[Prevalence and incidence of hospital infections at the University Polyclinic of Palermo].

Hospital infections (H.I.) were studied in the University Polyclinic of Palermo, a 1270 beds General Hospital. A prevalence study was carried out in four periods (December, February, June, October) on a sample of 590 patients from randomly chosen from medical and surgical departments. All patients present in the departments were investigated. Data were obtained from hospital records, using a form prepared according to the Italian Study Project of Hospital Infections (SIPIO), and through interview of the patients and/or of the medical staff. After excluding community infections, point prevalence was calculated; moreover, the day in which sign or symptoms of infections appeared was assessed. Cumulative incidence of H.I. and some risk factors associated were estimated using Kaplan-Meier method and Cox model for multivariate analysis.