Assessment of causality:
DILI was pre-defined in this study as >x3 ULN ALT in the presence of symptoms or >x5 ULN in absence of symptoms (Aithal et al. , 2011). The ULN for ALT was 50 IU/L in the populations studied, as defined by local clinical practice. The Roussel Uclaf Causality Assessment Method (RUCAM) was used to determine formal causality between anti-TB medication and liver injury (Bethseda (MD): National Institute of Diabetes and Digestive and Kidney Diseases, 2017). The pattern of liver injury was determined using the R ratio, considering ALT and alkaline phosphatase activity. Further factors considered include time to onset, course of injury, risk factors (age and alcohol), concomitant drugs, the exclusion of non-drug causes of injury and previous information on drug hepatotoxicity.