Background Kenya has an estimated 86,300 children living with HIV (CLHIV) on antiretroviral therapy (ART) of whom a third are not virally suppressed. Looking at predictors of viral non-suppression guides the program in designing intervention strategies to abate inauspicious treatment outcomes.

Methods This was a cross-sectional study that used de-identified electronic medical records of the Christian Health Association of Kenya HIV/AIDS project database. We included all CLHIV aged 9 years and below who were active in care as of 30^thSeptember 2018 and had been on ART for at least 6 months with a recent viral load result. We defined non-suppression as a result of ≥1000 copies/mm³ which was our outcome of interest. Predictor variables studied were age, sex, duration prior to ART initiation and duration on ART, ART regimen, orphan status, baseline WHO staging and adherence. Bivariate analysis and multivariate logistic regression were used to establish determinants of non-suppression.

Results We included 1,066 CLHIV of whom 51.3% were female, median age was 7.5 years (IQR 5.7– 9) and a quarter were orphans. Median duration on ART was 51 months (IQR 31–79), 20.4% were on second line ART regimen with an overall viral suppression rate of 88%. Children who had been on ART for a longer duration (>5 years) were more likely to be suppressed [aOR=0.38, (95% (CI)=0.17–0.86), p=0.02]. A protease inhibitor containing regimen was associated with non-suppression on bivariate analysis [OR=2.43, (95% CI = 1.04–5.65), p=0.039] however this was not significant in multivariate analysis. Non-adherence to ART increased five-folds the odds of non-suppression [aOR=5.47, (95% CI = 1.12–26.69), p=0.035] whereas those who were orphans were more likely to be suppressed [aOR=0.56, (95% CI = 0.37–0.86), p=0.007].

Conclusion CLHIV within our study population had sub-optimal viral suppression. Innovative strategies to address adherence remains crucial in addressing non-suppression.

Disclosure No significant relationships.