The spatial variation of the relations between Medicare spending (MS), use and quality in the United States was investigated employing spatial regression. A focus of the study was whether, and to what extent, MS and use vary by service type. Employing different spatial regression designs based on Medicare regional data, the impact of the heterogeneous spatial effects of hospital readmissions on MS for the elderly at the aggregate level was examined. The results were followed up by investigation whether the effects of hospital readmissions are heterogeneous with regard to service type. It was found that poor quality indicators lead to increased MS at the aggregate level and thus higher costs per beneficiary, and that the quality effects are heterogeneous with variable impacts, both spatially and by type of medical service. The results shed new light on the relationship between quality and MS highlighting the pitfalls of global averaging models that hide the reality of a highly diversified and spatially stratified country. Reducing payments to high-spending areas and increasing payments to low-spending areas should reduce spending variability but the quality indicators of care become ambiguous and not easy to interpret.
Medicare; Quality; Elderly; Geographically weighted regression; Spatial econometric models; USA.