Drs Rabeneck and Wray1 deserve to be congratulated for an excellent summary of the "state of the prognostic art" in human immunodeficiency virus and acquired immunodeficiency syndrome (AIDS). They will be pleased to know that the system by Justice et al2 for predicting mortality in patients with AIDS has been independently validated3 in a 1987 sample from the Boston (Mass) Standardized Metropolitan Statistical Area. Stone and her colleagues3 used a geographic sample consisting of all intravenous drug—using men with AIDS, all women with AIDS, and a random sample of nonintravenous drug—using men with AIDS. The study demonstrated mortality rates (P<.01) as shown in the Table.
Thus, the "Justice system" has been shown to be valid and useful in a generalized setting (22 different hospitals in Boston) by independent investigators. While my colleagues and I are continuing to work on improving the prognostic power of this