Impact of a post-prescription audit and feedback antimicrobial stewardship intervention on inappropriate carbapenem prescribing: An interrupted time series analysis

Background Antimicrobial stewardship programs (ASPs) are essential to improve antibiotic prescribing. This study evaluated the impact of a post-prescription audit and feedback intervention on carbapenem prescribing appropriateness in a large university hospital. Methods This retrospective observational study utilized interrupted time series (ITS) analysis, employing an Autoregressive Integrated Moving Average (ARIMA) model, to assess carbapenem prescribing across three consecutive phases: a 12-month pre-intervention, a 6-month intervention, and a 6-month post-intervention follow-up. Carbapenem prescribing appropriateness was retrospectively evaluated using an in-house developed algorithm, based on international and national guidelines and institutional protocols. The intervention involved bedside consultations by infectious diseases specialists employing a post-prescription audit with face-to-face feedback. Results We evaluated 1825 carbapenem therapies, primarily prescribed for suspected/confirmed bloodstream infections (46%, 843/1825). Among these, 458 (25%) were deemed inappropriate, mainly due to unnecessarily broad-spectrum use (72%, 331/458). The ITS-ARIMA model showed an immediate 11% reduction in the rate of inappropriate prescriptions during the first month of intervention phase (P = 0.001), followed by a non-significant downward trend during the remaining intervention period. However, an immediate 14.9% increase in inappropriate prescriptions was observed at the onset of the post-intervention phase (P = 0.001), indicating a rebound effect after the withdrawal of the active stewardship intervention. Conclusions Implementing a post-prescription audit and face-to-face feedback intervention was associated with a short-term improvement in carbapenem prescribing appropriateness. We observed a reduction in the trend of inappropriateness, although this change was not statistically significant. Future studies should investigate strategies for implementing sustainable ASPs optimizing human resources and time investment.