Towards accurate prescribing analysis in general practice: accounting for the effects of practice demography.

The object of this study was to develop a more accurate method of analysing prescribing patterns in general practice than is currently provided by prescribing analysis and cost (PACT) data. In particular it was hoped to be able to develop a system to predict accurately both the prescribing activity related to practice population age-sex distributions, and the related prescribing activity in the individual therapeutic groups used in PACT reports. Data from the Vamp research data bank were combined with mean prescription costs from the Prescription Pricing Authority and the patient age-sex profile of each practice. The estimated cost of prescribing for each practice in each therapeutic group was derived, and subsequently compared with the PACT data. Large variations were found between predicted patterns and PACT 'averages' for both overall costs, and within therapeutic groups. This was most noticeable where a practice's age-sex profile varied markedly from the family health services authority mean age-sex distribution. Comparison of data for the different therapeutic groups showed the largest differential between the predicted patterns and PACT averages to be in the drugs for the cardiovascular system. Audit requires a baseline against which to measure, change and re-evaluate performance and this system of analysis provides a more accurate baseline against which to measure a practice's prescribing habits. Subsequent evaluation will provide a judgement of the quality of that prescribing. A practice population's age-sex structure is such a fundamental determinant of the practice's prescribing patterns that some modification of the current PACT weighting system should be undertaken so as to reflect more accurately average prescribing in different age groups.(ABSTRACT TRUNCATED AT 250 WORDS)