Abstract
Nosocomial outbreaks of gastroenteritis are a major burden on hospital inpatient services, costing an estimated £115 million annually to the English National Health Service. We actively followed-up 171 inpatient units from four major acute hospitals and 11 community hospitals in South-west England for one year. Outbreaks of gastroenteritis were ascertained through an active surveillance network using standard clinical definitions. Survival analysis Cox regression models using an outbreak of gastroenteritis as the endpoint were fitted to identify institutional and operational attributes related to increased outbreak rates at the level of the care unit. Greater number of beds in unit [hazard ratio (HR) 1.22 (per 10 additional beds), 95% confidence intervals (CI) 0.96-1.55] was associated with increased hazard, as were geriatric (HR 2.6, 95%CI 1.6-4.3) and general medical (HR 1.7, 95%CI 1.1-2.6) care units. The average length of stay on a unit was inversely associated with outbreak incidence [HR=0.89 (per additional week of stay), 95%CI 0.80-0.99]. Larger care units and those with higher throughput have increased rates of gastroenteritis outbreaks. These results should guide infection control policy and support the design of hospitals with smaller care units.
Original language | English |
---|---|
Pages (from-to) | 135-143 |
Number of pages | 9 |
Journal | Journal of Hospital Infection |
Volume | 60 |
Issue number | 2 |
DOIs | |
Publication status | Published - Jun 2005 |
Keywords
- Cox regression
- Gastroenteritis
- Hospital design
- Norovirus
- Nosocomial
- Regression analysis