Sustainability of US supported health programs in Honduras

or its predecessors since the 1940's. Findings are based on document review, statistical analysis, site visits, and interviews with Honduran and U.S. Government personnel. The projects examined include water system and health facililty construction, anti-malarial campaigns, and family planning efforts undertaken from the 1940's to the 1960's and primary care activities (nurse training, hand pump and latrine provision, and nutritional planning) implemented in the 1970's. Sustainability was defined as the continuation of project outputs and benefits after A.I.D. funding ends, and was divided into immediate outputs (e.g., trained personnel, or hand pumps and latrines installed during the project) and "replicating" outputs necessary to reproduce immediate outputs (e.g., schools, a construction agency). It was found that when A.I.D. funding stopped, some outputs could be sustained with national funds, but replicating outputs were usually sustained with funds from other donors. Analysis of three cases of high sustainability (nurse training, rural water supply, latrines and pumps) and three of low sustainability (family planning, malaria, and nutrition projects) indicated that: (1) a high national commitment to project goals is a necessary but not sufficient condition of sustainability; (2) projects that were perceived as effective during their life were more likely to be sustained, as were those characterized by cooperative A.I.D./Honduran planning and implementation (projects viewed as A.I.D.-imposed were not well-sustained); (3) vertically organized projects often generated institutional resentment which jeopardized sustainability, while integrated projects tended to be sustained but at lower levels of effectiveness (a matrix-like organization which combines vertical and horizontal components is suggested); and (4) sustainability can be jeopardized if several donors support the same project and terminate funding simultaneously - on the other hand, projects which receive other donor support after A.I.D. funding ends are likely to be sustained. Factors found not to be clearly related to sustainability were: size of A.I.D. funding, national assumption of recurrent salary costs, cost recovery, proportion of the national health budget devoted to hospitals. Nor were size and duration of TA, training components, or community participation significantly related. Therefore, A.I.D. should more clearly define its objectives for sustainability in future projects, and it should not expect all outputs to be sustained.