Impact of universal antiretroviral therapy for pregnant and postpartum women on antiretroviral therapy uptake and retention : AIDS

Secondary Logo

Journal Logo


Impact of universal antiretroviral therapy for pregnant and postpartum women on antiretroviral therapy uptake and retention

Abrams, Elaine J.a,b,c; Langwenya, Nontokozod,e; Gachuhi, Averiea; Zerbe, Allisona; Nuwagaba-Biribonwoha, Harrieta,c; Mthethwa-Hleta, Simangelef; Sahabo, Rubena; Lesosky, Maiad; Okello, Velephif; Myer, Landond,e

Author Information
doi: 10.1097/QAD.0000000000002027



Universal initiation of lifelong antiretroviral treatment (ART) for all HIV-positive pregnant and breastfeeding women (Option B+) was endorsed by the WHO in 2013 [1]. It was the first population-based WHO policy recommending universal ART, foreshadowing 2015 guidelines endorsing the ‘treat all’ approach for all individuals living with HIV [2,3].

The rationale for the shift to the Option B+ approach was predicated on the hypothesis that implementation of universal ART would overcome many of the impediments limiting effective implementation of Option A, where eligibility for treatment was determined based on immunologic and clinical status [2,4]. However, multiple challenges, including determining ART eligibility, initiating treatment, and effectively integrating prevention of mother-to-child transmission (PMTCT) and HIV care within maternal child health (MCH) services, diminished the efficacy of PMTCT services under Option A and MTCT rates remained high [5,6]. By providing a simplified approach for all HIV-positive pregnant and breastfeeding women, the Option B+ approach was hypothesized to streamline service delivery and ensure that all women receive effective therapy leading to improved maternal and infant health outcomes. The Option B+ approach has since been endorsed by 95% of low and middle-income countries (LMIC), and several countries have demonstrated rapid increases in the number of pregnant and postpartum women initiating ART with the transition from Option A to B+ [7,8].

Despite the expansion of Option B+, there is limited understanding of how this approach to universal ART for pregnant and breastfeeding women may influence key programmatic outcomes. Recent findings confirm superior efficacy of ART compared with zidovudine (ZDV) to prevent mother-to-child transmission among women with CD4+ cell counts greater than 350 cells/μl [9], but at the same time, significant concerns have emerged around retention among pregnant women initiating ART under the ‘test and start’ approach [10]. For example, pregnant women initiating ART under Option B+ in Malawi were almost five times more likely not to return for a follow-up visit and 1.6 times more likely to be lost in the first year compared with nonpregnant adults starting treatment [11]. There is an urgent need to examine retention outcomes with Options A and B+ that will provide important insights as countries transition their ART programs to provide universal ART to all individuals living with HIV.

The Kingdom of eSwatini has the highest HIV prevalence in Southern Africa: one in three pregnant women test HIV-positive [12]. In 2013, the national HIV program supported Option A but wanted to demonstrate the advantages of the Option B+ approach before changing national guidelines. This implementation science study was designed with the eSwatini Ministry of Health (MOH) to test the hypothesis that the change from Option A to Option B+ would improve ART uptake among pregnant women entering PMTCT, improve maternal retention during pregnancy and postpartum, and lead to fewer new paediatric HIV infections.


Study design

From 19 August 2013 to 30 October 2015, we conducted a stepped-wedge evaluation at 12 health facilities in eSwatini ( NCT01891799). Ten facilities providing Option A (the standard of care), were transitioned to offer Option B+ for HIV-positive pregnant women initiating PMTCT services. There was a staggered rollout with one facility transitioning routine PMTCT services from Option A to Option B+ approximately every 4 weeks. In addition, two facilities continuing Option A services were included as control sites to assess the impact of secular changes in service delivery over the duration of the study.


Working with the MOH, facilities were selected from two geographically contiguous regions, Manzini and Lubombo. To be eligible for inclusion, each facility provided PMTCT services, saw at least seven HIV-positive pregnant women not on ART each month, and was in the process of implementing the MOH's program of nurse-initiated ART services (NARTIS). Participating facilities included three hospitals, three large public health centres, and six small primary care clinics drawn from urban (n = 7), peri-urban (n = 2) and rural (n = 3) settings.

The order of site transition from Option A to Option B+ was selected by the MOH based on each facility's compliance with the NARTIS program. All health services including PMTCT and antenatal care (ANC) were provided by facility healthcare workers throughout the study period. The research team worked with facility staff to assist in the transition to Option B+ and stayed on-site for 4 weeks after the start of Option B+ to address implementation issues; thereafter the research team visited each site on a rotating schedule to provide mentoring and oversight.

During the study period, MOH guidelines endorsed the Option A approach to PMTCT services: ART [efavirenz (EFV) with lamivudine (3TC) with tenofovir (TDF)] during pregnancy and breast feeding for women meeting treatment eligibility criteria (WHO stage III/IV or CD4+ ≤350 cells/μl) with 6 weeks of daily nevirapine (NVP) for their infants and for healthier women (WHO stage I/II and CD4+ >350 cells/μl) twice daily zidovudine (ZDV) during pregnancy, single-dose NVP at delivery and 2 weeks ZDV with 3TC postpartum and daily infant NVP throughout breast feeding. Option B+ called for ART [EFV with 3TC or emtricitabine (FTC) with TDF] to all pregnant and breastfeeding women (irrespective of CD4+ cell count or WHO stage) with 6 weeks of daily infant NVP. Early infant HIV diagnosis (EID) using dried blood spots for DNA PCR was routinely performed for HIV-exposed infants (HEI) at 6 weeks of age or the earliest time thereafter. As part of the evaluation, at 6 months of age, an additional DNA PCR test was offered for all HIV-exposed infants receiving care at the 10 health facilities where Option B+ was implemented.

Participants and procedures

All pregnant women making their first ANC visit were eligible for inclusion if they were documented to be HIV-positive and not already using ART. For each eligible woman, routine patient-level data were abstracted from the first ANC visit through a minimum of 6 months postpartum; data on their infants were abstracted when infants could be traced to mothers on study. To assist with this abstraction, health workers at each site recorded in routine facility systems, which women initiated PMTCT services during the study period, and identified their infants in child health registers if possible. Using these registers and associated patient files, research staff abstracted data on ANC, HIV and ART care, postnatal care (PNC), and HEI services and EID results. Data included patient age, parity and gestation at first ANC visit, estimated date of delivery (EDD), previous HIV diagnosis, CD4+ cell count, WHO stage, antiretroviral use, and scheduled dates of facility attendance and future appointments. Quality of data was ensured through training and regular mentorship of data collection staff, database validation rules, monthly comparisons of key abstracted data with aggregate summary data for each facility, and routine structured assessments by study supervisors of agreement between paper and electronic study data and source data in facilities.


The primary outcome was maternal retention in care, defined as the proportion of women with any clinic attendance documented within 56 days of delivery or EDD (antenatal retention) and clinic attendance documented within 84 days of 6 months postpartum (postnatal retention). Definitions were based on 1 month after the longest possible ART-dispensing interval during the antenatal and postnatal periods (1 month and 2 months, respectively). Secondary outcomes included antenatal and postnatal retention analysed separately. In sensitivity analyses, we examined alternate retention definitions (STable 1,; alternate definitions did not alter results appreciably. Other outcomes included the proportion of women initiating antiretroviral medications during pregnancy, time from first ANC to ART initiation, and the proportions of infants testing HIV-positive.

Statistical analysis

Data were analysed using Stata (Stata Corporation, College Station, Texas, USA) and R (R Foundation, Vienna, Austria). Participant characteristics were summarized using proportions with 95% confidence intervals (CI) or medians with interquartile ranges (IQR); comparisons employed chi-squared and Kruskal–Wallis tests. In analysis, all eligible women making their first ANC visit at each site during each study month formed a cohort followed through at least 6 months postpartum. The main comparison of interest was between women enrolled under Option A vs. Option B+; data on those women enrolled during the transition month are reported in supplementary materials. The proportions of women retained at each site and study month were described using heatmaps without additional rescaling. Nonretention (loss to follow-up) from the first ANC visit was analysed using Kaplan–Meier methods with comparisons based on log-rank tests.

The frequencies of missing data on participant characteristics are described throughout. Multiple imputation using the bootstrap Expectation-Maximization (EM) algorithm was used to create 25 imputed datasets with a minimum chain length of 500; any variable transformations were carried out only after imputation [13]. To examine the impact of the change in PMTCT policy from Option A to Option B+, probit models using generalized estimating equations (GEE) with an exchangeable correlation structure were fit to each of the imputed datasets, with sites clustered on study month [14]. Sensitivity analyses were conducted using alternate correlation structures, and using only available data, and showed no substantive variation. From these models, coefficients and standard errors were pooled using Rubin's rules with results presented as risk ratios (RR) with 95% CI. Two-sided P values for coefficients were estimated from the pooled, imputed estimates via normal approximation.

Ethical approval

Ethical approval was provided by the Columbia University Medical Center Institutional Review Board, the eSwatini Scientific and Ethics Committee, and the University of Cape Town Human Research Ethics Committee. The requirement for informed consent from individuals was waived for abstraction of routinely collected data; mothers provided written consent for study-specific infant HIV testing.


Between 19 August 2013 and 29 August 2014, 10 094 women began ANC at 12 study health facilities; 4012 (40%) women were HIV-positive and 1180 (29%) were already on ART and thus ineligible. Of the remaining 2832 women (71%), 2518 (89%) were included on study (SFigure 1, Within this total, 171 women began ANC during a transition period and were excluded from the main analysis. Of the 2347 HIV-positive women included in the analysis, 55% (n = 1296) began ANC under Option A and 45% (n = 1051) under Option B+.

Table 1 describes women included overall and by PMTCT Option. At first ANC visit the mean age was 26.4 years, median gestation was 20 weeks (IQR, 15–24), and 51% of women tested HIV-positive for the first time. The median CD4+ cell count was 404 cells/μl and 34% of all women had a CD4+ cell count 350 cells/μl or less, meeting Option A ART eligibility criteria. Women enrolled under Options A and B+ were similar with respect to most demographic and clinical characteristics, but women under Option A were older with higher median CD4+ cell count, on average. Characteristics of the women entering care during transition months did not differ from women under Options A and B+ (STable 2,

Table 1:
Characteristics of HIV-positive women at first antenatal care visit by PMTCT approach at 12 health facilities in eSwatini, September 2013 to August 2014.

Overall, 95% of women received some form of antiretroviral drugs (Table 2). Cumulative ART initiation at any time from first ANC to 6 months postpartum was observed in 36% (n = 469) of women under Option A vs. 94% (n = 983) under Option B+ (P < 0.001). Most women (88%, n = 1137) under Option A initiated ZDV with subsequent switches to ART in approximately one-third (34%); 22% of these switches (n = 83) did not take place until after delivery. Under Option B+ ZDV use decreased to 2% of all women, and 5% (n = 51) received no therapy.

Table 2:
Antiretroviral use during pregnancy under Options A and B+, in all women and by CD4+ cell count.

Among women starting ART, the median time from first ANC visit to ART initiation decreased from 39 days (IQR, 12–112, range 0–436 days) under Option A to 0 days (IQR, 0–0, range 0–506 days) under Option B+; under Option B+, 86% of women initiated ART at the first ANC visit. Differences in ART uptake by PMTCT Option were similar in subgroups of women categorized by age and gestational age at first ANC visit (STable 3, Restricted to women with CD4+ cell count 350 cells/μl or less, the proportion ever initiating ART was 67% under Option A compared with 94% under Option B+ (P < 0.001; Table 2).

Levels of retention varied between sites and study months, but overall retention appeared higher under Option B+ compared with Option A (Fig. 1a). Table 3 shows retention during the overall evaluation period (from ANC enrolment through 6 months postpartum), the antenatal period, and the postnatal period for the entire study population, by PMTCT Option, and for subgroups by ART use and CD4+ cell count. Overall 39% (n = 912) of all HIV-positive women were retained from the first ANC visit to 6 months postpartum, 24% (n = 353) under Option A and 53% (n = 559) under Option B+.

Fig. 1:
Overall retention in care from first antenatal clinic visit through 6 months postpartum among HIV-infected women included in the evaluation of Option A vs. Option B+, by site and study month.Panel (a) shows retention outcomes for all women; Panel (b) shows outcomes restricted to women who initiated ART during the study. Sites A through J participated in the stepped-wedge evaluation; site K and L implemented Option A throughout the study period. The grey boxes forming a diagonal represent transition months from Option A (earlier in the study) to Option B+ (later in the study) at each site. The grey boxes in Panel (a), site K and Panel (b), sites D, J and K, indicate the months during which there were no new enrollments into antenatal care at the site.
Table 3:
Retention in care (primary definition) during total period, antenatal and postpartum periods.

During the antenatal period, 68% were retained in care through delivery; this proportion increased from 44% under Option A to 73% under Option B+. After adjusting for patient age, gestational age at first ANC, previous HIV diagnosis and CD4+ cell count, women under Option B+ were 32% more likely to be retained during the antenatal period compared with Option A (aRR 1.32; 95% CI 1.18–1.49; P < 0.001; Table 4). Retention worsened during the postnatal period (56% of all women were retained through 6 months postpartum), but a similar finding for increased retention under Option B+ (71%) vs. Option A (42%) was observed (Table 3). Differences in postnatal retention persisted in adjusted models (aRR 2.11; 95% CI 1.79–2.49, P < 0.0001; Table 4).

Table 4:
Results of univariable and multivariable modelling to predict maternal retention in care during the total evaluation period (from first antenatal are visit through 6 months postpartum) as well as the antenatal and postpartum periods.

These associations remained in subgroups of women with CD4+ cell count 350 cell/μl or less and did not appear to be modified by timing of HIV diagnosis, age, gestation at first ANC (STable 4,, or alternate definitions of retention (STable 5, and did not vary appreciably by clinic (SFigure 2A, Among other predictors examined, older age was consistently associated with increased retention during all periods (STable 6,

Associations between PMTCT Options and retention appeared to differ substantially when analyses were stratified by ART initiation during pregnancy (Fig. 1b). Under Option A, antenatal and postnatal retention appeared low among women receiving ZDV (52 and 10%, respectively) in comparison with women initiating ART during pregnancy (89 and 76%, respectively; Table 3). However, whereas the number and proportion of all HIV-positive women retained was substantially higher with Option B+, among women initiating ART, retention appeared lower under Option B+ compared with Option A overall (57 vs. 66%) and when restricted to the antenatal and postnatal periods (77 vs. 89% and 73 vs. 76%, respectively). These findings appeared consistent across sites (SFigure 2B, and after adjusting for women's demographic and clinical characteristics (aRR, 0.82; 95% CI 0.70–0.95; Table 4).

The record of an infant enrolled in care at a study facility was identified for slightly over half of the mothers included in the evaluation: 1324 (56%) infants were identified, 702 (53%) linked to women enrolled under Option A and 622 (47%) to Option B+ women. Infants were more likely to be identified among mothers who were retained in care: among women retained postpartum, 733 (62%) infants were identified, compared with 179 (39%) infants of women not retained postpartum (P < 0.001; STable 7, Overall 93% (1237) of the infants had a valid PCR test: 38 tests were positive (3% overall) with no differences between PMTCT approaches.


We evaluated the impact of changing from CD4+-guided ART eligibility, Option A, to universal immediate ART initiation, Option B+, among pregnant women in eSwatini. We found that transition to universal ART resulted in more than double the number of women initiating treatment and dramatically decreased the time to ART initiation from a median of 40 to 0 days. Furthermore, there was a marked improvement in retention with universal ART – the total number of women retained through 6 months increased by 150%, improving from 24% under Option A to 53% with Option B+. However, whereas fewer women initiated treatment under Option A, among those who did start ART, a higher proportion were retained compared with B+. Our findings provide valuable insights into the impact and potential limitations of universal treatment and ‘test and start’ as countries begin to implement and evaluate programs for universal ART initiation in all HIV-positive individuals.

The transition to Option B+ resulted in a rapid increase in pregnant women initiating ART from 36 to 94%. Option B+ ensured that all ART-eligible women (CD4+ cell count 350 cells/μl or less) began treatment, compared with only 67% under Option A. The practice of same day initiation dramatically reduced time to ART start increasing time on ART prior to delivery. Rapid ART initiation is particularly important during pregnancy as most in-utero infections occur late in gestation and viral load is the critical determinant of transmission risk [15]. Each additional week on treatment during pregnancy confers added protection against MTCT [16]. Option A eligibility screening and adherence preparation introduced systematic delays in ART initiation [5,6]. Option B+ eliminated these delays by offering same day initiation for all women resulting in better coverage, particularly for less healthy women at highest MTCT risk, and more days on treatment before delivery.

It is noteworthy that although the introduction of Option B+ improved maternal retention, retention was unacceptably low across both Options, particularly postpartum. Historically, retention in PMTCT care has been problematic and has emerged as a particular concern with Option B+ [10,11,17,18]. Although women initiating ART during pregnancy appear to be at higher risk for nonretention compared with nonpregnant ART-eligible adults, few studies have examined changes in retention rates and risks with the transition from Option A to Option B+ [18]. Here, the transition resulted in substantial improvements in women retained through 6 months postpartum, but overall retention remained alarmingly low. These findings were generally consistent across clinics and were not modified by maternal characteristics (timing of HIV diagnosis, age, gestation at first ANC visit, CD4+ cell count) or alternate definitions of retention.

We observed notable differences in patterns of retention across approaches. Both antenatal and postnatal retention were higher with Option B+ compared with Option A (73 vs. 44% and 71 vs. 42%, respectively) but when restricted to only those who initiated ART, retention was slightly better for Option A women (77 vs. 89% antenatally and 73 vs. 76% postnatally, for Options B+ and Option A, respectively). This is likely attributable to high rates of loss among women on ZDV and relatively low rates of ART initiation among ART-eligible Option A women. During Option A, one-third of women with CD4+ cell count 350 cells/μl or less did not start treatment, and only 5% of women on ZDV were retained in care through 6 months postpartum. Under Option A, women unprepared to initiate treatment drop out early in the testing to treatment cascade whereas with universal, same day ART initiation, the number of steps in the pretreatment cascade is reduced and opportunities for disengagement from care are delayed from pre-ART to post-ART initiation [19]. Finally, we also attribute the somewhat lower ART retention in B+ to the rapid increase in the number of women on ART. No new staff were added with the transition to B+, and clinics may have outgrown their capacity to effectively engage and retain all clients. In qualitative assessments, healthcare workers described increased workloads and documentation requirements associated with the transition to B+ [20]. These findings suggest that it will be critical to address increased demands on infrastructure and human resources as countries expand universal treatment to all individuals living with HIV.

Reasons for nonretention in PMTCT services are well described and include structural, social, and behavioural factors that influence a woman's decision to remain engaged [21]. Undocumented transfers and deaths may also account for some misclassification of retention status. In a study of women who missed PMTCT appointments in Malawi; among the 40% successfully traced, 4% had died; 30% had transferred to another clinic, and 54% had discontinued treatment [22]. In qualitative assessments, healthcare workers at our study facilities reported that women often migrate home from urban centres where the study was implemented to their rural homelands after delivery [20].

Only slightly over half of the women included in the evaluation had infants identified as enrolled in care at the study facilities. As might be expected, a higher proportion of infants were identified among women who were retained postpartum. The MOH reports that greater than 80–90% of HIV-exposed infants receive an HIV diagnostic test so we suspect that many children were tested at other facilities. The low rate of identification of infants born to study mothers reflects the complexity of trying to link mothers and babies using routine medical records in the absence of a formal linkage system [23].

By shifting the standard of care from Option A to Option B+, using a stepped wedge design, we were able to evaluate the impact of introducing a new model of care directly on maternal retention during pregnancy and postpartum. Women were not enrolled in the study and hence their engagement in care was not influenced by study participation. At the same time, we were unable to trace women no longer in care at the study facilities nor did we interview women to gain insights into their health-seeking behaviours or experiences with antiretroviral treatment. Future studies would benefit from including tracing and other approaches to account for undocumented transfers and healthcare engagement in facilities outside of study facilities. Other limitations of our study include the short duration of follow-up. In many settings, including eSwatini, breastfeeding and hence MTCT risk, continues for up to 2 years. Studies have demonstrated that periods of highest risk for nonretention are immediately after ART initiation and postdelivery, two periods included in this study. Finally, this study was conducted at 10 urban and peri-urban health facilities in a high HIV prevalence country with relatively well established HIV and PMTCT services and facilities were purposefully rather than randomly selected Therefore, our findings may not be generalizable to other settings.

In conclusion, our findings have important implications for health systems implementing universal treatment guidelines in high-prevalence contexts. Implementation of the ‘test and start’ approach using a once daily fixed-dose combination (FDC) ART regimen resulted in substantially more women initiating treatment, more time on treatment prior to delivery and more HIV+ women retained in care. The change in approach, however, only partially improved treatment outcomes; overall retention was well below global targets. There is an urgent need for additional interventions to improve retention and ensure that the full benefits of universal treatment are achieved.


The authors thank the eSwatini Ministry of Health, health facility staff, and study team for their support and contribution to the study.

E.J.A. and L.M designed the study. E.J.A., A.G., A.Z., H.N.B., V.O., and L.M. developed the protocol. E.J.A., A.G., A.Z., H.N.B., S.M.H., R.S. were responsible for study oversight. N.L., M.L., and L.M. conducted the data analysis. All authors were involved in the interpretation of the findings. E.J.A. with L.M. and N.L. wrote the first draft of the article. All authors commented extensively, revised the manuscript critically, and approved the final version.

Sources of support: This study is funded by the United States Agency for International Development (USAID), through the U.S. President's Emergency Plan for AIDS Relief (PEPFAR), USAID Award Number: AID-OAA-A-12–00020. This study was also supported in part by a research grant from Investigator-Initiated Studies Program of Merck Sharp & Dohme Corp.

Disclaimers: The opinions expressed in this article are those of the authors and do not necessarily represent those of the United States Government or Merck Sharp & Dohme Corp.

Conflicts of interest

There are no conflicts of interest.


1. World Health Organization (WHO). Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. Recommendations for a public health approach. June 2013. Available at:
2. World Health Organization (WHO). Early release guidelines on when to start antiretroviral therapy and on preexposure prophylaxis for HIV. September 2015. Available at:
3. World Health Organization (WHO). Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. Recommendations for a public health approach, 2nd ed. 2016 [cited 7 August 2017]. Available at:
4. 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; 8:473–488.
5. Wettstein C, Mugglin C, Egger M, Blaser N, Vizcaya LS, Estill J, et al. IeDEA Southern Africa Collaboration. Missed opportunities to prevent mother-to-child transmission: systematic review and meta-analysis. AIDS 2012; 26:2361–2373.
6. Sibanda EL, Weller IV, Hakim JG, Cowan FM. The magnitude of loss to follow-up of HIV-exposed infants along the prevention of mother-to-child HIV transmission continuum of care: a systematic review and meta-analysis. AIDS 2013; 27:2787–2797.
7. Kieffer MP, Mattingly M, Giphart A, van de Ven R, Chouraya C, Walakira M, et al. EGPAF Technical Directors Forum. Lessons learned from the early implementation of option B+: the Elizabeth Glaser Pediatric AIDS Foundation experience in 11 Africa countries. J Acquir Immune Defic Syndr 2014; 67 (Suppl 4):S188–S194.
8. Joint United Nations Programme on HIV/AIDS. Ending AIDS, progress towards the 90-90-90 targets, global AIDS update 2017. Available at:
9. Fowler MG, Qin M, Fiscus SA, Currier J, Flynn PM, Chipato T, et al. IMPAACT 1077BF/1077FF PROMISE Study Team. Benefits and risks of antiretroviral therapy for perinatal HIV prevention. N Engl J Med 2016; 375:1726–1737.
10. Knettel BA, Cichowitz C, Ngocho JS, Chumba LN, Mmbaga BT, et al. Retention in HIV care during pregnancy and the postpartum period in the Option B+ era: systematic review and meta-analysis of studies in Africa. J Acquir Immune Defic Syndr 2018; 77:427–438.
11. Haas AD, Tenthani L, Msukwa MT, Tal K, Jahn A, Gadabu OJ, et al. Retention in care during the first 3years of antiretroviral therapy for women in Malawi's option B+ programme: an observational cohort study. Lancet HIV 2016; 3:e175–e182.
12. UNAIDS. Global plan country fact sheet. 2017. Available at:
13. Rubin DB. Multiple Imputation for nonresponse in surveys. New York: John Wiley & Sons; 1987.
14. Hardin JW, Hilbe JM. Generalized estimating equations. 2nd ed.Boca Raton, FL: Chapman & Hall/CRC; 2012.
15. Kourtis AP, Lee FK, Abrams EJ, Jamieson DJ, Bulterys M. Mother-to-child transmission of HIV-1: timing and implications for prevention. Lancet Infect Dis 2006; 6:726–732.
16. Myer L, Phillips TK, McIntyre JA, Hsiao NY, Petro G, Zerbe A, et al. HIV viraemia and mother-to-child transmission risk after antiretroviral therapy initiation in pregnancy in Cape Town, South Africa. HIV Med 2017; 18:80–88.
17. Miller K, Muyindike W, Matthews LT, Kanyesigye M, Siedner MJ. Program implementation of Option B+ at a president's emergency plan for AIDS relief-supported HIV clinic improves clinical indicators but not retention in care in Mbarara, Uganda. AIDS Patient Care STDS 2017; 31:335–341.
18. Pfeiffer JT, Napúa M, Wagenaar BH, Chale F, Hoek R, Micek M, et al. A stepped wedge cluster randomized controlled trial to promote Option B+ retention in central Mozambique. J Acquir Immune Defic Syndr 2017; 76:273–280.
19. Chan AK, Kanike E, Bedell R, Mayuni I, Manyera R, Mlotha W, et al. Same day HIV diagnosis and antiretroviral therapy initiation affects retention in Option B+ prevention of mother-to-child transmission services at antenatal care in Zomba District, Malawi. J Int AIDS Soc 2016; 19:20672.
20. DiCarlo A., Gachuhi A., Peters Z.J., Zerbe A., Okello V., Mthethwa S., et al. Healthcare worker experiences with Option B+ for prevention of mother-to-child HIV transmission in Swaziland: Findings from a two-year follow-up study [Abstract: A_130]. 8th International Workshop on HIV Pediatrics 2016; 15–16 July
21. Colvin CJ, Konopka S, Chalker JC, Jonas E, Albertini J, Amzel A, et al. A systematic review of health system barriers and enablers for antiretroviral therapy (ART) for HIV-infected pregnant and postpartum women. PLoS One 2014; 9:e108150.
22. Tweya H, Gugsa S, Hosseinipour M, Sepight C, Ng’ambi W, Bokosi M, et al. Understanding factors, outcomes and reasons for loss to follow-up among women in Option B+ PMTCT programme in Lilongwe, Malawi. Trop Med Int Health 2014; 19:1360–1366.
23. Hamilton E, Bossiky B, Ditekemena J, Esiru G, Fwanba F, Goga AE, et al. Using the PMTCT cascade to accelerate achievement of the global plan goals. J Acquir Immune Defic Syndr 2017; 75 (Suppl 1):S27–S35.

HIV; Option B+; prevention of mother-to-child transmission; retention; universal antiretroviral therapy

Supplemental Digital Content

Copyright © 2019 Wolters Kluwer Health, Inc.