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.

Five cases of nosocomially acquired herpetic whitlow are presented; three cases developed following a common exposure. All infections were acquired despite the use of surgical gloves as recommended by existing guidelines on control of nosocomial infection. In two cases the mistaken diagnosis of pyogenic paronychia led to surgical intervention, with excessive morbidity and disability. Among the five cases, morbidity ranged from several days of unemployment to permanent sequelae necessitating a career change. Revised guidelines for infection control should include the use of gloves on both hands whenever patient care involves contact with a debilitated patient with herpes simplex labialis.

Between May 16 and July 1,1973, four definite cases and one possible case of clinical salmonellosis occurred in a 175-bed community hospital; there were no deaths. Three of the four patients with definite salmonellosis had had cholecystectomies done by the same general surgeon (A); the fourth was an intensive care unit nurse who cared for one of the ill patients during the diarrheal phase of illness before salmonellosis was diagnosed. Epidemiologic investigation implicated the plastic tubing of an intermittent-suction machine located in the recovery room as the environmental reservoir of the organism, and having a nasogastric tube in place postoperatively was the critical host factor related to illness. The salmonella organisms isolated from the suction machine tubing were identical in serologic reaction, biochemical test results, and bacteriophage susceptibility pattern to those recovered from the four patients with confirmed salmonellosis. Antimicrobial susceptibility patterns were similar but not identical.

This article details the appropriate microbiologic support that is critical to the successful investigation of nosocomial infection problems. The infection control team must have ready access to microbiologic data, and the laboratory should retain epidemiologically relevant bacterial isolates. Investigation of epidemics is facilitated by precise identification of bacteria and careful antibiotic susceptibility testing. In some situations, biotyping, serotyping, phage typing, bacteriocin typing, and other specialized techniques may be required. Plasmid analysis may be useful in the investigation of nosocomial infection problems caused by antibiotic-resistant bacteria.