Tremendous gains have been made worldwide to eliminate mother-to-child transmission of human immunodeficiency virus (HIV). From 2009 to 2015, the annual number of new pediatric HIV infections dropped by 60%; as a result, over 1.2 million new pediatric HIV infections were averted.1 The provision of universal antiretroviral therapy (ART) for pregnant and breastfeeding women—known as the “Option B+” approach—has played a critical role in the success of global programs for the prevention of mother-to-child HIV transmission (PMTCT). Initiated during pregnancy and breastfeeding and continued for life, ART can reduce vertical HIV transmission to less than 2%.2,3 There is strong evidence for improved maternal survival and decreased horizontal HIV transmission as well.4 Importantly, the Option B+ policy is aligned with international HIV treatment recommendations, which have moved toward universal ART for all HIV-infected adults and children.5 First implemented in Malawi in 2011,6 this strategy has since expanded to many settings of high HIV prevalence in sub-Saharan Africa.1
To date, assessments of Option B+ programs have focused largely on HIV-infected mothers. Many studies have emphasized programmatic indicators, such as uptake, retention and adherence.7–11 Others have focused on maternal health outcomes, including immune recovery.12 Surprisingly, there are comparatively few reports of infant health in the context of Option B+, especially for broader—and perhaps more informative—outcomes such as HIV-free survival.13 To address this gap in the medical literature, we evaluated the population-level impact of Option B+ services in Zambia via a large cohort of HIV-exposed infants and their HIV-infected mothers.
We targeted 33 sites in the peri-urban and rural settings of Zambia’s Lusaka Province. All participated in the Better Health Outcomes through Mentoring and Assessment (BHOMA) project, an initiative to improve the quality of primary care through community- and facility-based interventions.14 Like many African countries, PMTCT policy in Zambia has changed rapidly. In 2010, the Ministry of Health endorsed the World Health Organization’s “Option A” strategy: antenatal zidovudine monotherapy, peripartum nevirapine (with a week-long tenovofir/emtricitabine “tail”) and infant nevirapine prophylaxis during breastfeeding.15 Because of logistical challenges—and a growing desire to align PMTCT policy with general adult HIV guidelines—the Ministry of Health announced a policy shift to Option B+ in January 2013, with roll-out beginning the following year.16 Our study coincided with the start of Zambia’s adoption and implementation of Option B+.
We conducted a prospective cohort study as part of a larger initiative to measure PMTCT impact at the population level. Trained community health workers from our target facilities identified potential candidates for the study. HIV-infected women and their HIV-exposed newborns (≤30 days of life) were recruited at the health facility or via the community-based networks from the BHOMA project.14 Maternal HIV infection was confirmed through review of medical records (eg, antenatal card) or via on-site rapid HIV antibody testing by trained personnel. Eligible mother–infant pairs were given information about the study objectives and procedures; those wishing to participate provided written informed consent.
At enrollment, we collected baseline demographic and medical information. At subsequent visits—scheduled at 6 weeks, 6 months, 12 months and 18 months of infant life—mothers gave an update on their own health and that of their child, including HIV treatment status. Infants were tested for HIV via DNA polymerase chain reaction at enrollment, 6 weeks, 6 months and 12 months of life. At 18 months of life, in accordance with Zambian national guidelines, rapid HIV antibody tests were used. Children who initially tested HIV-positive had a repeat test done; those with confirmed HIV infection were immediately referred to the nearest HIV treatment facility. In an effort to maximize participant retention, study visits were conducted either at the clinic or at the household, based on the participant’s stated preference.
Our original target was 40 HIV-infected/-exposed mother–infant pairs per site or 1320 mother–infant pairs overall. This sample size was based on practical considerations (eg, duration of enrollment) and on what was deemed an acceptable level of precision within estimates of HIV-free survival at the community level (±11%). As the study began enrolling, however, we became aware of the significant heterogeneity in clinic volumes and HIV prevalence among sites. In some communities, full enrollment was completed in a matter of months; in others, only a few participants could be enrolled over the course of a year. Ultimately, we closed enrollment after 19 months (before reaching our original target) because of financial and time constraints related to project funding. Our study protocol was reviewed and approved by the University of Zambia Biomedical Research Ethics Committee (Lusaka, Zambia) and the University of North Carolina Institutional Review Board (Chapel Hill, NC).
The primary outcome of this cohort study was HIV-free survival among HIV-exposed infants, a metric that considers the direct (ie, preventing vertical HIV transmission) and indirect (ie, improving child survival) benefits often associated with PMTCT programs.13 In our study, infants were categorized as having met the study outcome if they died or were diagnosed with HIV (see above). Infant HIV-free survival represents the proportion of infants who did not meet these study endpoints—that is, they were alive and HIV-uninfected. We originally proposed to measure infant HIV-free survival at 18 months; however, over the course of the study, this was changed to 12 months to allow for extended recruitment at slowly enrolling sites.
We estimated overall infant survival and HIV-free survival using Kaplan–Meier analyses. We pooled data from all sites and then stratified our analysis by reported maternal ART use at enrollment and by district. We also calculated estimates for infant HIV-free survival for each participating community. Because of their relatively small sample sizes, we excluded 4 sites that enrolled fewer than 10 participants [Chongwe I (n = 6), Chongwe J (n = 7), Rufunsa C (n = 8) and Rufunsa E (n = 2)]. Further, we investigated potential factors associated with infant HIV infection or death, the 2 study endpoints used to generate infant HIV-free survival. We identified exposure variables that were associated with infant HIV infection or death at a P ≤ 0.05 level in univariate analyses. These were used to construct a multivariable Cox proportional hazards model. Robust standard errors were used to adjust for within-person clustering; maternal ART use was treated as a time-varying exposure.
Our secondary outcome was use of maternal antiretroviral drugs at 12 months, which was based on participant self-report. We calculated overall and site-specific proportions of reported antiretroviral drug use at each visit; missing values were considered nonuse. Similar to our primary outcome, we calculated a pooled overall estimate and then stratified by district and by community. Finally, we plotted site-level estimates for infant HIV-free survival and reported maternal antiretroviral drug use to evaluate the linear relationship between these 2 outcomes. All statistical analyses were conducted using Stata version 14.1 (College Station, TX).
From June 2014 to November 2015, we enrolled 827 HIV-infected mothers and their HIV-exposed infants across 33 communities in Lusaka Province (Fig., Supplemental Digital Content 1, https://links.lww.com/INF/D112). One hundred ninety-nine mother–infant pairs were enrolled in Chilanga District (5 sites), 325 were enrolled in Chongwe district (14 sites), 253 were enrolled in Kafue District (9 sites) and 50 were enrolled in Rufunsa District (5 sites). Of 827 enrolled mother–infant pairs, 36 (4.3%) were lost to follow-up or had withdrawn, 23 (2.8%) infants had died and 22 (3.0%) infants had a confirmed HIV infection in the first year of life (Fig., Supplemental Digital Content 2, https://links.lww.com/INF/D113). Nearly 80% reported an institutional delivery, either at a government facility or at a mission hospital or clinic. Almost all (99.4%) were breastfeeding at time of enrollment (Table, Supplemental Digital Content 3, https://links.lww.com/INF/D114). High rates of breastfeeding were reported throughout the follow-up period: 98.1% [95% confidence interval (CI): 96.8%–98.9%] at 6 weeks; 96.3% (95% CI: 94.6%–97.5%) at 6 months; and 84.8% (95% CI: 81.9%–87.4%) at 12 months.
In pooled Kaplan–Meier analysis, survival among HIV-exposed infants was 99.6% (95% CI: 98.9%–99.9%) at 6 weeks, 97.8% (95% CI: 96.5%–98.6%) at 6 months and 97.1% (95% CI: 95.7%–98.1%) at 12 months. HIV-free survival among HIV-exposed infants was 99.0% (95% CI: 98.0%–99.5%) at 6 weeks, 97.5% (95% CI: 96.1%–98.4%) at 6 months and 96.3% (95% CI: 94.8%–97.4%) at 12 months. In stratified analyses, we found that infants whose mothers reported ART use at enrollment had higher subsequent rates of infant HIV-free survival than those who did not (97.4% vs. 89.0%; P = 0.01; Table 1). Significant differences were also observed at the district level (P = 0.01; Table 1), with HIV-free survival ranging from 93.2% (95% CI: 89.1%–95.8%) in Kafue to 99.5% (95% CI: 96.4%–99.9%) in Chilanga. Infant HIV-free survival by community is shown in Figure 1 (P = 0.01).
In univariate analysis, several factors were associated with infant HIV infection or death, the two adverse outcomes used to generate HIV-free survival; these were included in our multivariable Cox proportional hazards model. Two continuous variables were associated with these study end points: years of education [adjusted hazard ratio (AHR): 0.89, 95% CI: 0.80–0.98] and parity (AHR: 0.81, 95% CI: 0.71–0.92). Among categorical variables, women who reported an institutional delivery were less likely to report infant HIV infection or death compared with those who delivered outside of a health facility (AHR: 0.48, 95% CI: 0.23–0.99). Marriage (AHR: 0.81, 95% CI: 0.32–1.35) and reported use of maternal ART (AHR: 0.52, 95% CI: 0.21–1.27) both trended towards a protective effect, but neither was statistically significant.
At the time of enrollment, 724 of 827 (87.6%) women reported using antiretroviral drugs. This figure declined gradually over time: 85.7% (95% CI: 83.2%–88.0%) at 6 weeks, 82.6% (95% CI: 79.8%–85.1%) at 6 months and 80.5% (95% CI: 77.7%–83.2%) at 12 months. In stratified analyses, the 12-month estimates for antiretroviral use was highest in Chilanga District (84.4%, 95% CI: 78.6%–86.4%), followed by Chongwe (82.4%, 95% CI: 77.9%–86.4%), Kafue (77.1%, 95% CI: 71.4%–82.1%) and Rufunsa Districts (70.0%, 95% CI: 55.4%–82.1%). At the community level, reported use of maternal antiretroviral drugs varied significantly at 12 months (P = 0.001), ranging from 55.6% (Chongwe L) to 100% (Chongwe E; Fig. 1). We were unable to identify a linear relationship between infant HIV-free survival and reported maternal antiretroviral drug use at 12 months (P = 0.65, Fig. 2).
In this rural, community-based cohort study, we provide empiric data about infant outcomes in the era of universal ART for pregnant and breastfeeding women. We found HIV-free survival to be generally high at 12 months of infant life. We observed significant heterogeneity in maternal antiretroviral drug use at 1 year postpartum but, at the site level, usage rates did not correlate with infant outcomes.
We measured infant HIV-free survival in our primary outcome, a metric that considers the effectiveness of PMTCT interventions alongside indirect infant survival benefits that may accompany enhanced health services.13 At 12 months of life, our pooled estimate was 96.3%; by community, they ranged from 93.2% to 99.5%. These rates are comparable to those observed in survey studies from sub-Saharan Africa. Buzdugan et al,17 for example, measured population HIV-free survival between 9 and 18 months of life across 157 communities in Zimbabwe. Before the introduction of the 2010 Option A strategy for PMTCT, HIV-free survival was estimated at 89.6% (95% CI: 87.1%–92.1%); 2 years after implementation, this had increased to 95.1% (95% CI: 93.6%–96.6%). In a nationally representative sample in Rwanda—before the implementation of more efficacious PMTCT regimens—HIV-free survival was measured at 91.9% (95% CI: 90.4%–93.3%) between 9 and 24 months.18 In a pilot program of universal combination antiretroviral regimens for HIV-infected pregnant women—conducted in Zambia’s Kafue District in 2009–2010—we observed an HIV-free survival of 89% (95% CI: 83%–94%) at 24 months.19 These 4 communities were also part of the current study.
Our findings are consistent with prior cohorts as well. In a meta-analysis by Chikhungu et al,20 pooled estimates from 10 studies showed a 12-month HIV-free survival of 89.8% (95% CI: 86.4%–93.2%). When limited to the 3 studies of lifelong maternal ART—including 2 in the setting of universal HIV treatment—the estimate was slightly higher at 91.8% (95% CI: 87.7%–95.9%). Similar to other studies,21,22 the high rates of continued breastfeeding observed in our cohort (84.8% at 12 months) likely contributed to the overall HIV-free survival.
At approximately 80%, our reported antiretroviral use at 12 months postpartum is similar to other regional studies, but below the “90-90-90” thresholds endorsed worldwide.23 In the Malawi national PMTCT program, for example, 76.8% (95% CI: 76.3%–77.3%) were retained in care at 12 months after ART initiation; at 24 months, this figure was estimated at 70.8% (95% CI: 70.3%–71.4%).8 Myer et al24 described similar trends in a cohort of South African women who initiated ART during pregnancy and attained virologic suppression (<50 copies/mL). At 12 months postpartum, 70% maintained virologic suppression; however, 8% had minor virologic episodes (<50–1,000 copies/mL), and 22% had at least one major episode (>1000 copies/mL).
We recognize the potential for biases with our self-reported measure of ART adherence. Despite attempts to minimize such reporting biases (eg, employment of community health workers, home study visits), participants may have inaccurately reported ART use out of perceived social desirability. Our study question (“are you taking antiretroviral drugs now?”) could have led to misclassification if antiretroviral drugs were confused with other prescribed medicines (eg, trimethoprim-sulfamethoxazole). Generally, such factors could lead to an over-reporting of ART use. They could also contribute to the observed differences in reported ART use in pregnancy and at time of enrollment. It should be noted that participants were asked the same question about ART use at each of the study visits. This approach enhanced the internal validity of the outcome, particularly when analyzing trends over time.
At the community level, we observed a significant heterogeneity in ART continuation rates in the 12 months after delivery. While the Option B+ policy was included in the Zambian PMTCT guidelines in 2013–2014, the implementation of services was not immediate, particularly in peri-urban and rural sites. Similar to other settings, the model of service delivery may have varied.25 In addition, implementation may have been slowed by existing health infrastructure constraints, often exacerbated by the increased demands of the Option B+ services. In an analysis of 42 health facilities in Lusaka Province, Mutale et al26 showed the variation in health systems performance across different domains (ie, patients and communities, human resources, service capacity, finance, governance, service provision, overall vision). Understanding ART use in the context of health system capacity could help to identify high-leverage points for implementation support.
At the individual level, use of combination antiretroviral regimens during pregnancy leads to improved infant HIV-free survival.19,27 As part of our analysis, we investigated whether a similar association could be demonstrated at the population level. Interestingly, in aggregate, we did not show relationship between these programmatic outcomes (ie, continued maternal ART use) and health outcomes (ie, infant HIV-free survival). With only 29 facilities and surrounding communities—our unit of analysis—we likely had insufficient statistical power, especially given the narrow range of HIV-free survival at the facility/community level. This highlights the challenge of translating individual-level associations to meaningful population-level metrics for monitoring and evaluation. Our reliance on self-reported adherence measures—and its potential for misclassification—may have contributed to this lack of association observed between maternal ART use and infant HIV-free survival.
We recognize several strengths of our study. In line with published guidance from the World Health Organization,28 we employed a prospective cohort approach to minimize recall and reporting biases. We maintained high patient retention (>95%) to address an important potential weakness of this design. We sought a “real world” perspective from nonurban settings; to accomplish this, we streamlined study activities and engaged participants at both the facility and community levels. We acknowledge important limitations as well. First, despite intensive efforts to identify and enroll eligible mother–infant pairs, we faced substantial recruitment challenges in some of the smaller and more remote facilities. Participant outcomes from these sites were underrepresented in our pooled analysis, which we addressed by reporting community-level estimates. Second, for logistical reasons, we truncated our primary outcome from 18-month to 12-month HIV-free survival. Lengthier follow-up could provide definitive early childhood outcomes for HIV, particularly if HIV testing were performed after breastfeeding cessation. Third, our population was likely enriched with women who sought institutional antenatal care and delivered at health facilities. Although we recruited from communities, we did not sample at the household level because of the intensive nature of such an approach. Finally, we did not conduct detailed adherence evaluations (eg, pill counts, pharmacy record reviews) nor did we obtain more objective measures of drug ingestion (eg, direct observation, drug levels), even in a subset of participants. Haas et al7 has shown the decreasing probability of maintaining adequate adherence (defined as ≥90%) over time—even among those reporting taking ART—and this should be considered when interpreting our results.
In summary, we demonstrated encouraging infant outcomes in the era of Option B+ in peri-urban and rural Zambia. Our estimates for infant HIV-free survival remained relatively stable, despite a wide range of maternal ART continuation at 12 months. Our study highlights the ongoing need for program monitoring and evaluation and its role for improving Option B+ services. Alternative approaches, including household surveys, may be more efficient and less time-consuming, but such methodologies require validation.
We thank Dr. Charles Msiska for his contributions to the study. This work was supported by the U.S. National Institutes of Health (R01 HD075131, K24 AI120796, K01 TW010857, T32 HD075731, P30 AI050410). No potential conflicts of interests have been declared by co-authors.
1. Joint United Nations Programme on HIV/AIDS. On the fast-track to an AIDS-free generation. Available at: http://www.unaids.org/sites/default/files/media_asset/GlobalPlan2016_en.pdf
. Accessed July 27, 2017.
2. Shapiro RL, Hughes MD, Ogwu A, et al.Antiretroviral regimens in pregnancy and breast-feeding in Botswana. N Engl J Med. 2010;362:22822294.
3. Fowler MG, Qin M, Fiscus SA, et alIMPAACT 1077BF/1077FF PROMISE Study Team. Benefits and risks of antiretroviral therapy for perinatal HIV prevention. N Engl J Med. 2016;375:17261737.
4. Cohen MS, Chen YQ, McCauley M, et alHPTN 052 Study Team. Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med. 2016;375:830839.
5. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach. 2016. 2nd edition, Available at: http://apps.who.int/iris/bitstream/10665/208825/1/9789241549684_eng.pdf
. Accessed August 27, 2017.
6. Schouten EJ, Jahn A, Chimbwandira F, et al.Is Option B+ the best choice? Lancet. 2013;381:12721273.
7. Haas AD, Msukwa MT, Egger M, et al.Adherence to antiretroviral therapy during and after pregnancy: Cohort Study on Women Receiving Care in Malawi’s Option B+ Program. Clin Infect Dis. 2016;63:12271235.
8. Haas AD, Tenthani L, Msukwa MT, et al.Retention in care during the first 3 years of antiretroviral therapy for women in Malawi’s option B+ programme: an observational cohort study. Lancet HIV. 2016;3:e175e182.
9. Chimbwandria F, Mhango E, Makombe S, et al.Impact of an innovative approach to prevent mother-to-child transmission of HIV—Malawi, July 2011–September 2012. MMWR Morb Mort Wkly Rep 2013;62:148151.
10. Atanga PN, Ndetan HT, Achidi EA, et al.Retention in care and reasons for discontinuation of lifelong antiretroviral therapy in a cohort of Cameroonian pregnant and breastfeeding HIV-positive women initiating ‘Option B+’ in the South West Region. Trop Med Int Health. 2017;22:161170.
11. Schwartz SR, Clouse K, Yende N, et al.Acceptability and feasibility of a mobile phone-based case management intervention to retain mothers and infants from an Option B+ program in postpartum HIV care. Matern Child Health J. 2015;19:20292037.
12. Miller K, Muyindike W, Matthews LT, et al.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;31:335341.
13. Stringer EM, Chi BH, Chintu N, et al.Monitoring effectiveness of programmes to prevent mother-to-child HIV transmission in lower-income countries. Bull World Health Organ. 2008;86:5762.
14. Stringer JS, Chisembele-Taylor A, Chibwesha CJ, et al.Protocol-driven primary care and community linkages to improve population health in rural Zambia: the Better Health Outcomes through Mentoring and Assessment (BHOMA) project. BMC Health Serv Res. 2013;13(suppl 2):S7.
15. Ishikawa N, Shimbo T, Miyano S, Sikazwe I, Ghidinelli MN, Syakantu GField effectiveness of WHO PMTCT guidelines in preventing postnatal HIV transmission in resource-limited settings: operational barriers and complexities related to the implementation of extended infant prophylaxis [Abstract TUPE177]. 19th International AIDS Conference, 2012.Washington DC,
16. Ishikawa N, Shimbo T, Miyano S, et al.Health outcomes and cost impact of the new WHO 2013 guidelines on prevention of mother-to-child transmission of HIV in Zambia. PLoS One. 2014;9:e90991.
17. Buzdugan R, Kang Dufour MS, McCoy SI, et al.Option A improved HIV-free infant survival and mother to child HIV transmission at 9-18 months in Zimbabwe. AIDS. 2016;30:16551662.
18. Ruton H, Mugwaneza P, Shema N, et al.HIV-free survival among nine- to 24-month-old children born to HIV-positive mothers in the Rwandan national PMTCT programme: a community-based household survey. J Int AIDS Soc. 2012;15:4.
19. Chi BH, Musonda P, Lembalemba MK, et al.Universal combination antiretroviral regimens to prevent mother-to-child transmission of HIV in rural Zambia: a two-round cross-sectional study. Bull World Health Organ. 2014;92:582592.
20. Chikhungu LC, Bispo S, Rollins N, et al.HIV-free survival at 12-24 months in breastfed infants of HIV-infected women on antiretroviral treatment. Trop Med Int Health. 2016;21:820828.
21. Marazzi MC, Nielsen-Saines K, Buonomo E, et al.Increased infant human immunodeficiency virus-type one free survival at one year of age in sub-Saharan Africa
with maternal use of highly active antiretroviral therapy during breast-feeding. Pediatr Infect Dis J. 2009;28:483487.
22. Cournil A, Van de Perre P, Cames C, et alKesho Bora Study Group. Early infant feeding patterns and HIV-free survival: findings from the Kesho-Bora trial (Burkina Faso, Kenya, South Africa). Pediatr Infect Dis J. 2015;34:168174.
23. Joint United Nations Programme on HIV/AIDS. 90-90-90: An ambitious treatment target to help end the AIDS epidemic. Available at: http://www.unaids.org/en/resources/documents/2017/90-90-90
. Accessed July 28, 2017.
24. Myer L, Dunning L, Lesosky M, et al.Frequency of viremic episodes in HIV-infected women initiating antiretroviral therapy during pregnancy: a cohort study. Clin Infect Dis. 2017;64:422427.
25. van Lettow M, Bedell R, Mayuni I, et al.Towards elimination of mother-to-child transmission of HIV: performance of different models of care for initiating lifelong antiretroviral therapy for pregnant women in Malawi (Option B+). J Int AIDS Soc. 2014;17:18994.
26. Mutale W, Godfrey-Fausset P, Mwanamwenge MT, et al.Measuring health system strengthening: application of the balanced scorecard approach to rank the baseline performance of three rural districts in Zambia. PLoS One. 2013;8:e58650.
27. Kesho Bora Study Group. Triple antiretroviral compared with zidovudine and single-dose nevirapine prophylaxis during pregnancy and breastfeeding for prevention of mother-to-child transmission of HIV-1 (Kesho Bora study): a randomised controlled trial. Lancet Infect Dis 2011; 11(3): 17180.
28. World Health Organization. A short guide on methods: measuring the impact of national PMTCT programmes. Available at: http://apps.who.int/iris/bitstream/10665/75478/1/9789241504362_eng.pdf
. Accessed July 27, 2017.