Abstract
BACKGROUND: The dispersible fixed-dose combination drug has been recommended as the mainstay of treatment for TB in children. However, more needs to be known about its effect on treatment. This study aimed to assess the effectiveness of the formulation on treatment adherence among children with TB. METHODS: A historical cohort design was used to assess and compare adherences of old loose non-dispersible and new dispersible fixed-dose anti-TB drugs, using a convergent parallel mixed-method approach for data collection. Determinants of treatment adherence were assessed using binary logistic regression. RESULTS: The proportion of children with good treatment adherence was higher in the new dispersible formulation group (82 [64.6%]) relative to the proportion among the loose non-dispersible formulation group (29 [23.4%]). Reports of forgetfulness, travelling and pill burden were significantly higher among those with poor adherence in the loose non-dispersible formulation group. Significant predictors of treatment adherence were acceptability (adjusted OR [AOR]=4.1, p=0.013, 95% CI 1.342 to 12.756), travelling from treatment areas (AOR=8.9, p=0.002, 95% CI 2.211 to 35.771) and forgetfulness (AOR=74.0, p<0.001, 95% CI 23.319 to 234.725). CONCLUSIONS: The determinants of treatment adherence are multifactorial. In addition to ensuring universal access to the drug, flexible referral in case of travelling and ensuring treatment partners' participation to minimise forgetfulness to take pills, are essential.