To the Editor.
—Both Rogers et al1 and von Reyn et al2 in their investigations of human immunodeficiency virus (HIV) transmission by infected physicians make implausible statistical claims about the maximum likely true rate of transmission.Rogers et al, based on the absence of HIV transmission in 369 person-hours of surgical exposure, regard it as unlikely that transmission occurs more often than once per 1000 person-hours of such exposure. The specific power to detect more than once per 1000 person-hours is given as 91% in the text, and the general claim that such a rate is unlikely is repeated in the abstract.On its face, it is difficult to see how an event occurring once in 1000 hours could have even a 50% probability of occurring in 369 hours, let alone a 91% probability; yet, that is the implication of the authors' assertion.Assuming the 369 person-hours to
Gracely EJ. Risk Calculations for HIV Transmission From Infected Health Care Workers. JAMA. 1993;270(13):1543. doi:10.1001/jama.1993.03510130049016
Customize your JAMA Network experience by selecting one or more topics from the list below.