Between 1979 and 1990, Columbia University conducted 26 family planning operations research (OR) projects in 13 sub-Saharan African countries. Most of these projects were implemented in settings where family planning service delivery had not yet been initiated or was new and poorly developed. In keeping with program needs in the early stages of development, the majority of the OR projects were based on demonstration or diagnostic designs. Only four of the 23 projects were comparative or quasi-experimental in design. Projects demonstrated the growing demand for family planning, and the feasibility and acceptability of a range of service delivery models. Sixteen of the projects have been sustained or replicated by national or local governments and institutions following the initial OR phase, replication of two others is planned, and another four resulted in policy formulation or improvements in existing national programs. OR in Africa played an important role in generating political and medical support for contraceptive services, and served to improve project management. OR proved more useful in the implementation and strengthening of individual programs than in generating universal lessons regarding service delivery. In order to maximize OR's utility in Africa, simple research designs and methods should be adopted.
PIP: This overview of 11 years of operations research (OR) conducted in Africa by the Columbia University Center for Population and Family Health (CPFH) includes 26 OR projects in 13 sub-Saharan countries. The experience is summarized with emphasis on the successes and limitations. OR, as an interface between social science and program functioning, provides data on issues of direct interest to managers; i.e., sociocultural information to increase program acceptability, measures of the effectiveness of program components, information on the feasibility of new program strategies, and guidance for cost reduction. Designs include diagnostic and evaluation projects, demonstration and pilot projects, comparison tests, and quasi-experimental projects. the CPFH projects between 1979-89 are tabled by country location, project goal, sector, research design, general description of project or service delivery model, and current status. Topics in the article are identified as the African context for family planning OR, lessons learned from OR in Africa (demand for family planning exists, family planning programs are acceptable, and motivation to use family planning is growing), lessons learned about OR in Africa (primary uses of OR in Africa in the past decade, support for initial family planning programs, enhancement of the political and medical acceptance of FP, OR as a management tool, the importance of host institutions and personnel, selection of OR designs, data collection methods, institutionalizing or, and the limitations of OR), and discussion. Included in the discussion of OR designs is he suggested general approach to the selection of research designs based on preexisting levels of contraceptive use (0-10%, 5-25%, and 20%). Thus in Africa, CPFH's projects have been primarily demonstrations (58%), then diagnostic and evaluative (27%), and comparison studies or quasi-experiments (15%). This contrasts with the quasi-experimental projects in Latin America which constitute just under 60%, as appropriate to the more advanced level of existing services in that region. Determining the most appropriate use of OR has at times strained relationships between donors and project staff. OR has been useful in developing, testing, and replication of delivery models and fine tuning individual program components, but has been less useful in the development of universal models for program components, or new projects. It is not particularly useful in predicting whether 1 delivery strategy will yield a greater number of acceptors than another, or in projecting what prevalence levels can be achieved through any one strategy. The definition of OR itself has sometime created problems. A major challenge is to convince institutions to maintain their research and evaluation capabilities, and to meet the needs of underserviced groups such as lactating women, men, and teenagers.