Antiretroviral therapy (ART) has fundamentally changed health in countries with high HIV prevalence.1–4 The unprecedented success of ART scale-up as measured by people being prescribed ART, people with virologic suppression, and reduced mortality has led to increasingly ambitious approaches to reduce HIV morbidity and transmission. Over the past decade, the CD4 threshold for ART initiation recommended by World Health Organization and most national guidelines has steadily increased. In addition, other indications for ART have broadened and generally now include pregnancy5. In countries that have the capacity for implementation, World Health Organization currently recommends “Option B+,” an approach that includes ART initiation for any HIV-infected pregnant woman not already on ART with lifelong continuation after delivery.
Implementation of Option B+ by national HIV programs is uneven. For example, in 2013, South Africa introduced Option B; with this strategy, ART is initiated for all pregnant women, but treatment is discontinued at cessation of breastfeeding among women with a nadir CD4 T-cell count greater than 350 cells per cubic milliliter; in January 2015, the policy shifted to Option B+.6,7 The Option B+ strategy leverages prenatal care during pregnancy to engage women in long-term HIV management. Implementation of Option B and Option B+ have exposed weaknesses in systems to provide sustained ART delivery during care and life transitions.8,9 Postpartum, women need to make the transition from ART received as part of prenatal care to care in a general adult ART clinic. In addition, there is the personal transition to being the mother of a newborn.10 Each transition may increase the risk for impaired ART adherence, which can result in HIV treatment failure and acquired drug resistance.11
Multiple studies have evaluated limitations in delivering interventions for prevention of mother to child transmission (PMTCT) of HIV during pregnancy.12–14 Several studies have also highlighted postpartum challenges including disengagement from care and virologic failure.13–16 Retention in care and adherence are clearly challenges during the postpartum period. Among those women who do stay engaged in care, it is important to assess treatment response in programmatic settings and explore consequences of treatment failure. We assessed virologic outcomes 12 months postpartum in participants who initiated ART before delivery and continued to receive ART through the 12-month postpartum period. We explored potential contributors to HIV treatment failure and consequences of failure on immunologic response (CD4 T-cell count) and HIV drug resistance.
Tshepiso is a prospective cohort study evaluating maternal and infant outcomes among HIV-infected pregnant women with and without tuberculosis (TB). HIV-infected pregnant women with TB disease and matched controls without TB were enrolled from 10 prenatal clinics in Soweto, South Africa, between January 2011 and January 2014. Pregnant women diagnosed with TB disease, aged ≥18 years, with documented HIV infection, and who were at 13–38 weeks gestation were enrolled as cases. For each case, 2 controls were enrolled: pregnant women with HIV infection who did not have TB were matched by age (within 5 years), gestational age (within 2 weeks), and planned delivery location (hospital or local midwife obstetric unit). All women were followed for pregnancy-related, TB, and HIV outcomes with scheduled study visits during pregnancy, at delivery, and at 1 week, 6 weeks, 6 months, and 12 months postpartum. Visits included questionnaires and blood testing. Depression was assessed using the Edinburgh Postnatal Depression Scale17 at most study visits. HIV RNA and CD4 T-cell count were routinely assayed at each visit; efavirenz serum concentrations were measured at 6 weeks postpartum through serial drug level assessment 12, 14, 16, and 18–20 hours after self-reported timing of the evening efavirenz dose. HIV drug resistance testing was performed retrospectively on stored samples from the 12-month postpartum visit among women with an HIV RNA level greater than 1000 copies per milliliter. Women who initiated ART before delivery and who had a 12 month postpartum follow-up visit and reporting that they were still receiving ART were included in this analysis.
Women received all their prenatal, HIV, and TB care at public sector clinics in Soweto, South Africa. ART was provided through the prenatal care program for women initiated on ART during pregnancy. Women already on ART continued care at their previous ART clinic. After delivery, all women received HIV care at adult ART clinics (usually in the same facility as they received prenatal care). All women provided written informed consent. This study was approved by the Johns Hopkins Medicine institutional review board and the University of the Witwatersrand Human Research Ethics Committee (Medical).
We selected the definition of virologic suppression as an HIV RNA level of <400 copies per milliliter, to be consistent with previous reports and to avoid classifying virologic blips as virologic failure.18 For women receiving efavirenz, analyses used a categorical modeled efavirenz trough concentration of undetectable, detectable but <1, and ≥1 μg/mL. The threshold of 1 μg/mL was based on previous reports suggesting that lower concentrations may be associated with increased risk for virologic failure,19,20 although many individuals with trough values somewhat less than 1 μg/mL maintain excellent virologic control.21,22CYP2B6 loss-of-function polymorphisms were classified into 2 groups: the very slow, slow, and intermediate efavirenz metabolizers were compared to women who were extensive efavirenz metabolizers. This binary classification was selected because extensive metabolizer status is the most likely to be associated with virologic failure. HIV drug resistance to nucleoside reverse transcriptase inhibitors, nonnucleoside reverse transcriptase inhibitors (NNRTIs), and protease inhibitors was based on the primary mutation list published by the IAS-USA.23 Depression was defined as present if the score on the Edinburgh Postnatal Depression Scale was 10 or higher on any study visit.17
CD4, HIV RNA, and resistance testing were performed at an accredited commercial laboratory using dual-platform technology for CD4 T-cell count and the Cobas Ampliprep/Cobas TaqMan real-time polymerase chain reaction system for HIV RNA quantification. HIV drug resistance was assessed by consensus sequencing of amplified real-time polymerase chain reaction products using the Abbott Viroseq system. Postpartum efavirenz plasma concentrations were determined by liquid chromatography–tandem mass spectrometry and then interpreted using population pharmacokinetic modeling.20 Trough concentrations were estimated using individual Bayesian post hoc estimates from the pharmacokinetic model. The method was linear over the range of 0.02–20 μg/mL with a lower limit of quantification of 0.02 μg/mL. Genotyping of three CYP2B6 loss-of-function polymorphisms that predict increased efavirenz exposure [516G→T (rs3745274), 983T→C (rs28399499), and 15582C→T (rs4803419)] was performed at Vanderbilt University as previously described.20,21
Population characteristics are presented as proportions or medians with interquartile ranges (IQRs). We compared proportion of women with an HIV RNA <400 copies per milliliter prepartum and at 12 months postpartum using McNemar test. We assessed for associations between postpartum efavirenz level and CYP2B6 polymorphism using the χ2 test. We used logistic regression to assess factors associated with HIV suppression 12 months postpartum. Odds ratios from logistic regression were converted to relative risk, using the method of Zhang and Yu.24 We included all variables with a P ≤0.1 in univariable analysis for multivariable model building. To perform a single multivariable analysis that included women on efavirenz and non-efavirenz containing regimens, we created missing value levels for efavirenz concentrations and efavirenz metabolizer status for women not receiving efavirenz or with missing values in these fields. Chi-square testing was used to assess for associations with HIV drug resistance among women with treatment failure at 12 months.
Among 116 women who enrolled in Tshepiso between January 2011 and January 2014 and who started ART before delivery, 103 completed the 12-month follow-up visit and were included in these analyses. The 13 women who lacked 12 months of follow-up on ART either moved from the study area (3), discontinued ART before the 12-month postpartum visit (6), or were lost from study follow-up (3). The women with and without 12 months of follow-up on ART were similar in terms of age, gestational age at enrollment, CD4 count, TB treatment, and receiving ART before pregnancy (all P values, P > 0.2). Of the 103 women in the study population, median age was 29 years (IQR: 26–32), gestational age at enrollment was 29 weeks (IQR: 25–34), and enrollment CD4 T-cell count was 317 cells per cubic millimeter (IQR: 218–385; Table 1). Forty-three (42%) women were either receiving TB treatment at the time of enrollment or were diagnosed with TB during enrollment screening. Twenty-three (22%) women were already on ART before pregnancy and 80 (78%) were started during pregnancy. ART was started later in pregnancy in participants with TB coinfection (median, 22 weeks before delivery; IQR: 18–26) than among women with HIV alone (median, 28 weeks before delivery; IQR: 22–94; Fisher exact P = 0.007), and the CD4 count at enrollment was lower among woman with TB coinfection, 278 cells per cubic millimeter (IQR: 144–385) and 348 per cubic millimeter (IQR: 263–394), respectively (Fisher exact P = 0.01). There were no differences in terms of age, parity, or ART regimen (all P > 0.2). At 12 months postpartum, most women were receiving a nucleoside backbone of tenofovir and lamivudine (95, 92%). The remaining women were receiving zidovudine (3, 2.9%) or stavudine (5, 4.8%). The additional agent was efavirenz for 86 (83%), nevirapine for 10 (9%), or lopinavir/ritonavir for 7 (7%). Edinburgh Depression Scale scores ranged from 0 to 16 with a median of 3 (IQR: 0–7); 12 women (12%) met criteria for depression at one or more study visits.
Efavirenz Pharmacokinetics and Pharmacogenetics
Genotype data for CYP2B6 were available for 80 women. Of these women, 2 (2.5%) were very slow metabolizers of efavirenz, 17 (21%) were slow, 42 (52%) were intermediate, and 19 (24%) were extensive metabolizers. Estimated efavirenz trough concentrations were available for 75 of 86 women dispensed efavirenz postpartum and ranged from undetectable (<0.02 μg/mL) to 17 μg/mL with a median of 1.8 μg/mL (IQR: 1.2–3.1). Eight (11%) women had undetectable efavirenz levels, 9 (12%) had detectable levels of less than 1 μg/mL, and 58 (77%) had a 6-week postpartum efavirenz concentration ≥1 μg/mL. The remaining 28 either were not receiving efavirenz or did not have pharmacokinetic studies performed. Women with efavirenz concentrations that were detectable but <1 μg/mL were more likely to be extensive metabolizers (6 of 9, 66%) than either women with undetectable concentrations (2 of 8, 25%) or women with concentrations >1 μg/mL (9 of 58, 16%; χ2P = 0.003).
Before or at delivery, 90 women (87%) had an HIV RNA <400 copies per milliliter, whereas at 12 months postpartum, 73 (71%) women had an HIV RNA <400 copies per milliliter (difference from prepartum to 12 months postpartum; McNemar test; P < 0.001). In univariable analysis, age ≥30 years [relative risk 1.3; 95% confidence interval (CI): 0.98 to 1.4], detectable postpartum efavirenz estimated trough concentration (compared to undetectable efavirenz, detectable but <1 μg/mL relative risk: 5.3; 95% CI: 0.80 to 35, and >1 μg/mL relative risk: 6.6; 95% CI: 1.0 to 41), prepartum HIV RNA <400 copies per milliliter (relative risk: 5.1; 95% CI: 2.8 to 6.2), and a depression score <10 (relative risk: 1.8; 95% CI: 1.1 to 4.3) were associated with a suppressed 12 month HIV RNA. In multivariable modeling, the P value for depression increased to 0.1; all other P values remained ≤0.05; all factors with univariable significance were included in the multivariable model (Table 2). Compared to the 60 women not on TB treatment, the 43 women receiving TB treatment during pregnancy were no different in 12-month postpartum suppression (P = 0.8). Self-reported timing of HIV diagnosis, prepartum CD4 T-cell count, parity, and CYP2B6 genotype were not significantly associated with virologic suppression at 12 months in univariable analysis.
We assessed the effect of virologic failure on CD4 T-cell count decline. From delivery to 12 months postpartum, women with a suppressed HIV RNA had a median CD4 T-cell count increase of 180 cells per cubic millimeter (IQR: 80–276) to 480 cells per cubic millimeter (IQR: 374–625) compared with a median CD4 T-cell count decline of 13 cells per cubic millimeter (IQR: −66–140) to 278 cells per cubic millimeter (IQR: 159–443) among women without virologic suppression (Kruskal–Wallis; P < 0.001).
Plasma samples from 29 of 30 women with virologic failure at 12 months underwent consensus HIV sequencing. Of the 29 samples, 26 were able to be amplified. The median duration on ART at the time of resistance testing was 502 days (IQR: 464–579). The ART regimens at the time of resistance testing were tenofovir and lamivudine for 23, stavudine and lamivudine for 2, and zidovudine and lamivudine for 1 combined with efavirenz for 21, nevirapine for 3, and lopinavir/ritonavir for 2. Twelve (46%) had major resistance mutations: all 12 had NNRTI resistance (K103N, M106V, V108I, Y188H, and/or P225H), and 4 (15%) had an M184V mutation, of which one also had a K65R mutation (Table 3). The 2 women with drug resistance results who were receiving lopinavir/ritonavir had previously been prescribed an NNRTI; both had NNRTI resistance. Among the 26 women with virologic failure and resistance testing results, resistance was associated with never achieving an HIV RNA <400 copies per milliliter while pregnant (χ2P = 0.002) but was not associated with duration on ART, receiving TB treatment, CYP2B6 genotype, age, or parity (all χ2P > 0.2).
We observed a 29% rate of virologic failure among HIV-infected women retained in care 12 months postpartum, including many women who achieved HIV suppression during pregnancy (when including those lost from the study as virologic failures, the proportion is 36% with failure). Furthermore, treatment failure had a severe, negative impact on CD4 cell count recovery and was associated with development of HIV drug resistance. Although multiple factors could plausibly be associated with postpartum treatment failure in this population, we identified undetectable postpartum efavirenz drug concentrations, failure to achieve virologic suppression while pregnant, and depression as the most important. Virologic failure was not explained by rapid efavirenz metabolizer status, pointing to poor adherence as the main cause for subtherapeutic levels. Depression may have contributed to virologic failure through decreasing adherence. Unfortunately, virologic failure resulted in HIV drug resistance and clinically important CD4 T-cell count reductions.
The decline in HIV RNA suppression from 87% prepartum to 71% at 12 months postpartum is substantial and concerning. It is unclear what factors are most important in explaining why women who had excellent treatment response during pregnancy are unable to maintain adherence postpartum. One factor could be the added stresses and responsibilities of caring for a newborn.10 Another possibility is the transition from receiving ART through prenatal care to routine adult ART care. Depression may have contributed to poor adherence, as has been previously reported.11,25,26 Financial constraints, employment, stigma, and disclosure may also have reduced adherence but were not evaluated.10,27 Notably, we did not observe an association with either TB treatment or having initiated ART before pregnancy and treatment failure. Undetectable concentrations were almost certain because of missed doses, whereas low but detectable concentrations were likely due to rapid efavirenz clearance in CYP2B6 extensive metabolizers (an undetectable efavirenz concentration suggests that efavirenz had not been taken for at least one or 2 weeks before sampling).28
Several studies have reported virologic outcomes in adult HIV treatment cohorts29 and in PMTCT cohorts.30 In an analysis from South Africa, 9.9% of the patients who achieved virologic suppression on first-line ART during pregnancy developed virologic failure, a median of 16 months from ART initiation.27 This finding from a single-site nongovernmental organization managed adult ART, and PMTCT clinic is lower than our finding of 16% failure among those with prepartum virologic suppression. It is plausible that having more resources than are generally available in public sector primary care clinics contributed to the difference. A study from Kenya reported a total of 21% with virologic failure at 24-weeks postpartum but did not provide further analysis by prepartum suppression.30 These findings are slightly better than the 29% failure we observed at 12 months. Notably, these outcomes are worse than recent reports from general adult ART clinics ranging from 9% to 16% virologic failure.31–33 Other pregnancy-related studies have reported a high rate of loss from follow-up but have not included virologic outcomes.14,16,34
Antiretroviral drug resistance is expected with virologic failure. The observed proportion of women with treatment failure with identified resistance mutations (46%) is on the lower side of what has been reported regarding resistance during treatment failure in general adult ART populations. Studies reporting on resistance from African ART programs in which either virologic monitoring was available or cross-sectional 12 month studies were performed have reported NNRTI resistance among 44%–100% of patients with treatment failure and the M184V mutation among 37%–64% of patients with virologic failure.35–40 The difference in resistance in our study compared to other reports could be a result of a more abrupt discontinuation of ART after pregnancy as opposed to intermittent adherence or a result of a shorter duration of undiagnosed treatment failure in our cohort.
Our estimate of depression among 12% of the study cohort is consistent with some, but not all studies of depression among people living with HIV in sub-Saharan Africa. A large instrument validation study reported that 12.6% of HIV-infected individuals in care met criteria for depression.41 Other studies have reported between 8% and 38% with depression.42–44 It is unclear whether instrument, rater, or population accounts for this wide variation in estimated depression.
Our study's major strength is that it was a prospective cohort study with regular study visits that included sample storage. Limitations of the study include a cohort enriched in women with TB who may not reflect the general population of HIV-infected pregnant women. In addition, we lacked good measures of adherence and lacked data from clinic visits and adherence interventions at the primary care clinics from which the participants received PMTCT and ART care; we also lacked precision when virologic failure occurred because of the availability of HIV RNA data from discrete time points. Furthermore, for this analysis, we intentionally focused on women reporting continued ART care. Thus, our virologic suppression analysis does not include treatment failure due to disengagement from care, potentially underestimating the total decline in ART success between the pregnancy and the postpartum period.
ART as part of prenatal care can be successfully delivered with a high level of virologic suppression. However, sustaining success in the postpartum period is a challenge that requires specific attention and research focus so that evidence-based interventions to improve outcomes can be developed. Care transitions, such as from prepartum to regular adult ART care, are inevitable. However, as presently configured, this transition is unsuccessful among a large proportion of women, as evidenced by treatment failure, immunologic decline, and HIV drug resistance. There is a need to implement effective retention in care strategies that address patient, clinic, and systems-level factors that currently lead to poor outcomes. These strategies may include managing structural barriers, time constraints, and improved diagnosis and management of depression. PMTCT programs need improved approaches to manage the transition of care for women postpartum to ensure the well-being of the mothers (and infants) postpartum.
1. Johnson L. Access to antiretroviral treatment in South Africa, 2004-2011. South Afr J HIV Med. 2012;13:22–27.
2. Johnson LF, Mossong J, Dorrington RE, et al.. Life expectancies of South African adults starting antiretroviral treatment: collaborative analysis of cohort studies. PLoS Med. 2013;10:e1001418.
3. Larson E, Bendavid E, Tuoane-Nkhasi M, et al.. Population-level associations between antiretroviral therapy scale-up and all-cause mortality in South Africa. Int J STD AIDS. 2013;25:636–642.
4. Bendavid E, Holmes CB, Bhattacharya J, et al.. HIV development assistance and adult mortality in Africa. JAMA. 2012;307:2060–2067.
5. Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection. Geneva, Switzerland: World Health Organization; 2014.
6. The South African Antiretroviral Treatment Guidelines; PMTCT Guidelines Revised March 2013. South African National Department of Health: Pretoria, South Africa; 2013.
7. Motsoaledi A. Health Budget Vote Speech. Department of Health SA. ed, Pretoria, South Africa: Government of South Africa; 2014.
8. Ahmed S, Kim MH, Abrams EJ. Risks and benefits of lifelong antiretroviral treatment for pregnant and breastfeeding women: a review of the evidence for the Option B+ approach. Curr Opin HIV AIDS. 2013;8:474–489.
9. Clouse K, Schwartz S, Van Rie A, et al.. “What they wanted was to give birth; nothing else”: barriers to retention in option B+ HIV care among postpartum women in South Africa. J Acquir Immune Defic Syndr. 2014;67:e12–e18.
10. Ngarina M, Popenoe R, Kilewo C, et al.. Reasons for poor adherence to antiretroviral therapy postnatally in HIV-1 infected women treated for their own health: experiences from the Mitra Plus study in Tanzania. BMC Public Health. 2013;13:450.
11. Nachega JB, Uthman OA, Anderson J, et al.. Adherence to antiretroviral therapy during and after pregnancy in low-income, middle-income, and high-income countries: a systematic review and meta-analysis. AIDS. 2012;26:2039–2052.
12. Wettstein C, Mugglin C, Egger M, et al.. Missed opportunities to prevent mother-to-child-transmission: systematic review and meta-analysis. AIDS. 2012;26:2361–2373.
13. Kaplan R, Orrell C, Zwane E, et al.. Loss to follow-up and mortality among pregnant women referred to a community clinic for antiretroviral treatment. AIDS. 2008;22:1679–1681.
14. Tenthani L, Haas AD, Tweya H, et al.. Retention in care under universal antiretroviral therapy for HIV-infected pregnant and breastfeeding women (“Option B+”) in Malawi. AIDS. 2014;28:589–598.
15. Wang B, Losina E, Stark R, et al.. Loss to follow-up in a community clinic in South Africa–roles of gender, pregnancy and CD4 count. S Afr Med J. 2011;101:253–257.
16. Phillips T, Thebus E, Bekker LG, et al.. Disengagement of HIV-positive pregnant and postpartum women from antiretroviral therapy services: a cohort study. J Int AIDS Soc. 2014;17:19242.
17. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression
. Development of the 10-item Edinburgh Postnatal Depression
Scale. Br J Psychiatry. 1987;150:782–786.
18. Hoffmann CJ, Fielding KL, Johnston V, et al.. Changing predictors of mortality over time from cART start: implications for care. J Acquir Immune Defic Syndr. 2011;58:269–276.
19. Marzolini C, Telenti A, Decosterd LA, et al.. Efavirenz plasma levels can predict treatment failure and central nervous system side effects in HIV-1-infected patients. AIDS. 2001;15:71–75.
20. Dooley KE, Denti P, Martinson N, et al.. Pharmacokinetics of efavirenz and treatment of HIV-1 among pregnant women with and without tuberculosis Coinfection. J Infect Dis. 2014;211:197–205.
21. Holzinger ER, Grady B, Ritchie MD, et al.. Genome-wide association study of plasma efavirenz pharmacokinetics in AIDS Clinical Trials Group protocols implicates several CYP2B6 variants. Pharmacogenet Genomics. 2012;22:858–867.
22. Haas DW, Severe P, Jean Juste MA, et al.. Functional CYP2B6 variants and virologic response to an efavirenz-containing regimen in Port-au-Prince, Haiti. J Antimicrob Chemother. 2014;69:2187–2190.
23. Wensing AM, Calvez V, Gunthard HF, et al.. 2014 update of the drug resistance mutations in HIV-1. Top Antivir Med. 2014;22:642–650.
24. Zhang J, Yu KF. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280:1690–1691.
25. Nel A, Kagee A. The relationship between depression, anxiety and medication adherence among patients receiving antiretroviral treatment in South Africa. AIDS Care. 2013;25:948–955.
26. Collins PY, Holman AR, Freeman MC, et al.. What is the relevance of mental health to HIV/AIDS care and treatment programs in developing countries? A systematic review. AIDS. 2006;20:1571–1582.
27. Clouse K, Pettifor A, Shearer K, et al.. Loss to follow-up before and after delivery among women testing HIV positive during pregnancy in Johannesburg, South Africa. Trop Med Int Health. 2013;18:451–460.
28. Haas DW, Gebretsadik T, Mayo G, et al.. Associations between CYP2B6 polymorphisms and pharmacokinetics after a single dose of nevirapine or efavirenz in African Americans. J Infect Dis. 2009;199:872–880.
29. Barth RE, van der Loeff MF, Schuurman R, et al.. Virological follow-up of adult patients in antiretroviral treatment programmes in sub-Saharan Africa: a systematic review. Lancet Infect Dis. 2010;10:155–166.
30. Okonji JA, Zeh C, Weidle PJ, et al.. CD4, viral load response, and adherence among antiretroviral-naive breast-feeding women receiving triple antiretroviral prophylaxis for prevention of mother-to-child transmission of HIV in Kisumu, Kenya. J Acquir Immune Defic Syndr. 2012;61:249–257.
31. Denison JA, Koole O, Tsui S, et al.. Incomplete adherence among treatment-experienced adults on antiretroviral therapy in Tanzania, Uganda and Zambia. AIDS. 2015;29:361–371.
32. Fox MP, Cutsem GV, Giddy J, et al.. Rates and predictors of failure of first-line antiretroviral therapy and switch to second-line ART in South Africa. J Acquir Immune Defic Syndr. 2012;60:428–437.
33. Rusine J, Asiimwe-Kateera B, van de Wijgert J, et al.. Low primary and secondary HIV drug-resistance after 12 months of antiretroviral therapy in human immune-deficiency virus type 1 (HIV-1)-infected individuals from Kigali, Rwanda. PLoS One. 2013;8:e64345.
34. Centers for Disease C, Prevention. Impact of an innovative approach to prevent mother-to-child transmission of HIV–Malawi, July 2011-September 2012. MMWR Morb Mortal Wkly Rep. 2013;62:148–151.
35. Barth RE, Wensing AM, Tempelman HA, et al.. Rapid accumulation of nonnucleoside reverse transcriptase inhibitor-associated resistance: evidence of transmitted resistance in rural South Africa. AIDS. 2008;22:2210–2212.
36. Marconi VC, Sunpath H, Lu Z, et al.. Prevalence of HIV-1 drug resistance after failure of a first highly active antiretroviral therapy regimen in KwaZulu Natal, South Africa. Clin Infect Dis. 2008;46:1589–1597.
37. Hoffmann CJ, Charalambous S, Sim J, et al.. Viremia, resuppression, and time to resistance in human immunodeficiency virus (HIV) subtype C during first-line antiretroviral therapy in South Africa. Clin Infect Dis. 2009;49:1928–1935.
38. Hamers RL, Sigaloff KC, Wensing AM, et al.. Patterns of HIV-1 drug resistance after first-line antiretroviral therapy (ART) failure in 6 sub-Saharan African countries: implications for second-line ART strategies. Clin Infect Dis. 2012;54:1660–1669.
39. Ramadhani HO, Thielman NM, Landman KZ, et al.. Predictors of incomplete adherence, virologic failure, and antiviral drug resistance among HIV-infected adults receiving antiretroviral therapy in Tanzania. Clin Infect Dis. 2007;45:1492–1498.
40. Orrell C, Walensky RP, Losina E, et al.. HIV type-1 clade C resistance genotypes in treatment-naive patients and after first virological failure in a large community antiretroviral therapy programme. Antivir Ther. 2009;14:523–531.
41. Bhana A, Rathod SD, Selohilwe O, et al.. The validity of the patient health questionnaire for screening depression in chronic care patients in primary health care in South Africa. BMC Psychiatry. 2015;15:118.
42. Peltzer K, Szrek H, Ramlagan S, et al.. Depression and social functioning among HIV-infected and uninfected persons in South Africa. AIDS Care. 2015;27:41–46.
43. Yeji F, Klipstein-Grobusch K, Newell ML, et al.. Are social support and HIV coping strategies associated with lower depression in adults on antiretroviral treatment? Evidence from rural KwaZulu-Natal, South Africa. AIDS Care. 2014;26:1482–1489.
44. Abas M, Ali GC, Nakimuli-Mpungu E, et al.. Depression in people living with HIV in sub-Saharan Africa: time to act. Trop Med Int Health. 2014;19:1392–1396.