Abstract
Reports on dengue outbreaks at hospitals are extremely rare. Here the authors analyze a dengue outbreak at the Teaching Hospital-Kandy (THK), Sri Lanka. Our hypothesis was that the present outbreak of dengue was due to nosocomial infections. Our objectives were to illustrate epidemiological evidence for nosocomial dengue infections among THK workers and comparison of dengue incidence of hospital workers of wards that treat dengue patients with workers of other wards, to ascertain whether most nosocomial dengue incidences occur closer to where dengue patients are treated and vector larvae were detected, and to draw the attention of the medical community to the significance of hospital outbreaks, making suggestions on how to improve dengue preventive work at the THK. We calculated weekly dengue incidences for the hospital workers and for the surrounding Kandy district population, plotted epicurves, and compared them. We also compared these with the temporal changes of numbers of patients who were admitted for other illnesses and then diagnosed with dengue and the numbers of containers with vector mosquito larvae found on hospital premises. Dengue incidence of the hospital workers for the 24-week study period (2388 per 100000 population) was significantly high when compared to incidence of the district (151 per 100000 population). Peaks of dengue incidence in hospital workers, the numbers of patients hospitalized for other illnesses contracting dengue, and numbers of containers with vector larvae occurred in the same week. The peak dengue incidence of the Kandy district happened six weeks later. There was no evidence to indicate blood contact causing dengue among hospital workers. The outbreak was controlled while dengue was rising in the district. This evidence indicates a probable nosocomial dengue outbreak. This outbreak adversely affected hospital workers, patients, and the community. We propose some measures to prevent such outbreaks.