Outcomes of antiretroviral therapy in Vietnam: results from a national evaluation.

ABSTRACT

OBJECTIVES: Vietnam has significantly scaled up its national antiretroviral therapy (ART) program since 2005. With the aim of improving Vietnam's national ART program, we conducted an outcome evaluation of the first five years of the program in this concentrated HIV epidemic where the majority of persons enrolled in HIV care and treatment services are people who inject drugs (PWID). The results of this evaluation may have relevance for other national ART programs with significant PWID populations. DESIGN: Retrospective cohort analysis of patients at 30 clinics randomly selected with probability proportional to size among 120 clinics with at least 50 patients on ART. METHODS: Charts of patients whose ART initiation was at least 6 months prior to the study date were abstracted. Depending on clinic size, either all charts or a random sample of 300 charts were selected. Analyses were limited to treatment-naïve patients. Multiple imputations were used for missing data. RESULTS: Of 7,587 patient charts sampled, 6,875 were those of treatment-naïve patients (74.4% male, 95% confidence interval [CI] 72.4-76.5, median age 30, interquartile range [IQR] 26-34, 62.0% reported a history of intravenous drug use, CI 58.6-65.3). Median baseline CD4 cell count was 78 cells/mm(3) (IQR 30-162) and 30.4% (CI 25.8-35.1) of patients were at WHO stage IV. The majority of patients started d4T/3TC/NVP (74.3%) or d4T/3TC/EFV (18.6%). Retention rates after 6, 12, 24, and 36 months were 88.4% (CI 86.8-89.9), 84.0% (CI 81.8-86.0), 78.8% (CI 75.7-81.6), and 74.6% (CI 69.6-79.0). Median CD4 cell count gains after 6, 12, 24, and 36 months were 94 (IQR 45-153), 142 (IQR 78-217), 213 (IQR 120-329), and 254 (IQR 135-391) cells/mm(3). Patients who were PWID showed significantly poorer retention. CONCLUSIONS: The study showed good retention and immunological response to ART among a predominantly PWID group of patients despite advanced HIV infections at baseline.