describe efforts to improve tuberculosis control practices and document the consequences.

Published case reports often involve organizations with particularly interesting situations or problems such as an outbreak of multidrug-resistant tuberculosis in a hospital or a case of disease in an atypical setting. Stable, nonoutbreak situations are less interesting to researchers, government agencies, and journals. Moreover, from the committee’s personal experience and conversations with CDC staff, not all outbreaks are reported, and not all those reported are investigated. Of those investigated, not all result in published reports. Thus, published reports on outbreaks therefore cannot be treated as representative of all outbreaks, much less all employers.

Several articles have summarized information presented in case reports, and the committee’s reading of the individual reports is consistent with these summaries. Menzies and colleagues (1995), for example, reviewed 13 incidents (all before 1993) of single or multiple cases of occupationally acquired tuberculosis infection or disease in hospitals. The reports on these incidents usually associated outbreaks with delayed diagnosis of hospitalized patients with infectious tuberculosis, inadequate therapy, or unrecognized drug resistance. The reports also frequently cited inadequate ventilation and isolation practices. In addition, the review authors noted other reports describing low levels of health care worker compliance with treatment for tuberculosis infection, which reduces the benefits of tuberculin skin testing programs.

Dooley and Tapper (1997) summarize reports on 21 inpatient facilities with episodes of tuberculosis transmission to patients or workers in inpatient facilities all before 1993. Most (17) involved patients with undiagnosed, untreated infectious tuberculosis, and most of the sustained outbreaks involved people with human immunodeficiency virus (HIV) infection and suppressed immune systems who were exposed to patients with undiagnosed infectious tuberculosis or unrecognized drug-resistant disease. Some reports cited isolation practices that departed from recommendations. Such departures included the ending the isolation before a response to treatment was documented, failure to close doors to isolation rooms, and failure to keep patients confined to isolation rooms. Many reports also cited inadequate engineering controls including use of recirculated air with no or few air changes and isolation rooms with positive or essentially neutral pressure. In at least two episodes, transmission occurred despite frequent air changes. Most reports did not describe respiratory protection policies or practices, although three described transmission to workers who had worn surgical masks during contact with a patient with tuberculosis.

A review by Garrett and colleagues (1999) includes 23 pre-1993 episodes of transmission of M. tuberculosis to patients or workers, most of which are covered by Tapper and Dooley. The review also covers five



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement