To the Editor.
—The article by Dr Nelson and colleagues1 is a useful contribution to our understanding of the dynamics of human immunodeficiency virus (HIV) infection in Thailand. Especially helpful is the link between HIV status and hypothesized social and behavioral factors. One very surprising finding is that condom use with prostitutes "was not shown to be protective of prevalent HIV infection." The authors offer a number of possible explanations for this result, but do not address what I view to be the most likely factors: weak operational measures and a misspecified causal model. At stake is more than an arcane statistical argument, since a failure to demonstrate the protective benefits of condom use could lead to the impression that promotion of condom use with prostitutes is a misguided policy in Thailand's campaign against HIV. In fact, Nelson et al conclude by implying that programs to combat sexually transmitted