CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Methods: We conducted a retrospective analysis of retention at 6 months post-treatment initiation for all adults initiated on ART between October 2012 and July 2013, using aggregated facility-level data from iSanté, an electronic medical record system used by 89% of facilities delivering ART. Only facilities that had initiated individuals, both pregnant women and non-pregnant adults, on ART and had complete electronic data for the specified 6-month follow-up period were included in the analysis. Consistent with Ministry of Health definitions, 6 month- retention was defined as a medical appointment or pharmacy refill within 3 months of the 6 month post-ART initiation date. Using aggregate count data from facilities, we compared the cumulative incidence of 6 month retention between pregnant women and non-pregnant adults initiating ART. The Mantel-Haenszel method was used to adjust relative risk by type of health facility, sector (public, private, mixed), and location and to test for homogeneity of risk by these characteristics. Results: Between October 2012 and July 2013, 8262 patients initiated ART at 78 facilities; 1365 (16.5%) were pregnant women and 6897 (83.5%) were men and non-pregnant women (149 of whom became pregnant after ART initiation). Overall, 87.1% of pregnant women received ART. Retention at 6 months was lower among women who initiated ART during pregnancy than in the comparison group (74.4% vs. 81.5%, adjusted RR=0.91, p<0.001). Among facilities with at least 10 patients in the Option B+ group, retention rates ranged from 43.2% (95% CI: 28.3-59.0%) to 100% (95% CI: 82.4%-100%). Differences in relative risk of retention were found by sector (p<0.001), but not by health facility type or location. Conclusions: In the first year of Option B+ implementation, retention rates were lower and more variable for pregnant women initiating ART than for non-pregnant adults. Further investigation is needed to identify both structural and patient factors contributing to attrition among pregnant women in order to plan program interventions to strengthen retention.

THURSDAY, FEBRUARY 26, 2015 Session P-T4 Poster Session

Poster Hall

2:30 pm– 4:00 pm Health Outcomes of HIV- and ARV-Exposed Infants, Children, and Youth 876 Malnutrition Among HIV-Exposed Uninfected Children in Botswana Kathleen M. Powis 1 ; Quanhong Lei 2 ;Yvonne Chinyanga 3 ; EstherTumbare 4 ; Nealia Khan 5 ; Jacinta Sibiya 3 ; Erik vanWidenfelt 6 ; Joseph Makhema 6 ; Roger Shapiro 7

1 Massachusetts General Hospital, Harvard Medical School, Boston, MA, US; 2 Harvard School of Public Health, Center for Biostatistics in AIDS Research, Boston, MA, US; 3 Botswana Ministry of Health, Gaborone, Botswana; 4 Elizabeth Glaser Pediatric AIDS Foundation, Harare, Zimbabwe; 5 Harvard School of Public Health, Boston, MA, US; 6 Botswana Harvard AIDS Institute, Gaborone, Botswana; 7 Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, US Background: In resource limited settings, HIV-exposed infants who remain uninfected (HEU) experience higher morbidity and mortality compared with infants born HIV- uninfected women (HUU). In Botswana, where malnutrition is significantly associated with mortality in children under 5 years of age (U5), we sought to determine the prevalence between HEU and HUU children and determinants of malnutrition. Methods: We conducted a cross-sectional study in 5 health districts in Botswana with medium to high levels of malnutrition among U5 children. Malnutrition was defined as a weight-for-length or length-for-age z-score > 2 standard deviations below the median by 2006 World Health Organization Child Growth Standards. Caregivers and U5 children were recruited while attending well-child clinics at government health facilities, where food rations for children 6 to 60 months are also dispensed. Child weight and length/height were measured by trained study staff. Results: Of the 1,703 children enrolled, 1,109 (66.2%) were born to mothers reported to be HIV-uninfected, 432 (25.8%) to HIV-infected mothers, and 162 (9.5%) to mothers with unknown HIV status. Among HIV-exposed children, 396 (91.7%) were HEU, 7 (1.6%) HIV infected, and 29 (6.7%) were either never tested or unknown. The mean age of children did not differ between HUUs and HEUs (26 months vs 25 months; p=0.30). Prevalence of malnutrition was 26% among HUU children and 33% among HEUs (p = 0.005). Univariate differences between HEU and HUU infants are shown in Table 1. In multivariate logistic regression, birth weight (BWT) < 2.5 kg [aOR 3.3 (95% CI 2.4-4.6); p<0.0001], male child [aOR 1.5 (95% CI 1.2-1.9); p=0.002], absence of gas or electricity cooking source in the household [aOR 1.6 (95% CI 1.1-2.2); p=0.005], household food insecurity in the last month [aOR 1.3 (95% CI 1.0-1.8); p=0.05] and mother being unmarried [aOR 1.6 (95% CI 1.1-2.5); p=0.02] were associated with increased risk of malnutrition, but HEU was not [aOR 1.1 (95% CI 0.9-1.5); p=0.37]. Maternal and Infant Characteristics

Poster Abstracts

CI = Confidence Interval; Pula represents Botswana currency; student t-test used to compare means; Chi squared test used to compare proportions Conclusions: Low BWT and markers of poverty were associated with higher risk of malnutrition in U5 children in Botswana. In multivariate analyses, being born to an HIV infected mother did not place children at higher risk of malnutrition. However, HEU children were more likely to experience low BWT and to reside in socioeconomically deprived households. Interventions that improve BWT of HEUs and programs that address poverty eradication may minimize malnutrition among HEU children and overall in Botswana.

528

CROI 2015