Are men the losers of the antiretroviral treatment scale-up?

In this issue of AIDS, Hawkins and colleagues [1] report sex differences in the response to antiretroviral treatment (ART) in a large cohort of HIV-infected adults in Dar es Salaam, Tanzania. After adjusting for baseline CD4 cell count, age, clinical stage and other baseline factors, men had 19% increased risk of death and 43% increased risk of immunologic nonresponse compared to women. The study is important for two reasons. Firstly, previous reports from resource-limited settings have been inconsistent regarding the prognostic role of sex. A number of studies from sub-Saharan Africa have found a nonsignificant increased mortality in men receiving ART compared to women, but these studies were usually small and underpowered to detect any but large survival differences [2–4]. Only a few studies have found a statistically significant increased mortality in African men [5,6], the most important of which was a large study of more than 10 000 HIV-infected adults from Côte d'Ivoire, South Africa and Malawi recently published in The Lancet[7]. In this study women had 32% reduced risk of death in the first year of ART, after adjustment for baseline CD4 cell count, age, clinical stage and bodyweight, in line with the findings reported by Hawkins and colleagues. Secondly, over the past three decades large programs launched by the WHO and the United Nations have specifically targeted women's health, such as The Safe Motherhood Initiative and Millennium Development Goal 5: Improve Maternal Health. At the advent of these programs there was good reason to focus on women's health, and women still face a number of health problems that require targeted responses, such as pregnancy complications, gender-based violence and female genital mutilation. But the study by Hawkins and colleagues raises an important question: Are health services for men sufficient in the era of HIV/AIDS? HIV is mainly a disease of the reproductive age; most patients who enroll in HIV clinics in Africa are between 20 and 45 years old. Women in this age span are often accustomed to attend health services for antenatal care and other pregnancy-related issues, and HIV testing is usually an integrated part of these services. In general, therefore, women have a more timely access to HIV care and ART. Men of reproductive age, on the contrary, do not routinely access any health services, and their health-seeking behavior might be less favorable. Later detection of HIV infection in men, as reflected by their more advanced immunodeficiency at baseline in several studies [1,7], is probably an important explanation of men's poorer prognosis. The study by Hawkins and colleagues, however, found a sex difference even after adjusting for baseline disease severity. Certainly, there might have been prognostic factors not accounted for in their analysis, such as opportunistic infections, but sex differences in treatment adherence probably also contributed to men's poorer prognosis. Men were significantly more likely than women to be nonadherent or lost to follow-up, a tendency which has also been observed in other studies [6,8]. A possible explanation is that women are more accustomed to daily medication, such as contraceptive pills early in their reproductive years or iron/folic acid supplementation during pregnancy, whereas most men are unfamiliar with the concept of regular medication. Other psychological or stigma issues might also play a role, including alcohol abuse, which is more common in African HIV-infected men than women [9]. So, are the sex differences only attributable to behavioral differences? There is some laboratory evidence that sex-specific genetic factors might influence the course of AIDS [10]. Certain epidemiological studies in industrialized countries have also reported sex differences in the response to ART, including a large study of 6923 seroconverters from Europe, Australia and Canada, in which progression to AIDS and death was significantly slower in women compared to men within the same transmission category [11]. However, the authors were unable to conclude whether the observed differences were biological or behavioral in nature. Future studies, therefore, should aim to improve our understanding of the underlying causes of sex differences, so that effective strategies to improve men's response to ART in Africa can be identified. The study by Hawkins and colleagues has certain weaknesses, and some caution should be observed when interpreting the results. Most importantly, its observational nature renders it vulnerable to confounding factors not accounted for in the study, such as alcohol/substance abuse and other comorbidity. Furthermore, the substantial attrition, with 22.7% lost to follow-up, could have influenced the Cox proportional hazards models used to identify predictors of mortality. Finally, these results can not be generalized to all resource-limited settings since cultural and traditional sex issues vary greatly, and in settings in which women suffer discrimination these findings might be reversed. However, studies from diverse locations in sub-Saharan Africa support the conclusion that African men have a poorer prognosis than women after ART initiation. As a research community we have a responsibility to base our recommendations on evidence rather than political correctness. Evidence-based medicine in Africa looks different in 2010 than it did a few decades ago, and currently men appear to be the losers of the ART scale-up in Africa. Specific interventions targeting this group might be needed to improve men's treatment outcomes as we enter the second decade of scaling up ART in resource-limited settings. The Director-General of WHO, Margaret Chan, recently launched the report Women and Health: Today's Evidence – Tomorrow's Agenda. Will today's evidence translate into tomorrow's agenda for HIV-infected African men as well?