The "supply hypothesis" and medical practice variation in primary care: testing economic and clinical models of inter-practitioner variation

Medical practice variation (MPV) is marked, apparently ubiquitous across the health sector, well documented, and continues to be a focus of professional and policy interest. MPV have stimulated two paths of investigation, one economic in emphasis and the other more clinical in orientation; while health economists have stressed the potential role of income incentives in medical decision-making, health services research has tended to emphasise clinical ambiguity as a factor in practitioner decisions. Both sets of explanations converge in an implicit "supply hypothesis" that posits contextual practitioner and practice attributes as influential in clinical decisions. Data on inter-practitioner variation are taken from a large and representative regional survey of general practitioners in New Zealand, a country in which unsubsidised fee-for-service is the predominant mode of remuneration in primary care. The paper assesses the impact on three important areas of clinical decision-making -- prescribing, test ordering, request for follow-up -- of three key conceptual dimensions -- income incentives, physician agency, and clinical ambiguity (operationalised as local doctor density, practitioner encounter initiation, and diagnostic uncertainty respectively). Predictions are made about inter-practitioner variations in the rate of clinical activity in the three areas. The results of the analysis using multi-level statistical techniques are: 1. the extent of competition -- local doctor density -- seems to have no effect on the pattern of clinical decision-making; 2. doctor-initiated visits are, if anything, associated with lower rates of intervention; 3. diagnostic uncertainty is associated with higher rates of investigations and follow-up, both of which have clinical plausibility; 4. there is no significant interaction effect between density and uncertainty. It is concluded that, for the clinical activities studied and for the practitioner attributes as operationalised in this investigation, a clinical, rather than an economic, model of practitioner decision-making provides a more plausible interpretation of inter-practitioner variation in rates of clinical activity in general practice. The "supply hypothesis" requires further analytical refinement and empirical assessment before it can be applied as a generic explanatory framework for MPV.