Don't stop me now: The impact of retirement proximity on GP performance

Population ageing has been widely studied for its implications for healthcare demand. Yet much less attention has been paid to the parallel ageing of the medical workforce. Focusing on the Italian primary care setting, this study addresses this gap by examining whether GP performance changes as physicians approach the retirement age of 70. Drawing on human capital theory, retirement proximity may affect GP performance via two competing mechanisms: accumulated experience may enhance performance, while human capital depreciation, reflecting reduced effort or shifting focus, may offset these gains. We test the net effect of these mechanisms on selected utilization-based indicators of GP performance. Using administrative data from a large Italian Local Health Authority, we analyze type II diabetes patients continuously enrolled with GPs from 2018 to 2023. Exploiting variation in retirement proximity across GPs, we apply a staggered difference-in-differences approach to examine treatment effect dynamics. We find no evidence of systematic deterioration in potentially inappropriate use, hospitalizations, emergency department access, or diabetes-related avoidable hospitalizations. However, patients whose GPs are closer to retirement are less likely to receive at least two HbA1c tests per year, whereas their likelihood of receiving at least one HbA1c test remains unchanged. These findings suggest that, as GPs approach the retirement age of 70, utilization-based performance indicators are not systematically affected, although support for adherence to evolving chronic care monitoring standards may be needed.