CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

908

Comparison of Models of Care to Promote Postpartum Viral Suppression in South African Women Mustafa Shuaib 1 , Tamsin Philips 1 , Jasantha Odayar 1 , Thokozile R. Malaba 1 , Elaine J. Abrams 2 , Landon Myer 1 1 University of Cape Town, Cape Town, South Africa, 2 Columbia University, New York, NY, USA Background: Maintaining viral suppression (VS) in postpartum women on antiretroviral therapy (ART) is a major concern. There is significant interest in both integrated models of service delivery for maternal and child health (MCH), and in differentiated service delivery models (DSD) postpartum, but there are few rigorous data comparing these intervention strategies. Methods: We conducted a secondary analysis of individual patient trial data from Cape Town to compare head-to-head (i) an integrated MCH model with maternal and child care co-located and co-scheduled (MCHART; NCT01933477) versus (ii) a DSD model with mothers referred to community based "adherence clubs" for maternal ART dispensing (PACART; NCT03200054). Data for both interventions came from RCTs conducted in the same primary care community health service comparing each intervention to the local standard of care (SOC, referral to general adult ART services). Both trials measured demographic and behavioural covariates using the same tools; study viral load testing was conducted by trial personnel separate to routine antenatal ART (tenofovir+lamivudine+efavirenz). Analyses used frequentist network methods via generalised linear mixed models to compare VS (<50 copies/ml) under each model of care at 6m and 12m postpartum using the SOC as the reference; results are reported as odds ratios (OR) with 95% CI. Results: A total of 882 women (mean age: 29y; median time postpartum at enrolment, 1w [IQR, 0.6-1.9]) were included: 471 in MCHART and 411 in PACART. Follow-up through 12m was >85% in both trials; in women retained, VS was achieved by 299/375(80%) and 289/349(83%) women at 6m, and by 229/336(68%) and 231/329(70%) women at 12m, postpartum in MCHART and PACART, respectively. VS in integrated MCH at 6m and 12m was higher than in the DSD model (6m: 88% vs 87%; 12m: 80% vs 74%); VS under the SOC was higher in PACART than MCHART (6m: 79% vs 71%; 12m: 67% vs 55%, both p<0.05). In network comparisons, integrated MCH in MCHART was associated with significantly higher levels of VS compared to the DSD model in PACART at 12m postpartum (aOR 2.48; 95%CI 1.25-4.95; Table); results were consistent at VL<1000 copies/mL and robust across a range of sensitivity analyses. Conclusion: These novel data comparing two postpartum interventions in the same community suggest that integration of ART and postpartum MCH services achieved higher levels of VS compared to referral of mothers to DSD models of care in this setting.

concordance index. Multiple imputation with chained equations was used to address missing values. Predictors with nonzero coefficients in the LASSO regression model were selected for calculating predicted risk of a missed visit (>30 days) or elevated viral load ≥400 copies/ml). Backward selection with the imputed dataset was used to identify a final concise calibrated model. Model calibration and discrimination were assessed. Results: The final model resulted in 5 risk predictors of a future missed visit or elevated viral load, including baseline viral load> 1000 copies/ml, prior history of a missed visit, anticipated HIV stigma, low male partner support, and fair/ poor self-reported baseline ART adherence. The predicted risk scores ranged from 3.6% to 81.6%. The mean predicted probability of the outcome in the low-risk group, moderate-risk group, and high-risk group were 4.9%, 55.6%, and 68.0%, respectively. Model calibration was fair and intercept adjustment was done. The C-statistic by bootstrap cross validation indicated a strong model discrimination, 0.89 (95%CI 0.87, 0.91). The overall model performance was good (brier score = 0.14). Conclusion: The risk calculator will be validated in a separate sample. The risk calculator has the potential to be used in research and clinical care to identify PPWH at greatest need of interventions and enhanced support to achieve treatment success. Mentor Mothers! Community Model to Attain Elimination of Mother-to Child HIV Transmission, Tanzania Neema E Makyao 1 , Amani Maro 2 , Agnes Kosia 3 , Aaron Godwin 4 , Donatha Kayoza 2 , Amos Nyirenda 2 1 National Institute for Medical Research–Mbeya Medical Research Center, Dar es Salaam, United Republic of Tanzania, 2 Amref Health Africa, Dar es Salaam, United Republic of Tanzania, 3 San Francisco Department of Public Health, San Francisco, CA, USA, 4 Christian Social Services Commission, Dar es Salaam, United Republic of Tanzania Background: Despite the remarkable progress in Tanzania on ART provision to pregnant women living with HIV there are still significant number of new HIV infections among children. Data shows that progress in preventing HIV transmission from mother to child has almost stagnated. These can be linked with poor uptake of HIV testing, gaps in ART initiation, low retention rates and poor adherence to treatment. Social cultural factors associated with early pregnancy stages can also be linked to the above challenges. We aim to address these gaps through the use of a community peer model accepted and trusted by the community members. Methods: From January to June 2023 Amref implemented a community model using mentor mothers(MM) who are women living with HIV pregnant or breastfeeding champions identified from PMTCT clinics. Selection criteria included good adherence, willingness to disclose or have disclosed status and influential who can perform the task. We implemented it in 330 health facilities within 10 regions and 58 councils. A total of 947 MM who were trained, paired with Antenatal care supervisors in a health facility within their catchment areas to ensure early identification and linkage, throughout the cascade of care. Sites positivity rates of more than 5% in 2020 with HIV pregnant or breastfeeding attending PMTCT were identified. Each facility selected a total of 4 MM and one facility supervisor who were trained using MM National curriculum. Results: The implementation has increased early ANC bookings, couple counselling and HIV testing as well as increased number of women returned to care. From January to June 2023, we reached a total of 59,582 (93%) of pregnant women who were tested for HIV out of our target and 398 the positivity rate was 0.7%. We reached 51,818 women who tested for HIV. 540 were HIV+ which is 1.1%, their male partners reached were 36,888 and 395 were HIV positive which is 1.3% yield. Mother mentors were able to trace back a total of 219 out of 254 clients who were lost to follow up and linked back to services. To ensure retention a total of 1,839 beneficiaries from 289 groups are continuously supported and linked to psychosocial support groups and income-generating activities (PSAG). MM also identified 287 adolescents who are attending Conclusion: Mentor mothers model has improved follow up and retention to care for BFPW through out the continum of care, it is a community peer model accepted by the community and hence scale up will contribute to reaching eMTCT.

907

Poster Abstracts

909

Budget Impact Analysis of an Enhanced Retention Strategy for PMTCT Programs in Uganda Elly Nuwamanya 1 , Mohammed Lamorde 1 , Ronald M. Galiwango 1 , Tabitha Ayabo 1 , Dianah Namuddu 1 , Benjamin C Johnson 2 1 Infectious Diseases Institute, Kampala, Uganda, 2 The Johns Hopkins Hospital, Baltimore, MD, USA Background: Novel retention strategies have the potential to reduce mother to-child transmission and improve patient outcomes for people living with HIV. The enhanced retention strategy (ERS) of the DolPHIN-2 trial in Uganda achieved a retention rate of 92% among women receiving PMTCT services and reduced perinatal transmission (PT) from 2.8% (national rate) to 1.5% (DolPHIN cohort). This analysis built on the DolPHIN findings by estimating the budget impact of the ERS compared to the standard of care (SOC) approach for preventing PT among women initiating antiretroviral therapy (ART) in late pregnancy in Uganda. Methods: A budget impact analysis (BIA) was conducted from the payer (Uganda Ministry of Health) perspective with a 5-year time horizon. A

CROI 2024 284