Testing for tuberculosis in nursing homes – Adapted from the American Review of Respiratory Diseases, December 1993 – Tips from Other Journals
Residents of nursing homes are at higher risk for developing tuberculosis than are community-dwelling elderly persons. Reasons for the increased risk include the prolonged close contact that occurs in long-term care facilities, the large group of susceptible hosts, the reservoir of old infection in nursing home residents, delay in making the diagnosis, and staff who may come from countries with a high prevalence of the disease. Rosenberg and colleagues studied the problem of testing the residents and staff of nursing homes to identify those who need chemoprophylaxis and to reduce low-yield searches for infection.
Tuberculin testing was done on 272 residents and 218 staff members of a nursing home in Canada. Each subject was injected with 5 tuberculin units (0.1 mL) of purified protein derivative (PPD) following a standardized protocol. Reactions were recorded 48 hours after injection; the record specifically indicated the size of induration in millimeters. All subjects with induration of less than 10 mm were retested after one week. Those whose induration increased to 10 mm or more on the second test were called “boosters.” Subjects with induration of at least 10 mm on the first test were classified as “initial reactors.”
In addition, residents and staff who were able to answer a questionnaire provided information about age, sex, country of origin and year of arrival in Canada. Data concerning previous history of tuberculosis, positive PPD reactions and vaccination with bacille Calmette-Guerin (BCG) was also obtained. Charts of residents who were unable to answer the questionnaire were reviewed for this information.
More than half (56 percent) of the study subjects were born outside of Canada. BCG vaccination had been given to 34 percent of all staff members and no residents. BCG vaccination was the parameter most closely correlated with a positive initial reaction. Initial reactors also tended to be foreign-born and staff members. Of the subjects who received two injections, two main groups were identified. One group had a reaction of zero mm on both the first test and the second test (92 percent). The second group had initial reaction sizes of 1 mm to 9 mm. Of these subjects, 80 percent had second reactions of zero mm, 6 percent had boosted reactions after the second test and 13 percent had reactions that remained at 1 mm to 9 mm.
A history of BCG vaccination was the only variable strongly correlated with a positive second reaction. This correlation was determined in both initial and boosted reactors to be independent of country of origin. Further statistical analysis revealed that BCG vaccination and number of years in Canada were the only significant variables associated with positive initial tuberculin reaction, and that 72 percent of the positive tuberculin tests among the BCG-vaccinated staff of the nursing home could be attributed to the vaccine itself.
The authors disagree with the position of the American Thoracic Society and the Centers for Disease Control and Prevention that BCG vaccination does not necessarily cause a positive reaction and that if it does cause a reaction, the effect is small and not sustained. They conclude that in closed long-term care communities, positive booster reactions may not be related to age, as has previously been thought. They also question the usefulness of tuberculin testing in a group (such as staff workers) who have been vaccinated with BCG, since the majority of these will be reactive. Finally, the authors recommend that chemoprophylaxis not be undertaken in BCG-vaccinated reactors unless there has been a known exposure to tuberculosis. Further studies on the cost-effectiveness of screening in a long-term care facility are needed.
COPYRIGHT 1994 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group