Sector-wide approaches in developing countries
BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7254.129 (Published 15 July 2000) Cite this as: BMJ 2000;321:129
The aid given must make the most impact
- Paul Garner, head,
- Walter Flores, lecturer in international health,
- Shenglan Tang, lecturer in international health
Sector wide approaches (SWAPs) are a new development in the organisation of aid to developing countries where donors and lenders collectively contribute to funding the entire health sector.1 They are being widely promoted as a “good way to do business.” Donors are embracing sector wide approaches enthusiastically, and they have been introduced in the health sector in several countries, including Zambia, Ghana, and Bangladesh, but the realities have been little tested.2
In the past, donors provided support through project aid, resulting in many self contained projects which tended to fragment the health system, or through programme aid, where particular donors may provide funds through national budgets. In sector wide approaches donors agree to contribute to a single basket of funds, which in turn contributes to the country's national plan. The partner government and development agencies negotiate and agree policies and plans for development in the sector, including resource allocation, and aid is provided within this context.3 The potential benefit is that money is spent on priorities set by the country, not external agencies, and aid is more efficiently managed through the country's existing structures, with only one set of monitoring and accounting mechanisms.4 For poorer countries, with many donors contributing to poverty alleviation in the health sector, these intended benefits could dramatically improve the effectiveness of aid.
Yet sector wide approaches also have the potential to make things worse. The negotiations may fail; parts of the sector are sometimes excluded from the sector programmes; or spending on unproductive areas may continue or even increase. For example, national governments might use the funds to build hospitals with high recurrent costs or to support the salaries of staff working in unproductive areas. Similarly, once governments are in control of donors' resources, they may decide to reduce funding to programmes such as malaria and tuberculosis and other programmes likely to benefit a large proportion of the population.
Donors are introducing sector wide approaches in many countries, and the World Bank now regards them as integral to its strategy in Africa, based on the judgment that this is the appropriate next step for more effective aid management and health sector development.5 We have sought empirical evidence of benefit for sector wide approaches from published data using Medline, Sociofile, the World Bank web site, and reference lists from articles about this topic but have found none. An evidence based standpoint avoids decisions based on anecdote, expert consensus, or simple rationale, so can we therefore demand that policymakers provide “evidence” that sector wide approaches work before implementing them?
Disadvantages and risks
Some say no, as sector wide approaches are already happening now and seem to reflect a more mature relationship between governments and development agencies. Yet sector wide approaches do have disadvantages and risks, so there is an urgent need to establish methods of evaluation that are explicit, transparent, and scientific, and conducted in a form that encourages efficient communication between the stakeholders. This will help refine the approach, to the benefit of both the recipient, in ensuring effective healthcare provision, and to the donor, in improving value for money.
We believe there are three key principles in establishing effective evaluation of sector wide approaches within countries. Firstly, when sector wide approaches are negotiated a protocol for evaluation should be included, which specifies in advance criteria and methods to evaluate the degree of success implementing the strategy. This protocol should include a small number of specific criteria that apply to all evaluations of sector wide approaches, based on expected benefits and risks, such as proportion of funds spent on specialist hospitals. Secondly, the evaluation process should include people independent of the ministry and agencies. Thirdly, reports should follow the structure of the protocol, indicating any outcomes that the team intended to evaluate but were unable to. This process does not prevent researchers from investigating any new and relevant developments subsequently, but it keeps the focus on the primary intentions. Within countries this approach makes sense, but how can countries be compared? Every country is different, national policies vary, and the intention and contents of sectoral policies and strategies emerging from sector wide approaches will be unique. Thus what is required are standard terms and methods useful across countries to describe the setting, the policy framework, strategies, and allocations; the potential influences on implementation and impact; and types of indicator useful in evaluating implementation. There are already examples of country strategies that embody the principles of sector wide approaches, and these should be the starting points for developing a framework for evaluation.6 Given current international experience in research synthesis and the opportunity for updating afforded by electronic publication, these evaluations could be compiled together and updated as new information emerges.
Evaluation is essential
Sector wide approaches provide an opportunity for the World Bank and others to show that “components of [the evidence based medicine] concept are not new … for World Bank operations,”7 an assertion which some of us believe lacks substance.8 There are political and methodological challenges to be overcome. In particular, donor assistance evaluations are through expert appraisal, their consultancy reports from these experts rarely have a methods section, are usually unpublished, and their distribution is often restricted. In addition, countries or donors may resist full evaluation against the specified criteria, particularly if the evaluation is likely to be negative. Yet it is time for donors and countries to establish an open process to evaluate and synthesise evidence around sector wide approaches and other major policy decisions. This helps donors improve transparency and accountability to their taxpayers and, most important of all, to maximise the impact of aid.
Acknowledgments
We have all previously received salary replacement or research grants from donor organisations that are now actively promoting sector wide approaches.