(January 2011) Vouchers are frequently mentioned as a promising alternative finance mechanism to achieve a variety of goals in health systems and reproductive health services. Do vouchers work? Ben Bellows and co-authors reviewed the literature on vouchers to identify reproductive health programs and to determine the extent to which these programs have been evaluated. Findings were generally positive, but much remains unknown about program operations. How important are governance structures? What happens as voucher programs go to scale? How can challenges with fraud and poor information systems be overcome? Do programs reach the poor? Do providers really face competition for clients? In a PRB Discuss Online, Ben Bellows, associate with the Population Council and co-author of “The Use of Vouchers for Reproductive Health Services in Developing Countries: Systematic Review,” answered questions from participants about program design and performance for reproductive health vouchers.


Jan. 18, 2011 1 PM (EST)

Transcript of Questions and Answers

Joan Healy: Question One: In one country (Cambodia) using this system, providers complained about how long it takes for them to get reimbursed by the gov’t for the voucher after the service has been provided. This was in the early phase of introduction so this might be resolved now. Question Two: In India, community based workers are supposed to get reimbursed for bringing the woman in for the free delivery services. They do not get reimbursed until after the delivery. These community workers must then spend hours at the facility in order to get reimbursed. This is not feasible given these community workers oftentimes have small children who accompany them to the facility.
Ben Bellows: Hi Joan, Great questions. Yes, reimbursement is a key component in voucher projects and most projects experience some difficulties and delays as they learn to properly carry this out. In the Uganda STI voucher project for instance, some rural providers will delay filing claims, preferring to ‘batch’ them when transport to town is available. There’s mention of reimbursement issues in Melissa Densmore’s stakeholders paper (www.melissadensmore.com/papers/ictd09-mho-stakeholder.pdf). Delays on the voucher management agency side can reduce confidence in the project. In your India example, one can imagine that the situation could remain suboptimal if there are relatively few facilities for community workers to refer clients to. Community workers have little choice in where clients go and facilities have little incentive to be more responsive to clients and community workers. In voucher projects that use claims reimbursement, the review process needs to be timely for providers who operate on a cash basis. Claims review should be transparent and quick. In the Uganda STI project, we saw a reduction in processing time and fewer claims rejected as providers and VMA both learned from their experience over time.

Mary Lyn Gaffield: Do vouchers have an impact upon increasing access to family planning for women who have given birth within the past 12 months? And, are data available to show whether or not vouchers reduce unmet need among these women?
Ben Bellows: Hi Mary Lyn, Thank you for this question. In the Kenya Voucher project, there has been a noted increase in estimates of district level contraceptive prevalence among users of intrauterine device (IUD). I don’t know what proportion of these clients were women who had given birth within the past 12 months. If you’re interested, we could probably estimate that proportion based on subnational demographic data. Uganda is beginning a new long term and permanent methods voucher project and this will be something to measure on the claims form. One of the useful characteristics of many voucher projects is the use of a claims form that records a few variables of demographic, socioeconomic and spatial data that allows for a richer description of the client population.

Kennedy Ongwae: Hi Ben and colleagues, From your work with voucher schemes, do we know the main “drivers” of success of the schemes in achieving the desired impact? Issuing vouchers to users to access free or subsidised care? or more resources to the service providers? Depending on how you would define “going to scale”—do we have voucher schemes that have been scaled up and become part of the health care systems with substantial funding from Government budget?
Ben Bellows: Hi Kennedy, Great to see you on this forum and thanks for the question. National ownership of the program is a key driver for long term sustainability. As you know from the Kenyan experience, there are voucher champions who have made output-based initiatives in Kenya, like vouchers, part of the long term national development strategy (see “Vision 2030”).


Another key driver would be evidence that these projects can have an impact on utilization, out of pocket spending, facility quality, facility efficiency, and population health status. To date, those studies are few, but there are prospective studies planned or currently underway in Kenya, Uganda, Tanzania, Bangladesh, Cambodia, and several other countries and literature reviews of past studies (see the Council’s RH Voucher project website: www.rhvouchers.org). This PRB talk includes a link to a recent systematic review on reproductive health vouchers. DfID, 3ie and AusAID are funding other reviews on related healthcare finance issues and once these studies are available, we’ll have a much better idea where vouchers and other finance strategies offer value.


As for projects going to scale, the historic examples from Taiwan and South Korea in the 1960s and 1970s were both national initiatives with substantial funding from Government (the same Council systematic review we just completed draws evidence from the Taiwan and Korea programs).

Tanja Kiziak: 1. Are voucher programs an effective instrument to reach poor women in rural areas? 2. What are the best ways to distribute the vouchers—and how can it be ensured that only those entitled to the services (i.e. the poor) are given the vouchers?
Ben Bellows: These are great questions. Voucher programs can be a great mechanism to reach rural disadvantaged populations (including poor women) and voucher programs have been found to be successful in increasing healthcare utilization among rural populations. For example, the western Uganda program reaches rural communities across 20 districts and the two reproductive health voucher programs in Kenya have significant rural coverage. Voucher programs can include a transport reimbursement (Bangladesh and Cambodia are two examples) that help rural populations reach proper care.


There are three common strategies to distribute vouchers. Entire districts or counties or neighborhoods can qualify. This is called geographic targeting and has been used in western Uganda in several subdistricts and in Nicaragua in barrios near to schools (for adolescent health vouchers) and in red light areas to encourage commercial sex workers’ use of sexual health services. The second way is to use a simple standard poverty test or risk assessment. This selects beneficiaries based on standard criteria, such as the “below poverty line” (BPL) card in India. The third method involves locally determined criteria, using for instance measures of poverty that are agreed by the community. In all three strategies a decision is made to either distribute vouchers for free or charge some fee. The western Uganda program has sold STI treatment vouchers through drug shops. In Nicaragua, vouchers with a short term expiry were given for free. For the maternal care programs in Uganda and Kenya, community based workers sell the voucher at a very low price to clients who qualify. Only the poor are permitted to purchase the voucher. Community sales agents are required to conduct home checks to confirm the client does indeed qualify.

Awa Minteh: In my country there are not such programes. I would like to know if they have really worked in other countries where it is being implemented. I have developed interest in Reproductive Health, especially among teenagers. how do one identify reproductive health problems and design programs that can best address them.
Ben Bellows: Your question raises a good point. There is need for materials to guide needs assessment and setting up new programs. The World Bank produced a “Guide to Competitive Vouchers” in 2005. You can find a copy of the World Bank guide and related resources about a Nicaragua voucher program for adolescent sexual health on this website: www.icas.net/new-icasweb/english/en_pubs.html. The Population Council has managed a “Friends of Youth” program in Kenya for a number of years that has included adolescent sexual health ‘coupons’ redeemable at private providers in the community. More information is here: www.popcouncil.org/projects/61_KenyaFriendsOfYouth.asp.  

Kibet Sergon: What are the possible sustainable financing mechanisms for the voucher programs or other modalities of social health financing. How can government take up such programs?
Ben Bellows: This is a difficult question as financing dynamics involve often a good deal of politics, which will vary from place to place. Donor funding can be direct as is the case in the German Development Bank (KfW) in Uganda and Kenya or indirect through basket or Sector Wide Approach (SWAp) funding in Bangladesh. When talking about national governments, programs could be financed from social insurance or tax-based contributions from Treasury. Either way, the voucher service links financial disbursement with service delivery. More information on financing strategies can be found at the Center for Health Market Innovations (http://healthmarketinnovations.org). Governments are taking up voucher programs. In Bangladesh the demand side finance (DSF) voucher initiative has been very popular. In fact, in the selection of new upazilas for a recent program expansion, the program managers were encouraged to include an extra upazila to satisfy requests from Government. In Kenya, the output-based approach (OBA), as realized in the voucher program, is a popular component in the government’s multisectoral strategy to achieve significant growth over the next 20 years (“Vision 2030”).

Earl Grandstaff: Why hasn’t articles come out in magazines on the 2010 census count?
Ben Bellows: I’m sorry, but I don’t have enough information to know which census this refers to.

Pradeep Bohara: what is the exact meaning of reproductive health??
Ben Bellows: This a good question and important to have agreement on definitions. The World Health Organization says the following: Within the framework of WHO’s definition of health as a state of complete physical, mental and social well-being, and not merely the absence of disease or infirmity, reproductive health addresses the reproductive processes, functions and system at all stages of life. Reproductive health, therefore, implies that people are able to have a responsible, satisfying and safe sex life and that they have the capability to reproduce and the freedom to decide if, when and how often to do so. Implicit in this are the right of men and women to be informed of and to have access to safe, effective, affordable and acceptable methods of fertility regulation of their choice, and the right of access to appropriate health care services that will enable women to go safely through pregnancy and childbirth and provide couples with the best chance of having a healthy infant. From www.who.int/topics/reproductive_health/en. Wikipedia may be helpful here too: http://en.wikipedia.org/wiki/Reproductive_health.  

Bakrr: Can developing countries governments impose RH methods to halt ethic differences in TFR. In one community variance is so great 9 against 2.4 which raised concerns of out-numbering?
Ben Bellows: Thank you for your question. No, governments should not impose RH methods to level total fertility rates in different populations. It would generally be considered a gross violation of human rights. RH programs are designed to meet the needs of individuals who voluntarily seek RH services. For reference, please see numerous statements from the United Nations on sexual and reproductive health rights: www.un.org/ecosocdev/geninfo/women/womrepro.htm.  

Pradeep Bohara: In the developing country, people think that, Reproductive health is the Issue of the female only, how to minimize such type of thinking??
Ben Bellows: This is an important question addressing male involvement in reproductive health (RH) services and RH decisions. There are a good number of governments, donors, NGOs and others engaged in generating greater male involvement in family planning. USAID has funded many programs. I don’t have the papers available, there are a good number of papers on male involvement in pregnancy, family planning services and other RH issues. Two examples linked here: USAID www.usaid.gov/our_work/global_health/pop/news/amman.html and Population Council www.popcouncil.org/pdfs/frontiers/orsummaries/ors45.pdf. Vouchers can be a useful mechanism to get partners discussing RH issues. In the Uganda STI Voucher project, the voucher is sold to the client with half of the voucher available to the client’s sexual partner. This referral mechanism is very useful in settings were ‘contact tracing’ is not feasible response to STI epidemics. It can also encourage greater male involvement in cases where the woman bought the voucher.

Debbie Fugate: Could you provide citations for published studies on the issue of vouchers for RH that you would consider key?
Ben Bellows: The tables in this recent systematic review describe seven evaluations of 13 published voucher programs. The reference list is a rich source of citations: http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3156.2010.02667.x/pdf. If you’re interested in program design options, please see the World Bank 2005 “Guide to Competitive Vouchers in Health” linked near the top of this page (Instituto Centroamericano de la Salud), where you’ll find other citations: www.icas.net/new-icasweb/english/en_pubs.html  

Gillian Eva: Hi Ben, Could you clarify what the advantages are of using a voucher system over providing discounted or free services?
Ben Bellows: Hi Gillian, Great question. There are several advantages to using vouchers. On the demand side, clients have a physical reminder in hand to go to the facility. It’s the same reason retailers print coupons in newspapers. On the supply side, service use is linked to financial reimbursement. Even in ‘virtual voucher’ programs, the supply side benefits are available. In a good voucher program, information systems are in place to record medical and financial information about the clinical experience. Information systems are likely to be more complete in voucher programs that require clear and completed claims before payment is made. Perhaps the most useful distinction between voucher and free services is the targeting mechanism. For maternal delivery services, targeting poor women is important to reach those who may not otherwise consider a facility delivery.


For a good summary of a ‘virtual voucher’ program in Gujarat, India, please see this YouTube interview with economist April Harding: www.youtube.com/watch?v=eFfuGDDQreo.   

Karin Ringheim: I’m interested in the issue of integrating Family planning (FP) into other services, particularly maternal and child health (MCH) services. Have you come across examples of where vouchers have been successfully used to encourage mothers to access FP within recently integrated FP-MCH programs? Thanks.
Ben Bellows: Hi Karen, this is a good question and timely given the strong interest in integrated RH/ FP services. The Kenya voucher program offers a FP voucher, a separate maternal care (safe motherhood) voucher, and a third gender based violence recovery voucher. In a recent review of the program, it was recommended to integrate FP services in an expanded maternal RH voucher.

Carrie Ngongo: In a world of constrained resources where we would like to see fewer women dying and injured in childbirth, what would you say are the pros and cons of a country establishing a voucher program for maternity services (targeting the poorest or most rural populations) vs. announcing free maternity services for all, such as the policy decision in Sierra Leone in 2010? I’m especially curious about cost and public health impact considerations.
Ben Bellows: Hi Carrie, Thanks this is a very topical question. Announcing free maternity services for all—without changing the underlying reasons for low service utilization—takes place only at public facilities and risks leaving mothers disconnected from the healthcare system. We know from past studies that even in countries like Sierra Leone maternal death is disproportionately concentrated among the very poor and rural populations—the very populations least likely to benefit from removal of user fees. In contrast, vouchers are targeted to women who need the RH service. Vouchers can be used to provide transport subsidies as is done in Bangladesh and Cambodia. Vouchers can bring funding to public and private facilities based on client demand. The combination of stimulating client demand and linking that with financial disbursements to both public and private facilities is very different from removal of user fees at public facilities only.


An added benefit of vouchers is that targeting can be improved over time. Determining if people other than the intended beneficiaries are using the service (“leakage”) can be measured with simple checks at a sample of facilities. One disadvantage to vouchers is that healthcare management must improve if the services are to be effectively delivered. Prompt information feedback is important in most systems and that certainly holds for vouchers.

Marcia Gomez: Thank you for all the reference and resources being shared! My understanding is that many funds for RH are being minimized worldwide (especially for contraception). Are voucher programs something that has been looked at in areas where funding may be limited for these services; in other words, can these services continue in particular to poor families if a voucher method is instituted?
Ben Bellows: Hi Marcia, many thanks for the question. If rationing funds for healthcare to the poor is a priority, vouchers can be an effective mechanism to ensure that the services are properly targeted to the poor. Like any program, however, there is an administrative cost and decisions may have to be made on the desired level of verification versus percentage spend on service delivery.

Rose Reis: Has the Gujarat scheme Chiranjeevi Yojana been replicated in other Indian states?
Ben Bellows: Hi Rose, Good to see you on the forum and good question. I haven’t seen the Gujarat Chiranjeevi Yojana program replicated. As background for readers, Bhat and colleagues describe the program in this paper in 2009: www.banglajol.info/index.php/JHPN/article/view/3367/2814. Checking the Center for Health Market Innovations database, I didn’t find similarly titled programs mentioned there although a blog post from last year suggested that it would be a potentially useful model to try elsewhere. Some of the state level JSY programs in India have used vouchers for community referral purposes. The WHO produced a good 2010 summary on how vouchers fit with other demand side finance strategies with some discussion on vouchers in Indian healthcare finance reforms: www.who.int/healthsystems/topics/financing/healthreport/27DSF.pdf.  

Arin Dutta: What would be a good mechanism to integrate voucher schemes with non-governmental forms of health financing like community health insurance? The idea being that the voucher can still be used to generate choice and agency for the recipient, but the payments are settled from alternative sources. Are there any examples?
Ben Bellows: Hi Arin, Great question and thanks for posting. Vouchers use a targeting mechanism and give clients choice as you noted while contracting providers so vouchers could perhaps be used to expand uptake of community health insurance. I haven’t seen examples of this yet, but in Cambodia there are interesting health equity funds (HEFs) that have integrated vouchers with contracted health services. It’s perhaps the closest scenario in which vouchers are added to existing contracted services. Voucher-like cards will be used in Tanzania beginning later this year to encourage use of maternal health services in a pilot organized by the National Health Insurance Fund (NHIF) in the context of community health funds (CHFs). There may be disadvantages to vouchers in small community health insurance programs if voucher distribution or reimbursement bring more administrative costs.