Gender differences in presentation and early survival in an antiretroviral therapy programme in Gugulethu : South Africa, 2002-2007

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

By 2005, an estimated 500 000 people with HIV had initiated highly active antiretroviral therapy (HAART) in sub-Saharan Africa. However, disproportionately more women than men have accessed HAART in most developing countries including South Africa. While there has been considerable recent interest in the determinants of mortality among patients receiving HAART in developing countries, there is conflicting evidence about gender differences and survival in HAAR T programmes. This study explored whether there were gender differences in early mortality among 2 843 treatment-naive men and women entering care in a large South African HAART programme. The study was a secondary analysis of patient records covering three time periods: person-time from programme entry to the initiation ofHAART; person-time from HAART initiation to one year on treatment; and the total person-time from programme entry to one year on HAART. Cox' s proportional hazards regression ·was used to investigate crude and adjusted associations between basehne characteristics and mortality as we11 as loss-to-follow-up (LTFU). Using the Sobel test, the study explored whether the degree of disease ( according to CD4 count and WHO stage) played a mediating role in any association between gender and mortality. In all three time periods, the analysis found a strong crude associ~tiol). between male gender and mortality. Prior to HAART-initiation, there was a 31 % increase in the risk of mortality (crude Hazard Ratio (HR) 1.31, 95% CI, 0.93- 1.86; p=0. 131). In the period on HAART, this association strengthened (crude HR 1.57, 95% CI, 1.14-2.16; p=0.005). Overall, male gender increased the risk of mortality in the total cohort by 49% (crude HR, 1.49, 95% CI, 1.17-1.88 ; p=001). Adjustment for baseline characteristics, including CD4 count and WHO stage, attenuated these crude associations. After adjustment, there was no increase in risk associated with male gender in the period pre-HAART (HR 1.01, 95% CI, 0.67-1.51). On HAART, there was a 19% increase in risk (HR 1.19, 95% CI, 0.88-1.67). In the total cohort, this was slightly attenuated (HR 1.15, 95% CI, 0.93-1.50). There was evidence of mediation by degree of disease. In the preHAART period, the Sobel test found significant associations between mortality and CD4 count (p=0.044) as well as WHO stage (p=0.003). On HAART, too, CD4 count (p=0.045) and WHO stage (p< 0.001) appeared to mediate the effect of gender on death. Similarly, in the total cohort, there was evidence to support mediation by CD4 count (p=0.035) and WHO stage (p<0.001). There was a crude association between male gender and the risk of being L TFU (HR for L TFU during the total study period comparing males to females, 1.26, 95% CI, 0.89-1.78 ; p=0.194). This was strengthened by adjustment for age and monthly income (HR,1.35, 95% CI, 0.92-1.97). In this cohort, men appeared to have worse survival prospects than women due to more advanced HIV disease on programme entry. Previous studies have attributed the disproportionate access of women to HAART to gender differences in health seeking behaviour. This study argues that the prime obstacle might be the existing orientation of primary health care systems in developing countries towards the needs of women more than those of men. It suggests that women have better access to primary health services through the existing focus on maternal and child health. Women who are diagnosed and referred for HAART through these services are generally younger and healthier than men, who are diagnosed through services for tuberculosis (TB) and sexually transmitted infections (STis). This might explain why fewer men than women access HAART, and why they are diagnosed at later stages of disease progression. As a result, men may be disadvantaged in access to HAAR T in South Africa. The study suggests a number of short- and long-term solutions including : further research on obstacles to male access to HAART; changes in national policy; and the establishment of male-friendly services as an entry point for men into broader health services. Such approaches might facilitate the earlier diagnosis and treatment of men and improve their survival in HAART programmes.