I am awestruck to find myself in the midst of an argument with Harold Israel, who, with Arthur Myers, alerted the U.S. medical community of the occupational risk associated with caring for patients with tuberculosis. My attempts to reconstruct retrospectively the events in public health during the first part of the century cannot compare with the recollections of someone at the center of the action.

However, I disagree with Dr. Israel's implication that there is no real cause for concern. Mortality is only one measure of the effect of tuberculosis. At best, persons with drug-susceptible tuberculosis must receive a 6-month course of therapy—something more than a minor inconvenience. At worst, patients may have drug intolerance or resistant tuberculosis and may need courses of therapy extending from 12 to 24 months, with an uncertain outcome. I agree that the health care workers of the 1990s have less to fear from tuberculosis than their 1930s predecessors; however, I suspect that current workers compare risk not with that of 60 years ago but with that of other current occupations and diseases.

Finally, I agree with Dr. Israel that the risk for developing occupationally acquired diseases, including tuberculosis, necessarily goes with the territory of patient care and can never be reduced to zero. I am relieved by Dr. Nicholson's report that the Brompton Board of Governors finally determined that tuberculosis was indeed contagious and sought to control the problem with daily stout. The purported medical uses of stout, wine, and spirits are many and may themselves be worthy of a historical review.