Abstract
Antiretroviral therapy (ART) access in the developing world has improved, but whether increased access has translated to more rapid treatment initiation among those who need it is unknown. We characterize time to ART initiation across three eras of ART availability in Chennai, India (1996–1999: pregeneric; 2000–2003: generic; 2004–2007: free rollout). Between 1996 and 2007, 11,171 patients registered for care at the YR Gaitonde Centre for AIDS Research and Education (YRGCARE), a tertiary HIV referral center in southern India. Of these, 5726 patients became eligible for ART during this period as per Indian guidelines for initiation of ART. Generalized gamma survival models were used to estimate relative times (RT) to ART initiation by calendar periods of eligibility. Time to initiation of ART among patients in Chennai, India was also compared to an HIV clinical cohort in Baltimore, USA. Median age of the YRGCARE patients was 34 years; 77% were male. The median CD4 at presentation was 140 cells/µl. After adjustment for demographics, CD4 and WHO stage, persons in the pregeneric era took 3.25 times longer (95% confidence interval [CI]: 2.53–4.17) to initiate ART versus the generic era and persons in the free rollout era initiated ART more rapidly than the generic era (RT: 0.73; 95% CI: 0.63–0.83). Adjusting for differences across centers, patients at YRGCARE took longer than patients in the Johns Hopkins Clinical Cohort (JHCC) to initiate ART in the pregeneric era (RT: 4.90; 95% CI: 3.37–7.13) but in the free rollout era, YRGCARE patients took only about a quarter of the time (RT: 0.31; 95% CI: 0.22–0.44). These data demonstrate the benefits of generic ART and government rollouts on time to initiation of ART in one developing country setting and suggests that access to ART may be comparable to developed country settings.
Keywords:
Acknowledgements
The authors would like to thank the staff at YRGCARE especially the research nurses, clinicians, and data entry staff who were responsible for the data extraction and data entry processes over the years. They would also like to thank all our collaborators, both national and international, for their assistance over the years. Most importantly, they would like to thank the patients without whom these analyses would not have been possible. This work was supported in part by a grant from the National Institutes of Health (DA 11,602).