To the Editors:
We read with interest the recent article “Trends in Retention on Antiretroviral Therapy in National Programs in Low-Income and Middle-Income Countries”.1 We agree that there is an urgent need for accurate data on the outcomes of patients on treatment, particularly as large antiretroviral therapy (ART) programs continue to scale-up services. As noted in the article, South Africa-with the highest number of people living with HIV and the largest ART program worldwide2-did not report on program retention.
We have recently published a study on the outcomes of 44,177 adults in South Africa, approximately 10% of all those starting public sector ART nationally between 2002 and 2007.3 Based on routinely collected data, the study provides insight into the effectiveness of a large national ART program and has implications for other low-income and middle-income countries working toward universal ART access.
We found evidence of the rapid massive scale-up of ART: enrollment increased 12-fold over 5 years, and 63% of all patients were enrolled in 2006/2007. There were strong temporal trends in patient retention. Attrition was highest in the first year on ART but continued with duration on treatment: overall retention at 12, 24, and 36 months was 80%, 71%, and 64%, respectively. Over time, patient attrition was increasingly due to loss-to-follow-up (LTFU) compared with mortality, suggesting that there may be different risk factors for early and late attrition on ART. With each successive calendar year of enrolment, too, there was an increasing risk of appearing LTFU. For example, the risk of appearing LTFU at 12 months on ART was 12 times higher among patients enrolled in 2007 than those enrolled in 2002/2003 (adjusted hazard ratio: 11.9, 95% confidence interval: 6.4 to 22).
At a program level, it is important to characterize the patients defined as LTFU. Some will have been misclassified LTFU, either because they are dead4 or because they have interrupted treatment temporarily5; some patients recorded as LTFU will have been lost to care, having stopped treatment altogether, whereas another group will be alive and in care but considered LTFU because of administrative error including unrecorded visits and transfers. True retention in care may thus be underestimated.
By the end of 2009, South Africa had initiated nearly 1 million individuals on ART.2 The growing challenge for South Africa and other countries with large long-term ART programs is to find better ways of following and retaining as many of these patients as possible in life-long care.
Morna Cornell, MPH*†
Anna Grimsrud, MPH*
Andrew Boulle, MBChB, MSc, FCPHM(SA), PhD*
Landon Myer, MBChB, PhD*‡
*Centre for Infectious Disease Epidemiology & Research, School of Public Health & Family Medicine, University of Cape Town, Cape Town, South Africa
†Perinatal HIV Research Unit, University of the Witwatersrand, Johannesburg, South Africa
‡International Center for AIDS Care and Treatment Programs, and Department of Epidemiology, Mailman School of Public Health, Columbia University, New York.
1. Tassie J-M, Baijal P, Vitoria M, et al. Trends in retention on antiretroviral therapy in national programs in low-income and middle-income countries. J Acquir Immune Defic Syndr
2. UNAIDS. Republic of South Africa: Country Progress Report on the Declaration of Commitment on HIV/AIDS: 2010 Report. Available at: http://data.unaids.org/pub/Report/2010/southafrica_2010_country_progress_report_en.pdf
. Accessed March 3, 2010.
3. Cornell M, Grimsrud A, Fairall L, et al. Temporal changes in programme outcomes among adult patients initiating antiretroviral therapy across South Africa, 2002-2007. AIDS
4. Brinkhof MW, Pujades-Rodriguez M, Egger M. Mortality of patients lost to follow-up in antiretroviral treatment programmes in resource-limited settings: systematic review and meta-analysis. PLoS One
5. Kranzer K, Lewis JJ, Ford N, et al. Treatment interruption in a primary care antiretroviral therapy program in South Africa: cohort analysis of trends and risk factors. J Acquir Immune Defic Syndr
. 2010; 55:e17-23.