Inspections of 272 facilities were performed between May 1992 and October 1994 to determine compliance with applicable Occupational Safety and Health Administration (OSHA) requirements for prevention of tuberculosis (TB) transmission.

Retrospective record review of two data sources: (1) OSHA's Computerized Integrated Management Information System and (2) an inspector-completed questionnaire on inspection results.

Inspections of five types of facilities: healthcare institutions, correctional facilities, homeless shelters, long-term–care facilities for the elderly, and others, including drug treatment centers that the Centers for Disease Control and Prevention (CDC) identified as having a higher than expected rate of TB.

The OSHA Compliance Memorandum, based on the 1990 CDC Guidelines, which outlined elements of a TB prevention program, was used in performing 272 inspections of facilities between May 1992 and October 1994. Elements of compliance were recorded and reviewed from the IMIS database and inspectors' questionnaires.

Regulated facilities were not fully compliant with OSHA guidance. Generally, healthcare facilities performed better than other facilities. Most facilities (79%) were compliant with administrative elements of a comprehensive TB control program, such as early identification of known or suspected infectious TB patients and skin testing of workers. Only 29% of inspected facilities were found to have acceptable respiratory protection programs for the prevention of occupational TB.

Facilities have not been fully compliant with the OSHA memorandum describing protection of workers from TB. Facility compliance was better with some traditionally recognized TB infection control elements, but was weaker in the area of respiratory protection programs. This may reflect a lack of familiarity with the latter type of hazard protection.

To study compliance with preventive strategies at a university hospital during an outbreak of nosocomial influenza A during the winter of 1988, and the rates of vaccination of healthcare workers and of nosocomial influenza following changes in vaccine practices after the outbreak.

Retrospective review of employee health, hospital epidemiology, hospital computing, and clinical microbiology records.

A university hospital.

Unvaccinated personnel with exposure within the previous 72 hours to an unisolated case of influenza were offered influenza vaccine and 14 days of amantadine hydrochloride prophylaxis. Personnel with exposure more than 72 hours before evaluation were offered vaccine. A mobile cart was introduced for vaccinating personnel after the 1988 outbreak.

An outbreak of influenza with 10 nosocomial cases occurred in 1988. Only 4% of exposed employees had been vaccinated previously and 23% of exposed, unvaccinated employees agreed to take vaccine, amantadine, or both. A mobile-cart vaccination program was instituted, and annual vaccination rates steadily increased from 26.3% in 1989 to 1990 to 38% in 1993 to 1994 (P<.0001). The relative frequency of documented cases of influenza in employees with symptoms of influenza decreased significantly during this period (P=.025), but nosocomial influenza rates among patients did not change significantly.

A mobile-cart influenza vaccination program was associated with a significant increase in compliance among healthcare workers, but a majority still remained unvaccinated. The rate of nosocomial influenza among patients was not reduced by the modest increase in the vaccination rate, but influenza rates remained acceptably low, perhaps due to respiratory isolation of patients and furlough of employees with influenza.

To assess the effect on the rate of ventilator-associated pneumonia (VAP) of decreasing the frequency of ventilator circuit changes from three times to once per week.

Prospective, randomized trial.

Medical intensive care unit (MICU), a 12-bed, critical-care internal medicine unit, and neurosciences intensive care unit (NICU), a 21-bed, predominantly adult neurosurgical unit, of an urban university hospital.

All 447 patients requiring mechanical ventilation during October 1992 through June 1993.

Patients were allocated randomly on the basis of permanent medical record numbers: those with odd numbers had circuits changed three times per week, those with even numbers once per week. Intensive-care-unit surveillance was conducted in accordance with definitions and methods of the National Nosocomial Infections Surveillance System.

In the MICU, the one-change-per-week group had a VAP rate of 7.3 per 1,000 ventilator days, versus 5.9 for the three-per-week group (P=.6). In the NICU, the one-change-per-week group had a rate of 12.2 per 1,000 ventilator days, versus 12.6 for the three-per-week group (P=.9). Considering patients in both units ventilated for no more than 7 days, the one-change-per-week group had a VAP rate of 5.9 per 1,000 ventilator days, versus 9.0 per 1,000 for the three-changes-per-week group (odds ratio [OR], 0.65; 95% confidence interval [CI95], 0.25 to 1.69). Including patients in the two units maintained on mechanical ventilation for more than 7 days, the one-change-perweek group had a VAP rate of 13.2 per 1,000 ventilator days, versus 9.6 per 1,000 for the three-changes-per-week group (OR, 1.37; CI95 0.71 to 2.65).

Decreasing the frequency of ventilator circuit changes from three times to once per week had no adverse effect on the overall rate of VAP. Less frequent ventilator circuit changes may decrease the incidence of VAP among patients ventilated for no more than 1 week. However, the incidence of VAP may be higher among patients with once weekly circuit changes ventilated for more than 1 week.