Analyzing the impact of health services: project experience from India, Ghana and Thailand
Faruqee, R.
World Bank Staff Working Paper, International Bank for Reconstruction and Development 546: 44
1982
Document Number: 387058
Measuring health and its progress -- the emphasis of the health discipline -- is essential for economic analysis. In addition, economic analysis, which emphasizes comparing benefits with costs, is needed to ensure efficiency in the use of health resources. For both disciplines, therefore, it is important to find reliable health indicators that will show the impact of health services. Four categories of indicators are pertinent: environment (socioeconomic conditions and health policies); inputs (services offered); outputs (services received); and outcomes (changes in mortality, morbidity, and nutritional status). This paper reviews what three important controlled experiments reveal about these indicators: Narangwal in India, Danfa in Ghana, and Lampang in Thailand. The experiences of three non-experimental projects -- with no control groups -- are also reviewed. Outcome measures naturally are best for indicating the impact of services. But the three controlled experiments reviewed here show that collecting data on outcome variables is expensive, time consuming, and faced with difficulties of controlling for many non-project influences. All the projects -- experimental and non-experimental -- showed that output indicators served well in indicating the impact of services. At Narangwal, such indicators seem to have captured what the outcome variables showed about the impact of health services. The appraisal and evaluation of health projects still is a long way from the goals of economic analysis -- comparing benefits and costs or assessing whether an approach is the most cost-effective. This paper shows, however, that a beginning can be made in that direction by using outputs of health services as surrogates for outcomes.