Does a voucher program improve reproductive health service delivery and access in Kenya?

Results-based financing has emerged as a potentially significant strategy to finance health and other social protection services in the health sector. It is mainly used for expanding the quantity and quality of targeted subsidized health services [1–4]. Results-based financing include initiatives such as output-based aid (OBA), pay for performance, cash on delivery and performance-based financing [5–7]. It includes a range of health initiatives aimed at achieving health targets such as improvement in health outcomes, greater output of specific health services and changes in health related behaviors with increased use of health services by individuals [7–10].

The theoretical context of the OBA or voucher mechanism is found in the basic economics theories of supply and demand with the aim of using market mechanisms to efficiently subsidize health services for individuals who would likely go without the service in the absence of the voucher. The main rationale behind subsiding health care is the inequitable distribution of wealth and health [11]. Voucher programs target disadvantaged populations and contract healthcare providers to see the voucher clients, stimulating both supply and demand for services. The goal of OBA is often to increase access to and use of key services among the disadvantaged (e.g. poor or socially excluded) by offering them a subsidy to help purchase health services, and preferably by being able to choose a provider from among a number of alternatives [9, 11, 12]. Voucher programs have the potential to reduce gaps in equitable healthcare utilization by subsidizing low-income individuals’ purchasing power [11]. Voucher programs provide an economic incentive to accredited facilities by reimbursing them for services offered [13]. By doing so, the program can stimulate a market for the services and may generate greater competition between providers and indirectly motivate improvements in quality of services.

OBA programs provide incentives to clients and healthcare workers and subsidize specific health care services based on pre-determined prices [7, 11, 14–16]. When a client needs the services, s/he redeems the voucher for the specified service at one of the accredited facilities. The provider is then reimbursed service cost or paid an incentive upon submission of a claim and supporting evidence to the voucher management agency (VMA).

In Kenya since 2006, the OBA program funded by the German Development Bank (KfW) and the Government of Kenya delivers targeted subsidies for safe motherhood (SM), family planning (FP) and gender-based violence recovery services (GBVRS) to promote quality care and maximize healthcare utilization by people living in poverty. The SM and FP vouchers are sold through distributors to poor women in rural districts and low-income areas of Nairobi for a highly subsidized price, while the gender-based violence recovery services (GBVRS) vouchers are provided for free in GBVRS accredited facilities regardless of socio-economic status. The program was launched to generate health sector experience in targeting, accreditation, claims, reimbursement and quality for the proposed National Social Health Insurance Fund. The program has operated in Kisumu, Kitui, Kiambu and Kilifi counties, and Korogocho and Viwandani informal settlements in Nairobi. These sites are seen as representative of different regions of Kenya with poor populations found in both urban and rural settings.

Globally, voucher programs have been associated with increased skilled birth attendance, uptake of long-acting family planning methods, reduction in out-of-pocket expenditure, improved quality of care and improved access to care [17–21], but little is known about the facilities’ response to OBA programming. Drawing from interviews with healthcare providers and facility managers, this study assesses how well they understand the OBA voucher program, their attitudes toward the program, and their observations on the program’s benefits and challenges.

The paper draws from qualitative data collected in 2010 as part of a larger quasi-experimental design aimed at evaluating the impact of the Kenyan OBA voucher scheme on increasing access to, and quality of, selected reproductive health (RH) services [22]. These data were derived from in-depth interviews (IDIs) with facility managers and healthcare providers from program sites in Kitui, Kiambu, Nairobi, and Kisumu. The IDIs took place alongside quantitative health facility assessments and population surveys aimed to measure the impact of the voucher program on access to health facilities and quality of services. Details of the larger evaluation and population surveys are discussed elsewhere [22].

The in-depth interviews aimed at gaining a deeper understanding of the perceptions and priorities of the health care providers regarding the voucher program. The interview guide was amended appropriately after pre- testing the tools. The in-depth interview guide focused specifically on: (i) technical knowledge of services offered using the voucher (ii) attitudes towards the voucher program iii) benefits and (iv) challenges of the voucher program, processes and the approach (v) and perceptions on program implementation. A total of 69 interviews with facility managers (n = 30) and healthcare providers (n = 39) working at 30 OBA accredited public, private, and faith-based facilities were conducted. These facilities were randomly sampled from the 54 voucher accredited facilities selected to participate in the first phase of the Kenya OBA program. We planned to interview the facility manager and healthcare providers working in the reproductive health units in the study facilities. However, in some facilities the facility managers were not present on the days of the interviews, therefore interviews were only conducted with the healthcare providers directly involved with reproductive health or maternal and child health services. Overall the study interviewed 46 females and 23 males from facilities ranging from dispensaries to hospitals.

The IDIs were tape recorded and transcribed into a Microsoft Word file. The transcribed texts were then transferred to NVIVO 10 qualitative data analysis software. Qualitative data analysis was done by two researchers to ensure reliability in the coding and results. The study hypothesized that a voucher program would improve reproductive health service delivery and access to target population. A combination of inductive and deductive analysis was conducted based on previous research and project documents.

Following coding, a full list of themes was available for categorization within a hierarchical framework of main and sub-themes. The thematic framework was then systematically applied to all of the interview transcripts. We looked for patterns and associations of the themes and compared and contrasted within and between the providers and facility managers in different regions, sectors and levels.

Ethical approval for the evaluation was granted by Population Council’s Institutional Review Board (IRB) No. 470 and Kenya Medical Research Institute (KEMRI) SCC 174. Informed consent was obtained prior to all interviews that were conducted in settings that ensured privacy and confidentiality. Participants were informed that they could withdraw from the research at any time. Data collectors were trained on ethical conduct.

This paper explored the views of facility managers and providers on the Kenya voucher or OBA program. The study findings identify factors affecting program function as well as gauged supply side perceptions of the voucher program effects on improving access and utilization of reproductive health services. Our study demonstrates a wide variation in providers’ knowledge and awareness of the SM, FP and GBVR voucher services across the different health sectors and facility levels. The GBVR, FP and the postnatal care (PNC) component of the SM voucher are least known by health providers from dispensaries and health centers. In addition the FP is least known by health providers working in faith-based facilities. However, the SM vouchers were most well-known across all sectors and health care levels while the GBVR voucher was the least known. The low levels of awareness of FP vouchers in faith-based facilities could be attributed to the fact that most of these facilities do not support FP initiatives. This variation in knowledge appears to have affected the functioning of the OBA program design at the lower level facilities. For example, PNC services were sometimes not offered as part of the SM vouchers as providers did not know that PNC is covered by the voucher. In some facilities short term FP methods were provided under the voucher scheme although the program does not cover such services and such claims were further not reimbursed. The poor awareness of GBVRS voucher was accompanied by reported poor technical knowledge on the provision of GBVRS by majority of providers and managers interviewed. The variation in the knowledge and awareness of the benefit package of the different reproductive health vouchers could be attributed to limited training on specific RH issues as well as staff transfers and movements in and out of voucher accredited facilities. This poor level of awareness of the vouchers benefit package could also be attributed to: the lack of training updates on the voucher design and functioning; and existing gaps in program implementation with respect to marketing the voucher to health providers in accredited facilities. This points out to the need to improve information sharing between the VMA and accredited facilities on the key program designs. The knowledge gaps in relation to FP methods covered within the voucher program among mission facilities and lower level facilities could also be explained by the fact that, long-term FP methods were not offered at catholic (mission) and out-patient facilities; and majority of lower level facilities offer out-patient services only. At the dispensaries and health centers which mainly provide outpatient facilities, there was limited staff who would, for example, insert an IUD or implant. Still, the Catholic-based facilities could not provide the FP services due to their religious beliefs and values. These variation on knowledge and use of the SM, FP and GBV vouchers has also been demonstrated among targeted clients in the same program setting with a similar trend of better knowledge and use of SM and least knowledge and use of GBVR vouchers [23]. This variation of knowledge and use of the vouchers in both health workers and among the targeted clients could be indicative of the need to improve both facility and community sensitization of the voucher program and the benefit packages.

Our findings demonstrate poor skills, competence and knowledge to offer (LARC-PM) FP and GBVRS among providers especially those from health centers and dispensaries. This has implications on the quality of services offered. This calls for an urgent need to build providers capacity in all OBA accredited facilities to provide basic GBVRS and long acting and reversible contraception (LARC) and permanent methods (PM) (LARC-PM) Methods. There is a need to avail the recently updated national guidelines on management of sexual violence in Kenya and the national FP guidelines to all healthcare providers. Several other capacity building opportunities exist such as medical schools, nursing programs, clinical officers and laboratory technicians reviewing their curriculums to endow health workers with the relevant knowledge and continuous on job refresher training.

Although the OBA program was implemented to run within the existing health system, the public facilities performance was affected by this assumption since the inherent health system challenges in Kenya include lack of in-service training to update provider’s knowledge and skills. Overall, the low technical knowledge in key services provided within the voucher program highlights gaps in quality assurance that need to be strengthened. The effect of the voucher on the client load has implication on the implementation of the program. The private sector is able to adapt well by increasing the staffing levels and improving structures to accommodate the increased client load. However the increased workload in the public sector as a result of increased demand clearly indicates significant pressure on the current health system structure in government facilities. Care should be taken to ensure such facilities do not compromise the quality of services offered such as the rights and dignity of the clients in an attempt to meet the increased demand.

Majority of the facilities reported that they were affected by the delay of the VMA to reimburse the claims on time. Since the process and the benefit packages were not well understood, facilities submitted claims that were not reimbursable. Secondly, the VMA sometimes failed to provide feedback on reasons claims were not payable or for possible reason for delay in reimbursement further generating dissatisfaction with the claims and reimbursement process. It was evident that OBA training were non-continuous and paired with regular provider transfers and this impacted negatively specifically on the claims and reimbursement system. This resulted to poor working relationship of the VMA and some accredited facilities.

An important aspect in project implementation is communication and feedback between the various actors. Communication within the program between the providers and VMA appears to have been inadequate especially in rural facilities. Providers felt that automated systems and quicker reporting and reimbursement systems are key to better functioning of the program. If the OBA program is to function more efficiently there is need to explore the opportunity of decentralizing the claims process so as to offer more prompt services and feedback. There is also need to automate the claims and reimbursement system to reduce time and other expenses such as transport involved in claiming for the services provided.

The OBA program was affected by the existing health system challenges within the public sector including lack of drugs, understaffing, lack of equipment and supplies, poor referral systems, poor quality of care such as bed-sharing and poor hygiene. There is dire need to strengthen the public sector health system, and there are opportunities with the voucher program to utilize reimbursed funds to improve the facility especially the public sector. However, care should be taken to ensure facilities are not over dependent on the voucher reimbursed funds for expansion and management. Over dependence of the facilities on reimbursement from voucher programs poses a concern regarding the sustainability of the facilities in the absence of the voucher program. There is also an opportunity for the OBA program to strengthen supply chain management of supplies and commodities within accredited facilities to reduce stock outs. However, to improve such systems in the public sector, there is a need to advocate for change in policy environment such as allowing the public sector to utilize the reimbursed funds for expansion and improving the services provided. Further the OBA program is facing challenges in the rural areas associated with poor road network, limited transportation affecting access and the referral system. Although improving road infrastructure is beyond the scope of the program, it provides an opportunity for the OBA program to partner with other government ministries responsible for infrastructure.

The study found that managers and healthcare providers saw an impact of the voucher program on the accredited facilities and the targeted community. Majority of the providers and facility managers within the private sector felt that the voucher program is a profitable business model. The voucher program was also associated with increased income to the health facilities through the claims and reimbursement process, however this effect was reported more by the private sector compared to the public and faith based facilities. The revenues from the voucher program have been used by facility managers to upgrade the facilities, purchase drugs, supplies, equipment’s, provide incentives to clients such as improve hygiene and nutrition during admission, purchase of beds, provide incentives to providers to deliver quality services and compensate long working hours [24]. In the public facilities that were able to utilize the funds, the revenues have been utilized to improve health education on competition among health facilities in the region across the sectors which has led to an overall improvement of quality of care in the facilities. Private facilities appear to have used these revenues to expand and improve the quality of services as compared to public facilities. It was evident that the business model was more appreciated by the private and faith-based sector facilities compared to the public sector; this was mainly influenced by the facilities’ ability to make decisions on the utilization of the revenues. This perhaps implies the need of the County health departments within the devolved system to allow public facilities to utilize the funds to improve their facilities and make them competitive. This could be explained by the fact that privately owned facilities had the freedom to invest their funds in infrastructure, equipment, staff and client comfort, which satisfied and attracted more clients. In this setting, competition has been associated with improved quality of care and efficiency and suggest that results-based financing is associated with increased use and quality of care [9, 10, 20, 22].

The majority of the facility managers and providers felt that the voucher program increased access to services and reduced inequities by making RH services affordable to the poor. A population survey conducted in the same context to assess utilization of reproductive health services showed substantially higher proportions of voucher clients compared to poor non-voucher clients obtained LARC-PM, delivered at health facility and received skilled birth care and post-natal care services [25] This resonates with previous findings that demonstrate that voucher programs are associated with improving access and utilization of health services [21, 26]. By accrediting many health facilities at different levels, the voucher increased client choice and promoted competition among healthcare providers for the clients. The program’s competitiveness threatened the public facilities because given the freedom of choice by the voucher, many users had more confidence in, and utilized private facilities compared to public facilities.

Majority of facility managers and healthcare providers felt that the program was associated with a reduction of maternal and newborn mortality. There is need to conduct a prospective study that measures the impact of the voucher program on the reduction of maternal and neonatal deaths to verify this perception. In addition it was felt that the general health seeking behaviors of mother had improved. For example, majority of providers felt that mothers began attending ANC clinics early and that the proportion of mothers attending ANC clinics had significantly increased. It was also felt that the voucher had also led to increased facility deliveries within this setting. However, majority of providers felt that the vouchers had not been used to access PNC services compared to other services. Similar findings from other studies showing the association of voucher programs with increased facility deliveries and ANC uptake have been published [9, 19, 21, 23, 27]. A recent study published in the same context shows that PNC services were least known by women and that they reported very low PNC services [26]. The program needs to develop strategies to increase PNC uptake within this context.

The study identified some key lessons. For the successful program functioning there is need to ensure that the benefit package and criteria for targeting beneficiaries are well understood by managers and healthcare providers in the accredited facilities. Moreover, continuous clinical and program related training is required to emphasize both the quality in-service delivery as well as administrative matters such as reimbursement and claims process.

Income earned through the voucher program can be used to improve quality of care to satisfy clients’ needs and ultimately improving health seeking behaviors of the target population. However within the public sector, there are constrains on utilizing such funds and there is a need to explore ways that allow the public sector facilities to directly use the funds to reflect their client’s needs. The sustainability of the voucher program is greatly influenced by the client volume for the participating providers and the facility readiness to provide the services. Services that are stigmatized and critical such as GBVRS are not well addressed by voucher programs. Services that are provided by a wide range of health facilities are better accessed and utilized compared to services that requires referral for specialized services.

In this context, the vouchers programs are increasingly acknowledged as a promising health financing mechanism enabling access and improvement of quality of health services.