6% occurred between 6 and 12 months and 2.9% after Z-VAD-FMK 12 months. Among seroconverting patients who initiated HAART after enrolling into care, the median

time to seroconversion of the seronegative partner was 73 days after initiating therapy. Patients in relationships that seroconverted within 6 months of enrolling into care had significantly higher PVLs than patients in discordant relationships (178 251 vs. 88 456 copies/mL) (P=0.001). Patients in relationships that seroconverted within 6 months of enrolment were less likely to use condoms with their primary partners than patients in discordant relationships (41.8%vs. 50.9%) (P=0.047). Similar to the patients in relationships that seroconverted within 6 months of enrolment, patients in relationships that seroconverted between 6 and 12 months after enrolment had significantly higher PVLs than patients in discordant relationships (125 865 vs. 115 858 copies/mL) (P=0.035). Patients in relationships that seroconverted between 6 and 12 months after enrolment were diagnosed more often with genital Herpes simplex than patients in discordant relationships

(46.2%vs. 3.6%) (P=0.001). Among patients in discordant relationships, one patient developed syphilis and another patient developed vaginal candidiasis between 6 and 12 months. Patients in relationships that seroconverted between 6 and 12 months after enrolment reported less condom Sucrase use with their primary partners than patients in discordant relationships (61.5%vs. 74.9%) (P=0.035). More patients in relationships AZD6244 that seroconverted between

6 and 12 months after enrolment reported alcohol consumption than patients in discordant relationships (30.8%vs. 7.2%) (P=0.044). Table 3 summarizes the baseline demographic, behavioural and clinical correlates associated with seroconversion. In the univariate logistic regression, HIV-infected patients with a PVL >100 000 were 1.82 times more likely to transmit (95% CI: 1.1–2.8), HIV-infected patients who did not disclose their HIV status were 5.5 times more likely to transmit (95% CI: 4.3–6.2) and HIV-infected patients who did not use condoms were 2.8 times more likely to transmit (95% CI: 2.4–3.6) infection. These factors remained significant in the multivariate logistic regression analyses. The current study documents a substantial risk for heterosexual HIV transmission within South Indian discordant couples, and identifies several preventable behavioural and biological factors associated with HIV transmission. Patients who had not initiated HAART were more likely to transmit the virus to their partners. Men were more likely to transmit HIV to their wives, which reflects the continued risk of HIV transmission to married women [2].