A Prospective Quasi-Experimental Study of Multifaceted Interventions Including Computerized Drug Utilization Evaluation to Improve an Antibiotic Stewardship Program.

Objective: This study aimed to determine and evaluate interventions that improve an antibiotic stewardship program (ASP) in reduction of the amount of antibiotic use and errors in drug utilization evaluation (DUE) form completion.

Methods: A prospective quasi-experimental study was conducted to evaluate the effectiveness of the multifaceted interventions (preauthorization with computerized DUE, self-supervised checklists, and short active reminders of the antibiotic guidelines recommendations) incorporating an ASP among hospitalized internal medicine patients in a Thai tertiary care center. The preintervention period was from January to December 2018, and the postintervention period was from February 2019 to January 2020.

Results: A total of 9188 and 9787 patient data were analyzed during the preintervention and postintervention periods, respectively. The overall median antibiotic use (defined daily dose/1000 bed-days) was significantly decreased in the postintervention period compared with the preintervention period (867.68 versus 732.33; P < 0.001). The reduction in antibiotic use was predominantly observed among broad-spectrum antibiotics, including cefoperazone/sulbactam (15.35 versus 9.75; P = 0.04), cefepime (13.09 versus 6.37; P = 0.003), ciprofloxacin (32.99 versus 17.83; P < 0.001), piperacillin/tazobactam (99.23 versus 91.93; P = 0.03), meropenem (2.01 versus 0.99; P < 0.001), imipenem/cilastatin (9.53 versus 4.59; P = 0.04), and colistin (74.70 versus 22.34; P < 0.0001). The overall rate of errors in DUE form completion was significantly decreased after the intervention (40% versus 28%; P < 0.001). The incidence of carbapenem-resistant Enterobacteriaceae infections/colonization was significantly lower in the postintervention period (0.27 versus 0.12/1000 bed-days; P = 0.02).

Conclusions: The study multifaceted interventions incorporating ASP were associated with significant reduction in the amount of broad-spectrum antibiotic use, DUE form completion errors, and incidence of carbapenem-resistant Enterobacteriaceae infection/colonization.