Health services research in the English-speaking Caribbean 1984-93: a quantitative review

Evaluating the effectiveness and efficiency of health services is important for all countries, especially those with limited resources. This study aimed to evaluate the volume and quality of health services research (HSR) conducted in one developing region, the English-speaking Caribbean. Data were abstracted from all 770 abstracts describing presentations at the annual scientific meetings of the Commonwealth Caribbean Medical Research Council for the decade 1984 to 1993. Of these, 341 abstracts were judged to report health services research and were from the English-speaking Caribbean. Hospital services were evaluated in 240 (70%) reports while primary health services were evaluated in only 90 (26%). Most hospital-based studies evaluated the use and outcome of medical and surgical services through the collection of case series and cohorts of cases, with a median sample size of 104 (interquartile range 38 to 320). Evaluations at primary level were more likely to evaluate need or demand for services, were more likely to report cross sectional surveys or randomized trials and included larger numbers of subjects (median 343, interquartile range 121 to 661). Patient-based measures of health status and measures of resource use were not often reported at either primary or secondary level. Estimation and hypothesis testing were infrequently employed in data analysis. A large proportion of the research presented could be classified as HSR but measures are needed to increase the motivation for research into primary care and to improve skills in HSR study design, conduct and analysis among those presently conducting research.