Programme no. 222-OP
Health Economics
The extent and effect of socioeconomic factors on the distribution of resources in primary care in the region of Västra Götaland, Sweden
Erik Andersson*1, Sofia Dalemo2, Karin Starzmann3
1R&D Centre,Skaraborg Primary Care,Skövde,Sweden, 2R&D Centre,Skaraborg Primary Care,Skövde,Sweden;Dept of Public Health and Community Medicine/Primary Health Care ,Institute of Medicine, Sahlgrenska Academy at the University of Gothenburg,Göteborg,Sweden, 3R&D Centre,Skaraborg Primary Care,Skövde,Sweden;Dept of Public Health and Community Medicine/Primary Health Care,Institute of Medicine, Sahlgrenska Academy at the University of Gothenburg,Göteborg,Sweden
* = Presenting author
Objectives:

The objective is to explore whether socioeconomic factors, measured by the CNI, correlate with the ACG, as a measure of the primary care burden in Västra Götaland.

Background:

Earlier studies have shown that low socioeconomic status increases the risk of disease and premature death, leading to high costs for society. The Care Need Index (CNI) is a socioeconomic tool for measuring health care needs. However, many regions in Sweden use the diagnosis-based Adjusted Clinical Groups (ACG) model to allocate resources in primary care. In October 2009, the primary care service in the Västra Götaland region in Sweden introduced a new reimbursement system, based mainly on ACG and capitation. The ACG model has not previously been studied when used for resource allocation. Furthermore, it is not known whether there is any correlation between the ACG and the CNI.

Results: We found no statistically significant correlation between the CNI and the ACG (p=0.71). The spread in ACG between health care centres decreased over time and approached the mean. In contrast, the differences in the CNI increased over time. Metropolitan areas have slightly lower ACG averages compared with urban (p=0.016) and rural (p <0.001) areas, but there were no statistically significant differences for the CNI (p> 0.1) or between public and private health centres (p> 0.1).
Material/Methods: This study is an observational cross-sectional study of official primary care data from Västra Götaland, with 1.6 million inhabitants and a spending plan of € 550 million. CNI and ACG data for all health centres (n=204) in the region were retrieved three times (11/2009, 06/2010 and 06/2014) from the central registry and analysed for statistically significant correlations.
Conclusion: In contrast to what could be expected, health care centres in areas with low socioeconomic status, measured as high CNI, made no simultaneous high demands on health resources, measured as the ACG score. Furthermore, the difference in ACG between health care centres decreased over time, while the CNI increased.
Points for discussion:

The ACG system is based on patient diagnoses and can easily be influenced by physicians. Could this contribute to the decreasing ACG differences between health centres, despite the increase in the CNI? Could more of the resources be allocated through the CNI, which cannot be influenced by health centre staff?