OBJECTIVE: To characterize the value of total lymphocyte counts in predicting risk of death among patients initiating triple combination antiretroviral therapy. METHODS: Study subjects included antiretroviral-naive persons aged 18 years or older who initiated treatment with triple combination therapy between August 1 1996 and September 30 1999 in a population-based observational cohort of HIV-infected individuals. Total lymphocyte counts as well as CD4 count and plasma viral load were assessed at baseline. Separate Cox proportional hazards models were devised to evaluate the effect on survival of total lymphocyte count in lieu of or with CD4 count after adjustment for other prognostic factors including plasma viral load. RESULTS: A total of 733 antiretroviral-naive persons initiated triple drug combination antiretroviral therapy over the study period with a median follow-up of 29.5 months. In the first analysis, only baseline CD4 cell counts of 50-199 cells/microl or less than 50 microl were associated with an increased risk of mortality [adjusted relative risk (ARR) 2.90; 95% CI: 1.40, 5.98] and (ARR 6.30; 95% CI: 2.93, 13.54), respectively. When CD4 counts were excluded from the analysis as if unavailable, total lymphocyte count of between 0.8 and 1.4 G/I, and less than 0.8 G/I were both significantly associated with an increased risk of mortality (ARR 2.36; 95% CI: 1.16, 4.78) and (ARR 6.17; 95% CI: 2.93, 13.01), respectively. CONCLUSION: Total lymphocyte count may provide a simple and cost-effective alternative for prioritizing therapy initiation in resource-limited settings. Our results suggest that, if appropriately validated, judicious application of total lymphocyte counts could overcome one of the practical obstacles to more widespread provision of antiretroviral therapy in resource-poor settings.