Epidemiologic characteristics of health care–associated outbreaks and lessons learned from multiple outbreak investigations with a focus on the usefulness of routine molecular analysis

Abstract

Single outbreaks have often been reported in health care settings, but the frequency of outbreaks at a hospital over time has not been described. We examined epidemiologic features of all health care-associated outbreak investigations at an academic hospital during a 5-year period. Health care-associated outbreak investigations at an academic hospital (2012-2016) were retrospectively reviewed through data on comprehensive hospital-wide surveillance and pulsed-field gel electrophoresis (PFGE) analysis. Fifty-one health care-associated outbreaks (annual range, 8-15), including 26 (51%) outbreaks in intensive care units (ICUs), and 263 infected-colonized patients involved in these outbreaks were identified. The frequency of pathogens varied by affected location, specifically multidrug-resistant organisms (20/26 outbreaks, 77% in ICUs vs 2/25 outbreaks, 8% in non-ICUs; P < .0001) and gastroenteritis because of Clostridium difficile, norovirus, or adenovirus (1/26 outbreaks, 4% in ICUs vs 17/25 outbreaks, 68% in non-ICUs; P < .0001). Outbreaks occurred in approximately one-third of all units (37%) with some repeated instances of the same pathogens. Of 16 outbreaks caused by a bacterial pathogen evaluated by PFGE, 12 (75%) included some indistinguishable strains, suggesting person-to-person transmission or a common source. This study demonstrated epidemiologic characteristics of multiple outbreaks between ICUs and non-ICUs and the value of molecular typing in understanding the epidemiology of health care-associated outbreaks.