Factors associated with viral load suppression among orphans and vulnerable children and adolescents living with HIV in Kenya

**Background:** Despite global progress toward UNAIDS 95-95-95 targets, children living with HIV—particularly orphans and vulnerable children (OVC)—lag behind in viral load suppression (VS). In sub-Saharan Africa, evidence on specific interventions within OVC programs to support durable VS is limited. This study aimed to identify correlates of optimal VS among CALHIV in a large Kenyan OVC program.

**Methods:** This cross-sectional study used anonymized secondary programmatic data from USAID PEPFAR-funded OVC implementing partners (MWENDO, CASE OVC, AMPATHplus, COGRI) across 34 counties in Kenya (2 urban, 32 rural). Data were extracted from Child Protection Management Information System databases for CALHIV aged ≤18 years actively receiving ART with documented viral load (VL) results between October 2019 and September 2020. Of 65,816 CALHIV extracted, 31,291 had complete data for analysis (7,918 urban; 57,898 rural). Viral suppression was defined as <1000 RNA copies/mL per WHO guidelines. A multivariate mixed-effects logistic regression model (glmer in R) was used, with county as a random effect. Variables significant at bivariate level (chi-square, p<0.05) were included in the final model; those causing non-convergence or multicollinearity were excluded.

**Key Results:** Overall, 26,887 (85.9%) CALHIV had suppressed VL and 4,404 (14.1%) were unsuppressed. Female CALHIV had higher suppression (86.7%) than males (85.0%). Children aged 1–4 years had the lowest suppression (80.2%). CALHIV actively in school had higher suppression (87%) than those inactive (80%). PSS group members had higher suppression (87%) than non-members (84%). Caregiver VSLA membership was associated with higher suppression (87% vs. 83%). In the adjusted model, factors significantly associated with higher VS odds were: active education status (aOR=1.33; 95% CI: 1.07–1.65), child PSS membership (aOR=1.26; 95% CI: 1.15–1.38), and caregiver VSLA membership (aOR=1.23; 95% CI: 1.13–1.33). Factors associated with lower VS odds were: male sex (aOR=0.88; 95% CI: 0.83–0.94), male caregiver (aOR=0.91; 95% CI: 0.83–0.997), and high caregiver household vulnerability (aOR=0.81; 95% CI: 0.71–0.92). Child age group, caregiver education, caregiver PSS membership, CPARA completion, and positive parenting training were not significant in the final model.

**Clinical Implications:** This study identifies modifiable individual- and caregiver-level factors associated with VS among CALHIV in a large Kenyan OVC program. Keeping children actively in school, enrolling them in psychosocial support groups, and providing economic empowerment opportunities for caregivers (e.g., VSLA membership) may improve VS outcomes. Male CALHIV and those with male or highly vulnerable caregivers may need additional targeted support. Programs should integrate educational support, peer psychosocial interventions, and caregiver economic strengthening to close the VS gap for OVC living with HIV. Limitations include cross-sectional design (limiting causal inference), exclusion of children without documented VL, and lack of community/societal-level variables.