In the study by Brown et al,1 the authors did not mention several limitations that might have an important impact on the interpretation of the results. First, the study only included men, who were mostly white (approximately 86%). Therefore, the results cannot be extrapolated to women and other ethnic groups. Second, the family history of diabetes was not ascertained in the study participants, and this could have confounded the magnitude of differences in prevalence and incidence of diabetes between the 2 study groups. Third, the reported prevalence and incidence rates of diabetes were likely to be overestimated because the diagnosis of diabetes was not confirmed by a repeated measurement of fasting plasma glucose.
Mikhail N, Cope D. Antiretroviral Therapy and the Prevalence and Incidence of Diabetes. Arch Intern Med. 2005;165(21):2536-2537. doi:10.1001/archinte.165.21.2536-c