CHW program and to retain workers requires incentives? What are the most important additional investigation. It is unfortunate that differences among communities that affect despite the vast experience with CHWs, CHW programs, and what is the best way to relatively little scientific evidence is available assess these differences? How can planners to answer some of the basic questions: What efficiently tailor their programs to meet local are the current attrition rates? What are needs? Y R A M M U S

E V I T U C E X E

xiii

Section 1

CHWs: The Context

ommunity health workers (CHWs) are not a new concept. Health Cprograms have recruited and trained CHWs to carry out a variety of health promotion, case management, and service delivery activities at the community level for several decades. CHWs can serve as a bridge between professional health staff and the community and help communities identify and address their own health needs. They can provide health system managers with information that may otherwise never reach them and can encourage those in the health system to understand and respond to community needs. CHWs can help mobilize community resources, act as advocates for the community, and build local capacity (Center for Policy Alternatives 1998).

Background paradigm shift in the management of sick Yet for many public health practitioners, children and organization of health facilities. especially donor agencies, CHWs are a failed And with the advent of the community concept. Both the effectiveness of CHWs as component of IMCI, CHWs have been sought change agents and the feasibility of out as effective community agents with a role implementing and sustaining large-scale CHW to play in the prevention of disease, the programs have been called into question. promotion of healthy behaviors, and, in some After being touted earlier as a key component places, the case management of sick of the strategy of Health for All by the Year children. 2000, CHW programs have frequently failed to These dramatic changes in the public live up to the expectations of a dynamic health context have created new opportunities grassroots movement. for CHW programs. Now is a critical time to Despite this perception, CHW programs review past experience and draw lessons for are worth reviewing. Over the past two the future. Although there may be important decades, a number of studies have shown lessons to learn from other community-based that CHWs can help reduce morbidity and development workers, this paper focuses on mortality in certain settings. The overall community-based workers who provide some environment of international public health has type of health or nutrition service. changed dramatically with health sector CHW programs face many problems, reform and decentralization, giving local including poor training, inadequate supervision, T X governments greater autonomy and authority lack of supplies, and poor relationships with E T to develop and finance health solutions communities. One of the most frustrating N O C appropriate to their locales. In child survival elements of many CHW programs is their high E

programs, the Integrated Management of attrition rate. This paper examines recent H T

Childhood Illness approach has created a experiences with CHW programs to determine : S W H C

1 the factors that contribute to high motivation Walt (1998) provides the following and low attrition of CHWs. Identifying these common definition of CHWs as: incentives will help program planners develop sustainable CHW programs. ... generally local inhabitants given a limited amount of training to provide Attrition in CHW Programs specific basic health and nutrition Attrition rates for CHWs of 3.2 percent to 77 services to the mothers of their percent are reported in the literature, with surrounding communities. They are higher rates generally associated with expected to remain in their home volunteers (Walt 1989). One review (Parlato village or neighborhood and usually and Favin 1982) found attrition rates of 30 only work part-time as health workers. percent over nine months in Senegal and 50 They may be volunteers or receive a percent over two years in Nigeria. CHWs who salary. They are generally not, depend on community financing have twice however, civil servants or professional the attrition rate as those who receive a employees of Ministry of Health. government salary. In the Solomon Islands, attrition was attributed to multiple causes in addition to inadequate pay, including family Table 1. Alternative Titles for CHWs reasons, lack of community support, and upgrading of health posts (Chevalier 1993). Title Country High attrition rates cause several Activista Mozambique problems. Frequent turnover of CHWs Anganwadi India means a lack of continuity in the Animatrice Haiti relationships established among a CHW, Barangay health worker Philippines community, and health system. Basic health worker India Considerable investment is made in each Brigadista Nicaragua CHW, and program costs for identifying, Colaborador voluntario Guatemala screening, selecting, and training the CHW Community health agent Ethiopia rise with high attrition rates. When CHWs Community leave their posts, the opportunity is lost to health promoter Zambia build on their experience and further develop Community their skills over time through refresher health representative various countries training. The very effectiveness of CHW Community work usually depends on retention. health volunteer Malawi Community Definitions of Key Words nutrition worker India Community Community Health Workers resource person Uganda Since the role of the CHW was re-emphasized Female multipurpose during the Alma Ata conference in 1978, there health worker Nepal have been several variations and definitions of Health promoter various countries this term. The specific roles and Kader Indonesia responsibilities of CHWs vary greatly among Monitora Honduras T

X countries, depending on people’s access to Outreach educator various countries E

T health care and the presence of other cadres of Promotora Honduras N Rural health motivator Swaziland O health workers. Table 1 shows the vast array of C titles used for CHWs. In this report the term Sevika Nepal E

H CHW refers to all of these titles, with the local Village health helper Kenya T

: titles used to refer to specific examples. Village health worker various countries S W H C

2 Section 1

A distinction should be made between Motivation: Desire to serve and perform CHWs and other professional health staff to effectively as a CHW explain the unique role and background of CHWs. Table 2 presents an example from Retention: Length of time that an Latin America comparing two groups of health individual CHW actively workers—professional and volunteer. This performs appropriate comparison applies in most other countries community primary health and regions. care tasks Many definitions of CHWs also reflect expectations of their roles. For example, Sustainability: A continuing system of “…members of the communities where they recruitment, training, and work; should be selected by the communities; supervision of a cadre of should be answerable to the communities for volunteers in a community or their activities; should be supported by the district that meets its member health system but not necessarily part of its or group health care needs organization; and have a shorter training [period] than professional workers” (Frankel CHWs and Comprehensive 1992). CHWs usually provide services to Primary Health Care people living in specific catchment areas, During the Alma Ata conference in 1978, most often the areas where they themselves CHWs were identified as one of the live. cornerstones of comprehensive primary health care. CHWs had the potential to deliver Other Definitions equitable health services to populations living The following definitions and assumptions for in remote areas and to help fill the unmet the key concepts of CHW incentives, demand for regular health services in many motivation, retention, and sustainability are countries. Such workers could solve the used throughout the paper. problems of poor access to health care and the high cost of doctors and reduce the social and Incentives: Positive or negative, intrinsic cultural barriers to health care. While providing or extrinsic factors simple technical and educational interventions, influencing CHW motivation CHWs could also serve as an entry point into and volunteerism (Note: this the larger cultural, environmental, political, and paper focuses on incentives social factors that affect health. They could be that can be addressed agents of change stimulating community programmatically.) participation in efforts to resolve the causal

Table 2. Comparison of CHWs with Professional Health Staff

Auxiliary nurses or health technicians Health promoters or village health (professional health staff) workers (volunteers from the community)

■ Primary education plus 1–2 years ■ Third grade education plus 1–6

of training months of training T

■ ■ X

From outside the community From the community E

■ Employed full time ■ Employed part time T N ■ ■ Salary usually paid by the program Supported by farm labor or other O C

(not by the community) community help ■ May be traditional healers E H T

:

Source: Walt 1988 S W H C

3 factors of illness (Frankel 1992). In other homes, compared to a reduction of only 15 words, CHWs were thought to be the magic percent when it was available only through bullets of primary health care. pharmacists (Parlato and Favin 1982). In Nepal, After Alma Ata many countries initiated where CHWs were provided with antibiotics and programs to scale up local CHW initiatives to training in the diagnosis of pneumonia, an the national level. By the early 1990s, evaluation found that the case management by however, enthusiasm for CHWs was waning CHWs was correct in 80 percent of cases among ministries of health and donor (Dawson et al. 2001). A study in Ecuador found organizations for a variety of reasons. that CHWs were much more cost effective than hospital-based workers in vaccinating children ■ Countries could not replicate CHW (San Sebastian et al. 2001). Among populations programs and take them to scale while with limited access to basic health care, CHWs maintaining the original levels of altruism, can facilitate referrals. There is growing commitment, and effectiveness. evidence that CHW programs are an appropriate ■ The barefoot doctor movement in China, way to address concerns about equity among which was for many the model for scaling underserved populations, as well as to increase up CHW programs, declined in both the effectiveness of programs. Two impact coverage and effectiveness in the wake studies carried out in Colombia suggest that of economic reform and conversion of smaller-scale programs (within a neighborhood, much of the health sector to the fee-for- municipality, district, or ethnic community) service model. developed through community participation are ■ Attempts to create a global (one-size-fits- more effective in changing health-related all) approach to CHW programs led to practices than larger programs (Quinones 1999). inflexible programs and approaches. Along with other community resource people, ■ Use of the mass media and social CHWs can be critical to the promotion of key marketing seemed to be more cost preventive and treatment behaviors. effective than CHWs in promoting While CHWs may not be appropriate in all changes in health-related behaviors at the settings, they cannot be dismissed as a national level. programmatic option. Mass media and social ■ Attempts were initiated to upgrade the skills marketing approaches cannot by themselves of existing community-level health care provide the depth of interaction necessary to providers (pharmacists, private physicians, change complex health behaviors. For example, traditional healers) rather than to create new the media may promote eating foods rich in cadres of health workers who might cease vitamin A, but CHWs can work with individual activities once funds run out. families to develop acceptable recipes and use ■ The use of CHWs as agents for specific locally available foods. Similarly, approaches vertical programs, such as for diarrhea that rely exclusively on the private sector may and malaria, diminished the advantages have limited impact where commercial outlets of the holistic approach and often led to are rare. Healthy behaviors need to be multiple training with less time on the job. negotiated continually so that families can apply them to their own situations. CHWs can be a Despite these limitations, there is evidence practical and effective way to do this in many, T

X that CHWs can reduce morbidity and mortality though not all, situations. E

T under certain conditions and therefore provide N

O outreach and services that governments may CHWs in the Context of C not otherwise be able to deliver. In Egypt infant Health Sector Reform E

H mortality was reduced by 40 percent when oral Most countries in Africa and much of the rest T

: rehydration solution (ORS) was distributed to of the world are undergoing health sector S W H C

4 reform that involves the reorganization of the Governments that have devolved local entire health sector. Health sector reform decision making and authority to districts, usually includes decentralization, the municipalities, and villages seem to be introduction of fees for a variety of health successful in applying general government services, integrated health care packages, guidance flexibly to meet their local health and donor coordination. needs. Appropriate legislation and policies can The focus on health sector reform and help CHWs organize and retain members and decentralization has led to a renewed maintain a CHW system over time. Two emphasis on ways to extend coverage to examples are discussed on page 6. underserved areas as well as to increase local involvement in decision making regarding CHWs and Community IMCI health service delivery. The problems faced by In 1995 WHO and UNICEF launched the health systems are illustrated in a recent study Integrated Management of Childhood Illness of health services in Burkina Faso (Bodart et strategy. IMCI is an integrated approach to the al. 2001). The study documents a steady assessment, classification, and treatment of decline in the use of curative services, a sick children that combines aspects of nutrition, strong urban bias in public spending on health, immunization, disease prevention, and and high cost of care to patients. Policymakers promotion of growth and development. The are concerned about how to address these and strategy addresses the illnesses and health other issues to improve access to health care problems responsible for the majority of deaths among underserved populations. Health sector among children under five years of age and the reform, including decentralization of civil and fact that children are often ill from multiple health systems, provides a new opportunity to causes. IMCI ensures that a child who is revitalize CHW programs and develop local brought into a health center for diarrhea will also solutions to CHW incentives. be treated for malaria or pneumonia if needed. The role and tasks of CHWs vary by Until recently IMCI focused on improving district and community, and the type of health systems (including drug availability) and incentives needed to support the CHWs in the skills of health workers to assess, classify, these tasks need to be locally specific as and treat children accurately. In 1997 the well. As part of the movement to decentralize household and community (HH/C) component health services, ministries of health and donor of IMCI was launched (Lambrechts et al. 1999). organizations are turning increasingly to A framework has been defined for planning and nongovernmental organizations (NGOs)1 as implementing Community IMCI (Steinwand natural partners for extending coverage and 2001; Winch et al. 2001) that includes the three scaling up interventions. CHWs frequently following elements: play a key role in the operations of NGOs at the village and district levels. 1. Improving partnerships between health Because programs funded by facilities and services and the governments can often change or disappear communities they serve from one administration to the next, many 2. Increasing appropriate and accessible large-scale government CHW payment care and information from community- programs have not been sustained in the long based providers T term (Frankel 1992). The decentralized 3. Integrating promotion of key family X E framework provides new opportunities for the practices critical for child health and T N support and motivation of CHWs. nutrition O C

E H T 1. In this document NGO refers to U.S.-based private voluntary organizations and their international and local :

nongovernmental partners. S W H C

5 The three elements are to be implemen- CHW programs. Two examples of such ted by working across many sectors (a restructuring are described below. “multisectoral platform”) in order to address In 1995 the Philippine government enacted the social, economic, and environmental the Barangay Health Workers Act of 1995, factors that facilitate or hinder the adoption of which granted benefits and incentives to key family practices. Each of these elements accredited barangay health workers (BHWs). addresses a point of influence critical to The act included such provisions as appropriate child health care. At each point of subsistence allowance, career enrichment influence, CHWs are often involved in programs, recognition of years of primary health delivering services and messages and in care, special training programs, and preferential mobilizing the community (Table 3). access to loans. Experience from a pilot The role of a CHW varies with the strategy advocacy campaign showed that with adequate chosen to implement each element that is information and motivation, local government appropriate for a specific setting. Clearly, units were willing to provide financial and however, CHWs of some sort are critical to the logistical support, including a transportation success of Community IMCI. Implementation of allowance and budget for regular upgrade the community component of IMCI requires a training, as well as formal recognition of the cadre of workers to deliver needed services to roles of BHWs. BHWs have now organized peripheral areas, promote child wellness and themselves at the barangay level and in many good nutritional status, prevent child illness, and areas are federated at the municipal and link communities with health facilities that may provincial levels and are becoming a significant be underutilized. Although private practitioners political force for child survival (Paison 1999). of some kind may be available even in remote A program in Ceara, Brazil, found that a areas, the fee-for-service approach does not decentralized approach using paid health agents lend itself to preventive services or to nutrition- (HAs) could improve access to health care. The related interventions. Organizations such as the HAs had to have lived in the community for the Pan American Health Organization (PAHO) are previous five years. They also had to be over therefore promoting CHWs as the entry point for 18, able to work eight hours a day, and the implementation of Community IMCI. committed to social service. Each HA visits 75 households (225 in urban areas) once a month Policy Environment and to provide health education and minor curative Decentralization treatment. Nurses from the nearest clinic The overall health policy environment can supervise them. The agents earn the equivalent dramatically affect CHW programs. In Indonesia of US $112 a month (twice the average local the kader system survived on the tradition of monthly income), which is paid out of tax funds volunteerism and the support of the PKK, a from the state government to insulate the HAs national organization of the wives of political from local politics. To ensure local support for leaders (Favin 2001). In Mozambique the HAs, municipal governments must use community members who had been happy to some of the newly decentralized funds to volunteer under socialism are now demanding employ the nurse supervisors before the state cash incentives with the move toward funds can be released. The results have been a capitalism (Favin 2001). Policies often determine well-trained cadre of health workers and T

X who is eligible to become a CHW and whether dramatic improvements in child health, with an E

T the CHWs can administer antibiotics or receive infant mortality reduction of 32 percent. N

O cash incentives. With health sector reform, Unfortunately, attrition rates are not reported C some ministries of health are restructuring their (Svitone et al. 2000). E H T

: S W H C

6 Table 3. Roles of CHWs in the Implementation of the Three Elements of the HH/C IMCI Framework

Element CHW role

1. Improving partnerships between health ■ Help health facilities conduct community facilities or services and the communities outreach. they serve ■ Involve community members in planning and implementing health programs and services. ■ Raise awareness in the community about improvements to health services. ■ Educate community members about danger signs requiring care at health facilities. ■ Participate in data collection for community health information systems. 2. Increasing appropriate, accessible care and ■ Provide effective basic care (e.g., oral information from community-based rehydration therapy, antipyretic drugs) for providers sick children. ■ In some areas, treat sick children with other first-line drugs, such as chloroquine and cotrimoxazole, and advocate against harmful practices, such as injections. ■ Refer sick children to appropriate health facilities when advanced care is required. ■ Serve as a bridge to other providers (private sector and traditional healers). 3. Integrating promotion of key family practices ■ Engage communities in selecting behaviors critical for child health and nutrition to be promoted and identifying actions to be taken. ■ Promote key family practices for enhanced physical growth and mental development, prevention of disease, appropriate home care, and appropriate care-seeking behavior through individual counseling and community meetings. T X E T N O C

E H T

: S W H C

7

Section 2

Methodology

he renewed interest in CHWs described on pages 5-6 is leading T policymakers and program managers to examine critically past experience with CHW programs. In light of all the problems brought by high attrition, managers are asking what types of incentives are needed to motivate involvement of CHWs, retain them once they have been trained, and sustain their performance at acceptable levels. This review examines experience with various incentives for CHWs and their impact on the retention of CHWs and the sustainability of CHW programs.

Organization of This Review Group in the fall of 1999. BASICS and NGO Section 3 of this paper reviews the typical staff had observed that minimal tokens of characteristics of CHWs. Section 4 provides recognition could make a world of difference an overview of the wide range of their in enhancing community participation and functions. Section 5 looks at the various ways increasing volunteer retention. By looking cash can serve as both an incentive and a more closely at incentives, BASICS II hoped disincentive and reviews the use of in-kind to learn how they affect the motivation, payments. Section 6 discusses other program retention, and sustainability of CHWs. features that can motivate CHWs to continue This paper is based on a review of the their work, including relationships with other literature and interviews with program staff health staff, sense of personal growth and from many organizations. A literature search accomplishment, training opportunities, and was conducted using the Internet databases peer support. Section 7 examines the critical PubMed, Medline, and Popline. The key words relationship between the CHW and the motivation, incentives, sustainability, CHWs, community. Finally, Section 8 reviews the volunteers, and developing countries were contribution of the overall policy environment used in different combinations during the and the mix of incentives that can contribute search. Several books and articles to CHW retention. recommended by practitioners working with CHWs were also used for the paper. Since Methodology much of the information on CHWs is in the A review of incentives for CHWs was unpublished (“gray”) literature, this literature identified as a priority at a meeting between was identified through personal contacts and the staff of the Basic Support for through a request for information that included Institutionalizing Child Survival II (BASICS II) a questionnaire sent to the CORE Child Project and the Child Survival Collaboration Survival Community Group list serve and the 2 and Resources Group (CORE) IMCI Working MSH Community Health list serve. The Y G O L O

2. A network of more than 35 U.S. NGOs working together to improve primary health care programs for women and children and the D communities in which they live. O H T E M

9 questionnaire generated over 30 responses. in Annex 1 and focused on specific examples Some of these responses led to further and experiences from work with CHWs. interviews with over a dozen practitioners. A draft of this paper was circulated to all Two broad categories of literature were BASICS II technical field and headquarters identified and reviewed. The first category staff and through the CORE list serve in included papers reviewing the history and August 2000 to solicit additional information experience of CHW programs, and the second and comments. Further interviews were category included specific examples from conducted with project staff to elicit the programs working with CHWs. One of the key BASICS experience with CHWs. documents used in this paper is the 1982 Semistructured interviews were conducted paper Progress and Problems: An Analysis of with staff in the Ecuador, El Salvador, 52 AID-assisted Projects, by Parlato and Honduras, Madagascar, Nigeria, and Zambia Favin. Although almost 20 years old, this programs. The interview guide in Annex 2 paper is unique in systematically reviewing covers a wide range of topics, including the the experience of a wide range of primary selection of CHWs, support from the health health care programs. The authors provide system, training opportunities, and monetary important details about the work of CHWs in and nonmonetary incentives. On the basis of these programs, although they include little these comments and additional experiences, information about incentives. substantial revisions were undertaken and the Initial group and private interviews were final paper completed. conducted with BASICS II staff, NGO staff The literature review and interviews active in child survival programs, academics, focused primarily on CHWs working in child and current and former staff from programs health, rather than other types of development funded by the United States Agency for workers, such as family planning workers, International Development (USAID). These traditional birth attendants, or agricultural interviews were based on the interview guide outreach workers. Y G O L O D O H T E M

10 Section 3

Who Are CHWs and What Do They Do?

ong before considering money, t-shirts, or other external L incentives, programs must decide whom to select as CHWs. The characteristics of the CHWs are usually far more important for their ability to function effectively than their external incentives.

Characteristics of CHWs CHWs, 23 percent women, and 37 percent No prescription for the ideal CHW exists, but both men and women. Many programs require programs must understand the role and that CHWs be literate (primary school status of the people who work as CHWs in educated) so that they can record health order to plan appropriate incentives. In many information and use written materials. Other cultures men cannot visit and talk with programs have developed ways to record women, and women cannot travel alone to information for nonliterate people, such as other communities or talk with people in using color-coded cards or pebbles in boxes unfamiliar households. The ethnic group, (Storms 1979). Literacy requirements often religion, or language skills of CHWs are often affect the age of the selected CHWs: literate critical to their ability to work effectively. people tend to be younger. There is some Most non-Western cultures place greater evidence, on the other hand, that older CHWs emphasis on ascribed characteristics (those are more respected in their communities inherent in the person, such as age or (Ofosu-Amaah 1983). gender) than on achieved characteristics, such as special training. Thus, in some Duties of CHWs cultures young unmarried women are not The responsibilities of CHWs, as in any job, viewed as people with health expertise even are tied to the need and expectation of when they have received extensive training. various incentives. The specific duties and Similarly, communities respond differently to functions of CHWs can dramatically CHWs from inside the community than to influence their effectiveness and motivation ?

those from outside. Some community to stay on the job. A CHW may assist the O D members may feel that because insiders are health system in improving access to health, Y

“just like me” they have no special promoting preventive health messages, E H T knowledge. Others may feel that insiders providing nutritional counseling or curative understand their situation far better than care, and helping community residents find O D

outsiders. Such “insider” CHWs “…are armed other health care options through referrals. T A with knowledge that no professional can Some CHWs work only in health promotion H W match: an intimate knowledge of their own and have no curative functions. All national D culture” (Quinones 1999). health programs operate with financial N A

Most CHWs are from the communities constraints and limited trained staff. As a S where they work, but their personal result, these programs tend to add tasks and W H C

characteristics vary widely among countries. functions continually to the duties of existing E

One review of 38 projects (Parlato and Favin staff, especially CHWs. Although there is no R A 1982) found that 40 percent enlisted men perfect equation for the combination of CHW O H W

11 duties, recent experience sheds light on scientific evidence of the optimal number and some of the issues surrounding the role of mix of CHW functions and tasks. Programs the CHW. must carefully monitor CHWs’ workloads and their effects on motivation as additional tasks Single or Multiple Focus are added. The Alma Ata Declaration enumerated the following tasks expected of CHWs: “home Mix of Curative and Preventive visits, environmental sanitation, provision of Services water supply, first aid and treatment of simple Whether CHWs have a single or multiple and common ailments, health education, focus, the balance between curative and nutrition and surveillance, maternal and child preventive care has been identified as an health and family planning activities, issue. Prevention is extremely hard to sell in communicable disease control, community all public health programs. When curative care development activities, referrals, record- is offered, it is generally more welcomed and keeping, and collection of data on vital events” appreciated by the residents (Frankel 1992; (Ofosu-Amaah 1983). This long list of Heggenhougen et al. 1987; Walt et al. 1989; responsibilities would seem unreasonable to Curtale et al. 1995). A report from Tanzania demand of volunteer workers. Some programs noted that “CHWs have expressed frustration have trained several people and divided the at not being able to provide the quality of functions among them. For example, in Nepal services demanded by the community and the village health worker does the basic therefore want further training in curative preventive and curative work, while the medicine” (Heggenhougen et al. 1987). With community health leader motivates the disappointment on the part of the villagers community to participate in special campaigns and feelings of inadequacy among the CHWs, (Parlato and Favin 1982). A team of workers the relationship has been “characterized by a often exacerbates the problem of incentives, lack of support from the community….” The however, because more people require support. Tanzania report states that “unless the Many programs have trained CHWs to community’s expectations change, the lack of work in a single area, such as diarrhea, support for the CHWs will be aggravated if the malaria, or nutrition. For example, in many preventive role predominates over their parts of Latin America, volunteer collaborators curative activities….” conduct treatment and surveillance for “Credibility of CHWs is highly dependent malaria. They make home visits to people with on the workers’ curative role,” find Parlato and symptoms of malaria, complete patient Favin (1982). In Nepal community health reports, take blood smears, and administer volunteers who were able to treat acute doses of chloroquine (Ruebush et al. 1994). respiratory infection (ARI) greatly increased The advantage of CHWs with a single focus is their credibility among the village population. that they can be trained and monitored to (Curtale et al. 1995). A lack of curative skills perform a manageable set of tasks. The main may be a disincentive for CHWs, disadvantage is frequent training and compromising their standing in the retraining in various vertical programs, with no community. (Gilson et al. 1989). Given a opportunity for integration. Experience in choice between preventive and curative care, Madagascar suggests that a CHW can community members demand more curative manage only three or four themes at the most care (Walt et al. 1989), and problems often

Y (Gottert 2001). arise when CHWs cannot meet community G

O Apart from a consensus that no one demands. The relationship between the CHW L

O person can manage all the activities laid out and the community must be monitored and D

O in the Alma Ata Declaration, there is little supported to ensure an effective partnership. H T E M

12 Drugs with recruiting CHWs from practicing drug Closely linked to the importance of providing sellers and pharmacists has shown that this curative care are CHWs’ access to and supply strategy makes drugs available, gives the of drugs. The kind of medicines CHWs should drug sellers greater prestige, and greatly be allowed to administer has been the subject reduces attrition (Ishan 2001). of much debate. Many are concerned that In Nigeria’s Gongola State, village health treatment with antibiotics and antimalarials, in workers (VHWs) were trained to work in particular, might lead to overuse and misuse remote villages to treat common diseases of these medicines and eventual increases in with basic drugs and provide health education. drug resistance. Those who advocate An operations research study conducted to inclusion of these drugs in CHW kits argue determine what contributed to the high VHW that they are readily available from local attrition rate found that one of the main pharmacists and drug sellers and that trained reasons was villagers’ dissatisfaction with the CHWs may be able to promote proper usage. VHWs’ limited curative role. The VHWs’ lack The respect and status of CHWs in their of training or licenses to give injections communities unquestionably increases when created a discrepancy between what the they have drugs at their disposal. In their community wanted and what the VHW could review of 52 projects, Parlato and Favin (1982) provide (Gray and Ciroma 1987). found that “CHWs’ credibility suffers when drug supplies are irregular.” Recent experience Y G O L O D O H T E M

13

Section 4

Monetary Incentives and Disincentives

y definition a CHW is not usually a full-time salaried employee of B the ministry of health (MOH) or other organization. The primary reason is the belief that the MOH cannot afford to pay CHWs over the long term. Compensation of CHWs for their services, however, is a recurrent issue in many programs. CHWs often work long hours, even full time, alongside salaried employees, which inevitably leads to demands by CHWs for regular compensation for services provided. While full-time salaried CHWs are relatively rare, many CHWs receive some type of cash incentive. This section of the paper reviews cash incentives of all types.

Money as an Incentive month in 1984). Men with lower monthly There are many advantages to providing incomes worked two years and women with CHWs with cash incentives. From the program lower incomes worked one year, while men perspective, paid CHWs can be asked to work with higher pay stayed an average of 3.25 longer hours to achieve specific objectives years and higher paid women stayed 1.5 within a specified time frame. When agents years. Small salaries were mentioned most are paid, rigorous supervision can be often as the reason VHWs found the work exercised, programs can be implemented difficult. In a system established in Ethiopia’s rapidly, work routines can be standardized, Gumer District, each household contributed and service quality can be maintained one birr (US$0.15) a year to support the

(Phillips 1999). Negative reinforcers such as community health agents (CHAs) and S E V firing or punishment can be used to encourage traditional birth attendants (TBAs). This I T desired performance. Payment is also seen as contribution was enough to cover a modest N E helping to build some economic equity in a stipend for all trained CHAs and TBAs, and C N minimally literate or economically the attrition rate fell from 85 percent a year to I S I D

disadvantaged population. zero (Wubneh 1999). The main programmatic advantage to From the CHW perspective, appropriate, D N A cash incentives is the apparently lower respectful, and regular compensation is a sign S attrition rate among paid CHWs. In Gongola of acknowledgment and approval that allows E V I

State, Nigeria, the Rural Health Program of them to earn a living or supplement other T the Christian Reformed Church found that income. Cash incentives may come in several N E

VHWs left their posts after one to three years forms. CHWs may be part of the civil service C N I

(Gray and Ciroma 1987). The VHWs worked and be paid a salary. They may also be given Y R

one or two hours a day and received a small a small stipend. CHWs are often given per A T salary (the equivalent of US$13 to $27 a diem and travel allowances to attend training E N O M

15 or make field visits. Cash incentives may also interventions and whose children were fully be tied to drug sales. immunized and participating in growth The source of CHW payments can be the monitoring were eligible to receive low-interest community (contributions from individual loans for income-generating activities. Each households), the government, an NGO, or group of mothers paid an annual fee for a even a for-profit company. The source of health card, and the funds were used to funds may affect the role and allegiance of support the CHW. These funds have been the CHW. Several NGOs have tried to create matched by a one-time grant from the community revolving drug funds or other institution sponsoring the CHW program. types of community-based credit funds Mothers had an economic incentive to learn specifically for health incentives. When health interventions in order to have access to associated with profits that are the “incentive” the loans (Augustin and Pipp 1986). for the CHW, few of these schemes have been successful or achieved any level of Problems with Using Money as scale (Edison 2000; Henderson 2000). When an Incentive compensation is tied to drug sales, CHWs While paying CHWs regularly can solve many tend to focus on curative care, while CHWs problems, experience in many countries has with salaries maintain both preventive and shown that such payment can have unforeseen curative activities (Parlato and Favin 1982). negative consequences, depending on how it is Fee-for-service schemes often result in an handled. Money can be a divisive factor for increase of curative over preventive activities CHWs and can undermine their commitment and the overprescription of medications and the relationships they have with their (Davis 2000). Some NGOs report misuse of communities. Volunteers often cite lack of the funds through “borrowing” from the remuneration as a key factor causing their proceeds of the sales. attrition, but they also cite other critical reasons, There is some indication that such as lack of community support and lack of decentralization increases the flexibility of the supervision (Wubneh 1999). Payment is difficult local government to respond to issues of to disaggregate from other reasons because CHW remuneration. In the Philippines an they are often interconnected. When using cash increasing number of honoraria, or travel incentives, program managers should calculate allowances, have been provided to community how long such a payment scheme can be volunteer health workers (CVHWs) from both funded. When the government or an NGO offers municipal governments and village monetary support, special effort is needed to

S development councils. The honoraria, which compensate for possible distrust or heightened E

V range from US$.50 to US$50 a month, are expectations in the community. Some examples I T possible because of the devolution of health of the negative consequences of paying CHWs N

E services from the provincial level to the are described below. C N

I municipality and village levels. At each level S I local support for the health programs is The Money Is Never Enough D

D funded out of the government’s respective The first problem with money is that workers N

A revenue allocation (Paison 1999). inevitably demand more money, benefits, and

S Some countries have experimented with opportunities for promotion. In the Solomon E V

I insurance plans. In Haiti a combination of a Islands 38 percent of nonworking VHWs left T

N prepaid scheme, existing community groups, because of irregular remuneration. Ninety-two E

C and revenue-generating activity has been percent of the 66 working VHWs surveyed N I

used to motivate CHWs to provide preventive thought their allowances were inadequate and Y

R services. Groups of mothers who could wanted them doubled from US$13.67 to A

T demonstrate their knowledge in child survival US$27 a month (Chevalier 1993). In E N O M

16 Swaziland CHW salaries did not change over receive any remuneration (Quinones 1999). a decade in which the local currency gained Monitoras who work in Honduras along the 400 percent in buying power (Green 1996). If border with El Salvador frequently complain CHWs do not consider their salaries that their Salvadoran counterparts are paid adequate, their performance and retention while they are not (Griffiths 2001). levels may by negatively affected. Are CHWs Accountable to the Sustainability of Payments Community or to the Government? The second problem with monetary CHWs who receive a salary or stipend may compensation is that payment is often see themselves as employees of the irregular and may end altogether when government or NGO rather than as servants project funding runs out. Using other terms of the community. Financial incentives can for payments, such as “field allowance,” destroy the spirit of volunteerism and work “transport allowance,” or “per diem,” can against the volunteer philosophy of a sense of have advantages in some circumstances community (Alonzo and Hurtate 2000). Even a because they create fewer expectations. tiny allowance can reinforce the community’s The sustainability of such incentives, perception that the CHWs are government however, ultimately may depend on their employees and lead to expectations that they source. give even more freely of their time and Payments are often linked to specific personal resources (Taylor 2000; Hilton 2000). training sessions. In Zambia the NGO A community may become less willing to Adventist Development and Relief Agency support the volunteers in other ways. For (ADRA) provided small “meal allowances” to example, communities in Mozambique that CHWs when they brought their monthly thought activistas would receive a monetary reports to the health centers (Edison 2000). To incentive from an NGO or the MOH withheld circumvent salary issues, an NGO in Bolivia their in-kind support (Snetro 2000). When gave very small financial incentives to CHWs people distrust the government, they distrust for discrete tasks that were easy to measure CHWs who are perceived to be a part of the and track (Shanklin 2000). Staff had to spend government system. time evaluating over-reporting and double reporting among CHWs, however. In-kind Payments as an Effective “Compromise” Inequity among Workers Paying CHWs in kind rather than in cash has

Comparison of their salaries with those of other advantages. In-kind payments are less prone S E

workers may lead CHWs to call for salary to comparison with levels of compensation of V I increases or benefits such as pensions and salaried employees because their exact value T N health care. Payment is rarely consistent may be difficult to quantify. CHWs can be paid E C N among cadres of workers such as CHWs and in kind with cooking, food, housing, and help I S community-based distributors of contra- with agricultural work and child care. Most I D

ceptives, who may work side by side and successful in-kind payments are planned and D N

perform similar duties. Such discrepancies can implemented by the community. Beneficiary A

result in jealousy and enmity. If some but not families in Peru have taken turns working for S E V all CHWs or other community workers are paid, free on the farms of the volunteers in I T

tension can result between the paid and unpaid recognition of their important contribution N E

groups. In Colombia CHWs who receive (Buenavente 2000). C N I financial compensation for their work have Another type of in-kind payment is Y

generated tension and envy among other material items provided by NGOs. Such R A

CHWs and community leaders who do not items are often, though not always, related T E N O M

17 to the CHWs’ job functions. Successful in- Preferential Treatment kind payments provided by NGOs have Several programs demonstrate appreciation included bags to carry supplies, agriculture for CHWs’ work through preferential treatment, tools, raincoats, backpacks, supplies for such as access to credit programs, literacy home improvement, educational materials, classes, or first-in-line treatment at health herbal plants, and fruit trees. Alonso and posts. For example, CHWs in Guatemala were Hurtarte (2000) have found, however, that exempt from military service (Parlato and incentives given too often or in too many Favin 1982). In India the CHW must show forms are unsuccessful and demotivating in success with an income-generating activity to the long term. The Shishu Kabar Hearth gain recognition as a health worker. Rather nutritional program in Bangladesh gave each than receiving a salary or wage, the Indian volunteer mother a set of dishes at the end CHW is given access to credit for income- of the sessions, avoiding a cash payment. generating activities through a bank loan This incentive helped the volunteers feel (Arole 2000). Other NGO programs give appreciated and made it easier for trainers CHWs, especially women, priority for to recruit mothers in new communities inclusion in other development programs, (Wollinka 1997). In some instances food such as group-guaranteed lending and supplements have been used as payments, savings programs. In the Ashanti Region of but CHWs have been reluctant to continue Ghana, members of the village health working when the food supplements have committees (VHCs) receive identity cards that ended. ADRA’s experience has shown that allow them to be seen quickly at clinics. When any kind of financial support or subsidy, there is a death in the family of a VHC despite its positive short-term impact, is member, a small cash donation is given to the problematic for long-term sustainability family, and the district health management (McHenry 2000). Selectively giving payment team (DHMT) is represented at the funeral or food to some communities and not to (Leonard 2000). In all such cases, preferential others can generate animosity among treatment of CHWs must be monitored communities and jealousy among families carefully to ensure that community members (Shanklin 2000). do not resent special treatment of CHWs. S E V I T N E C N I S I D

D N A

S E V I T N E C N I

Y R A T E N O M

18 Section 5

Nonmonetary Incentives and Disincentives

ven when monetary or in-kind incentives are provided to CHWs, E they are not sufficient to maintain and retain CHWs’ motivation. Other types of incentives, often intangible, are critical to job satisfaction and fulfillment. These incentives include a good relationship with health staff, personal growth and development opportunities, training, and peer support. Perhaps the most important nonmonetary incentive, a good relationship with the community, is discussed in the following section.

Supervision or Recognition as a result (Elder 1992). In Kitwe, Zambia, CHWs occupy a unique position in the health where the community health promoters system. They are usually not full-time salaried (CHPs) had no contact with the health health workers, yet they are the pivotal bridge system, frequent visits by outsiders (donors between the community and the health and NGOs) helped them maintain their system. Compared with other health workers, commitment and motivation (Steel 2001). they tend to have the lowest status because Sometimes the MOH sends letters of of their low levels of education and poor appreciation to the CHWs and their families, economic status. To ensure that the CHWs although such expressions of appreciation are remain “of the community,” ministries of health not the norm. are usually reluctant to treat them as another Typically, after the initial training a CHW’s cadre of health worker, while the CHWs are relationship with the rest of the health system S

often eager to be identified with the prestige of is limited to what is usually called supervision. E V I

the health system. These competing and Supervisors can give the CHW opportunities T N

contradictory tensions create a host of to discuss problems, exchange information, E C

problems related to a CHW’s sense of and take advantage of continuing education. N I S inclusion in and support from the health Supervisory visits help reduce the feelings of I D system. isolation that often accompany a CHW’s D

The MOH can help CHWs feel supported occupation. To be effective, supervisory visits N A

and appreciated in many ways. In Indonesia a should be regular and based on a common S E

radio-based health communication campaign understanding of the purpose of the visit. V I T

motivated the kaders by publicly praising CHWs appreciate good supervision given with N them as “volunteers who work without the honest intention of capacity building and E C N I

compensation for our children in our village for mentoring. In Guatemala supervised CHWs Y

the sake of the future.” After the campaign had attrition rates two to three times lower R A

mothers and village headmen complimented than those of unsupervised CHWs because T E

the kaders and attended the health posts their link with outside experts gave them N O

more often. Retention improved significantly higher status (Parlato and Favin 1982). M N O N

19 Weak, inadequate, and inconsistent hospitals felt that their work was undervalued supervision is cited frequently as a cause of and that they were treated differently from the low rates of CHW retention (Frankel 1992; other health workers and assistants, even Ofosu-Amaah 1983; Heggenhougen et al. though they performed the same tasks. This 1987; Walt et al. 1989; Curtale et al. 1995; perception was seen as a major demotivater Ojofeitimi 1987; Schaefer 1985). Problems and reason for attrition (Quinones 1999). with supervision range from the logistical When supervision is inconsistent, CHWs difficulties of reaching remote communities to may not feel supported by the health system. interactions that are more punitive than Ensuring that supervisors are trained to supportive. Ofusu-Amaah (1992) summarizes supervise and soliciting the community’s many of the problems of CHW supervision of involvement in supervision can increase by health professionals in the following list: retention of CHWs and help ensure their long- term sustainability in the community. A study ■ heavy clinical and other responsibilities of in Colombia (Robinson and Larsen 1990) health professionals found that the community had more influence ■ inappropriate training of health on the CHWs than the health system, professionals in primary health care contrary to widely held assumptions. If the ■ inaccessibility of villages community and not the health system is the ■ multiple uncoordinated supervision visits primary reference group for CHWs, then by different health personnel working with feedback from the community has a the CHWs significant influence on motivation and ■ lack of vehicles or petrol performance. This study suggests that the ■ lack of per diem supervisor should ask, “How can my contact ■ general shortage of health personnel with the CHW contribute to further development of the relationship with the In Sri Lanka the frequency and duration of community?” These findings indicate that a supervision was inadequate because of major health facility or NGO supervisor should foster shortages of supervisory staff at all levels, more positive interactions and dialogue with and especially at the central MOH level, community members on pertinent issues. where a third of the positions were vacant (Ofosu-Amaah 1983). In Tanzania CHWs Identification and Job Aids received some supervision from the village One of the commonest and easiest ways to S

E leaders and village council but were not strengthen a CHW’s affiliation with the MOH or V I

T familiar with CHW training or job descriptions supporting organization is to provide some form N

E (Heggenhougen et al. 1987). Poor of identification. Identification cards, badges, or

C management can also affect the quality of diplomas can provide security in politically N I S

I supervision at all levels of peripheral health volatile situations and are status symbols in the D services and primary level services. Gilson et community. Many NGOs have given CHWs t- D

N al. (1989) describe this effect as “the shirts, notebooks, caps, ponchos, and bags A reluctance to supervise which comes from with identifying logos that promote group S

E lack of incentives, lack of confidence in solidarity and facilitate entry into households V I

T supervisory techniques, and lack of during a project (Pearcy 2000; Rubardt 2000). N

E objectives and targets for which to work.” Some programs provide bicycles or motorcycles C

N While often beneficial, close contact with for CHWs to use but usually not own. People I

Y health staff can create problems when CHWs who completed the Ghana Red Cross training R

A compare themselves with professional health program were allowed to purchase and wear the T

E workers. For example, CHWs in Colombia Red Cross smock or t-shirt. The Red Cross N

O affiliated with health institutions such as symbol identified them as Red Cross volunteers M N O N

20 and provided recognition and respect from their curative services and nutritional interventions communities and from the MOH (Leonard 2000). rather than from preventive services. In Haiti Job aids are materials that help a CHW volunteer mothers, or animatrices, are perform the required tasks. While providing a motivated by seeing their lethargic children sense of affiliation and enhancing the CHW’s with no appetite become “bright, energetic authority, appropriate job aids also strengthen children who eat ravenously” (Wollinka et al. skills and are invaluable in increasing 1997). This dramatic change convince the confidence. Job aids have included animatrice “that her efforts have had an medicines, health education materials such impact, which appears to strengthen her as counseling cards, first aid kits, pots for commitment to the program, the balanced demonstrating preparation of weaning foods, menu, and the more frequent feeding pattern pens and pencils, flipcharts, notebooks, and the program recommends.” In Ghana volunteer boxes to store records. These frequently cited mothers are motivated by the health of their incentives are important to CHWs’ self- children and their desire to help other mothers esteem and ability to fulfill their role have healthy children (Leonard 2000). (Henderson 2000). Mothers’ support groups meet regularly with the support of health workers to discuss Personal Growth and breastfeeding and help new breastfeeding Development Opportunities mothers solve problems. The volunteers often Personal growth and development is use their own children as examples of healthy, mentioned consistently in the gray literature exclusively breastfed babies. as a major incentive for CHWs. Acquisition of CHWs can derive a sense of knowledge and skills is seen as a stepping accomplishment at a collective level as well as stone to future employment and a necessary from seeing changes in individual children, component in meeting community health which is often difficult. CHWs who collect and needs. Their jobs put CHWs living in rural use health information can monitor and feel areas with little chance of employment on the proud of their own progress. In Bolivia CHWs, path of lifelong learning. Ongoing skill health care providers, and community development (acquisition and promotion of members meet monthly to discuss community- preventive messages, basic curative collected health data and plan action based on services, problem analysis, and problem- the data (Howard-Grabman 2000). This process solving skills) is viewed as important to job has resulted in increased community S

satisfaction. awareness, more concern for maternal and E V I

CHW posts have been an entry into child health issues, and positive attitudinal T N

government employment in some situations, but changes in the community and among health E in many other situations the training and job care providers. CHWs are seen as the bridge C N I S

duties provided are too minimal to prepare for these empowering meetings between health I D CHWs for such employment, where it exists. In care providers and the community. Vaccination, D

Solapar District in India, 93 percent of vitamin A usage, and growth monitoring N A volunteers were not satisfied with their duties programs increased in the Bolivian S because they believed they would be a stepping communities using this “integrated community E V I stone to future government jobs that never epidemiological system,” and women in pilot T N materialized (Kartikeyan and Chaturvedi 1991). communities were 2.2 times more likely to E C

breastfeed within one hour postpartum. N I

Personal Accomplishment Y R

Training A Witnessing positive change is a strong T

Lack of general and skills-based training is E motivator for CHWs. A sense of quick N accomplishment often comes from providing frequently mentioned as a barrier to effective O M N O N

21 CHW performance (Walt et al. 1989; Gilson et with the communities while enhancing their al. 1989; Kaseje et al. 1987; Robinson and standing as they try to meet community Larsen 1990). Most observers of community - needs. based contraceptive distribution programs Without the ability to provide treatment or agree that the quality and intensity of agents’ prevention, a CHW can lose standing in the training is the most important single community. The volunteers in the Sri Lanka determinant of program quality and impact study state, “Often we have to go to the (Phillips 1999). Training can provide CHWs public health nurse midwife to get an answer, with the opportunity to learn skills, receive and then tell the householder. When this education, interact with higher levels of happens, the community loses faith in us and professional staff, and obtain other benefits refuses to accept any advice we give them” that they would not be able to obtain (Walt et al. 1989). Those designing training otherwise. Learning skills is one of the main should consider the way material is taught, reasons CHWs volunteer. the place where training is carried out, and Training is essential if CHWs are to carry relevant skills that strengthen CHWs’ ability to out their work effectively. Training covers not educate community members (Ofusu-Amaah only providing preventive, curative, or other 1983; Gilson et al. 1989; Kaseje et al. 1987; relevant services to the community, but also Robinson and Larsen 1990). Problem-solving teaching and communicating with community skills are a critical part of the training needed residents. In Nepal more training allowed the to promote behavior change rather than community health volunteers (CHVs) to knowledge accumulation. identify causes and treatment of night blindness and to recognize fast breathing as a Training Methods Make a Difference major sign of ARI (Curtale et al. 1995). Their Many training methods are inadequate. The ability to deliver treatment increased their methods tend to be too theoretical, too motivation. classroom based, and too complicated To be effective, training has to be done (Gilson et al. 1989). Such methods can be a regularly and continuously, with the needs of disincentive to CHWs who are learning the community in mind (Gilson et al. 1989; unfamiliar information. The following lessons Kaseje et al. 1987; Robinson and Larsen 1990; learned for participatory training have Walt et al. 1989). VHWs in Gongola State, emerged through NGO work with CHWs Nigeria, said in interviews that they felt that (LeBan 1999): S

E further health care training would allow them V I ■ T to advance to professional health care work CHW functions need to be clearly defined N

E and receive higher pay (Gray and Ciroma before training.

C 1987). ■ Curricula, tools, and methods must cover N I S

I The right combination of skills can help a each specific CHW task, with ample D CHW become a more qualified worker. Having opportunity for hands-on management of D

N skills that the community values raises the real cases. A status of a CHW in the community. In ■ Role modeling and one-on-one tutorial S

E Colombia and Tanzania training strategies training approaches work extremely well. V I

T were based on community surveys completed ■ Adult participatory learning methodologies N

E by CHW candidates before training began and problem-solving approaches help C

N (Robinson and Larsen 1990). The skills the CHWs assume the role of change agent I

Y CHWs learned were directly related to the at the community level. R

A health issues in the communities. Robinson T

E and others further explain how this training The training venue is also important to N

O orientation solidifies the CHWs’ connection the CHW’s ability to learn. Training is more M N O N

22 effective in a setting that matches CHWs’ frustrated as they became more proficient in places of residence, whether urban or rural. identifying disabilities. They realized that they Robinson reports that most training should needed additional education to handle more take place in the community. Time spent in complex rehabilitation problems (Lysack and hands-on activities increases visibility and Frefting 1993). In Kenya continuous training reinforces the relationship with the community provided enough motivation for the village (Robinson and Larsen 1990). Trainers and health helpers to continue working even CHWs should go together to the rural or urban without financial support (Kaseje 1987). In setting to work and assess skills in real Mozambique monthly refresher training situations (Gilson et al. 1989). CHWs should featured a specific health theme, which be trained in the closest health facility by allowed activistas to emphasize that theme health facility staff trainers to better link the during the following months’ health education formal health care system with the community (Koepsell et al. 1999). (LeBan 1999). After training, the awarding of Sometimes training alone is enough to certificates to CHWs or a community keep motivated workers going. La Leche celebration or recognition ceremony can be League mothers in Guatemala continued to invaluable in recognizing the CHWs’ provide counseling and referrals four years accomplishments (Hilton 2000; Henderson after the end of the project grant (Rasmuson 2000; Edison 2000; Pearcy 2000). et al. 1998). Their motivation was attributed to Using other CHWs to assist with the a combination of refresher training, annual training can help ensure that it is relevant to workshops, peer support, and visible change. the local situation. In an adaptation of the Training, however, has a negative side. training-of-trainers approach in Mozambique, CHWs, especially effective ones, tend to be lead activistas (the strongest CHWs) were targeted by a variety of vertical health excellent auxiliary trainers, and their programs (such as tuberculosis, malaria, and assistance reduced the time and costs of onchocerciasis) and taken frequently for training (Koepsell et al. 1999). training in these topics. Usually the CHWs enjoy the additional perks of training (a Refresher Training chance to leave the community, travel Continuous training has been cited as “an allowances, interaction with peers, learning essential prerequisite for an effective CHW new skills), but the communities are left program” (Frankel 1992) and an important without CHWs during the training. In El S

factor in retaining the motivation of workers, in Salvador CHWs were found to spend more E V I

light of the short training periods available and time in training than on the job (Contreras T N

the low levels of education of most CHWs 2001). E

(Ofosu-Amaah 1983). Refresher training Training is clearly a critical and ongoing C N I S

allows the CHWs to learn new skills, take on part of any CHW program. The NGO CARE I D new challenges, and interact with peers, (2000) describes a comprehensive training D

keeping the job interesting and promoting strategy in Nyanza, Kenya, that has the N A personal development. Little information is involvement of both the MOH and the S available on best practices associated with community. CARE combined a training and E V I the periodicity of refresher training or the supervision strategy for CHWs using NGO, T N sequencing of messages and skill MOH, and community trainers and E C development. In most situations refresher supervisors. Community health committees N I

training depends on budgets and is often cut (CHCs) recruited CHWs to serve 20 families Y R when resources are scarce. each. The district health office (DHO) provided A T

Experienced community-based health facilities for training, quarterly in- E N rehabilitation kader in Indonesia became service training, and referral. Practical training O M N O N

23 took place at the hospital. CARE and MOH support, and allowed for local exchange of staff provided additional incremental training information (Marsh et al. 1999). for three months following the initial training. Group meetings can provide motivation The DHO and the CHCs assumed overall for CHWs through peer support. Findings from supervisory responsibility for the performance Colombia, Mozambique, Nepal, and Uganda of the CHWs. Within two years, 319 CHWs show that peer support is as important to were managing sick children at the CHW performance as supervisory feedback community level. The clinical proficiency of (Snetro 2000; Robinson and Larsen 1990; the CHWs was equal to or better than that of Taylor 2000; Oriokot 2000). Successful the MOH clinicians in that setting (LeBan programs have brought CHWs together in 1999; Steinwand 2001). The motivation of the monthly meetings and used these meetings to CHWs, which had been high early in the promote CHW bonding, as well as provide in- project, began to subside after a few years. service training and supervision (Robinson The community elders, already engaged in and Larsen 1990). In the Shishu Kabar Hearth supervision, promised to take a more nutritional program in Bangladesh, trainers aggressive role in solving the problems of had the freedom to be creative and make CHW incentives in their areas. suggestions to improve the program. Their ideas were incorporated into the program to Peer Support and CHW give them a sense of ownership and Networks involvement in decision making. Their Interaction with other CHWs can be a critical contributions were recognized at weekly motivator for people who often work with little meetings. The trainers were encouraged to supervision or tangible evidence of their discuss successes and solve problems effectiveness. Peer support comes in several among themselves to exchange information forms. Several NGO programs have and create a supportive environment successfully paired CHWs so that they can (Wollinka 1997). work together and support each other. In the Examples of supporting groups of CHWs Atención Integral a la Niñez (AIN) program in in forming CHW associations exist in many Honduras, for example, monitoras worked in countries. In Peru CARE has effectively groups of three. Working in teams allows mobilized community volunteers into local CHWs to divide their work and reduces the committees that cover specific geographic sense of isolation and complete areas. Representatives of these committees S

E responsibility for a geographic area. In Liben organize themselves into district associations. V I

T District, Ethiopia, Save the Children The committees meet monthly to discuss N

E mobilized communities to form “bridge to experiences and mutually reinforce

C health” teams (BHTs). Each BHT included a commitment. They raise funds to cover their N I S

I wisewoman, a wiseman, and a young own activities, organize training events, and D traditional apprentice. Two-thirds of the BHT advocate for health with government and the D

N members were influential, respected TBAs, MOH. This arrangement has resulted in A bone setters, herbalists, or circumcisers. dedicated, well trained, and active CHWs who S

E Most traditional healers in the district were have strong ties to the MOH but are not V I

T elected as BHTs. The teams decreased the dependent on it (McNulty 2000). N

E isolation of the BHTs, provided mutual C N I

Y R A T E N O M N O N

24 Section 6

Relationship with the Community

orking with the community gives health workers a platform from Wwhich to strengthen their relationship with the community and receive community feedback, as well as a structure for regular interaction with health facility staff. Community participation is an integral part of CHWs’ motivation. Without involvement, communities lack interest and expectations, leaving CHWs without a support system.

Enhancing Relationships complexity. Communities are not homogenous between CHWs and groups of people who always work well Communities together. Like all communities, those in In many programs the potential of the CHW developing countries are made up of various has not been realized because of a poor groups of people based on such criteria as relationship with the community. In the religion, ethnicity, and economic status. With Solomon Islands, 32 percent of the notable exceptions, the most marginalized and nonworking village health workers surveyed powerless groups—women and the very poor— left their posts because of a lack of need CHW services the most, yet rarely have community support (Chevalier 1993). real involvement in CHW programs. Programs Programs and organizations that do not that do not recognize the complexity of local engage communities actively in CHW communities or ensure that the marginalized programs from their inception generally are given a voice may find that their CHWs are experience low morale among their CHWs. pawns of the local elites. One way to guard This lack of shared ownership generally against this is to work through existing triggers a separation and distance from active community groups and to increase the total community participation in the CHW program, number of CHWs in a community. resulting in high attrition. All poor communities are not alike. If If CHWs are to serve as the bridge programs make any distinction among between the health system and the communities, it is the distinction between community, the relationship with the rural and urban communities. Planning tends Y T

community must receive great attention. Yet to be inflexible in responding to the diversity I this relationship is often given plenty of among communities. Of course, communities N U rhetoric but few if any financial or technical differ considerably. Some have more M M

resources. Two misconceptions of the resources than others. Some have access to O C community have been common: a naive health facilities, markets, and cash crops, E concept of community and an assumption while others are on the verge of famine, with H T

that all communities are alike. Much current little food or other resources. The more H T understanding of the complexity of stressed communities cannot and should not I W

communities comes from the work of social be expected to provide labor, money, or other P I scientists and the field experiences of NGOs. resources to support CHWs. Communities H S

Programs have tended to oversimplify the with access to better job markets may have a N O I

“community” and underestimate its social hard time recruiting volunteers, as Catholic T A L E R

25 Relief Services found in El Salvador (see p. at all times of the day, even when they were 32). Communities may also have different busy” (Ruebush 1994). Candidates should also epidemiological profiles, as in Ethiopia, where be selected on the basis of their demon- malaria is a problem only in some altitudes. strated involvement in and commitment to the Clearly, the relationship between a CHW and community (Robinson and Larsen 1990). community varies with the characteristics of Community members are often much more the community. Little has been documented aware of the characteristics that will ensure on approaches to differentiating communities. CHW retention.

Selection of CHWs Process of Selection Ideally, a community should be involved in all Once criteria for selection are established, the aspects of a CHW program, including “community” is usually asked to nominate selection, training, and supervision, but candidates. Because this process is often a community members may not have the time black box to outsiders who know little if and resources to invest in all these areas. anything about the internal social dynamics of Community involvement in selecting CHWs, the community, many problems can occur at as well as in using their services and this stage. Communities may not be contributing in-kind payments, appears to be organized to choose CHWs representative of critical to CHW programs. the majority of residents, or they may not fully understand the functions of the CHWs. Criteria for Selection Communities that do not understand the role Selecting a CHW involves many steps. of the CHWs are less likely to give the CHWs Criteria must be established, candidates the necessary support and may not identified, and a final selection made, perhaps understand their own role in improving their after a trial period. Although the community health. In Saradidi, Kenya, the responsibilities could have a role in each step, usually it is of the village health helper were discussed in involved only in proposing candidates who open community meetings and formal and meet criteria established by program staff. As informal exchanges (Kaseje 1987). In many discussed in Section 4, the personal cases, however, selection of CHWs is characteristics of CHWs play an important completed well before the community has a role in their relationships with the community clear understanding of what they do. CHWs’ and their continued motivation. ability to carry out their tasks effectively can When given the opportunity, communities be enhanced when communities are invested are often able to develop criteria that ensure in trying to improve their own health. CHWs stay on the job. In trying to revive the Communities understanding of their own Community Health Agent (CHA) program, the role in changing their health status can help Y

T MOH in Ethiopia established literacy as the sustain the CHWs’ activities. Community I

N only criterion for recruitment. When asked members should be informed of the job U

M what criteria they would use to select CHAs, description, capabilities, and commitment of M

O however, local communities listed 16 CHWs (Frankel 1992; Ofosu-Amaah 1983; C characteristics, including selection by the Heggenhougen et al. 1987; Walt et al. 1989). If E

H community, married status (so the CHWs the communities understand what the CHWs T would not leave the community), and no are trained to do, there is less chance that H T I addiction to chat, an herbal stimulant residents’ expectations of a CHW will go W (Bhattacharyya et al. 1997). In Guatemala unmet. Community understanding will also P I

H local residents thought the volunteer reduce inappropriate demands and frustrations S

N collaborators should be “responsible (Heggenhougen 1987). O I individuals” and “able to take care of patients Ideally, a CHW should be chosen with the T A L E R

26 input of the community so that residents’ the MOH is able to respond with additional health needs are considered and they respect training and support. This is the ultimate goal and feel comfortable interacting with the CHW of many CHW programs. In Mozambique, for their health services. In some cases, where activistas have worked for many years, however, CHWs have been selected by village their roles seem to have matured, and leaders who choose relatives or friends or by communities seem to accept their work to a village committees that disregard community high degree. Community members now input. A survey-style CHW evaluation by approach activistas for family planning UNICEF in 1989 reported that 45 percent of services and other types of support (Snetro the CHWs surveyed were related to the local 2000). When sevikas in Nepal were asked why chief or subchief. The percentage would have they continued their volunteer work, they said, been higher if other kinship ties and filial “Our neighbors won’t let us resign; they insist connections with members of the chiefs’ we continue because their children’s health councils had been included (Green 1996). A depends on us” (Taylor 2000). brief evaluation in Swaziland found that Several programs have mentioned the nepotism and self-interest determined the support of the community as an incentive for choice of the CHW by the local chief and his CHWs. Trust, prestige, mobility, and social council, with no consideration of the interaction are other factors that are favorably candidates’ interest in or qualifications for the mentioned (Walt et al.; Kaseje 1987; Lysack job (Green 1996). and Krefting 1993; Ruebush 1994). Many Extensive field experience and long CHWs volunteer because they enjoy serving association with specific communities have the community. helped many NGOs find ways to ensure that the CHW selection process is fair and Visible Benefits representative of marginalized groups. For Communities that have directly and visibly example, CHVs in Senegal were chosen during benefited from CHW programs are the most community meetings by community leaders willing to support the continued presence of who did not include women or representatives CHWs. The community’s interest in sustaining of all community groups. Although the chosen a CHW program is based in large part on CHVs were motivated in the beginning, 60 evidence of positive changes in health status percent had abandoned their jobs after two because of the CHW or on benefits such as years, and many of those who remained were effective referrals to health facilities. Visible no longer motivated to carry out community change is limited by the predominance of health activities. During the following phase of preventive health in a CHW’s work. With the the project, volunteers were chosen instead exception of nutritional rehabilitation or the through in-depth discussion with community use of ORS for dehydrated children, few Y

members and village health committees dramatic changes in health are visible to T I established by World Vision. Five years later community members. N U

few CHVs had left their posts, an achievement One way CHWs and communities can M M

attributed to official community recognition of create visible change is to monitor simple O C their roles and to moral encouragement from health indicators over time. In Kitwe, E their communities (Aubel et al. 1999). Zambia, communities monitor the number of H T children who gain weight in the past month H T Community Recognition of as an overall indicator of child health (Steel I W CHW Work 2001). In Eastern Province, Zambia, P I

Community recognition and appreciation of communities use risk maps to monitor H S

the work of CHWs can have a snowball effect indicators such as immunization status and N O as communities demand more services and availability of latrines. Risk maps identify I T A L E R

27 households using color codes to mark Status indicators, such as green for fully Being identified as a CHW and affiliated with immunized children and red for those not the health system is usually, though not fully immunized (Bhattacharyya and Murray always, a status symbol that generates power 1997). The collection and analysis of health and respect within the community. CHWs in information to chart changes in behavior (for Colombia ranked “having influence in the example, more use of health services) or community” as the most important extrinsic health status (for example, fewer cases of reward affecting their performance. Informal dehydration) allows CHWs to show the observations indicated that this influence communities the results of their work. made the CHWs opinion leaders on a variety Encouraging communication and of issues of concern to the community. The interactions between CHWs and community fact that influence in the community is highly members is critical to building an valued by community-based workers adds a understanding of the CHWs’ role and support dimension to their role that few other health for their work. There are a number of workers, particularly those based in examples of community supervision of CHWs. institutions, can share (Robinson and Larsen While community health committees cannot 1990). Identification badges, uniforms, and be expected to do clinical supervision, they relationships with “outsiders,” as in the Kitwe can monitor CHW performance at the example, can increase the status of a CHW in community level. For example, the barefoot a community. doctors in China were accountable to the Praise and respect from community villages and were given technical supervision residents and peers can motivate CHWs by the health centers (Frankel 1992). positively and increase their length of service. The appreciation of the people they serve is a Individual Interactions strong incentive that is often cited as impor- Negative CHW behavior has a negative effect tant to CHWs’ job satisfaction. Minnesota on community support of CHWs and the International Health Volunteers has trained messages they promote. Health workers in about 2,000 community volunteers in Uganda Niger did not treat mothers respectfully or for a variety of tasks. Community recognition patiently and did not counsel them on has proved to be a valuable tool in motivating possible side effects of the vaccinations and retaining community volunteers by given to their children. The mothers reported increasing their status in the community. that this negative behavior was a major barrier About 70 percent of them have been elected to their using the CHWs’ health services to various positions on their local councils (Boyd and Shaw 1995). since becoming volunteers (Mullins 2000). To be effective change agents, CHWs Especially for women, public recognition Y

T need to demonstrate the positive effects of outside the family can generate self-respect I

N new practices in their own homes. In and empowerment to act in the community. U

M Mozambique CHWs who had pit latrines in Poor women in Dhaka, Bangladesh, who M

O their homes stimulated community interest in served as CHWs for ten years were seen as C their messages promoting the use of such valuable members of their communities E

H latrines, while those who did not discouraged (Silimperi 2000). The activistas in Mozambique T such interest (Snetro 2000). In Honduras, repeatedly emphasized the importance of H T I Save the Children and the MOH posted the community value and support, demonstrated W photos and names of CHWs on the health by their neighbors’ increased respect, P I

H post wall, bringing public recognition to the reciprocal gestures of help, and acceptance of S

N CHWs and increasing their visibility and health behavior change messages; community O I retention (Amendola 1999). leaders’ understanding of and support for their T A L E R

28 role; frequent visits by NGO staff; and emphasis behaviors and community opportunities to learn (Snetro 2000). mobilization. Health action committees (HACs) of ten to 12 elected members were Community Organizations that trained to support the four to six BHTs in Support CHW Work each kebele (village) and to support TBAs in In addition to community recognition, the safe delivery and danger sign recognition. formation of community organizations or village Mothers and primary health caregivers development committees has been cited as believed that BHT health messages came useful in supporting and sustaining the role of from traditionally respected sources. BHT CHWs. Some form of viable community members were motivated by training, organization is necessary to establish an effectiveness in the community, and support operational relationship between the from local leadership in the HACs (Marsh et community and the government. In Gongola al. 1999). State, Nigeria, the support and encouragement When support from community groups is of the village health committee emerged as an missing, CHWs face an uphill battle in gaining important factor in VHW job satisfaction. The the respect of the community. CHWs from average length of service for 13 former VHWs Cochabamba, Bolivia, felt that the community who had met monthly with their local was unsupportive and unaware of their committees was three years, while that of 14 activities. They saw themselves as divorced others who had never met with their local from important decision-making organizations. committees or met with them rarely was 1.3 They also felt that institutional support from years (Gray and Ciroma 1987). highly visible community leaders would In another example, Save the Children increase their motivation and their credibility and the MOH trained BHTs in Ethiopia in with the villagers (Gonzalez 1987). Y T I N U M M O C

E H T

H T I W

P I H S N O I T A L E R

29

Section 7

Putting It All Together: Multiple Incentives

uccessful CHW programs depend on a framework of incentives S at the individual, community, and health system levels that together can motivate people to become CHWs and continue in this capacity for a few to several years, as well as motivate communities or ministry health offices to maintain and support CHWs and replace them over several years. Successful projects generally use multiple incentives simultaneously to motivate CHWs. Understanding the functions of various incentives can help programs combine these incentives effectively.

Behavioral Model to support CHWs at different levels of the Several models in the behavioral science system (Table 5). literature apply to motivation in a workplace setting. A model by Pareek (1986) identifies Examples of Multiple Incentives six primary needs or motivators relevant to As the rest of this paper shows, appropriate understanding the behavior of people in incentives depend on the social status of organizations. This model identifies key CHWs, their duties, and other opportunities in motives that contribute to employee the community. This section reviews several satisfaction and fulfillment (Table 4). It examples of programs that have used multiple

suggests that incentives that positively affect incentives. More detail about each program S E and reinforce each of these motives would can be found in Annex 1. V I contribute to higher motivation and retention T N of CHWs, and that incentives that exacerbate Catholic Relief Services, El Salvador E C N “fears” would likely lead to higher attrition. Catholic Relief Services (CRS) has used the I

E

The third column in Table 4 categorizes Pareek behavioral model to plan and L P the incentives that have been reviewed in this implement multiple incentives for CHWs in El I T paper according to the motivation model. This Salvador. This is the only example found of an L U M type of categorization can help program organization using an explicit behavioral : planners choose several incentives that model to plan CHW incentives and tracking R E reinforce positive aspects of all six motives attrition rates to evaluate the effects of the H T simultaneously. incentives. The El Salvador experience yields E G

The same list of CHW incentives, several lessons. First, the use of multiple O T

including the key disincentives mentioned in incentives based on the model was critical not L L the literature and in personal communications, only in reducing attrition but also in involving A

T I can be organized into a systems approach the community (Rosales et al. 2000). Second, G

that shows an implementer what can be done attrition rates fell by 18 percent over all three N I T T U P

31 Table 4. Motivation Model

Motive Definition CHW incentives

Achievement Concern for excellence; setting of ■ Possibility of future employment challenging goals ■ Personal growth and development ■ Acquisition of skills

Affiliation Concern for establishing and ■ Peer support maintaining close, personal ■ CHW associations and relationships networks ■ Community involvement ■ Identification (badges, shirts, etc.)

Extension Concern for others; urge to be ■ Community recognition of and relevant and useful to larger respect for CHW work groups ■ Successful referrals

Influence Concern with making an impact ■ Status in the community on others; desire to change ■ Accomplishment matters and develop others ■ Visible changes

Control Concern for orderliness; desire to ■ Clear role be and stay informed; urge to ■ Job aids monitor and take corrective action ■ Feedback to the MOH and when needed community ■ Support from the health system ■ Policies or legislation that support CHWs

Dependency Desire for the help of others in ■ Satisfactory remuneration one’s own self-development; urge (monetary and nonmonetary) to maintain an “approval” ■ Training and refresher training relationship ■ Supervision ■ Preferential treatment

S Source: Pareek 1996 E V I T

N communities studied and by 54 percent in two the BASICS Project since 1995. The AIN E

C of the communities. Third, as discussed program has a very strong group of monitoras N I above, not all communities are alike. The who weigh children under two years old each E L establishment of garment factories in one of month and counsel mothers whose children P I

T the communities led to very high attrition of have not gained weight adequately. Several L

U CHWs who took jobs in the factories. The factors appear to be critical to the monitoras’ M

: main lesson of the CRS experience may be success. First, they work in groups of three R

E that creating a cadre of volunteer workers and are free to divide their tasks any way they H

T may not be the best approach to establishing like. Second, the monitoras, health center E

G CHWs in areas with growing opportunities for nurses, and program staff all focus on the O

T paid employment. same indicator: adequate child growth in the

L

L previous month. Because of this unified goal, A Atención Integral a la Niñez, Honduras every actor in the program knows which T I The MOH of Honduras has implemented the children are not growing adequately and why, G N

I AIN program with technical assistance from as well as what actions have been taken to T T U P

32 Table 5. CHW Incentives and Disincentives Organized by a Systems Approach

Motivating factors Incentives Disincentives

Monetary factors that ■ Satisfactory remuneration; ■ Inconsistent remuneration motivate CHWs material incentives; financial ■ Change in tangible incentives incentives ■ Inequitable distribution of ■ Possibility of future paid incentives among different employment community workers

Nonmonetary factors ■ Community recognition and ■ CHWs from outside community that motivate CHWs respect ■ Inadequate refresher training ■ Acquisition of valued skills ■ Inadequate supervision ■ Personal growth and ■ Excessive demands or time development constraints ■ Accomplishment ■ Lack of respect from health ■ Peer support facility staff ■ CHW associations ■ Identification (badge, shirt) and job aids ■ Community status ■ Preferential treatment ■ Flexible and minimal hours ■ Clear role

Community factors that ■ Community involvement in ■ Inappropriate selection of motivate CHWs CHW selection CHWs ■ Community organizations that ■ Lack of community involvement support CHW work in CHW selection, training, and ■ Community involvement in support CHW training ■ Community information systems

Factors that motivate ■ Visible change ■ Unclear role and expectations communities to support ■ Contribution to community (preventive versus curative and sustain CHWs empowerment care) ■ CHW associations ■ Inappropriate CHW behavior

■ Successful referrals to health ■ Failure to take community S E V

facilities needs into account I T

Factors that motivate ■ Policies or legislation that ■ Inadequate staff and supplies N E

MOH staff to support support CHWs C N ■ I and sustain CHWs Visible change

■ E

Government or community L P funding for supervisory I T

activities L U M

: R improve the children’s growth. Third, the Isika project, which implements community- E H program has used a variety of small based IMCI and community-based nutrition T E incentives to encourage and support the interventions using an integrated G O monitoras. communication strategy. The Madagascar T

L

approach to using community volunteers, or L A Jereo Salama Isika, Madagascar animateurs, is unique in several respects. The T I

In Madagascar BASICS has provided program expects that 50 percent of the G N technical assistance to the Jereo Salama animateurs will drop out after 12 to 18 months. I T T U P

33 After the animateurs leave the program, project trained over 2000 basic health workers however, they are still viewed by the (BHWs) to provide preventive and curative community as important sources of health care. While the project faces the unique information. For this reason, the program challenges of war, cross-border traffic, and celebrates its graduates rather than worrying large numbers of refugees, it shares with about its dropouts. The program’s goal is to other CHW projects the issues of selecting train as many people as possible (at least 1 BHWs and providing them with incentives. percent of the population) in a two-day training The BHWs were selected carefully to workshop. The technical focus of the program ensure that they were committed to improving is on promoting “small, doable actions” that health services in Afghanistan and to serving are illustrated on counseling cards (for the resistance movement. Their work was example, taking a child to be immunized). All monitored by checking administrative records, NGOs and donors in the country use the reviewing reports of border crossings, and same illustrations in a wide range of making annual (in some cases triannual) materials, resulting in consistency and visits. After their training the BHWs received continuity to health communication. diplomas, which were considered prestigious. The Jereo Salama Isika program also The project decided to pay the BHWs to give highlights the importance of considering the them an incentive to stay in Afghanistan interaction of the role of the volunteers (in rather then move to Pakistan, as many people Madagascar they are health promoters only, were doing. AHSSP demonstrated tight with no curative functions) and their accountability by cross-matching three data incentives. Finally, the word supervision is not sources, an approach that the BHWs used to describe the relationship between appreciated because they felt the project was health staff and volunteers. Instead, able to identify “cheaters.” The BHWs also communities and health staff celebrate the valued their contact with foreign agencies. The achievements each year in a health festival. attrition rates were fairly low, with an average of 5 percent, and the average time of Afghanistan Health Sector Support participation was just over two years. These Project (AHSSP) statistics did not change much even when The Afghanistan Health Sector Support salaries were cut by 50 percent. The BHWs Project (HSSP) began in 1986 to provide who served in areas that bordered Pakistan

S basic health care to the rural Afghan had much higher dropout rates, probably E V

I population scattered among small villages because of family connections and better T

N separated by natural barriers or war. The opportunities in Pakistan. E C N I

E L P I T L U M

: R E H T E G O T

L L A

T I

G N I T T U P

34 Section 8

Conclusions and Recommendations

he experiences with CHW programs reviewed in the previous T section show that no one CHW program will work for all communities in all countries. Nevertheless, because program planners find it easier to develop one program and apply it globally, they are tempted to move in that direction. If a global CHW program is impracticable, how should programs be tailored and adapted to specific situations? What issues should be considered? How should decisions about incentives be made?

Perhaps the most important conclusion of this demands. Policies on CHW distribution of review is already known: there is no tidy antimicrobials and antimalarials can have package of three incentives that will ensure tremendous effects on their relationship with motivated CHWs who continue to work for the community. years. Instead, a complex set of factors affect At a micro level, the position of CHWs in CHW motivation and attrition, and the way their communities influences their motivation these factors play out varies considerably and retention. The inherent characteristics of from place to place. Program planners do not CHWs, such as their age, gender, ethnicity, need to start from scratch, however: they can and even economic status, affect the way draw on the public health community’s they are perceived by community members extensive experience with CHW programs. and their ability to work effectively. In summary, CHWs do not exist in a The specific tasks and duties of CHWs vacuum. They are part of and are influenced affect their motivation and retention. Given too by the larger cultural and political environment many tasks, CHWs may feel overwhelmed in which they work. The process of health with information or spend so much time in sector reform, the IMCI strategy, and the training that they rarely practice what they achievements of community-based nutrition have learned. Often the catchment areas are S programs have generated renewed interest in too large, making it difficult for CHWs to find N O I

the potential contribution of CHWs. Health the time or transportation to visit all the T A

sector reform has changed the supervisory households. Many CHWs are restricted to D N

structure within health systems and given preventive and promotive roles that leave E more autonomy to peripheral health facilities. them unable to respond to community M M

Reform has also decentralized the control of demands for curative care (and usually O C health funds, allowing greater flexibility in medicines). E R spending for various types of CHW incentives. Monetary incentives can increase D

The IMCI strategy includes a training retention. CHWs are poor people trying to N A

curriculum for the assessment and treatment support their families. But monetary S N

of mild and moderate child illnesses. This incentives often bring a host of problems: the O I strategy allows CHWs to play a curative role, money may not be enough, may not be paid S U which is usually what the community regularly, or may stop altogether. Monetary L C N O C

35 incentives may also cause problems among incentives over time to keep CHWs different cadres of development workers, motivated. A systematic effort that plans for some of whom are not paid. Nevertheless, multiple incentives over time can build a some programs have paid CHWs success- CHW’s continuing sense of satisfaction and fully, and many have used in-kind incentives fulfillment. Identifying the functions of each of effectively. the incentives would be useful to clarify the Nonmonetary incentives are critical to the critical functions and how those might vary success of any CHW program. CHWs need to based on the CHW role and type of feel through supportive supervision and community. appropriate training that they are part of the health system. Relatively small tokens, such as Support the CHW’s Relationship identification badges, can give CHWs a sense with the Community of pride in their work and increased status in Surprisingly, the fact that the effectiveness of their communities. Appropriate job aids, such as the work of CHWs depends almost entirely on counseling cards and regular replenishment of their relationship with the community is often supplies, can ensure that CHWs feel competent overlooked. Many programs end up focusing to do their jobs. Peer support can come in many on clinical training, supervisory checklists, forms, such as working regularly with one or two and logistics (all of which are extremely other CHWs, receiving frequent refresher important) to the exclusion of activities that training, or joining CHW associations. support the community relationship. In the end a CHW’s effectiveness depends Effective programs have (explicitly or on his or her relationship with the community. implicitly) oriented the whole program to Programs must do everything possible to support and strengthen every interaction of strengthen and support this relationship. First, CHWs with community members. The paper program planners must recognize the social includes many examples of such orientation, complexity and diversity of communities. from public recognition of CHW work by Different communities need different types of supervisors to job aids that support the incentives, depending on other local job ongoing dialogue between community opportunities, prior experience with CHWs, the members and CHWs. Programs must ask economic situation, and other factors. continually what they can do to promote Unfortunately, little experience or beneficial CHW-community interactions. guidance on differentiating communities is available. Programs should involve Use Multiple Incentives communities in all aspects of the CHW In most of the programs reviewed in this paper, program, but especially in establishing criteria incentives were implemented ad hoc rather S for CHWs and making the final selection. than as part of a systematic program. While N O I Programs can provide opportunities for quick, multiple incentives are used in successful T

A visible results that promote community programs, new incentives are often proposed in D

N recognition of CHWs’ work. CHWs must be reaction to a crisis of low morale rather than as E

M trained in appropriate and respectful part of an overall program effort to maintain M

O interactions with all community members and high morale. Programs should consider a C in appropriate responses to difficult people or systematic effort to plan for multiple incentives E R

situations. Community-based organizations over time to build CHWs’ continuing sense of D

N such as religious groups or youth clubs can satisfaction and fulfillment. A provide support to CHWs and lessen their Programs might find it useful to identify S

N load significantly by taking on health the functions of each of the incentives using O I

S education activities. Pareek’s model or some other model to U

L Many successful programs use multiple understand the critical functions and how C N O C

36 those might vary based on the CHW role and community-based workers’ ability to provide type of community. Intrinsic incentives work preventive services and some curative to promote a sense that the work is treatment of childhood illnesses. For this worthwhile, while extrinsic incentives include element CHWs will use IMCI concepts and salary, increased status in the community, tools to classify and treat illnesses and also and the support of colleagues. Both intrinsic to provide health education. In this and extrinsic incentives clearly should be arrangement, which is similar to that in the implemented and monitored. An incentives examples from El Salvador and Honduras plan could address the multiple motives of described on pages 31-33, the main problem achievement, affiliation, extension, influence, with high attrition is the training costs. control, and dependency, as presented by Programs therefore should plan incentives Pareek. Alternatively, the plan could combine and develop realistic expectations of the incentives targeted at different parts of the length of service. systems—monetary or nonmonetary factors The third element of the Community IMCI that affect the individual CHW, community framework involves promoting the key family factors that encourage and support CHWs, practices critical for health and nutrition. When and health system factors that support CHWs work as health promoters, as in the CHWs. Ideally, an incentive plan would Madagascar example on pages 33-34, include a combination of both approaches. programs should consider maximizing the number of training graduates so that healthy Match Incentives with Duties behaviors are spread widely. CHWs continue to play an important role in many international primary health care Employees or Volunteers? programs. While continuing their preventive In general CHWs are not paid salaries and community mobilization tasks, CHWs are because the MOH or donors do not consider increasingly becoming involved in community- salaries to be sustainable. Yet CHWs are often based case management of prevalent held accountable and supervised as if they childhood diseases. While CHWs’ success were employees. CHW programs must rate is often lauded in the early stages of a recognize that CHWs are volunteers, even if new and exciting project, their motivation they receive small monetary or nonmonetary diminishes over time unless frequent steps incentives. They are volunteering their time to are taken to maintain their enthusiasm for serve the community. Too often CHWs are their essential but voluntary role. treated as inferior employees instead of The Community IMCI framework lays out helpful volunteers. three elements of implementation and Perhaps high attrition is not a problem, describes the types of communities in which but an opportunity to involve more community S N O those elements are appropriate. The role of members in promoting good health and I T

CHWs, and consequently their incentives, will nutrition. The example from Madagascar A D vary among the elements. The first element shows ways to change the way people think N E emphasizes building strong partnerships of CHWs: to celebrate graduates rather than M M

between health facilities and communities, worry about dropouts, plan for high turnover, O which depend to a great extent on the CHWs’ and hold shorter, more frequent training C E R acting as bridges. In such situations, where sessions. D personal relationships are critical, incentive N A packages should be developed to avoid Importance of Monitoring S frequent turnover. Many programs do not understand why their N O I

The second element of the Community CHWs drop out. Such programs would be well- S U

IMCI framework focuses on improving served by monitoring some of the most L C N O C

37 important factors that affect CHWs’ motiva- Topics for Research tion and desire to stay on the job. Such Unfortunately, despite the vast experience monitoring need not be complex or with CHWs, relatively little scientific evidence quantitative. Much could be done qualitatively, is available to answer some of the basic through interviews with and observations of questions: What are current attrition rates? CHWs and community members during What attrition rates are realistic? What is the routine field visits. most efficient way to monitor CHW programs? Perhaps the most important issue that How can CHW attrition be reduced should be monitored by CHW programs is successfully and retention increased? What whether the programs are able to stay abreast financing strategies can be used to ensure of the “competition” for CHWs; that is, other that CHWs are paid regularly and sustainably? jobs and opportunities. Another issue for What critical functions are achieved by monitoring is how CHWs manage the increased different incentives? What important workload when new tasks or functions are differences among communities affect CHW added. Yet another is the interaction of programs, and what is the best way to assess community members and CHWs. What these differences? How can planners demands do community members make on the efficiently tailor their programs to meet local CHWs? How are CHWs treated by their needs? supervisors and other staff in the health These issues, touched on in this paper, system? Do the CHWs have the training and require further investigation. Much research is job aids they need to be effective and feel still needed to determine the best ways to competent in their jobs? What monetary or sustain long-term CHW programs and retain nonmonetary incentives would increase their volunteer health workers. motivation and support their work? S N O I T A D N E M M O C E R

D N A

S N O I S U L C N O C

38 Section 9

References

Alonzo, L, and S Hurtarte. Catholic Relief Center for Policy Alternatives. Community Health Service. Personal communication, March 21, Workers: A Leadership Brief on Preventive 2000. Health Programs. Washington, DC: Amendola, L. Save the Children. Personal Georgetown University Law Center, Harrison communication, June 1999. Institute for Public Law, Civic Health Institute, Arole, RS. JAMKHED Comprehensive Rural Codman Square Health Center, 1998. Health Care Project. Personal communication, Chevalier, C, et al. Why Do Village Health June 1999. Workers Drop Out? World Health Forum Aubel, J, et al. Strengthening Partnerships 14(3):258–61, 1993. between Communities and Health Workers for Curtale, F, et al. Improving Skills and Utilization of Child Health. In B Burkhalter and V Graham, Community Health Volunteers in Nepal. Social eds. Presented Papers: High Impact PVO Child Science & Medicine 40(8):1117–25, 1995. Survival Programs, Volume 2. Proceedings of Dawson, P. Partnerships to Increase Access, an Expert Consultation, Gaulladet University, Quality, and Scale of Childhood Pneumonia Washington, DC, June 21-24, 1998. Published Treatment through Female Community Health for the PVO community and the U.S. Agency for Volunteers. In R Steinwand. Reaching International Development by the Child Communities for Child Health: Advancing PVO/ Survival Collaboration and Resources (CORE) NGO Technical Capacity and Leadership for Group, Washington, DC, and Basic Support for Household and Community Integrated Institutionalizing Child Survival Project Management of Childhood Illness (HH/C IMCI). (BASICS II), Arlington, VA, 1999: 73-87. Proceedings of a workshop, January 17-19, Augustin, DA, and MM Pipp. Alternative Methods 2001, Baltimore, Maryland. Washington, DC: of Motivating Community Health Workers Child Survival Collaboration and Resources (abstract). Chevy Chase, Maryland: PRICOR/ (CORE) Group and Arlington, Virginia: Basic University Research Corporation, 1986. Support for Institutionalizing Child Survival Bhattacharyya K, P Freund, W Made, and D Project (BASICS II), 2001. Teshome. Community Demand Study for the Davis, T. Food for the Hungry International. Essential Services for Health in Ethiopia Personal communication, April 14, 2000. Project. BASICS Technical Report. Arlington, Edison, J. Adventist Development and Relief Virginia: BASICS Project, 1997. Agency. Personal communication, March 2000. Bhattacharyya, K, and J Murray. Community Elder, J, et al. HealthCom Indonesia: The Use of Assessment and Planning: A Participatory Radio Spots to Improve Performance and Approach. BASICS Technical Report. Arlington, Motivation of the Kader. International Quarterly Virginia: BASICS Project, 1997. of Community Health Education XI/4 Bodart C, G Servais, YL Mohamed, and B (December), 1992. Schmidt-Ehry. The Influence of Health Sector Favin, M. CHANGE Project. Personal Reform and External Assistance in Burkina communication, June 2001. Faso. Health Policy and Planning 16(1):74–86, Frankel, S, and MA Doggett. The Community 2001. Health Worker: Effective Programmes for Boyd, B, and WD Shaw. Unlocking Health Worker Developing Countries. New York: Oxford Potential: Some Creative Strategies from the University Press, 1992. Field. Washington, DC: Academy for Gilson, L, et al. National Community Health Educational Development/HealthCom, 1995. Worker Programs: How Can They Be Buenavente, L. Project HOPE. Personal Strengthened? Journal of Public Health Policy

communication, March 3, 2000. 10(4):518–32, 1989. S E

CARE Kenya. Community Initiatives for Child C N

Survival: Siaya (CICSS-II) Detailed E R

Implementation Plan. March 2000. E F E R

39 Gonzalez, DR. Community Financing to Reduce Koepsell J, et al. The Strength Project: Attrition of Community Health Workers and Strengthening District and Community Teams Increase Health Service Coverage in Rural for Mother and Child Health in Northern Bolivia (summary). Bethesda, Maryland: Mozambique. Project application. Nacala, PRICOR/University Research Corporation, Mozambique and Washington, DC: Save the 1987. Children, 1999. Gottert, P, R Roland, and R Alban. Streamlining Lambrechts, T, et al. Integrated Management of Community-based IMCI: Six Guiding Principles. Childhood Illness: A Summary of First Project paper BASICS/JSI/LINKAGES/ Experiences. Bulletin of the World Health Madagascar, 2000. Organization 77(7):582–94, 1999. Gray, DHH, and J Ciroma. Attrition among Village LeBan, K, ed. Workshop Proceedings: Reaching Health Workers in Nigeria (summary). Communities for Child Health: Partnering with Bethesda, Maryland: PRICOR/University PVOs in Integrated Management of Childhood Research Co., 1987. Illness (IMCI). Washington, DC: Child Survival Green, EC Indigenous Healers and the African Collaboration and Resources (CORE) Group, State: Policy Issues Concerning African 1999. Indigenous Healers in Mozambique and Leonard, SG, ed. Community IMCI Workshop, Southern Africa. New York: PACT Publications, March 29 and 30, 2000. Consultancy to the 1996. Ministry of Health, Ghana. Arlington, Virginia: Griffiths, M. The Manoff Group. Semistructured BASICS II Project, 2000. interview, 2001. Lysack, C, and L Krefting. Community-based Griffiths, M. The Manoff Group. Personal Rehabilitation Cadres: Their Motivation for communication, July 12, 2000. Volunteerism. International Journal of Griffiths, M, and V De Alvarado. Honduras: The Rehabilitation Research 16(2):133–41, 1993. AIN Community Experience (unpublished Marsh, D, et al. Enlisting Traditional Healers for monograph). The Manoff Group, 1999. Rapid Change in Community Health Norms in Heggenhougan, K. Community Health Workers: a Conservative Rural Ethiopian District. Child The Tanzanian Experience. New York: Oxford Survival XIII Midterm Evaluation. Save the University Press, 1987. Children, October 1999. Henderson, A. World Vision. Personal McHenry, M. Adventist Development and Relief communication, April 10, 2000. Agency. Personal communication, March 4, Hilton, D. Emory University. Personal 2000. communication, April 13, 2000. McNulty, J. CARE International. Personal Howard-Grabman, L. Community Epidemiology communication, March 19, 2000. Information System. Save the Children and Mullins, J. Minnesota International Health Johns Hopkins University presentation at Volunteers. Personal communication, July 11, BASICS II, Arlington, Virginia, 2000. 2000. Ishan, T. Save the Children. Personal Ofosu-Amaah, V. National Experience in the Use communication, June 6, 2001. of Community Health Workers. A Review of Kartikeyan, S, and RM Chaturvedi. Community Current Issues and Problems. WHO Offset Health Volunteers: A Waste of Precious Publication 71:1–49, 1983. Resources? World Health Forum 12:341–342, Ojofeitimi, EO, et al. Increasing the Productivity of 1991. Community Health Workers through Kaseje, DC, et al. Characteristics and Functions Supervision in the Rural Areas of Nigeria’s Ife- of Community Health Workers in Saradidi, Ijesha Zone (summary). Bethesda, Maryland: Kenya. Annals of Tropical Medicine and PRICOR/University Research Co., 1987. Parasitology 81 suppl. 1:56–66, 1987. Oriokot, F. AMREF. Presentation at a meeting with government and NGOs contributing to improvement in child health. Kampala, S

E Uganda, February 9, 2000. C N E R E F E R

40 Paison, NG. Policy-level Impact of the Save the Robinson, SA, and DE Larsen. The Relative Children–Department of Health Partnerships Influence of the Community and the Health on the Community Volunteer Health Workers System on Work Performance: A Case Study of Program. In Presented Papers: High Impact Community Health Workers in Colombia. PVO Child Survival Programs, Volume 2. Social Science & Medicine 30(10):1041–8, Proceedings of an Expert Consultation, 1990. Gaulladet University, Washington, DC, June Rosales, A, S de Mena, and J de Mazariego. 21-24, 1998. Published for the PVO community Community Support for Improving Child Health: and the U.S. Agency for International Its Promotion and Network Sustainability. In R Development by the Child Survival Steinwand. Reaching Communities for Child Collaboration and Resources (CORE) Group, Health: Advancing PVO/NGO Technical Washington, DC, and Basic Support for Capacity and Leadership for Household and Institutionalizing Child Survival Project Community Integrated Management of (BASICS II), Arlington, VA, 1999: 183–191. Childhood Illness (HH/C IMCI). Proceedings of Pareek, U. 1986. Motivational Analysis of a workshop, January 17-19, 2001, Baltimore, Organizational Behavior (MAD-B). In JW Maryland. Washington, DC: Child Survival Pfeiffer and LD Goodstein, eds. The 1986 Collaboration and Resources (CORE) Group Annual: Developing Human Resources. San and Arlington, Virginia: Basic Support for Diego, California: University Associates, 1986. Institutionalizing Child Survival Project Parlato, M, and M Favin. Progress and Problems: (BASICS II), 2001. An Analysis of 52 A.I.D. Assisted Projects. Rubardt, M. Personal communication, March 22, Washington, DC: American Public Health 2000. Association, 1982. Ruebush, TK, et al. Qualities of an Ideal Volunteer Pearcy, K. Mercy Corps International. Personal Community Malaria Worker: A Comparison of communication, April 19, 2000. the Opinions of Community Residents and Phillips, JF, WL Greene, and EF Jackson. National Malaria Service Staff. Social Science Lessons from Community-based Distribution of & Medicine 39(1):123–31, 1994. Family Planning in Africa. Population Council San Sebastian M, I Goicolea, J Aviles, and M Policy Research Division Working Paper No. Narvaez. Improving Immunization Coverage in 121. New York: The Population Council, 1999. Rural Areas of Ecuador: A Cost-effectiveness Quinones, IdPL. Community Health Workers: Analysis. Tropical Doctor 31(1):21–4, 2001. Profile and Training Process in Colombia. Sanghvi, T. BASICS Project. Semistructured Washington, DC: Pan American Health interview, 2001. Organization and World Health Organization, Schaefer, M, and Reynolds, J. Operations 1999. Research Issues: Community Health Workers. Rasmuson, M, et al. (eds.). Community-based Bethesda, Maryland: PRICOR/University Approaches to Child Health: BASICS Research Co., 1985. Experiences to Date. Arlington, Virginia: Shanklin, D. Andean Rural Health Care. Personal BASICS Project, 1998. communication, March 9, 2000. Reis, T, et al. An Examination of the Performance Silimperi, D. University Research Corporation. and Motivation of Indonesian Village Health Personal communication, April 17, 2000. Volunteers. International Quarterly of Snetro, G. Vision for Change: Community Health Community Health Education 11(1):19–27, Promoters in Mozambique (draft). Save the 1990. Children Health, Population and Nutrition Unit, 2000. S E C N E R E F E R

41 Solat, S. Africare. Personal communication, July Walt, G, et al. Are Large-scale Volunteer 12, 2000. Community Health Worker Programmes Steel, A. BASICS Project. Semistructured Feasible? The Case of Sri Lanka. Social interview, 2001. Science & Medicine 29(5):599–608, 1989. Steinwand, R.W. Reaching Communities for Werner, D. The Village Health Worker: Lackey or Child Health: Advancing PVO/NGO Technical Liberator? World Health Forum 2(1):46–68, Capacity and Leadership for Household and 1981. Community Integrated Management of Winch, P, K LeBan, and B Kusha. Reaching Childhood Illness (HH/C IMCI). Proceedings of Communities for Child Health and Nutrition: a workshop, January 17-19, 2001, Baltimore, NGO Contributions to Community IMCI. Maryland. Washington, DC: Child Survival Calverton, Maryland: The Child Survival Collaboration and Resources (CORE) Group Technical Support Project, Arlington, VA: Basic and Arlington, Virginia: Basic Support for Support for Institutionalizing Child Survival Institutionalizing Child Survival Project Project (BASICS II), and Washington, DC: Child (BASICS II), 2001: 45. Survival Collaboration and Resources (CORE) Stinson, W, M Favin, and B Bradford. Training Group, 2001. Community Health Workers. An issue paper Wollinka, O, et al. Hearth Nutrition Model: prepared for UNICEF by the World Federation Applications in Haiti, Vietnam, and of Public Health Associations, 1983. Bangladesh. Arlington, Virginia: Basic Support Storms, D. Training and Working with Auxiliary for Institutionalizing Child Survival (BASICS) Health Workers: Lessons from Developing Project, 1997. Countries. APHA Monograph Series 3. Wubneh, H. Gumer Child Survival Project and the Washington, DC: American Public Health Horses: An Alternative and Sustainable Association, 1979. Approach to Child Survival Programs. In B Svitone, EC, R Garfield, and MI Vasconcelos. Burkhalter and V Graham, eds. Presented Primary Health Care Lessons from the Papers: High Impact PVO Child Survival Northeast of Brazil: The Agentes de Saúde Programs, Volume 2. Proceedings of an Expert Program. Pan American Journal of Public Consultation, Gaulladet University, Health 7(5):293–302, 2000. Washington, DC, June 21-24, 1998. Published Taylor, C. Health Equity in Nepal. (draft, January for the PVO community and the U.S. Agency for 29, 2000). Johns Hopkins School of Public International Development by the Child Health. Survival Collaboration and Resources (CORE) Walt, G. Community Health Workers: Policy and Group, Washington, DC, and Basic Support for Practice in National Programs. EPC Institutionalizing Child Survival Project Publication No. 16. London: The Evaluation (BASICS II), Arlington, VA, 1999: 124-133. and Planning Centre, London School of Hygiene & Tropical Medicine, 1988. S E C N E R E F E R

42 Annexes

Annex 1 Examples of the Use of Multiple Incentives

Catholic Relief Services, disaggregated by community, two of the El Salvador communities had lower rates (Santiago de The Catholic Relief Services/CARITAS child Maria fell from 85 percent to 15 percent and survival project began in El Salvador in 1995. San Vicente from 67 percent to 33 percent), Community health committees, health while the third community, Zacatecoluca, promoters, and health collaborators were increased its attrition rate from 40 percent to elected during community meetings. The role 60 percent. The project found that garment of the health committees was to maintain a factories recently opened in Zacatecoluca had link with health facilities and promote drawn the volunteers away for paid intersectoral collaboration. The health employment (Rosales 2000). promoters collected local health information and supervised the health collaborators. The Atención Integral a la Niñez, health collaborators were responsible for Honduras holding monthly meetings and making home The Atención Integral a la Niñez project is visits to mothers to discuss breastfeeding, found in about three-fourths of the health vaccination, and diarrhea management. areas in Honduras. It is implemented by the The midterm evaluation of the project MOH with technical assistance from BASICS. found a high attrition rate among the All health centers (HCs) in those areas collaborators. The project used the Pareek participate and are phasing in the number of motivational model to plan multiple incentives communities served, for a total of about 1500 for the volunteers. The achievement motive communities. While the numbers vary was addressed by using participatory training considerably, three monitoras cover 25 to 35 that reflected community needs, included the households with children under two years old. development of community organizing skills, HC nurses select communities with the and took place in the community. Forming lowest health indicators on the basis of a women’s support groups and health health needs assessment. Communities are committees, providing ID badges, and asked whether they would like to participate in promoting volunteer networks addressed the the program. Two communities have refused. affiliation motive. The extension and influence The nurses hold community meetings in which motives were achieved when the community community members are asked to nominate and MOH recognized and expressed support two to five people to be trained as monitoras. for the volunteers through “achievement days” The only criterion for selection is that one of every six months. The dependency motive the nominees must be highly literate. About a was attained by periodic supervision by health third of the volunteers have previous staff and preferential treatment of volunteers experience as midwives or health promoters. at health facilities. The control motive was Most are women with children of all ages. achieved through the use of the health Literate people are often younger and without information system. children. Many are among the better off in the The final evaluation showed that the community, and all tend to be extroverts and attrition rate had dropped from 59 percent to have a “presence” in the community. 41 percent (a difference of 18 percent) among The monitoras’ duties include the 1

three communities. When the data were following: X E N N A

43 ■ Weigh all children under two each month and commitment to volunteer work, of which and provide individual nutritional there is a long history in Honduras. Their counseling based on the child’s weight work is celebrated in a yearly party, which gain in the previous month. usually includes several hours of training. ■ Identify and seek out all children under Initially, the program laid out a year’s worth two who do not attend the weighing of incentives, including a letter to the session and keep the lists of children up monitoras’ families from the MOH thanking to date. them for allowing the monitoras to work, ■ Keep track of children who are not gaining certificates of achievement, identification weight. badges, and t-shirts. ■ Make follow-up home visits to children Anecdotal evidence shows a fairly low who were sick or did not gain weight in dropout rate among the monitoras. Those who the previous month. do drop out tend to be younger, literate people ■ Hold community meetings three times a who find paying jobs. Some of the monitoras year to report the results of the monthly who work on the border with El Salvador weighing sessions. border often ask for payment because their ■ Be available at set times to treat sick Salvadoran equivalents are paid a salary. children under five using a modified IMCI Monitoras were very upset when their algorithm and, if needed, provide oral reimbursement for travel costs for training rehydration solution or antibiotics. (The was delayed. monitoras do not make follow-up Monitoras report monthly on the results of household visits for this component, the weighing sessions. Sometimes the HC which was added recently and is being nurses attend the weighing sessions to phased in). vaccinate children. Increasingly, the nurses recognize the critical role played by the The monitoras spend about a day or half monitoras in decreasing their workload. a day a month conducting the weighing Sometimes monitoras help at the HC during sessions and perhaps another two days national immunization days and other health making home visits. Three monitoras usually events. Attempts are being made to make the work in each community. Working as a group municipalities more responsive to community is critical to their success. The groups can needs, such as improved water supply decide how to divide their work. All seem to (Griffiths 2000; Griffiths and De Alvarado work well together. The program has not heard 1999). of any problems within the groups. The initial training lasts five days, after Jereo Salama Isika, Madagascar which the monitoras are given three days of The Jereo Salam Isika program began in two training for disease treatment. For this training pilot districts in 1997 and is now going to they receive a minimal travel allowance. The scale in 20 districts, with a total population of frequency of refresher training and review 4.5 million people. When the decision was meetings varies greatly, usually depending on made to go to scale, a number of changes the interest and initiative of the health center were made in the pilot program, including nurse. About 60 percent of monitoras receive expanding the topics to include family some type of refresher training, primarily planning, dropping one cadre of health focused on counseling. volunteers (the amis de santé), and deciding The monitoras receive no financial to use only existing community groups rather compensation for their work. Their main than create new ones. incentives seem to be their recognition by The current program has two levels of 1

X the community and their sense of altruism community volunteers: encadreurs and E N N A

44 animateurs. The encadreurs are selected by of village animation committees proved to be the MOH from existing community leaders. a labor-intensive process. The current program The animateurs are volunteers who are works only with existing community groups. nominated by the community. The only criteria The program also uses mass media, for selection are that they want to serve as broadcasting 45-second radio spots and short animateurs and are nominated by the rural radio programs frequently over ten community. The program is so popular that stations. Volunteers are not supervised, but people knock on the doors of the encadreurs are supported and celebrated through health asking to become animateurs. In each festivals (Sanghvi 2001; Gottert et al. 2000). community served, ten encadreurs support 30 animateurs. The main tasks of the animateurs Afghanistan Health Sector are organizing village theatre 20 to 30 times a Support Project (AHSSP) year, using the counseling cards with groups In October 1986 a cooperative agreement was of people, and planning health festivals. signed in Peshawar, Pakistan, between Sometimes the animateurs go to the health Management Sciences for Health and USAID center to help organize people when crowds to begin the Afghanistan Health Sector gather for occasions such as immunization Support Project (AHSSP). Very early on the days. The animateurs are trained in groups of project developed an accelerated strategy to 20 in two-day workshops, where they learn the train a high volume of community or primary use of counseling cards and village theatre health care workers, called basic health techniques. workers. The BHWs would provide preventive, Most of the animateurs are women with promotive, and simple curative services and young children with a range of literacy, would be expected to enroll in ongoing 12-day although some are men. When asked why refresher courses when they went to replenish they want to be animateurs, most say they their supplies. want to improve the health of their own The BHWs were selected according to the families and increase their respect in the following criteria: community. The program expects that 50 percent of the animateurs will stop working 1. ability to read and write after 12 to 18 months. The animateurs give 2. equivalent sixth class education two reasons for stopping their work: 1) their 3. age of at least 16, preferably 20 to 30 child is seven years old, and there is nothing 4. residence in the assigned work location more to learn and 2) they can have jobs and 5. immediate family inside Afghanistan still be seen as resources in the community. 6. no employment in Pakistan A number of principles and lessons 7. willingness to participate in the resistance learned from the pilot program have benefited movement inside Afghanistan after the current program (see Gottert et al. 2000 completion of the course for the complete description). “Small, doable 8. Muslim religion and previous participation actions” rather than increasing knowledge, are in the resistance movement the focal points of the overall strategy. These actions were agreed on by all MOH and NGO The first three criteria established the partners, ensuring complementary and candidates’ capacity to complete training consistent collaboration. Counseling cards successfully. The next three were aimed at were developed to show each of the small, selecting candidates with a personal interest doable actions, and the same images were in seeing improved health services inside used in a variety of other materials. The two- Afghanistan. The seventh criterion reinforced day workshops allow many more people to be the expectation that the BHW would work 1

trained. During the pilot program, the creation inside Afghanistan: serving the resisting X E N N A

45 population was part of the resistance the life of the project lost their lives while movement. The last criterion tried to eliminate traveling. The BHWs were not reimbursed for candidates who had not yet served their their travel to Pakistan for the initial training, regular term as mujahideen and consequently but they were often sponsored by the might be called to serve in the armed forces community inside Afghanistan. Travel from of the resistance. Pakistan to the locality of assignment and From 1987 to 1993, 2,242 BHWs were transport of medical supplies were advanced trained and 2,190 served for some period. at going rates. With very few exceptions, all BHWs trained Cross matching of three data sources, as were male because of the travel requirements. well as having monitoring reports Every BHW’s location inside Afghanistan was countersigned by local community leaders, visited at least once a year, and many up to gave the BHWs a sense of being treated three times a year. Visiting project monitors fairly. Particularly those who stayed in the took pictures, obtained signed statements of project for several years appreciated the military and civil authorities on the BHWs’ project’s ability to pick up and single out performance, and recorded structured “cheaters.” In many communities the BHWs interviews with patients. Quality control was received recognition as “doctors” because no the most difficult aspect of the project: few other source of Western medical care was highly skilled health workers, such as available. Many BHWs expressed their physicians and nurses, were willing to risk satisfaction with being able to converse traveling around the country for technical directly with the agency implementing the supervision visits. program rather than depending on Afghan BHWs were given several incentives. authorities, who were often perceived as Upon graduation from the initial training and partisan and corrupt. The BHWs—even those completion of each refresher course, each excluded from the program for their poor BHW received a written document confirming performance—expressed appreciation for his new skills. This document was valued getting a “fair deal.” highly for the prestige it gave the BHW and The average time of active participation the accountability it ensured to community of all the BHWs who dropped out of the leaders for the time spent in Pakistan. project was 25 months, with a minimum of Although aware of the possible negative long- less than a month and a maximum of 67 term effects of providing salaries, the AHSSP months. The average (and median) attrition decided to pay the BHWs because salaries rate was 5 percent of the total BHWs enrolled were seen as “cash for work” and gave Afghan a quarter, with a minimum of 1 percent and a families still living in rural Afghanistan an maximum of 8 percent a quarter between incentive to stay there. The amount was set at September 1988 and September 1993. 870 Pakistani rupees, whose value against In October 1992 the project decided to cut the U.S. dollar fell steadily to about half over salary supplementation to 25 percent of the the life of the project. Each BHW received an original amount by April 1992 and later to 50 initial kit of medical supplies and basic percent. The contents of medical kits were equipment that could be resupplied every revised and quantities reduced to reflect the three to six months. actual average patient load of the BHWs. No Moving between their assigned working increase in dropout rates was seen over the places and Pakistan for training, supplies, and 12 months following the initial cuts. salary payment was hazardous for the BHWs Interestingly, BHWs active in the because of the geography and infrastructure provinces bordering Pakistan had a much of the country and the political and military higher dropout rate (63 percent of all trained) 1

X instability. Most of the 22 BHWs killed during than those from other provinces (32 percent of E N N A

46 all trained). Reasons proposed to explain this ■ The border provinces were occupied by a difference include the following: mosaic of political factions and to a large extend made up the former “tribal belt” of ■ Refugee families in Pakistan were Afghanistan, making it difficult for the originally mostly from the border areas BHWs to link with possible referral health and often had “split families,” part of facilities in the hands of other whom still occupied land inside commanders. In the nonborder provinces, Afghanistan. BHWs from such families larger regional political entities had had less incentive to work permanently emerged by 1989, offering more local inside Afghanistan. support to BHWs and in some cases an ■ The location of most cross-border health opportunity to fit into regional health projects in the border provinces created systems. opportunities for more gainful employment of BHWs once they were trained, as well (This appendix was drafted by Paul Ickx from as competition with more skilled health project reports and discussions with Laurence workers. Ickx-Laumonier, former field operations officer of the AHSSP.) 1

X E N N A

47

Annexes

Annex 2 Questionnaire for E-mail and Initial Interviews

1. Please give examples of incentives that were successful or unsuccessful in retaining CHWs on a long-term basis and keeping them motivated.

2. Please relate any specific positive or negative experiences regarding the use of the following types of incentives for CHWs:

a) Public recognition b) Income-generating activities c) Management of a first aid, drug, or commodity fund or kit d) Training e) Supervision f) Personal development opportunities g) Provision of food h) Provision of a monetary stipend i) Provision of a bicycle or moped j) Mentoring k) Other

3. Have you found any differences between regions of the world (Africa, Latin America and the Caribbean, Asia, the Newly Independent States) and the types of incentives that were successful in promoting long-term retention and motivation of CHWs? Please give examples.

4. Do you track information regarding retention rates of CHWs and the cost of rehiring and retraining?

5. Can you tell us how to obtain documentation on your project experiences with CHWs?

6. Who else would you recommend that we speak to regarding incentives for CHWs? How can we reach them through e-mail or by phone?

7. Is there any additional information on CHWs that would be useful to you if collected and analyzed? 2

X E N N A

49

Annexes

Annex 3 Interview Guide for BASICS Examples

Learning about CHWs: Semi-structured Interview Guide 1. Background: ■ What is the population served by your program? ■ What is the total number of CHWs active at any one time? ■ How many CHWs leave the program or become inactive each year? ■ How many CHWs do you recruit each year?

2. CHW selection: How were CHWs selected? What criteria were established? Who established them? What role did the community play?

3. CHW characteristics: Could you describe some of the characteristics of the CHWs (e.g., sex, literacy level, previous experience)?

4. Duties: What are the specific duties of the CHWs? What is the relative balance between curative and preventive activities? How many households does each CHW cover? How much time is a CHW expected to spend on his/her duties?

5. Drugs: What drugs, especially antibiotics or antimalarials, were CHWs supplied with? How are they resupplied? How common are stock-outs?

6. Training: What resources do you put into training of newly recruited CHWs each year? How many training courses for new CHWs are there in a year? How many CHWs are trained each time? What was the length of training that CHWs received? What is the frequency and duration of in-service (or refresher) training? Were per diem or travel allowances provided?

7. Incentives: What incentives did the program plan for the CHWs? What do you think motivated the CHWs to work? Can you give examples of incentives that were successful in keeping CHWs motivated over a period of time? Please relate any specific positive or negative experiences regarding the use of the following incentives: ■ Cash in any form, including sale of drugs ■ In-kind incentives such as farming help, dishes, or t-shirts ■ Job aids, special identification ■ Community recognition

8. Multiple incentives: Is there more than one incentive for the CHWs to work? If so, did these evolve? Were they planned? How do they work together?

9. Disincentives: Have any CHWs dropped out? What do you think are the main reasons for dropping out? Do you feel that the program has a problem with high turnover or dropouts or attrition? Why or why not? 3

X E N N A

51 10. CHW/facility link: What is the relationship between the CHW and the health facility? Is the facility staff supposed to supervise the CHWs? How does this work? What difficulties are there?

11. CHW/community link: What is the status of CHWs in the community? Are they perceived as government health workers or volunteers? What CHW functions does the community value most? How are CHWs linked to community groups, including health committees?

12. Peer support: Do the CHWs work alone? In pairs? In groups of three? How are they meant to work together? What opportunities are CHWs given to interact with other CHWs from other areas?

13. Information on CHWS: What information do you routinely collect on CHWs (e.g., retention rates, costs of training, costs of incentives)? Is this type of information important to you? If so, how do you use this data (e.g., to modify incentives)?

14. Documentation: Do you have any documentation of experiences with CHWs that you can send to us? 3

X E N N A

52