Implementation of an evidence-based guideline to reduce duration of intravenous antibiotic therapy and length of stay for patients hospitalized with community-acquired pneumonia: A randomized controlled trial

PURPOSE: Patients with pneumonia often remain hospitalized after they are stable clinically, and the duration of intravenous antibiotic therapy is a rate-limiting step for discharge. The purpose of this study was to determine whether implementation of an evidence-based guideline would reduce the duration of intravenous antibiotic therapy and length of stay for patients hospitalized with pneumonia. METHODS: In a seven-site, cluster randomized clinical trial, we enrolled 325 control and 283 intervention patients who were admitted by one of 116 physician groups. Within site, physician groups were assigned randomly to receive a practice guideline alone (control arm) or a practice guideline that was implemented using a multifaceted strategy (intervention arm). The effectiveness of guideline implementation was measured by the duration of intravenous antibiotic therapy and length of stay; differences in the rates of discontinuation and hospital discharge were assessed with proportional hazards models. Medical outcomes were assessed at 30 days. RESULTS: Intravenous antibiotic therapy was discontinued somewhat more quickly in the intervention group (hazard ratio [HR] = 1.23; 95% confidence interval [Cl]: 1.00 to 1.52; P = 0.06) than in the control group. Intervention patients were discharged more quickly, but the difference was not statistically significant (FIR = 1.16; 95% Cl: 0.97 to 1.38; P = 0.11). Fewer intervention (55% [157/283]) than control (63% [206/325]) patients had medical complications during the index hospitalization (P = 0.04), with no differences in other medical outcomes, including mortality, rehospitalization, and return to usual activities, between treatment arms. CONCLUSION: The multifaceted guideline implementation strategy resulted in a slight reduction in the duration of intravenous antibiotic therapy and a nonsignificant reduction in length of stay, without affecting patient outcomes. (C) 2003 by Excerpta Medica Inc.