The construction of a decision tool to analyse local demand and local supply for GP care using a synthetic estimation model

Willemijn A De Graaf-Ruizendaal,Dinny H De Bakker

doi:10.1186/1478-4491-11-55

Abstract

BackgroundThis study addresses the growing academic and policy interest in the appropriate provision of local healthcare services to the healthcare needs of local populations to increase health status and decrease healthcare costs. However, for most local areas information on the demand for primary care and supply is missing. The research goal is to examine the construction of a decision tool which enables healthcare planners to analyse local supply and demand in order to arrive at a better match.MethodsNational sample-based medical record data of general practitioners (GPs) were used to predict the local demand for GP care based on local populations using a synthetic estimation technique. Next, the surplus or deficit in local GP supply were calculated using the national GP registry. Subsequently, a dynamic internet tool was built to present demand, supply and the confrontation between supply and demand regarding GP care for local areas and their surroundings in the Netherlands.ResultsRegression analysis showed a significant relationship between sociodemographic predictors of postcode areas and GP consultation time (F [14, 269,467] = 2,852.24; P <0.001). The statistical model could estimate GP consultation time for every postcode area with >1,000 inhabitants in the Netherlands covering 97% of the total population. Confronting these estimated demand figures with the actual GP supply resulted in the average GP workload and the number of full-time equivalent (FTE) GP too much/too few for local areas to cover the demand for GP care. An estimated shortage of one FTE GP or more was prevalent in about 19% of the postcode areas with >1,000 inhabitants if the surrounding postcode areas were taken into consideration. Underserved areas were mainly found in rural regions.ConclusionsThe constructed decision tool is freely accessible on the Internet and can be used as a starting point in the discussion on primary care service provision in local communities and it can make a considerable contribution to a primary care system which provides care when and where people need it.

Highlights

Responsive primary care There is a growing academic and policy interest in the appropriate provision of primary healthcare services to the population of local areas to increase health status and decrease healthcare costs [1-3]
The regression analysis showed a significant relationship between the predictors of the model and general practitioners (GPs) consultation time (F [14, 269,467] = 2,852.24; P
The results revealed that 11 variables were significant predictors of GP consultation time

Summary

Introduction

Responsive primary care There is a growing academic and policy interest in the appropriate provision of primary healthcare services to the population of local areas to increase health status and decrease healthcare costs [1-3]. Spatial microsimulation models have a long history in economics and are increasingly used in epidemiology as an alternative to local health surveys [12]. Such models construct large synthetic micro data at the small area level on the attributes of individuals or households by combining different sources of information to ‘estimate geographical distributions of variables which were previously unknown’ [13], p 1128. The research goal is to examine the construction of a decision tool which enables healthcare planners to analyse local supply and demand in order to arrive at a better match

Methods

Results

Discussion

Conclusion