Multimodal intervention to reduce acquisition of carbapenem-non-susceptible Gram-negative bacteria in intensive care units in the National Referral Hospital of Indonesia: An interrupted time series study.

Abstract

To evaluate a low-cost multimodal intervention on the acquisition of carbapenem-non-susceptible Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa by patients in low-resource intensive care units. We performed a quasi-experimental study in a referral hospital in Jakarta, Indonesia: pre-intervention phase 1 (2013-2014), intervention phase 2 (2014-2015) and post-intervention phase 3 (2015-2016). The intervention was hand hygiene promotion and environmental cleaning and disinfection combined with patient disinfection and cohorting. The primary outcome was acquisition of resistant bacteria per 100 patient-days at risk, which was assessed by active microbiological surveillance and analysed with a multilevel Poisson segmented regression model. In phase 1 (387 patients), the acquisition rate was 4.3/100 days for carbapenem-non-susceptible A. baumannii versus 1.1/100 days for both K. pneumoniae and P. aeruginosa. There was a significant step change from phase 1 to phase 3 (361 patients) in the acquisition of carbapenem-non-susceptible strains, the incidence rate ratio (IRR) was 0.343 (99%CI: 0.164-0.717). This significant change was mainly due to reduced acquisitions of resistant A. baumannii (IRR 0.4, 99%CI: 0.181-1.061). Negative confounding was observed. A multimodal intervention to prevent acquisition of resistant pathogens is feasible and may be effective in ICUs in lower-middle income countries.