Control-group selection importance in studies of antimicrobial resistance: examples applied to Pseudomonas aeruginosa, Enterococci, and Escherichia coli.

Abstract

We aimed to illustrate the importance of control-group selection on the results of risk factor analysis for (1) imipenem-resistant Pseudomonas aeruginosa, (2) vancomycin-resistant enterococci (VRE), and (3) ampicillin-sulbactam-resistant Escherichia coli. Case patients were compared with 2 different control groups: patients with the susceptible form of the organism (type 1), and control patients among whom the case patients arose during the same period as the case patients (type 2). Comparison of case patients who had imipenem-resistant P. aeruginosa with type-1 control patients identified use of imipenem (odds ratio [OR], 27.1) and quinolones (OR, 3.25) as a risk factor for selection of antimicrobial resistance, and comparison of the same case patients with type-2 control patients identified imipenem (OR, 6.34). When case patients with VRE were compared with type-1 and with type-2 control patients, use of vancomycin was identified as a risk factor (OR, 4.38 and 2.77, respectively). Comparison of case patients who had ampicillin-sulbactam-resistant E. coli compared with type-1 control patients identified ampicillin-sulbactam (OR, 2.71) and quinolones (OR, 2.72), and comparison with type-2 control patients identified ampicillin-sulbactam (OR, 1.68). The selection of control patients from the potentially suboptimal control type 1 can falsely identify certain antibiotics and overestimate the OR of the resistance-defining antibiotic.