Another report on the experience of a Chicago hospital suggests that personal respirators do not compensate for inadequate engineering controls. Kenyon and colleagues (1997) reported an outbreak of multidrug-resistant tuberculosis in a facility that had provided and fit-tested workers with high-efficiency particulate respirators but that had no isolation rooms that met CDC criteria. Delays in recognizing and treating infectious patients also contributed to the outbreak. Three of the 11 previously skin test negative workers who converted their tuberculin skin test (including one ward secretary with no patient care responsibilities) had no contact with the source case patients. The authors conclude that a respiratory protection program alone cannot protect all workers. In the absence of appropriate isolation rooms, air that escapes from rooms housing infectious patients can infect those outside the room.
Also pertinent is a survey of 52 former house staff who served in the tuberculosis facility associated with the University of Virginia between 1979 and 1987 (Jernigan et al., 1994). The 52 individuals had experienced a total of 70 6-week rotations in the facility, which had negative-pressure isolation rooms and ultraviolet germicidal radiation. A simple surgical mask (no fit-testing program) was the only form of respiratory protection used. Those surveyed reported no skin test conversions associated with the rotation.
In another survey, the Society for Healthcare Epidemiology (SHEA) surveyed member hospitals for tuberculosis control practices from 1989 to 1992 and evaluated responses from 210 hospitals (Fridkin et al., 1995a,b). Four control practices described in the 1990 CDC guidelines were evaluated: (1) placing known or suspected tuberculosis patients into a single room (or a room shared by two such patients); (2) negative-pressure ventilation of the isolation room, (3) at least 6 or greater air changes per hour in isolation rooms; and (4) air exhaust directly to the outside. For the sub-group of hospitals that admitted at least six or more tuberculosis patients per year and met all four criteria, the tuberculin skin test conversion rate among health care workers was significantly less for those hospitals compared to those for others (0.6 versus 1.89 percent; p = 0.02). Conversion rates were not associated with type of respiratory protection. The survey did not cover fit testing.
In all of the reports cited above, the implementation of tuberculosis control measures was associated with low levels of tuberculosis transmission among health care workers in hospitals where tuberculosis was prevalent. These data, although imperfect and limited, support CDC’s emphasis on administrative controls and suggest the lesser contribution of a respiratory protection program in the hierarchy of tuberculosis infection control. Admittedly, the data lack sufficient power to support firm conclusions. Such conclusions would require well designed, prospective controlled studies to investigate specifically the independent contribution