To the Editors: Wang et al1 provide retrospective data regarding highly active antiretroviral therapy coverage, HIV incidence, and behavioral, clinical, and quality-of-life risk factors for HIV transmission among serodiscordant couples from Henan Province, China.1 The study found that HIV seroconversion rates were generally low over time. HIV seroconversion was associated with not always using condoms, having sexual activity 4 or more times per month, not switching antiretroviral treatment regimens, and having a high quality-of-life score on the psychological domain. Use of antiretroviral therapy by the seropositive member of the couple was not found to be protective against HIV seroconversion. Although these findings are reaffirming of how psychological and sexual behaviours are implicated in seroconversion, they do not add meaningfully to the discussion regarding the role of antiretroviral treatment as prevention. We are surprised that this is not highlighted in the accompanying editorial.2 Quite simply, and as acknowledged by Wang et al,1 HIV transmission is directly associated with the level of virus present in the HIV-positive partner,3 and they are unable to assess this relationship in this analysis, as they do not have data on longitudinal plasma HIV-1 virel load. As they note in the discussion, they do not even have data on adherence levels among those on therapy. Furthermore, the relationship described in this study between seroconversion and not switching antiretroviral treatment suggests that virological failure may indeed be playing an important role as a driver of HIV transmission. Complete monitoring of relevant variables, most critically plasma viral load, is essential to address the relationship between antiretroviral therapy coverage and HIV transmission.4-8 Julio Montaner, MD* Robert Hogg, PhD† *Department of Medicine, University of British Columbia, Vancouver, BC †Faculty of Health Sciences, Simon Fraser University, Burnaby, BC