Life expectancy of persons receiving combination antiretroviral therapy in low-income countries: a cohort analysis from Uganda

Abstract

Background: Little is known about the effect of combination antiretroviral therapy (cART) on life expectancy in sub-Saharan Africa.

Objective: To estimate life expectancy of patients once they initiate cART in Uganda.

Design: Prospective cohort study.

Setting: Public sector HIV and AIDS disease-management program in Uganda.

Patients: 22 315 eligible patients initiated cART during the study period, of whom 1943 were considered to have died.

Measurements: All-cause mortality rates were calculated and abridged life tables were constructed and stratified by sex and baseline CD4 cell count status to estimate life expectancies for patients receiving cART. The average number of years remaining to be lived by patients who received cART at varying age categories was estimated.

Results: After adjustment for loss to follow-up, crude mortality rates (deaths per 1000 person-years) ranged from 26.9 (95% CI, 25.4 to 28.5) in women to 43.9 (CI, 40.7 to 47.0) in men. For patients with a baseline CD4 cell count less than 0.050 × 10(9) cells/L, the mortality rate was 67.3 (CI, 62.1 to 72.9) deaths per 1000 person-years, whereas among persons with a baseline CD4 cell count of 0.250 × 10(9) cells/L or more, the mortality rate was 19.1 (CI, 16.0 to 22.7) deaths per 1000 person-years. Life expectancy at age 20 years for the overall cohort was 26.7 (CI, 25.0 to 28.4) additional years and at age 35 years was 27.9 (CI, 26.7 to 29.1) additional years. Life expectancy increased substantially with increasing baseline CD4 cell count. Similar trends are observed for older age groups.

Limitations: A small (6.4%) proportion of patients were lost to follow-up, and it was imputed that 30% of these patients had died. Few patients with a CD4 cell count greater than 0.250 × 10(9) cells/L initiated cART.

Conclusion: Ugandan patients receiving cART can expect an almost normal life expectancy, although there is considerable variability among subgroups of patients.

Primary funding source: Canadian Institutes of Health Research.