Share this article on:

Higher rates of AIDS during the first year of antiretroviral therapy among migrants: the importance of tuberculosis

The Antiretroviral Therapy Cohort Collaboration (ART-CC)

doi: 10.1097/QAD.0b013e32835faa95
Epidemiology and Social

Objective: In lower-income countries rates of AIDS-defining events (ADEs) and death are high during the first year of combination antiretroviral therapy (ART). We investigated differences between foreign-born (migrant) and native-born (nonmigrant) patients initiating ART in Europe, the US and Canada, and examined rates of the most common ADEs and mortality during the first year of ART.

Design: Observational cohort study.

Methods: We studied HIV-positive adults participating in one of 12 cohorts in the Antiretroviral Therapy Cohort Collaboration (ART-CC).

Results: Of 48 854 patients, 25.6% were migrants: 16.1% from sub-Saharan Africa, 5.6% Latin America, 2.3% North Africa/Middle East, and 1.6% Asia. Incidence of ADEs during the first year of ART was 60.8 per 1000 person-years: 69.9 for migrants and 57.7 for nonmigrants [crude hazard ratio (HR) 1.18; 95% confidence interval (CI) 1.08–1.29], adjusted HR (for sex, age, CD4, HIV-1 RNA, ART regimen, prior ADE, probable route of infection and year of initiation, and stratified by cohort) 1.21 (95% CI 1.09–1.34). Rates of tuberculosis were substantially higher in migrants than nonmigrants (14.3 vs. 6.3; adjusted HR 1.94; 95% CI 1.53–2.46). In contrast, mortality was higher among nonmigrants than migrants (crude HR 0.71; 95% CI 0.61–0.84), although excess mortality was partially explained by patient characteristics at start of ART (adjusted HR 0.91; 95% CI 0.76–1.09).

Conclusions: During the first year of ART, HIV-positive migrants had higher rates of ADEs than nonmigrants. Tuberculosis was the most common ADE among migrants, highlighting the importance of screening for tuberculosis prior to ART initiation in this population.

Supplemental Digital Content is available in the text

*Writing committee listed at end of manuscript.

Correspondence to Bryan E. Shepherd, Department of Biostatistics, Vanderbilt University School of Medicine, 1161 21st Avenue South, S2323 MCN, Nashville, TN 37232-2158, USA. E-mail:

Received 6 September, 2012

Revised 10 January, 2013

Accepted 31 January, 2013

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

© 2013 Lippincott Williams & Wilkins, Inc.