CORRELATES OF PHYSICIAN UTILIZATION - WHY DO MAJOR MULTIVARIATE STUDIES OF PHYSICIAN UTILIZATION FIND TRIVIAL PSYCHOSOCIAL AND ORGANIZATIONAL EFFECTS

This paper examines some major discrepancies in the literature on illness behavior involving psychosocial and organizational factors affecting physician utilization. At issue is why the large-scale multivariate studies find that such factors explain only small amounts of variation, whereas qualitative and more restricted quantitative studies find these predictors to be more influential. Among the factors accounting for discrepancies in results are the interpretation of 'illness' measures, differences in concept measurement and data aggregation, and the crosssectional versus processual analytical approaches to studying this issue. It is suggested that the advantages of both types of studies can be combined, facilitating improved theory and prediction.