Jung 2015

Methods	Multicenter, randomised, open‐label, non‐inferiority trial conducted across 7 centres in Republic of Korea 2 treatment arms: CBZ and LEV
Participants	Children aged 4‐16 years with newly diagnosed focal epilepsy, no previous anti‐epileptic therapy and "above borderline" intelligence Number randomised: CBZ = 64, LEV = 57 (ITT population) 69 male participants (57%) 100% of participants with partial epilepsy Mean age (SD): CBZ = 8.05 (3.02), LEV = 9.28 (3.37) years
Interventions	Monotherapy with CBZ or LEV 4‐week dose titration period to a minimal target dose of CBZ = 20/mg/kg/d or LEV = 40/mg/kg/d Trial duration: 52 weeks, range of follow‐up not stated
Outcomes	Neuropsychological outcomes; change from baseline to 52 weeks in neurocognitive (Korean‐WISC‐III or Korean‐Wechsler Preschool and Primary Scale of Intelligence‐III), behavioural (Korean‐CBCL), and emotional (Children's Depression Inventory and Revised Children's Manifest Anxiety Scale) function assessments Mean percentage change in seizure frequency from baseline Seizure‐freedom rates Incidence of adverse events
Notes	IPD could not be provided for the trial due to restrictions on data sharing from the Korean Food and Drug Administration (information provided by corresponding author). Outcomes chosen for this review were not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomised independently at each centre using a computerised random code assignment based on stratified permuted block randomisation that were designed separately and independently for each participating centre
Allocation concealment (selection bias)	Low risk	At each centre, allocation concealment was carried out by the pharmacy in order to blind those assessing outcomes from the trial medication
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label trial for participants and personnel
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Those assessing outcomes were blinded to trial medication (pharmacy allocation)
Incomplete outcome data (attrition bias) All outcomes	High risk	7 randomised participants did not take any trial medication so were not included in ITT population. Results for neuropsychological outcomes recorded only for those who completed the trial ‐ 81/121 participants (67%)
Selective reporting (reporting bias)	Low risk	All neuropsychological, efficacy, and tolerability outcomes specified in the methods sections were reported well in the results section. No protocol was available. Outcomes chosen for this review were not reported
Other bias	Low risk	None identified