A prospective three-month study of the hospital costs associated with nosocomial infections was done in a 390-bed, 30-bassinet community-teaching hospital early in 1978. All hospital charges for diagnostic and therapeutic services required by nosocomial infections, and added length of stay due to the infections, were recorded. During the study period 123 infections occurred in 104 patients (a 2.7% incidence, considering the 4,485 patients discharged during this time). The average length of stay was prolonged 3.0 days for all nosocomial infections; this accounted for about half of the $636 average hospital charges for each nosocomial infection. The 65 nosocomial urinary tract infections prolonged the length of stay an average of 0.6 days and the total hospital charges were $146 for each infection, leading us to believe that the proportion of nosocomial infections at any given site may greatly alter the “average” cost of nosocomial infections. Our data from a community hospital show a shorter prolongation of length of stay and lower hospital costs when compared with the few other studies in the literature.

Data available for the monitoring of nosocomial infections, including mandatory state reports of communicable diseases, Morbidity and Mortality Weekly Report, National Nosocomial Infections Study reports, and ad hoc epidemiologic investigations are discussed in this article, and their usefulness, validity, limitations, and representativeness are compared. Problems with interpretation of hospital infection “rates” are described, and solutions are proposed.