RESUMEN
OBJECTIVE: To know the trend of nosocomial infections rates at a third level hospital since the beginning of the infection control program until the present and to evaluate the impact in mortality and length of hospital stay. MATERIAL AND METHODS: A study descriptive, retrospective and retrolective was done in a reference hospital providing tertiary care in Mexico City. Recorded data included rates, type of nosocomial infections and distribution of nosocomial infections from the database of Hospital Epidemiology Division between 1991 to 1996. In every patient the degree of severity disease was evaluated according to the McCabe-Jackson's scale, we also collected age and the length of stay. Statistical analysis was done with chi 2 test for tendencies of different evaluated parameters. This study was divided in three periods: before remodel (1991-1993), to remodel (1994-1995) and after remodel (1996). Compared the before remodel period against after remodel period. RESULTS: In the study period, mean nosocomial infections ratio was of 8.6 by 100 discharges, showing a decrease of 20% (p < 0.01). Hospital area with the highest infections rate was ICU (26.9 by 100 discharge), followed by general ward of hospitalization with shared rooms (9.47) and private rooms (7.5). Urinary tract infections was the most frequent (26.6%), followed by surgical wound infection (24.4%), pneumonia (12.1%) and primary bacteremia (9.5%). Rates of urinary tract infections and primary bacteremias had decreased significantly (p < 0.05 and p < 0.00001 respectively), while surgical wound infections and pneumonias increased (p < 0.005 and p < 0.00001 respectively). Associated mortality diminished in 36% (p < 0.00001). The length of stay diminished 42.8%. There was not differences in the severity of disease along this period. CONCLUSIONS: Since the establishment of the nosocomial infections surveillance and control program at the INNSZ on 1985, nosocomial infections rate (56%) and the associated mortality (36%) have diminished. These changes are consequence of the establishment of a nosocomial infections control program and the decrease of length of stay and does not seems to be related to the age or to the severity of the disease of the hospitalized patients.