Cohorting to prevent acquisition of multidrug-resistant bacteria: An interrupted time series study.

ABSTRACT

BACKGROUND: Grouping patients who acquired resistant microorganisms within a single area (cohorting) has been used to prevent cross-transmission. We aimed to assess cohorting effectiveness in the absence of an outbreak. METHODS: An interrupted time series study was performed in a general hospital considering patients admitted to wards. In the first year, patients who acquired multidrug-resistant (MDR) bacteria were isolated without physical transfer. In the second year, cohorting was implemented, and patients with mixed MDR bacteria were transferred to individual rooms in a specific isolation unit. Cultures were requested upon clinician orders, and surveillance or routine cultures were not performed. The effect of cohorting on the incidence density of MDR bacteria acquisition was assessed using segmented regression analysis. RESULTS: In the first and second years, 2.0 and 2.8 cases per 1,000 patient-days acquired MDR bacteria. The length of hospitalization and mortality rate were similar between phases. There was a linear increase of the monthly incidence densities of MDR bacteria acquisition in the first year (ß1 0.11; 95% confidence interval [CI] -0.02 to 0.24), though without an immediate impact of cohorting (ß2 -1.32; 95% CI -3.81 to 1.16) or a change in the temporal trend (ß3 0.04; 95% CI -0.14 to 0.23) from the first to second phase. CONCLUSION: Cohorting may not reduce the incidence density of MDR bacteria acquisition in the absence of an outbreak.