VALIDITY OF A DECISION TREE FOR PREDICTING ACTIVE PULMONARY TUBERCULOSIS

The recent outbreaks of multidrug-resistant strains of M. tuberculosis in health care facilities has increased concern over its transmission in health care facilities. Isolation has been recommended for all pat ients suspected to have tuberculosis even though the feasibility and t he cost of this recommendation can be substantial. We have developed a classification tree using clinical and radiographic data from 277 iso lation episodes in patients admitted between August 1992 and March 199 4 who required isolation for suspicion of tuberculosis. The classifica tion tree was developed with a sensitivity and negative predictive val ue of 100% by binary recursive partitioning to predict those patients who are unlikely to require isolation. The predictor variables were up per zone disease on chest radiograph, a history of fever, weight loss, and CD4 count. The tree was validated in a separate cohort of 286 iso lation episodes between April f994 and December 1995. In this validati on cohort, no erroneous prediction was made of not isolating a patient with active pulmonary tuberculosis. The classification tree had a sen sitivity of 100% (95% confidence interval [CI]: 92.5 to 100%), a speci ficity of 48.1% (95% CI: 43.8 to 52.4%), and a negative predictive val ue of 100% (95% CI: 98.5 to 100%). We estimate that the use of the tre e could have reduced the number of patients requiring isolation by mor e than 40% without increasing the risk of cross infection.