The effectiveness of contracting-out primary health care services in developing countries: A review of the evidence

MAIN FINDINGS:

The authors included evidence from 16 studies assessing the effectiveness of 13 contracted-out health-service interventions in Sub-Saharan Africa, Latin America and the Caribbean, Europe, South Asia and East Asia and the Pacific. The results of the review are heterogeneous. On the one hand, the authors deduce that contracting-out health services significantly improves access and availability, especially within under-served regions. On the other hand, they fail to find conclusive evidence concerning the impact of contracting-out on other dimensions of health-system performance such as equity, quality and efficiency. Indeed, the authors report that these factors are addressed in only a very few studies, and they highlight the need for more rigorous studies exploring the impact of contracting-out health services on the equity, quality and efficiency of health systems. The authors suggest that the effectiveness and success of contracting-out depend on a range of contextual and contract-design factors. More specifically, they observe that contractual financial incentives and the application of a payment-by-performance scheme have been demonstrated to be key determinants in the success of an intervention.Contextual factors related to the capacity of the public sector to develop the contracted-out services and the degree to which these services complement or replace existing services are also important determinants of the impact of contracted-out interventions on the equity, access, quality and efficiency of health systems. Finally, the authors highlight the need for more research on possible side effects of contracting-out interventions, which so far have been only barely assessed.

BACKGROUND:

During the past decade, improving the performance of health systems has become a key objective for the governments of many developing countries. Contracting-out is defined as a “contractual arrangement by which the government provides compensation to private providers in exchange for a defined set of health services for specified target population”. By providing incentives to health-care professionals and by fostering competition among health-service providers, contracting-out health-service interventions can improve the quality and efficiency of health-care services, leading to improvements in the performance of the health system. The findings of previous reviews on the effectiveness of contracting-out health services are mixed, and this systematic review aims to shed light on the effectiveness of contracting-out on the four main dimensions of health-system performance: access, equity, quality and efficiency.

RESEARCH OBJECTIVES:

To assess and synthesise the evidence on the impact of contracting-out interventions on health-system performance in developing countries.

METHODOLOGY:

The authors included all studies assessing the effectiveness of contracting-out health-care services to private providers in developing countries. As quality criteria, the authors only included studies that used one of the following study designs: experimental, non-randomised controlled designs, before-and-after designs without controls and cross-sectional studies with controls. They conducted a systematic search in published and unpublished literature for relevant studies. They searched the electronic database PubMed, and used the search engine Google. They also searched the websites of relevant organisations, including the World Bank and the World Health Organization, and checked previous reviews in the field and other literature for additional studies.Finally, the authors grouped the results by health-system performance dimension – access, equity, quality and efficiency – and synthesised the results in a narrative format.

QUALITY ASSESSMENT:

This systematic review has clearly defined inclusion criteria and use appropriate methods for analysis of results. It has some limitations however. The search is not sufficiently comprehensive, and it is not clear whether language bias is avoided. Moreover, although the inclusion criteria in terms of methodology are wide, the authors do not report any systematic assessment of the quality of included studies. The authors acknowledge the weaknesses of the evidence base and do not provide and strong policy conclusions, and this mitigates some of the limitations of the review.