From Option B+ to Universal “Test and Treat” in Cameroon: Identification and Evaluation of District-level Factors Associated with Retention in Care

1. Introduction

The remarkable success of the global implementation of lifelong antiretroviral therapy (ART) to pregnant and breastfeeding women living with HIV (WLHIV) regardless of CD4 count and clinical stage (Option B+ or “test and treat” for pregnant and breastfeeding women), has raised prospects for the elimination of pediatric HIV infections.1 The number of pregnant and breastfeeding women on lifelong ART has rapidly increased from 47% in 2010 to 93% in 2017.2 Despite the success of Option B+ for the prevention of mother-to-child transmission of HIV (PMTCT), there are persistent concerns that poor retention of mothers and their infants in PMTCT services may compromise the gains made in eliminating mother-to-child transmission of HIV. The 2020 report by the Joint United Nations Programme on HIV/AIDS (UNAIDS) revealed that the interim milestones for achieving epidemic control by 2030 were not met.3 For example, the 2020 target of reducing the number of new HIV infections among children to fewer than 20,000 was missed. According to the UNAIDS and partners, 1.7 million children aged 0-14 years were living with HIV in 2020, among which were 150,000 new HIV infections.4 These data emphasize the need for a greater focus on key populations that are most at risk of HIV infection. This prompted UNAIDS, the United Nations Children’s Fund (UNICEF), and the World Health Organization (WHO) to launch a new Global Alliance to end AIDS in children by 2030, with Cameroon as one of the 12 starting members. One of the four focus areas for the Alliance is to reduce the treatment gap for pregnant and breastfeeding adolescent girls and WLHIV. Adolescent girls and young women in sub-Saharan Africa (SSA), the region with the highest global HIV burden, have been shown to be vulnerable to new HIV infections.5,6 It is likely that the poor retention of women in PMTCT services contributed to the missed 2020 targets. To support this claim, a systematic review of studies in Africa found an estimated retention rate of 76.4% among pregnant and breastfeeding women 12 months after initiating Option B+.7 Also, a cross-sectional survey of women in PMTCT care in four SSA countries, including Cameroon, showed that only 36.4% of pregnant and breastfeeding women completed all steps of the cascade.8 Other studies have examined factors associated with retention in PMTCT care, but few comparative studies have been conducted at the district level.9-11 Factors influencing retention in PMTCT services are reported to differ by the referral levels and settings of healthcare facilities, including urban and rural settings.9 In this study, we evaluated factors associated with retention in an urban and rural health district of Cameroon.

2. Methods

Study design, setting, and participants. This was a retrospective study of the early Option B+ program implemented in Cameroon from October 2013 through July 2014 in 22 health facilities. The health facilities included 12 in the Bamenda urban and 10 in the Kumba rural health districts of the Northwest and Southwest regions, respectively. This study used a convenience sample and a retrospective design that is prone to missing data, therefore limiting the generalizability of the results. The Option B+ pilot program and health facilities were described previously.14

Data collection. From September 1, 2015, to February 29, 2016, the first author collected longitudinal data from program and laboratory registers of mothers and their HIV-exposed infants. De-identified sociodemographic and clinical characteristics, ART initiation dates, and study exit dates when an infant attained 18 months were abstracted. The same abstractor abstracted data directly into a standardized Microsoft Excel spreadsheet, using client and facility-level identifiers for coding. The program data analyst cleaned the dataset and cross-examined it before further analysis.

We ensured the reliability (retest) of data collection by using the same abstractor to abstract data from uniform longitudinal program registers into a standardized spreadsheet.

Exclusion and inclusion criteria. All pregnant and breastfeeding WLHIV who were ART naïve prior to ART initiation were included in the study. We also included women transferred out, presumed to be receiving care in another facility.

3. Results

Out of the 560 reviewed records of pregnant and breastfeeding women, 174 (31.1%) were not in care at 12 months post-ART initiation. Of these, 26 (4.6%) stopped treatment mostly due to ARV side effects, 19 (3.4%) died, and 129 (23.0%) were LTHF. The analytical sample included 35 women (6.3%) who were transferred out.

3.2. Main Variable Results

Overall, the retention rates among the 560 women were 82.0% (95% CI: 0.79-0.85) and 68.9% (95% CI: 0.65-0.73) through 6 and 12 months post-ART initiation, respectively. The mean age-adjusted survival curves revealed higher, albeit non-significant (LRT test p-value = 0.187), retention rates among women in the Bamenda Urban Health District compared to the Kumba Rural Health District (Figure 1).

Close

Figure 1

Cox Regression mean age-adjusted survival curves of retention probabilities; a. Overall estimated retention probability b. Estimated retention probabilities of women by health district.

Export to PPT

4. Discussion

Overview. Overall, we found retention rates among pregnant and breastfeeding women in PMTCT services of 82% and 70% through 6 and 12 months after ART initiation, respectively. However, reports from the same cohort of women in a prospective study found higher retention rates of 90% and 79% through 6 and 12 months after ART initiation, respectively.14 In both studies, there was a progressive decline in rates over time from 6 to 12 months which is in line with reported attrition along the PMTCT cascade.16

Comparison of retention rates and study design. The study design may partly explain the differences in retention rates between studies: retrospective studies are prone to missing data. This may have been the case, seeing our smaller analytical sample of 560 women compared to 669 in the prospective study. In addition, researchers may use different methods to calculate retention rates, which might give different values. For example, our denominator included women who transferred out; whereas, in the prospective study, the authors excluded women who transferred out from their analysis. Similarly, another prospective study of the Option B+ approach in the Kumba health district of Cameroon reported a decline in retention rates from 88% to 81% at 6 and 12 months, respectively. This study examined a much smaller sample of 268 women.17 Attrition along the PMTCT cascade remains a major threat to achieving the elimination of MTCT of HIV in SSA, and LTHF appears to be the main attrition type.18,19 In our study, LTHF accounted for 23.0% of all cases of attrition. Our rate was lower than the 30.5% reported in Uganda at 18 months post-delivery among mother-baby pairs in PMTCT services.20

Collectively, the extant literature highlights additional challenges in comparing retention rates between studies because of differences in the duration of follow-up and setting — urban or rural. Some studies in Cameroon have shown differences in retention rates among women in PMTCT services between urban and rural settings.21 In this study, we report higher but non-significant statistical differences in retention rates in the Bamenda urban compared to the Kumba rural health district. Other authors in Uganda found that the proportion of mother-infant pairs on lifelong Option B+ was five times greater in urban (20.3%) than in rural (4.5%) health facilities.22 In contrast, a study in Rwanda assessing retention rates and their associated individual- and facility-level factors showed that women receiving PMTCT services in rural facilities were more likely to be retained in care than those in urban facilities.9 Health district setting could, therefore, influence retention rates.

Factors associated with retention. In our adjusted regression model, two factors remained significantly associated with higher retention in PMTCT services: the participant’s knowledge of their male partner’s HIV-negative status and early gestational age at ANC enrollment. A study of factors associated with retention in PMTCT care in Cameroon found lack of knowledge of the male partners’ HIV status among women as a barrier to retention in care.10 This is consistent with our findings of significantly higher retention of women who knew their male partners’ HIV-negative status in the urban health district. Therefore, not knowing their male partners’ HIV status may suggest a lack of engagement of male partners to facilitate disclosure of HIV status. Male partner involvement in ANC and PMTCT services has been widely recognized as crucial in strengthening PMTCT services; and yet, most reports show low male partner involvement in PMTCT services.23,24 Scaling up evidence-based interventions that enhance the involvement of male partners at ANC booking would be crucial to facilitating mutual HIV status disclosure and improving retention in PMTCT care.25 Similarly, booking early for ANC was strongly associated with retention in PMTCT care. Consistent with our results in the rural health district, a systematic review of studies in Africa showed that booking in ANC at a later gestational age led to poor retention in PMTCT care.7 These findings suggest the need for granular subnational data to inform Cameroon’s transition of PMTCT sites to universal lifelong HIV care centers.

Interestingly, our adjusted model (Table 3) did not find factors such as maternal age10 and distance from health facility,26 which are commonly associated with retention in PMTCT care. Contrary to our findings, a study in Uganda found that women above 30 years of age had a 39% higher retention rate in care at 18 months after delivery when compared to those below 30 years of age.27 Another study in an urban setting in Cameroon showed that living far from the health facility was associated with poor retention in PMCTC care.10 Likewise, a study in Zambia found a 10% decrease in PMTCT uptake for every kilometer increase in distance from the health facility.28

5. Conclusion and Global Health Implications

Overall, retention rates substantially declined from above three-quarters to nearly two-thirds at 6 and 12 months after initiating ART, respectively. This decline underscores the need for improved strategies to address retention in PMTCT services, which have significant implications for global efforts to eliminate mother-to-child transmission of HIV. These findings demonstrate the need for tailored interventions at the district level to enhance retention in PMTCT care in the era of universal “test and treat.” Identifying factors influencing retention at the district level is crucial for Cameroon and other countries with similar health systems and challenges, thereby contributing to more effective PMTCT services and policies worldwide.