Physicians’ Incentives, Patients’ Characteristics, and Quality of Care a Systematic Experimental Comparison of Fee-for-Service, Capitation, and Pay for Performance

How to effectively combine performance pay with the traditional payment methods capitation and fee-for-service to improve the quality of care is not well understood. To establish a systematic comparison of physicians’ behavior under these payment methods, we employ a parsimonious experimental framework with a symmetric mirror-design of incentives and a constant patient population with heterogeneous severities of illness. In a series of controlled experiments with physicians and students, we find, in line with theory, that performance pay significantly reduces non-optimal service provision and enhances the quality of care. The effect sizes depend, however, on patients’ severity of illness in that severely ill patients gain most from capitation-based performance pay while mildly ill patients do so under a fee-for-service-based system. Intermediately ill patients benefit from both payment systems in a similar way. Health policy implications are discussed, including a cost benefit analysis