University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Treatment Interruption
Treatment interruption in a primary care antiretroviral therapy program in South Africa: cohort analysis of trends and risk factors
Global Health Sciences Literature Digest
Published January 03, 2011
Journal Article

Kranzer K, Lewis JJ, Ford N, Zeinecker J, Orrell C, Lawn SD, Bekker LG, Wood R. 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 Nov 1;55(3):e17-23.

In Context

By the end of 2009, over 5 million people were receiving antiretroviral therapy (ART) in low- and middle-income countries.(1) Despite this success, retaining patients in care and treatment programs is a major challenge. A systematic review of 39 cohorts in sub-Saharan Africa reported that only 65% to 87% of patients remained in care at 24 months.(2) Among those who were lost-to follow-up, a substantial proportion may have died, leading to concerns about inaccurate estimates of survival in program evaluations.(3) Additionally, some patients may transiently interrupt treatment returning to care at a later stage; however, few studies have measured this in resource limited settings. Treatment interruptions, whether planned or not, have been associated with increased risk of opportunistic infections, and death.(4, 5, 6)


To investigate the frequency and risk factors for defaulting treatment and to identify factors associated with subsequent return to care in a long-term treatment cohort in South Africa.


Prospective cohort study in South Africa. HIV-infected adult patients accessing ART in a public sector primary care clinic.


All patients aged ≥15 years who accessed ART in this primary care clinic between March 2004 and December 2009 were included. Initial follow-up schedule for those initiating ART included visits at 2 weeks, monthly visits until month 3, and then every three months for those who were stable on ART. Patients defaulting treatment were defined as those who had not presented at the pharmacy for ART refills for more than 30 days, which included both patients who subsequently returned to care and restarted ART (defined as treatment interrupters) and patients who had not returned to care at the time of censoring (defined as loss to follow-up). Death on ART was defined as any death within 3 months of drug refill. Data were collected prospectively on death, transfer out, and loss to follow-up. Data on treatment interrupters were obtained from electronic pharmacy dispensing records and was verified through folder reviews. For 48% of patients, vital status, date of death, date of transfer, and date of migration was determined by home visits.

The first endpoint was time from ART initiation to the first time when all the drugs were stopped for at least 30 days (default). For this analysis, follow-up was censored at date of death, date of transfer, or study end (December 2009). The second endpoint was treatment resumption, defined as time from defaulting treatment for the first time to the time of restarting ART. For this analysis, follow-up was censored at date of death, date of transfer, date of migration or study end. The Kaplan-Meir method was used to estimate the rate of defaulting treatment at 1, 2, and 5 years after ART initiation and the rate of treatment resumption after defaulting for the first time in the first, second, and third year after stopping treatment. Multivariate Poisson regression models were used to determine risk factors associated with defaulting treatment and for resuming treatment after defaulting. Sensitivity analyses were conducted excluding persons with unascertained vital status.


A total of 1154 patients received ART, of who 611 remained in care, 188 transferred to other ART clinics, 64 died, and 291 defaulted ART at least once. The median duration of follow-up was 1.45 years. Sixty-five percent of participants were women, the median age was 31.9 yrs, the majority were in WHO stage 3 (51.1%) and 4 (25.1%), and the median CD4 count was 122/µl before treatment initiation. Of the 291 patients that defaulted treatment at least once, 96 resumed therapy (treatment interruption) and 195 did not resume therapy during follow-up (lost-to follow-up). Median time that patients failed to receive ART during their first episode of default was 228 days.

Overall rate of treatment default for the first time was 12.8 per 100 person-years (95%CI: 11.4,14.4), with the highest rate in the first six months of ART (18.2 per 100 person-years; 95%CI: 14.7,22.5) and then decreased by more than half after 2 years (8.8 per 100 person-years; 95%CI: 7.0,11.0). Probability of defaulting treatment for at least 30 days was 14.9% (95%CI: 12.7,17.4) by 1 year, 25.6% (95%CI: 22.7,28.8) by 2 years, and 41.0% (95%CI: 37.0,45.3) by 5 years from ART initiation. The adjusted risk of defaulting treatment was higher among men (adjusted HR [aHR] 1.51, 95%CI: 1.18,1.93), baseline CD4 count >200 cells/µL (aHR 1.39, 95%CI: 1.02,1.91), and for each calendar year of initiating ART (aHR 1.30, 95%CI: 1.17,1.44). Adjusted risks for defaulting were lower among patients on treatment for > 2 years (aHR 0.69, (95%CI: 0.48,0.98) compared with patients on ART for less than 6 months. Excluding those patients whose vital status could not be ascertained produced similar results (data not shown).

Overall rate of treatment resumption after defaulting was 21.4 per 100 person-years (95%CI: 17.5,26.2). Probability of resuming treatment was 26.7% (95%CI: 21.7,32.7) in the first year, 37.1% (95%CI: 31.1,43.9) in the second year, and 42.1% (95%CI: 35.2,49.7) in the third year after stopping treatment. Resuming treatment was less likely among men (adjusted IRR [aIRR] 0.67, 95%CI: 0.43,1.04), and those over 1 year from defaulting treatment (aIRR 0.40, 95%CI: 0.25,0.63) and more likely among patients >30 years old (aIRR 1.80, 95%CI: 1.13,2.86). Excluding those patients whose vital status was not ascertained, similar results were found, with the exception that men were no longer less likely to resume treatment (IRR 0.81,95%CI: 0.52,1.26).


This study demonstrated that although treatment interruption is common, the probability of ART patients who are lost to follow-up resuming therapy within three years was 42%. Resumption of treatment is most likely within the first year of interrupting therapy. Interventions to target retention in care should focus on risk factors for defaulting which include targeting men, those with higher CD4 counts and patients during the first six months of treatment.

Study Quality

Ascertainment of death for treatment defaulters was incomplete and may have led to misclassification of deaths as defaulters. The sensitivity analysis excluded those with unascertained vital status and found the same results. There were a substantial number of patients (16%) who transferred to other ART sites and their outcomes were not ascertained, which may have influenced overall default and resumption rates. The clinical, immunologic, and virologic impact of treatment interruption on morbidity and mortality was not measured.

Programmatic Implication

Interventions should be directed to improve retention rates in HIV care and treatment programs. These interventions should target those patient populations at greatest risk of defaulting. Resumption of ART is most likely in the first year of interrupting therapy and suggests that efforts to track patients and bring them back to care might be most successful early after defaulting treatment.


  1. World Health Organization: Towards universal access: Scaling up priority HIV/AIDS interventions in the health sector; Progress report 2010.
  2. Rosen S, Fox MP, Gill CJ. Patient retention in antiretroviral therapy programs in sub-Saharan Africa; a systematic review. PLoS Med 2007;4:e298
  3. Bisson GP, Gaolathe T, Gross R, et al. Overestimates of survival after HAART: implications for global scale-up efforts. PLoS One 2008;3:e1725.
  4. Seminari E, De Silvestri A, Boschi A, et al. CD4+ guided antiretroviral treatment interruption in HIV infection: a meta-analysis. AIDS Rev 2008;10:236-244
  5. DART trial team. Fixed duration interruptions are inferior to continuous treatment in African adults starting therapy with CD4 cell counts < 200 cells/microl. AIDS 2008;22:237-247.
  6. El-Sadr WM, Lundgren JD, Neaton JD, et al. CD4+ count-guided interruption of antiretroviral treatment. N Engl J Med 2006;355:2283-2296.