Abstract
Poor adherence is a main challenge to successful second-line ART in South Africa. Studies have shown that patients can re-suppress their viral load following intensive adherence counselling. We identify factors associated with failure to re-suppress on second-line ART.
The study was a retrospective cohort study which included HIV-positive adults who experienced an elevated viral load ≥400 copies/ml on second-line ART between January 2013–July 2014, had completed an adherence counselling questionnaire and had a repeat viral load result recorded within 6 months of intensive adherence counselling. Log-binomial regression was used to evaluate the association between patient characteristics and social, behavioral or occupational factors and failure to suppress viral load (≥400 copies/ml).
A total of 128 patients were included in the analysis, and of these 39% (n = 50) failed to re-suppress their viral load. Compared to those who suppressed, far more patients who failed to suppress reported living with family (44.2% vs. 23.7%), missing a dose in the past week (53.3% vs. 30.0%), using traditional/herbal medications (63.2% vs. 34.3%) or had symptoms suggestive of depression (57.7% vs. 34.3%).
These patient-related factors could be targeted for interventions to reduce the risk for treatment failure and prevent switching to expensive third-line ART.
Keywords: second-line antiretroviral therapy, resource limited setting, virologic failure, viral suppression, intensive adherence counselling
Introduction
In South Africa, an estimated 10% of patients started on antiretroviral therapy (ART) will switch to second-line ART, and a further 33–40% of patients are expected to fail treatment in the first 12 months of second-line ART (Fox, Brennan, Maskew, MacPhail, & Sanne, 2010; Onoya et al., 2016). This is similar to the high rates (21–38%) of second-line treatment failure described for other low- and middle-income countries (Ajose, Mookerjee, Mills, Boulle, & Ford, 2012). Poor adherence rather than drug resistance is the main driver for virologic failure on protease-inhibitor (PI)-based second-line (Ajose et al., 2012; Court et al., 2014) and reports show that two-thirds of patients can re-suppress following intensive adherence counselling (IAC) (Cox, 2015; Evans et al., 2016; Fox et al., 2016). With rapid expansion of the ART program in South Africa, understanding factors that contribute to second-line treatment failure is essential to the future success of the national treatment program and highlights the importance of improving adherence on second-line ART. A better understanding of the reasons for failure to re-suppress can lead to the design of targeted interventions that reduce the risk of treatment failure among these patients, and ultimately prevent switching to more expensive third-line regimens. This study aims to aims to identify factors associated with failure to suppress viral load (≥400 copies/ml) among a cohort of patients with an elevated viral load on second-line ART.
Methods
Study site
Themba Lethu Clinic (TLC) is a large, urban public-sector clinic located in Johannesburg, South Africa. Since the start of the national ART program in April 2004, more than 30,000 adults have been part of the TLC cohort, of whom more than 28,000 have started on first-line and over 3000 on second-line ART. The cohort has been described elsewhere (Fox et al., 2013, 2016). TLC follows the South African National Department of Health adult ART treatment guidelines which call for second-line treatment with the boosted protease inhibitor, lopinavir/ritonavir (Department of Health, 2013)
Study design and population
This was a retrospective cohort study. After July 2012 TLC introduced IAC as part of a standardized, targeted approach for all patients with an elevated viral load (≥400 copies/ml) on second-line ART. The IAC intervention for patients failing second-line has been described elsewhere (Fox et al., 2016). Briefly, patients are identified from the electronic medical record called TherapyEdge-HIV™ and are flagged by clinic staff for IAC at their upcoming clinic visit in 1–2 months. At this visit patients bypass the normal clinic queue and undergo detailed IAC with an experienced counsellor or social worker who completes a standardized questionnaire. The questionnaire consists of several open- and closed-ended questions relating to adherence and side-effects, social situation, behavioral characteristics, occupational situation and depression-related symptoms. All patients also complete a standardized adherence screen. The form was developed in the clinic and therefore has not been validated but does allow the provider to assess potential barriers to adherence (Fox et al., 2016).
After counselling patients meet with a senior clinician to discuss adherence and possible side-effects. Patients return in 2–4 months for a repeat viral load test. Patients who suppress their viral load <400 copies/ml continue with second-line treatment, those who do not undergo genotype resistance testing.
Inclusion criteria
All HIV-positive adults (≥18 years) with an elevated viral load ≥400 copies/ml on second-line ART at TLC between January 2013–July 2014 were included. The analysis was restricted to patients who underwent IAC and for whom an adherence counselling questionnaire could be located in their medical file and those with a repeat viral load within six months of intensive adherence counselling.
Data collection
For each patient, data from the adherence counselling questionnaire was linked to the patient’s electronic medical record (TherapyEdge-HIV™) and patient-level data including information on demographics, medications, laboratory test results and other clinical information were extracted. In our study, any symptom suggestive of depression was defined as reporting at least one of the following; feeling sad or anxious, crying frequently, feeling hopeless, having difficulty sleeping or using alcohol to feel better.
Statistical analysis
Our primary outcome of interest was failure to suppress viral load; defined as a repeat viral load ≥400 copies/ml within 6 months of IAC. Numerous studies have used a cut-off of <400 copies/ml to define viral suppression (Johnston et al., 2014; Levison, Losina, Lu, Freedberg, & Wood, 2011). We identified factors associated with failure to suppress using log-binomial regression to estimate relative risks (RR) with corresponding 95% confidence intervals. Factors identified in univariate models (p < 0.2) and variables a priori considered important (i.e., age and gender) were included in the multivariate model. All analyses were conducted using SAS version 9.3 (SAS Institute, Cary, North Carolina, USA).
Ethical considerations
The study protocol was reviewed by University of the Witwatersrand Human Research Ethics Committee (M141187), who approved the use of routine clinical data without informed consent. All patient records and information was anonymized and de-identified prior to analysis.
Results
For the study period, 298 patients had a single viral load ≥400 copies/ml on second-line ART. Of those, 154 (51.7%) patients had an adherence counselling questionnaire and could be linked to clinical data. Of the 154 patients, 128 (83.1%) had a repeat viral load, of which 61% (78/128) suppressed their viral load after IAC (Figure 1). More patients who failed to suppress had a viral load more than 100,000 copies/ml at their first elevated viral load on second-line ART (20% vs. 2.6%) (Table 1).
Figure 1.
Study enrollment and study outcomes.
Table 1.
Demographic and clinical characteristics of patients with an elevated viral load on second-line ART at Themba Lethu Clinic, stratified by the repeat viral load after intensified adherence counselling (n = 154).
| Suppressed (N = 78) | Not suppressed (N = 50) | No repeat VL (N = 26) | P value | |
|---|---|---|---|---|
| Gender, Male, n % | 35 (44.9%) | 28 (56.0%) | 10 (38.5%) | 0.317 |
| Age at first elevated viral load, years | ||||
| Median, IQR | 42.9 (37.7–49.2) | 38.7 (33.6–45.0) | 41.7 (36.6–47.0) | 0.826† |
| <45 | 48/76 (63.2%) | 39/50 (78.0%) | 21/26 (80.8%) | 0.230 |
| ≥45 | 28/76 (36.8%) | 11/50 (22.0%) | 5/26 (19.2%) | |
| CD4 count at first elevated viral load | ||||
| Median, IQR& | 454 (274–593) | 266 (201–472) | 351 (195–501) | 0.784† |
| <250 cells/mm3 | 13/60 (21.7%) | 21/47 (44.7%) | 4/15 (26.7%) | 0.703 |
| 250–500 cells/mm3 | 20/60 (33.3%) | 18/47 (38.3%) | 7/15 (46.6%) | |
| ≥500 cells/mm3 | 27/60 (45.0%) | 8/47 (17.0%) | 4/15 (26.7%) | |
| Viral load at first elevated viral load | ||||
| Median, IQR | 2780 (1721–10,442) | 4539 (1994–74,153) | 12,606 (1684–47,924) | 0.269† |
| 1000–10,000 copies/ml | 57/77 (74.0%) | 29/50 (58.0%) | 15/25 (60.0%) | 0.620 |
| 10,000–100,000 copies/ml | 18/77 (23.4%) | 11/50 (22.0%) | 8/25 (32.0%) | |
| ≥100,000 copies/ml | 2/77 (2.6%) | 10/50 (20.0%) | 2/25 (8.0%) | |
| Transfer-in from another facility on second-line ART | 9/78 (11.5%) | 8/50 (16.0%) | 3/26 (11.5%) | 0.809 |
| Second-line ART NRTI back-bone | ||||
| Abacavir | 3/76 (4.0%) | 3/49 (6.1%) | – | 0.183 |
| Zidovudine | 29/76 (38.1%) | 21/49 (42.9%) | 14/24 (58.3%) | |
| Tenofovir | 37/76 (48.7%) | 25/49 (51.0%) | 8/24 (33.4%) | |
| Stavudine | 7/76 (9.2%) | - | 2/24 (8.3%) | |
| Time on second-line ART prior to elevated viral load, months (IQR) | 21.3 (9.4–35.0) | 17.5 (7.8–25.7) | 18.1 (10.1–26.2) | 0.242† |
| Time to repeat viral load, months (IQR) | 4.6 (3.6–5.5) | 4.6 (3.7–6.5) | N/A | N/A |
| Final treatment outcome** | ||||
| Alive and in care | 69 (88.5%) | 42 (84.0%) | 19 (73.1%) | 0.020 |
| Dead | 0 (0.0%) | 1 (2.0%) | 0 (0.0%) | |
| Lost to follow-up | 1 (1.2%) | 3 (6.0%) | 6 (23.1%) | |
| Transfer | 8 (10.3%) | 4 (8.0%) | 1 (3.8%) |
NRTI; Nucleoside Reverse Transcriptase Inhibitor. IQR; inter quartile range.
Those without a repeat VL (n = 26) compared to those with a VL (n = 128) using Chi square test for proportions or as otherwise indicated using the Kruskal-Wallis test for non-normally distributed continuous data ().
Final treatment outcomes defined at the end of follow-up. These included alive and in care, dead (verified through linkage with the National Vital Registration System for patients with a valid South African national identification document (Fox, Ive, Long, Maskew, & Sanne, 2010; Shearer et al., 2014), lost to follow-up (defined as being ≥3 months late for the last scheduled visit with no subsequent visit) and transferred to another facility.
CD4 at first elevated viral load on second-line, extracted from the patient medical record and defined as the closest result, up to 180 days prior to but not more than 30 days after, the date the patient received the adherence intervention.
Bolded values are statistically significant at p value < 0.05.
In terms of missing data, characteristics were similar between those included in the analysis (n = 128) and those not included (n = 26; p < 0.05). Compared to those with a repeat viral load, patients without a repeat viral load were more likely to be lost from care (≥3 months late for the last scheduled visit; RR 7.38 95%CI 2.24–24.34), and explains the missing laboratory results (Table 1).
From univariate analyses, we identified factors associated with failure to suppress viral load after intensified adherence counselling (Table 2). These included reporting a missed dose in the past week (RR 1.78 95%CI 1.15–2.74) and using traditional/herbal medications (RR 1.84 95%CI 1.20–2.84). Living with family vs. separated from them (RR 1.87 95%CI 1.01–3.46) was the only social factor associated with failure to suppress. Any symptom suggestive of depression was associated with failure to suppress viral load (RR 1.68 95%CI 1.10–2.57). On their own, only feeling sad or anxious (RR 1.53 95%CI 1.10–2.34) and feeling hopeless (RR 1.73 95%CI 1.12–2.67) were associated with failure to suppress.
Table 2.
Factors associated with failure to suppress viral load after intensified adherence counselling at Themba Lethu Clinic, Johannesburg.
| Failure to suppress viral load (n = 50) | |||
|---|---|---|---|
| Characteristic | Proportion not suppressed/N (%) | Crude Hazard Ratio (95% CI) | Adjusted Hazard Ratio (95% CI) |
| Education | |||
| Below tertiary | 6/12 (50.0%) | 1.0 | |
| Tertiary and above | 40/108 (37.0%) | 0.74 (0.40–1.37) | |
| Employed | |||
| Yes | 30/82 (36.6%) | 1.0 | |
| No | 17/42 (40.5%) | 1.11 (0.70–1.76) | |
| Gender | |||
| Female | 22/65 (33.9%) | 1.0 | 1.0 |
| Male | 28/63 (44.4%) | 1.31 (0.85–2.03) | 1.08 (0.86–1.33) |
| Age | |||
| <45 years | 39/89 (43.8%) | 1.0 | 1.0 |
| ≥45 years | 11/39 (28.2%) | 0.64 (0.37–1.12) | 0.86 (0.70–1.07) |
| CD4 | |||
| <250 cells/mm3 | 21/34 (61.8%) | 1.0 | |
| 250–500 cells/mm3 | 18/38 (47.4%) | 1.09 (0.69–1.70) | |
| ≥500 cells/mm3 | 8/35 (22.9%) | 0.52 (0.27–1.03) | |
| Viral load at first elevated VL | |||
| 1000–10,000 | 29/86 (33.7%) | 1.14 (0.66–1.98) | 1.01 (0.83–1.23) |
| 10,000–100,000 | 11/29 (37.9%) | 1.0 | 1.0 |
| ≥100,000 copies/ml | 10/12 (83.3%) | 2.50 (1.69–3.69) | 1.21 (0.80–1.82) |
| Second-line NRTI backbone | |||
| d4T/AZT/ABC | 24/63 (38.0%) | 1.0 | |
| TDF | 25/62 (40.3%) | 1.06 (0.69–1.64) | |
| Time on second-line ART | |||
| <24 months | 4/9 (44.4%) | 1.22 (0.54–2.74) | |
| 24–48 months | 11/35 (31.4%) | 1.0 | |
| ≥48 months | 27/67 (40.3%) | 1.10 (0.70–1.75) | |
| Year initiated second-line | |||
| ≤2010 | 8/31 (25.8%) | 1.0 | |
| >2010 | 34/80 (42.5%) | 1.65 (0.86–3.15) | |
| Knows names of ARV’s | |||
| Yes | 20/58 (34.5%) | 1.0 | |
| No | 28/67 (41.8%) | 1.21 (0.77–1.91) | |
| Regimen taken as prescribed | |||
| Yes | 37/97 (38.1%) | 1.0 | |
| No | 9/26 (34.6%) | 0.91 (0.51–1.63) | |
| Missed pills in last week | |||
| No | 24/80 (30.0%) | 1.0 | 1.0 |
| Yes | 24/45 (53.3%) | 1.78 (1.15–2.74) | 1.10 (0.91–1.34) |
| Patient has interrupted ARV’s | |||
| No | 26/69 (37.7%) | 1.0 | |
| Yes | 23/57 (40.4%) | 1.10 (0.69–1.66) | |
| ARV side effects reported | |||
| No | 27/79 (34.2%) | 1.0 | |
| Yes | 21/46 (45.7%) | 1.34 (0.86–2.07) | |
| Takes other chronic medication | |||
| No | 32/85 (37.7%) | 1.0 | |
| Yes | 16/40 (40.0%) | 1.06 (0.67–1.70) | |
| Ever had TB treatment | |||
| No | 21/65 (32.3%) | 1.0 | |
| Yes | 28/60 (46.7%) | 1.44 (0.93–2.25) | |
| Currently taking anti-TB drugs | |||
| No | 47/121 (38.8%) | 1.0 | |
| Yes | 1/3 (33.3%) | 0.86 (0.17–4.32) | |
| Uses traditional/herbal medicine | |||
| No | 37/108 (34.3%) | 1.0 | 1.0 |
| Yes | 12/19 (63.2%) | 1.84 (1.20–2.84) | 1.20 (0.90–1.59) |
| Use of alcohol | |||
| No | 35/99 (35.4%) | 1.0 | |
| Yes | 14/26 (54.9%) | 1.52 (0.98–2.38) | |
| Use of drugs | |||
| No | 45/116 (38.8%) | 1.0 | |
| Yes | 4/10 (40.0%) | 1.03 (0.47–2.28) | |
| Financial support | |||
| Employed | 30/82 (36.6%) | 1.0 | |
| Social grants | 3/7 (42.9%) | 1.14 (0.46–2.81) | |
| Family | 14/35 (40.0%) | 1.07 (0.65–1.75) | |
| Other | 3/6 (50.0%) | 1.33 (0.57–3.12) | |
| Use tools to remember | |||
| Yes | 44/114 (38.6%) | 1.0 | |
| No | 4/10 (40.0%) | 1.04 (0.47–2.29) | |
| Difficulty reading or understanding instructions | |||
| No | 41/109 (37.6%) | 1.0 | |
| Yes | 5/14 (35.7%) | 0.95 (0.45–2.00) | |
| Married | |||
| Yes | 40/99 (40.4%) | 1.0 | |
| No | 10/29 (34.5%) | 0.85 (0.49–1.49) | |
| Partners | |||
| Single | 13/34 (38.2%) | 1.0 | |
| Multiple | 32/87 (36.8%) | 0.96 (0.58–1.60) | |
| Partner tested HIV | |||
| Yes | 26/70 (37.1%) | 1.0 | |
| No | 2/13 (15.4%) | 0.41 (0.11–1.54) | |
| Partner status | |||
| HIV positive | 21/53 (39.6%) | 0.94 (0.60–1.48) | |
| HIV negative | 2/11 (18.2%) | 1.0 | |
| Unknown | 5/18 (27.8%) | 0.66 (0.30–1.48) | |
| Use of condoms | |||
| Yes | 26/63 (41.3%) | 1.0 | |
| No | 9/31 (29.0%) | 0.70 (0.38–1.31) | |
| Desire for children | |||
| Yes | 16/42 (38.1%) | 1.0 | |
| No | 21/67 (31.3%) | 0.82 (0.49–1.39) | |
| Feel about clinic staff | |||
| Bad | 3/7 (42.9%) | 0.94 (0.38–2.32) | |
| Good | 18/41 (43.9%) | 1.0 | |
| Neutral | 25/73 (34.3%) | 0.75 (0.48–1.16) | |
| Disclosed to employer | |||
| Yes | 6/22 (27.3%) | 1.0 | |
| No | 22/49 (44.9%) | 1.65 (0.78–3.48) | |
| Difficulty getting time off for medical visits | |||
| No | 29/82 (35.4%) | 1.0 | |
| Yes | 2/6 (33.3%) | 0.94 (0.29–3.03) | |
| Frequently travels for work | |||
| No | 20/59 (33.9%) | 1.0 | |
| Yes | 10/21 (47.6%) | 1.40 (0.79–2.49) | |
| Afford transport costs to clinic | |||
| Yes | 42/113 (37.2%) | 1.0 | |
| No | 1/4 (25.0%) | 0.67 (0.12–3.73) | |
| Lives with family | |||
| No | 9/38 (23.7%) | 1.0 | 1.0 |
| Yes | 38/86 (44.2%) | 1.87 (1.01–3.46) | 1.09 (0.88–1.33) |
| Patient hides pills | |||
| No | 41/109 (37.6%) | 1.0 | |
| Yes | 5/13 (38.5%) | 1.02 (0.48–2.11) | |
| Takes ARV’s openly | |||
| Yes | 33/90 (36.7%) | 1.0 | |
| No | 13/33 (39.4%) | 1.07 (0.65–1.78) | |
| Religious practices that interfere with treatment | |||
| Yes | 26/76 (34.2%) | 1.0 | |
| No | 18/45 (40.0%) | 1.17 (0.73–1.88) | |
| Cultural practices that interfere with treatment | |||
| No | 35/100 (35.0%) | 1.0 | |
| Yes | 12/23 (52.2%) | 1.49 (0.93–2.39) | |
| Feel sad/anxious | |||
| No | 28/82 (34.2%) | 1.0 | |
| Yes | 22/42 (52.4%) | 1.53 (1.01–2.33) | |
| Cry frequently | |||
| No | 33/89 (37.1%) | 1.0 | |
| Yes | 17/36 (47.2%) | 1.27 (0.82–1.98) | |
| Feel hopeless | |||
| No | 32/96 (33.3%) | 1.0 | |
| Yes | 15/26 (57.7%) | 1.73 (1.12–2.67) | |
| Difficulty sleeping | |||
| No | 35/95 (36.8%) | 1.0 | |
| Yes | 14/29 (48.3%) | 1.31 (0.83–2.07) | |
| Use alcohol to feel better | |||
| No | 41/111 (36.9%) | 1.0 | |
| Yes | 7/12 (58.3%) | 1.58 (0.92–2.70) | |
| Any symptom suggestive of depression | |||
| No | 35/102 (34.3%) | 1.0 | 1.0 |
| Yes | 15/26 (57.7%) | 1.68 (1.10–2.57) | 1.05 (0.80–1.38) |
Bold text p < 0.05. NRTI; nucleoside reverse transcriptase inhibitor. d4T; stavudine.; AZT; zidovudine. ABC; abacavir. TDF; tenofovir. TB; tuberculosis. ART; antiretroviral therapy. ARV; antiretroviral. VL; viral load. HIV; human immunodeficiency virus.
After adjusting for gender, age and viral load at first elevated viral load on second-line ART, none of the variables were associated with failure to suppress viral load. Despite this, four variables were prominent among those who failed to suppress. Compared to those who suppressed, more patients who failed to suppress reported living with family (44.2% vs. 23.7%; p = 0.048), missing a dose in the past week (53.3% vs. 30.0%; p = 0.010), using traditional/herbal medications (63.2% vs. 34.3%; p = 0.006) or any symptom suggestive of depression (57.7% vs. 34.3%; p = 0.017).
Discussion
Results from this study show that 61% of patients could suppress their viral load after IAC, which is consistent with other reports (Cox, 2015; Fox et al., 2016; Garone et al., 2013; Khan et al., 2014). Higher viral loads among those who failed to re-suppress supports the view that unresolved poor adherence, not drug resistance, drives second-line failure (Boyd et al., 2015; Ndahimana et al., 2016).
The main factors we identified undermining adherence in patients on second-line ART were patient-related factors such as use of traditional/herbal medicine and depression. We found that far more patients who failed to suppress reported living with family – which could suggest fear of stigmatization or poor family and social support which compromise adherence (Wood, Tobias, & McCree, 2004). In line with other reports, we found that having any symptom suggestive of depression was associated with failure to suppress viral load (Langebeek et al., 2014; Marconi et al., 2013). Our study also found that far more patients who failed to suppress reported using traditional/herbal medicine. However, the kind of herbal/traditional medicine a person was using was not documented in this study nor the reasons for taking them. A previous South African study conducted in South Africa showed the main reasons for taking these traditional/herbal medications was for pain relief, as an immune booster or to stop diarrhoea (Peltzer et al., 2011). Commonly used herbal medicines can result in drug-interactions with ART or reduce ART adherence (Kiguba, Karamagi, Ssali, Mugyenyi, & Katabira, 2007; Mills et al., 2005). Our study found that patients using traditional/herbal medication commonly (>60%) report missing doses or interrupting treatment.
As expected, poor adherence, in our study measured by a missed dose in the past week, was more common among those who failed to suppress. Results from this study and in line with other reports which have demonstrated a connection between poor adherence and elevated viral load (Ajose et al., 2012; Lima et al., 2008; Martin et al., 2008).
The findings should be considered in light of the study limitations. First, we did not have an objective measure of ART adherence (e.g., pill counts) or information relating to ART adherence. For example we did not have information on alcohol or drug use and did not measure factors such quality of healthcare services (e.g., lack of confidentiality and privacy, distance to clinic, drug stock outs, staff shortages etc.). Second, there was limited data to evaluate depression as a risk factor of failing second-line ART. In our study, any symptom of depression reported was used to classify a patient as possibly depressed. Therefore, it is possible that this may have led to misclassification. Third, 48% (144/298) of patients eligible for the study did not have an adherence counselling questionnaire and close to 17% (26/154) of patients included in the analysis did not have a repeat viral load. There were however no major differences observed between those included and not included in the analysis. Fourth, since the study only considers patients engaging in care, the factors identified and the estimate of adherence to ART may be different for those who are no longer in care. Lastly, it should be noted that we did not include information on drug resistance testing.
Conclusion
We demonstrate that patient-related factors (e.g., use of traditional/herbal medicine, depression, family/social support) and not necessarily provider-related factors, health system factors or treatment related factors contribute to poor adherence on second-line ART. Based on our findings, there is a need for on-going counselling and education of patients on second-line ART. Strategies to support patients, improved adherence evaluation and intensified monitoring need to be considered and evaluated in order to reduce the risk of virologic failure. Competing priorities and life circumstances can play a role in adherence and suppression, and should be taken into consideration when thinking through strategies or interventions to address these.
Acknowledgements
We acknowledge the directors and staff of Themba Lethu Clinic (TLC), Clinical HIV Research Unit (CHRU), HE2RO and Right to Care (RTC) – a PEPFAR (US President’s Emergency Plan for AIDS Relief) funded NGO, the Gauteng Provincial and National Department of Health for providing for the care of the patients at Themba Lethu Clinic as part of the National Comprehensive Care, Management and Treatment (CCMT) of HIV and AIDS program. Lastly, we would like to sincerely thank the patients attending the Themba Lethu Clinic for their continued trust in the treatment and care provided at the clinic.
Funding
MPF, DE, TS, CG and RB were supported through the South Africa Mission of the United States Agency for International Development (USAID) under the terms of Cooperative Agreement USAID-674-A-12-00029 to the Health Economics and Epidemiology Research Office. SD received a scholarship from Swedish International Development Cooperation Agency (SIDA; http://www.sida.org).
Footnotes
Disclosure statement
No potential conflict of interest was reported by the authors.
Publisher's Disclaimer: Disclaimer
This study is made possible by the generous support of the American people through Cooperative Agreement AID 674-A-12-00029 from the United States Agency for International Development (USAID). The contents of the aricle are the responsibility of the authors and do not necessarily reflect the views of USAID or the United States Government. The funders had no role in the study design, collection, analysis and interpretation of the data, in manuscript preparation of the decision to publish.
References
- Ajose O, Mookerjee S, Mills EJ, Boulle A, & Ford N (2012). Treatment outcomes of patients on second-line antiretroviral therapy in resource-limited settings: A systematic review and meta-analysis. AIDS (london, England), 26(8), 929–938. [DOI] [PubMed] [Google Scholar]
- Boyd MA, Moore CL, Molina JM, Wood R, Madero JS, Wolff M, … Cooper DA (2015). SECOND-LINE study group. Baseline HIV-1 resistance, virological outcomes, and emergent resistance in the SECOND-LINE trial: An exploratory analysis. The Lancet HIV, 2(2), e42–e51. [DOI] [PubMed] [Google Scholar]
- Court LR, Stewart A, Sunpath H, Murphy R, Winternheimer P, Ally M, & Maartens G (2014). Short term adherence tool predicts failure on second line protease inhibitor-based antiretroviral therapy: An observational cohort study. BMC Infectious Diseases, 14(1), 664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cox V (2015). Addressing treatment failure amongst adults on second line antiretroviral therapy in Khayelitsha. 7th South African AIDS conference, Durban, South Africa, June 2015. [Google Scholar]
- Department of Health, Republic of South Africa. The South African Antiretroviral treatment guidelines 2013. Retrieved from http://www.kznhealth.gov.za/medicine/2013_ART_Guidelines.pdf [Google Scholar]
- Department of Health, Republic of South Africa. The South African Antiretroviral treatment guidelines 2010. Retrieved from http://www.uj.ac.za/EN/CorporateServices/ioha/Documentation/Documents/ART%20Guideline.pdf
- Evans D, Berhanu R, Moyo F, Nguweneza A, Long L, & Fox MP (2016). Can short term use of electronic patient adherence monitoring devices improve adherence in patients failing second-line ART therapy? Evidence from a pilot study in Johannesburg, South Africa. AIDS and Behavior, 20(11), 2717–2728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fox MP, Brennan A, Maskew M, MacPhail P, & Sanne I (2010). Using vital registration data to update mortality among patients lost to follow-up from ART programmes: Evidence from the Themba Lethu Clinic, South Africa. Tropical Medicine & International Health, 15(4), 405–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fox MP, Evans D, Steegen K, Firnhaber C, Ive P, Spencer D, … Berhanu R (2016). Intensive adherence counselling for HIV-infected individuals failing second-line ART in Johannesburg, South Africa. Tropical Medicine and International Health, 21, 1131–1137. [DOI] [PubMed] [Google Scholar]
- Fox MP, Ive P, Long L, Maskew M, & Sanne I (2010). High rates of survival, immune reconstitution, and virologic suppression on second-line antiretroviral therapy in South Africa. JAIDS Journal of Acquired Immune Deficiency Syndromes, 53(4), 500–506. [DOI] [PubMed] [Google Scholar]
- Fox MP, Maskew M, MacPhail AP, Long L, Brennan AT, & Westreich D (2013). Cohort profile: The Themba Lethu Clinical Cohort, Johannesburg, South Africa. International Journal of Epidemiology, 42(2), 430–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Garone DB, Conradie K, Patten G, Cornell M, Goemaere W, & Kunene J (2013). High rate of virological re-suppression among patients failing second-line antiretroviral therapy following enhanced adherence support: A model of care in Khayelitsha, South Africa. Southern African Journal of HIV Medicine, 14(4), 166–169. [Google Scholar]
- Johnston VCK, Wiesner L, Morris L, Ledwaba J, Fielding KL, & Charalambous S (2014). Viral suppression following switch to second-line antiretroviral therapy: Associations with nucleoside reverse transcriptase inhibitor resistance and subtherapeutic drug concentrations prior to switch. The Journal of Infectious Diseases, 209(5), 711–720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Khan S, Das M, Andries A, Deshpande A, Mansoor H, & Saranchuk P (2014). Second-line failure and first experience with third-line antiretroviral therapy in Mumbai, India. Global Health Action, 7 10.3402/gha.v7.24861 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kiguba RB-TJ, Karamagi C, Ssali F, Mugyenyi P, & Katabira E (2007). Discontinuation and modification of highly active antiretroviral therapy in HIV-infected Ugandans prevalence and associated factors. JAIDS Journal of Acquired Immune Deficiency Syndromes, 45, 218–223. [DOI] [PubMed] [Google Scholar]
- Langebeek N, Gisolf E, Reiss P, Vervoort S, Hafsteinsdottir T, & Richter C (2014). Predictors and correlates of adherence to combination antiretroviral therapy (ART) for chronic HIV infection: A meta-analysis. BMC Medicine, 12(1), e438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levison JHO, Losina C, Lu E, Freedberg Z, & Wood KAR (2011). Early outcomes and the virological effect of delayed treatment switching to second-line therapy in an antiretroviral roll-out programme in South Africa. Antiviral Therapy, 16(6), 853–861. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lima V, Harrigan R, Murray M, Moore D, Wood E, Hogg R, & Montaner J (2008). Differential impact of adherence on long-term treatment response among naive HIV-infected individuals. AIDS (London, England), 22, 2371–2380. [DOI] [PubMed] [Google Scholar]
- Marconi VC, Wu B, Hampton J, Ordonez CE, Johnson BA, & Singh D (2013). Early warning indicators for first-line virologic failure independent of adherence measures in a South African urban clinic. AIDS Patient Care and STDs, 27(12), 657–668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martin M, Del Cacho E, Codina C, Tuset M, De Lazzari E, Mallolas J,… Ribas J (2008). Relationship between adherence level, type of the antiretroviral regimen, and plasma HIV type 1 RNA viral load: A prospective cohort study. AIDS Research and Human Retroviruses, 24(10), 1263–1268. [DOI] [PubMed] [Google Scholar]
- Mills E, Foster BC, van Heeswijk R, Phillips E, Wilson K, & Leonard B (2005). Impact of African herbal medicines on antiretroviral metabolism. AIDS (london, England), 19(1), 95–97. [DOI] [PubMed] [Google Scholar]
- Ndahimana JD, Riedel DJ, Muhayimpundu R, Nsanzimana S, Niyibizi G, Mutaganzwa E, … Redfield RR (2016). HIV drug resistance mutations among patients failing second-line antiretroviral therapy in Rwanda. Antiviral Therapy, 21(3), 253–259. [DOI] [PubMed] [Google Scholar]
- Onoya D, Brennan A, Berhanu R, van der Berg L, Buthelezi T, & Fox MP 2016. Changes in second-line regimen durability and continuity of care in relation to national ART guideline changes in South Africa. Journal of the International AIDS Society, 19, 20675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Peltzer K, Preez N. d., Ramlagan S, Fomundam H, Anderson J, & Chanetsa L (2011). Antiretrovirals and the use of traditional, complementary and alternative medicine by HIV patients in Kwazulu-Natal, South Africa: A longitudinal study. African Journal of Traditional, Complementary, and Alternative Medicines, 8(4), 337–345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shearer K, Maskew M, Ajayi T, Berhanu R, Majuba P, & Sanne I (2014). Incidence and predictors of herpes zoster among antiretroviral therapy-naive patients initiating HIV treatment in Johannesburg, South Africa. International Journal of Infectious Diseases, 23, 56–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wood SA, Tobias C, & McCree J (2004). Medication adherence for HIV positive women caring for children: In their own words. AIDS Care, 16, 909–913. [DOI] [PubMed] [Google Scholar]