Predicting Activity and Workload in General Practice from the Demographic Structure of the Practice Population

Abstract

Objectives:

The managerial requirements of budget-setting and performance monitoring in general practice (primary care) in the UK require an understanding of the causal relationship between practice activities and the characteristics of both the practice and its patients. This study sought to model the determinants of three major components of general practice activities (consultations, prescribing costs and referrals to secondary care), paying particular attention to the influence of the demographic structure of the patient list.

Methods:

Stepwise regression analysis was carried out on data for 98 practices in the county of Lincolnshire using 12 independent variables pertaining to patient and practice characteristics plus 14 statistical measures derived from the demographic structure of the patients registered with the practice.

Results:

Robust statistical models were estimated for the three dependent variables, of which list size emerged as the most significant independent variable. In addition, six other independent variables, including the patients' unemployment rate, fundholding status and single-handed status, were statistically significant in one or more of the equations. Variables based on the demographic structure of the practice population also appeared in each model. The Jarman score and degree of urbanization did not achieve statistical significance.

Conclusions:

Activity and workload in general practice can be predicted from routine data. Such models are of particular value for planning and financial management when demographic change in practice populations is anticipated.