Tuberculosis in Health Care Professionals: Assessing and Accepting the Risk
C. Nolan
DOI: https://doi.org/10.7326/0003-4819-120-11-199406010-00009
IF: 39.2
1994-06-01
Annals of Internal Medicine
Abstract:Tuberculosis has long been an accepted hazard of work in the health professions [1]. For decades, however, physicians of the era after World War II did not take the hazard seriously. I recall an attending physician who, during hospital rounds on a patient with pulmonary tuberculosis in the late 1960s, stated that he would rather have tuberculosis than a broken arm, so easy and painless had the curative treatment of tuberculosis become. That casual attitude changed abruptly in the early 1990s. Multidrug-resistant tuberculosis, the most ominous sequel to the now decade-long resurgence of tuberculosis in the United States, arrived on the scene, along with the confluence of tuberculosis, human immunodeficiency virus (HIV) infection, homelessness, and drug abuse [2, 3]. Multidrug-resistant tuberculosis is at best difficult to cure, and its treatment involves long-term, costly, and toxic drug therapy and, frequently, surgery [4]. At worst, in patients with HIV infection, it is a rapidly progressive, untreatable, and fatal infection [2, 3]. Many health care workers have probably been infected with the multidrug-resistant form during nosocomial outbreaks; some have died [5]. Because of concern over this new occupational risk, the U.S. Occupational Safety and Health Administration (OSHA) recently included tuberculosis among workplace hazards that can seriously harm or kill employees and mandated that health care institutions take the steps needed to eliminate employees' risk for tuberculosis. The article by Fagan and Poland [6] in this issue offers disquieting information on the current risk for tuberculosis in one group of health care professionals: medical students. Responding to the authors' survey, U.S. medical schools reported that the mean annual tuberculin skin-test conversion rate for medical students is between 1.3% and 2.2%. A conversion implies that tuberculous infection was recently acquired. Thus, if the data of Fagan and Poland are accurate and generalizable, U.S. medical students, with their annual risk exceeding 1%, are identified as having an excessive rate of new tuberculous infections [7]. Although most of the medical schools responding to the survey reported low annual conversion rates among students, it is particularly disturbing that 12 (16%) of 75 responding schools reported rates of 5% and higher. The most recent estimate of the annual risk for tuberculous infection in the U.S. general population (1986 data) was 8.6 per 100 000 persons [8]. Thus, medical students in those 12 schools may have an annual risk of more than 500 times that of the 1986 risk in the U.S. population. Fagan and Poland should be complimented for their effort to compile and report these data. Because Mycobacterium tuberculosis is reemerging as a nosocomial pathogen, ascertaining the level of risk in health care professionals is a prerequisite to a successful national control strategy. Fagan and Poland have scratched the surface of this endeavor. The difficulty with their study is that there is no way to know whether the protocols for administering and reading serial skin tests in the various schools were similar and whether the reported conversion rates were accurate and comparable. In fact, because of the changing concepts of the tuberculin test [9], it would not be surprising to learn that various techniques and criteria, all based to some extent on methods used in recently published studies, were being used to interpret individual skin test results and conversions. Recent innovations in tuberculin skin testing [9, 10] have been introduced in an attempt to improve the utility of this historic diagnostic test. It is now apparent that several factors other than the presence of tuberculous infection affect the delayed hypersensitivity response to intradermal injection of tuberculin. Among the most important are the host's immunologic status [11]; previous BCG immunization or exposure to environmental mycobacteria [12]; how the test is administered and how the results are interpreted [10]; and, probably, the commercial brand of tuberculin that is used [13]. Further, because the test has a specificity of less than 100%, its positive predictive value deteriorates sharply when it is applied to populations that have a low prevalence of tuberculous infection [10]. Among the innovations introduced in recent years to maximize the probability that a single tuberculin test will correctly classify a given patient with respect to the presence or absence of tuberculous infection and that serial tests will correctly identify a newly infected person are the following: the concept of a cutting point between a negative and a positive test result that varies inversely with the prevalence of tuberculous infection in the population of the tested patient [14]; the use of two-step testing to detect the booster effect [10]; and more stringent criteria for a skin-test conversion [10, 15]. Because tuberculin testing is now being used widely in medical institutions, drug treatment centers, homeless shelters, correctional facilities, and long-term care facilities, the officials of these institutions would do well to familiarize themselves with innovations in tuberculin testing and, if necessary, to seek expert consultation. Among the reasons that the method of skin testing is important is that one of the useful purposes of a skin testing program, beyond the identification of recently infected persons who need further medical evaluation, is to monitor the quality of the institution's tuberculous infection control program. Falsely high conversion rates because of failure to detect boosting or the interpretation of small increases in serial test results as conversions could lead to needless personal concern, time-consuming epidemiologic investigations, and costly and unnecessary corrective measures. The medical profession faces, however, issues more momentous than these methodologic matters as it grapples with the reemergence of M. tuberculosis as a nosocomial pathogen. Health care professionals may no longer reasonably view their risk for tuberculosis as less intimidating than a broken arm, as suggested by my former teacher. But is it any more reasonable, on the other hand, to accept the premise implied in the OSHA directive on tuberculosis that there should be no personal risk in the practice of medicine? Are we prepared to accept the ethical implications of that position? A successful institutional infection control program should consist of various components, including early identification and isolation of infectious patients, effective engineering controls, a respiratory protection program, and employee education and screening. Such multifaceted programs have successfully controlled institutional outbreaks of tuberculosis [16, 17]. But the total elimination of risk for exposure to tuberculosis in health care facilities could probably be achieved only by the routine use, in certain high-risk settings, of personal respiratory devices or masks with rigorous protective specifications. Many personsand I am onewho have interviewed and examined patients while wearing such masks report that the experience is at best awkward and difficult and that it is disconcerting and occasionally humiliating for patients. Members of a profession dedicated to altruism, a regard for others as a principle of action [18], may prefer to weigh the benefit of wearing a maskthe last degree of personal protectionagainst the riskobstructing patient rapportwith each patient visit rather than to use it routinely in predetermined types of patient encounters. In asserting this I recognize fully that a physician's personal choice about the use of a mask may conflict with his or her hospital's policy, brought about by the OSHA regulations on protection of employees from tuberculosis. A 5% annual risk for tuberculous infection in medical students is unacceptable, but so is the stigmatization of patients. Our profession, which recently debated the ethics of personal choices about treatment of patients with AIDS [19], now confronts similar issues as it engages its old and new nosocomial foe, tuberculosis.