Do general practitioners in low medical density areas have different practices ?

In France, we observe currently an increasing number of deprived areas in terms of health care where General Practitioners (GPs) are not numerous enough to cover the needs of the population.

First, this paper documents correlations between medical density and some activities or prescription practice variables. Medical density is measured thanks to a Local Potential Accessibility indicator (LPA) and by splitting GPs into two groups: (i) those working in Priority Intervention Zones (PIZs), these areas being eligible to programs that are designed to retain GPs or to favor their setting up, and (ii) the others. We use a nationally representative panel of French Gps matched with administrative data.

We find that GPs practicing in deprived areas have a higher number of patients. They do not react to higher demand by increasing their labor supply at the intensive margin, i.e., in increasing their working hours: they provide with shorter medical consultations. Some prescriptions seem to be correlated with medical density. By instance, GPs from deprived areas tend to prescribe more opioids and less paramedical care. Lastly, we find that preventive care is less frequent in PIZs .

The paper also examines the PIZs selection procedure. This selection is partly based on a threshold of the LPA indicator; it is also based on the Regional Health Authorities (ARS)’ ability to select municipalities. Using variables that have been previously found to be correlated with medical density, we compare this selection with a counterfactual one that would have been entirely centralized and based solely on the LPA indicator. The actual selection seems to reflect more the heterogeneity of physicians' practices than the counterfactual selection.