We present the main results from a five-year research project on health Sector-Wide Approaches (SWAps) aimed at supporting the transition from project approach to SWAps. The principal hypotheses tested were that: (i) SWAp is a dynamic approach and not a blueprint model; (ii) SWAp has major advantages over the project approach; (iii) however, SWAp is not a panacea. A multidisciplinary (public health, economics) research group, financed by the Belgian cooperation, was set up in 2004, comprising teams from 3 Belgian universities and a focal point from the Dept. of Development Cooperation. Over the 5 years, the group performed several missions in sub-Saharan Africa (3 in Mali, 3 in Benin, 2 in Senegal, 4 in Rwanda, 5 in Tanzania and 1 in Uganda). An analysis grid and 3 types of questionnaires were used to assess and follow-up the evolution of the SWAp dynamics and its results, which showed that (i) SWAps are not a rigid model and is definitely a dynamic way of working together between a government and its development partners. The core elements of a SWAp are a common sector strategy and a country-led coordination mechanism. However, different elements of the SWAp mature over time in a context-specific way. SWAps should constantly evolve and be used as a means of making progress in sector policy, planning, service provision and evaluation. (ii) SWAps can quickly bring in some positive results, especially in terms of coherence and coordination of interventions; policy dialogue, trust building and common understanding between partners; and ultimately sector reforms and health outcomes. Thus they have major advantages over the project approach. (iii) However, SWAps are not a panacea and its means should not be confounded with its ends, that is, improving sector performance. SWAps entail risks (notably those of technocratisation, centralisation and disconnection from field realities) and need to be constantly fuelled by shared analyses and common reform projects aiming at lifting major sector constraints. The case studies helped accumulating a good knowledge of SWAp dynamics at central level, but the evaluation of the impact of SWAps on health outcomes – which was also analysed – is uneasy. The results from the case studies were used by the Belgian cooperation and other stakeholders to accompany the SWAp dynamics, especially in Mali where a technical assistant was appointed to support the process.