The true impact of the French pay-for-performance program on physicians’ benzodiazepines prescription behavior

Audrey Michel-Lepage, Bruno Ventelou

The French pay-for-performance (P4P) contract CAPI implemented by the national health insurance included a target-goal which aims at reducing benzodiazepines prescriptions. In this investigation, we would like to assess whether: (1) the general practitioners (GPs) having signed P4P contract obtain better results regarding the target-goal than non-signatories; (2) (part of) this progression is due to the CAPI contract itself (tentative measurement of a "causal effect"); (3) (part of) the money spent on this P4P incentive can be self-financed with the amount of pharmaceuticals saved. We matched cross-sectional and longitudinal data including 4622 French GPs from June 2011 to December 2012. A treatment effect model using instrumental variables was performed to take into account potential self-selection issue in signing. After having identified the NET impact of the P4P, we calculate the cost of an avoided benzodiazepines treatment. In our study, GPs who have signed the CAPI contract (36 % of the sample) are more numerous in achieving benzodiazepines target goal than non-signatories: 90.7 vs. 85.5 %. After controlling for the self-selection bias, the propensity of GPs to achieve the benzodiazepines target is only 0.31 % higher for signatories than for their non-signing counterparts-estimate for June 2012, which yields a statistically significant gap. Our economic analysis demonstrates that the CAPI contract does not allow savings, but presents in 2012 a NET cost of 93.6<euro> per avoided benzodiazepines treatment (291<euro> in 2011). The P4P contract has a positive but modest impact on the achievement of GPs regarding benzodiazepines indicator.