The aim of this thesis is to assess the effect of doctor ethnicity and country of qualification on prescribing cost and frequency, and generic prescribing; taking account of doctor personal and practice characteristics where possible. A new refined weighting factor (ASTRO-PU) has been used and two study designs employed to achieve this aim. As well as global prescribing, four therapeutic groups were chosen as these were thought to be commonly issued by Asian doctors. In addition, prescribing patterns of GPs are described by these personal and practice characteristics. The first study (Ecological) used aggregate prescribing data from the 90 Family Health Services Authorities (FHSAs) and found that being an Asian doctor and qualifying from the Indian sub-continent (ISC) were associated with increased frequency and prescribing but not cost. The second record linkage study used data collected from 155 single handed GPs; from FHSAs; from the 1991 Census and from the Prescription Pricing Authority. To adjust for potential confounding, analysis of covariance was used and parsimonious models developed by backward elimination for each dependent variable. This showed that, for global prescribing, Asian doctors qualified from the ISC do not issue items more frequently, issue expensive items, or fewer generic drugs than Asian and white doctors qualified from the UK. Further, no significant difference was found between the three groups of doctors for: 1) volume of paracetamol and non-steroidal drugs; 2) cost of antibiotics; expectorants; paracetamol and non-steroidal drugs; and 3) generic prescribing of antibiotics. However, there was a difference between the three groups of GPs for both the volume of antibiotics and expectorants and the cost/volume of paracetamol. It was noted that the parsimonious models explained little of the variation in the prescribing dependent variables. This thesis lays to rest, for global prescribing and specific therapeutic groups, the myths of Asian doctors being high volume and cost prescribers. The second study has also contributed to the understanding of prescribing variation at the individual doctor level.