Abstract
Background: Diabetes is a frequent chronic condition, which can lead to costly complications if not managed well in the primary care setting. Potentially avoidable hospitalizations (PAH) are considered as an indirect measure of primary care. However, the association between primary care use and PAH in diabetic patients has not been investigated in France.Methods: We investigate the association between primary care indicators and PAH at an individual level among persons with diabetes in a population-based cohort study on the French national health insurance database (EGB sample). PAH occurrence in 2013 was modeled as a function of primary care use and access, health status and socio-economic indicators over the exposure period 2011-12 using a cause-specific hazards model with death as a competing event.Results: We included 25 293 diabetics in our cohort, among which 385 (1.5%) experienced at least 1 PAH in 2013. After adjustment on health status indicators, primary care use had a protective effect against PAH. Diabetic patients who had seen a general practitioner (GP) 10-14 times had a reduced hazard of PAH compared to less frequent encounters (HR=0.49, P<0.001). The effect size decreased when the number of encounters increased, suggesting a remaining confounding effect of health status.Conclusions: For the first time in France, this study shows a protective effect of the number of GP encounters against PAH at an individual level and highlights the importance of a frequent monitoring of diabetic patients in the primary care setting to prevent PAH occurrence.