Prevalence and factors associated with HIV among children with tuberculosis in ten high-burden regions of Tanzania, 2018–2023: a retrospective analysis of national surveillance data

Abstract Background Children living with HIV who develop tuberculosis (TB) have higher case fatality and lower treatment success than their HIV-negative peers. Tanzania’s routine surveillance reports pediatric TB-HIV co-infection descriptively, without formal trend testing or modelling of associated factors, leaving the National Tuberculosis and Leprosy Programme without evidence to target HIV testing and TB preventive therapy. We determined the prevalence of TB-HIV co-infection, described its pattern by year, age group, sex and region of registration, and identified associated factors among children aged below 15 years registered with TB in Tanzania. Methods We conducted a retrospective analysis of Tanzania’s national electronic TB surveillance system (DHIS2-eTL). The analysis included all children aged below 15-years with documented HIV status registered with TB across ten high-burden regions between January 2018 and December 2023. HIV positivity was confirmed at TB registration. We tested the annual trend with the chi-square test for linear trend and estimated adjusted prevalence ratios (aPR) by modified Poisson regression with robust variance. Results Of 25,449 children registered with TB and having a documented HIV status, 3,667 were HIV-positive (14.4%, 95% CI:14.0–14.8). Prevalence fell in every successive year, from 28.5% (95% CI:26.8–30.2) in 2018 to 8.9% (95% CI:8.1–9.8) in 2023 (χ²₁=719.7, p < 0.001). It rose across age groups, from 9.0% (95% CI:8.5–9.5) below 5 years to 17.7% (95% CI:16.8–18.6) at 5–9 years and 22.6% (95% CI 21.6–23.7) at 10–14 years (χ²₁=684.3, p < 0.001) and did not differ by sex ( p = 0.814). It was highest in Njombe (29.7%) and lowest in Ruvuma (6.7%) ( p < 0.001). In the adjusted model, co-infection was independently associated with age 10–14 years (aPR = 2.40, 95% CI:2.22–2.59), registration in Njombe (aPR = 1.46, 95% CI:1.30–1.64) and relapse TB (aPR = 2.30, 95% CI:1.92–2.76). Conclusions TB-HIV co-infection affected approximately one-in-seven children notified with TB and declined by two-thirds between 2018 and 2023, but remained concentrated in older children, in previously treated children and in Njombe, Mbeya and Iringa. The NTLP should prioritize repeat HIV testing for all children aged 5 years and above presenting with TB and for all retreatment cases, with geographic intensification in these three regions.