A Framed Field Experiment on the Effect of Feedback on Antibiotics Provision in Pediatrics

Whether feedback is an effective means to improve the quality of health care is empirically not well understood. We analyze the causal effect of expert feedback on individual physicians' decisions on antibiotic therapies in pediatrics. We employ a novel randomized, controlled framed field experiment with pediatricians in tertiary pediatric care units in Germany (N=73). Pediatricians decide on the length of antibiotics therapy for a series of stylized routine cases. In the intervention group, pediatricians receive feed- back in form of the aggregate length of therapy recommended by directors of pediatric departments. Feedback significantly reduces the length of therapy, by about one day on average. We also find that the pediatricians' gender, clinical experience, personality traits, and risk attitudes significantly affect decisions on the length of therapy. The feedback effect is robust when controlling for individual characteristics