THE COST-EFFECTIVENESS OF PREVENTING TUBERCULOSIS IN PHYSICIANS USINGTUBERCULIN SKIN TESTING OR A HYPOTHETICAL VACCINE

Background: Tuberculin skin testing using the purified protein derivat ive is recommended as part of a tuberculosis control program for healt h care workers. However, compliance with skin testing programs has bee n poor and their cost-effectiveness is unknown. Methods: A Markov-base d decision analysis was performed to determine the cost-effectiveness of tuberculin skin testing over the entire lifetimes of physicians who are now in medical school. Assumptions were deliberately chosen to pr esent a conservative estimate of cost-effectiveness. Indirect costs we re not included. Results: Annual testing cost $29 000 per life-year sa ved and $39 000 per case of pulmonary tuberculosis prevented. In contr ast, particulate respirators have been shown to cost millions of dolla rs per case prevented. Skin testing every 6 months was cost-effective in a subpopulation at high risk of infection (greater than or equal to 1.8-fold). During their entire lifetimes, physicians now in medical s chool can expect to avert 137 cases of pulmonary tuberculosis, prevent 7 tuberculosis deaths, and save 182 life-years because of skin testin g programs. Improved compliance with annual skin testing and prophylac tic isoniazid could more than triple this benefit. If available, a mod erately effective vaccine would be even more cost-effective than tuber culin skin testing programs. Conclusions: Tuberculin skin testing is c ost-effective and should be an integral part of any tuberculosis contr ol program. Vaccination may one day be a feasible and cost-effective a lternative to skin testing programs.