To the Editor.
—Von Reyn et al1 have underestimated the upper limit of the 95% CI (0 to 4.37/100 000 patients). The conclusions of von Reyn et al minimize the risk. Nevertheless, the finding of 0% prevalence among patients recalled and tested in their article and in the companion articles2,3 is reassuring.Based an the binomial model, when testing a sample representing 50.7% of the study population (1174 tested of 2317 potentially exposed patients), the chance that there is one positive individual who is not included in the sample is high (50.2%). This probability is precisely calculated by the following: (1-1/n)0.507n, where n is the population size (2317).If one HIV-infected patient was missed, then the sero-prevalence in this population would be 4.3 per 10 000 patients (1/2317), or 10-fold higher than the 95% upper CI of von Reyn et al. At the critical 0.05 confidence level,
John RKS, Yan P. Risk Calculations for HIV Transmission From Infected Health Care Workers. JAMA. 1993;270(13):1544. doi:10.1001/jama.1993.03510130049018
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: