To describe the epidemiology of childhood tuberculosis (TB) in Kenya, assess the magnitude of TB/human immunodeficiency virus (HIV) co-infection and identify risk factors for mortality during TB treatment.

We conducted a retrospective analysis of the Kenyan national TB program data for patients enrolled from 2013 through 2015. A total of 23 753 children aged less than 15 years were included in the analysis. Survival analysis was performed with censorship at 9 months and mortality was the main outcome. We used Cox proportional hazards regression for assessing risk factors for mortality.

Childhood TB accounted for 9% (n = 24 216) of all patients with TB; 98% of the notified children (n = 23 753) were included in the analysis. TB/HIV co-infection was 28% (n = 6112). Most TB cases (71%; n = 16 969) were detected through self-referral. Treatment was successful in 90% (n = 19 088) and 4% (n = 1058) died. Independent risk factors for mortality included being HIV infected but not on antiretroviral therapy (adjusted hazard ratio [aHR], 4.84; 95% CI, 3.59-6.51), being HIV infected and on antiretroviral therapy (aHR, 3.69; 95% CI, 3.14-4.35), children aged less than 5 years (aHR, 1.25; 95% CI, 1.08-1.44), and being diagnosed with smear negative pulmonary disease (aHR, 1.68; 95% CI, 1.27-2.24).

Most childhood TB cases in Kenya were detected through passive case finding. TB/HIV co-infection is high among children on treatment for TB, and HIV is associated with an increased risk of death. There is a need to intensify active case finding among children. TB prevention interventions among HIV-infected children, early diagnosis of HIV, and early antiretroviral therapy initiation among children on TB treatment should be strengthened.