A summary of the voucher panel at Global Health Council’s annual meeting in DC May 27th. Cross posted from Jaspal’s original post on Global Health Ideas…
As with the last post, I’m copying the description of the session I just attended - Vouchers for Health (Session C3) - from the conference website:
Presenters Discuss: the potential of competitive vouchers in increasing access to priority services for currently underserved populations (Nicaragua); how subsidized, targeted vouchers can achieve the goal of improving health of poor women and children, elements that improve access to high quality health care services, how targeted voucher subsidies help meet the goal of contributing to development of a national social health insurance system that is accessible, affordable and acceptable (Kenya); a nationwide, high transaction distribution project for insecticide treated nets (ITNs) that has given rise to the application of technologies in the use of a relational database as a means to track each voucher transaction, ensuring security and traceability, providing spatial analysis using GPS coordinates, leveraging data from this project to others by providing a central data repository to government, and exploring SMS and mobile phones for automating voucher transactions and gathering patient information (Tanzania); and how poor pregnant women were selected for vouchers jointly by field workers and community support group members, how the institutionalization process for the voucher scheme was implemented, and how the health facilities were strengthened to provide quality maternal health care (Bangladesh).
Why should we care about voucher schemes? Three reasons from where I sit. These schemes provide:
- An attractive model for extending the reach of services without significant infrastructural investment
- Incentives for competition to improve the quality of service delivery
- A mechanism for reducing financial barriers for the poor
Despite this promise, my experience over the last few years - including stumbling through a poster presentation about vouchers in Uganda for Ben at APHA in Boston a couple years back - has taught me that the concept tends to be elusive to “outsiders”. The World Bank’s Private Sector Development Blog has a concise overview of output-based aid for those that aren’t familiar. I won’t try to explain myself, since I’ll probably make some errors.
This session brought together diversity in geography - the talks covered Latin America, Africa, and Asia - and in services - anti-malarial bednets, safe motherhood, family planning, STI prevention and treatment. What was most interesting to me then was not the details of any specific program, but what we might be able to learn from having these different experts in the same room.
So what were the common themes?
- Vouchers work. Nicaragua saw increases in service utilization, condom utilization, family planning uptake, and KAP. Much of this across different subgroup analyses - age, level of sexual activity, type of residence. The others similarly all saw positive gains in what they were trying to achieve.
- Institutions matter. In Tanzania, the market is being flooded with free ITNs (possibly LLINs) under a new policy to achieve universal coverage - the fate of the 7000 bednet retailers under the Tanzania National Voucher Scheme (TNVS) is unknown. In Kenya, Nairobi and Kiambu saw better results for the voucher scheme because of existing infrastructure. Nairobi actually had provided more voucher-based services than vouchers sold in Nairobi because of women coming in from outlying areas. Tanzania benefits from a manufacturing base that can produce ITNs and a system that encourages people to pay for a portion of their health care.
- The system will be gamed. The much higher reimbursement for providers of C-sections than vaginal deliveries has led in some cases to higher rates of C-sections than is necessary. April Harding who was moderating said that multiple schemes have been used/proposed for dealing with this, including capitation (e.g. for 100 women, X% will be reimbursed for C-sections). Fraud is also a problem. While nobody discussed incidences of counterfeiting, the Tanzania program has taken big steps to prevent counterfeiting, including watermarks, microprinting, and a bar code. The bigger issue with fraud is related to ineligible participants obtaining and using vouchers intended for other populations, whether by income or geography. Which leads us to the next point…
- Equitable distribution is hard. Ineligible participants seem to be a problem with all these programs. In Bangladesh they responded by bypassing governmental decisions about who would get vouchers and relied on CSGs (community support groups) to make the decisions. Another trend is that those who are more poor tend to utilize vouchers less - it is unclear if this due to cost, education, and interaction among the two, or something we’re not thinking about.
- Understanding redemption is complicated. This was my one question. In both Tanzania and Kenya - the others didn’t present these numbers or I didn’t catch them - the rate of redemption (number of people receiving services under the voucher scheme divided by the total number of vouchers distributed) - was around 80%. After I asked the question and before the panelists responded, my neighbors said that: (1) 80% is pretty good, and (2) the Tanzanians unlike the Kenyans don’t pay for their vouchers until they use them. In Tanzania then, the reasons are grounded speculations: perceived value of the ITNs, access, cost. In Kenya, the majority of the outstanding 20% is due to accounting. The real utilization is expected to be much higher since the cutoff for the evaluation excluded women who had received vouchers, but had not yet delivered (safe motherhood program). Still a minor portion of this is due to women who are unable to deliver at contracted facilities due to a lack of transportation or the timing of the pregnancy.