Brodie 1999.

Methods	Randomised, multicentre, double‐blind, parallel‐group trial conducted in the UK 2 treatment arms: LTG and CBZ randomised in a 2:1 ratio
Participants	Adults over the age of 65 with newly diagnosed epilepsy with 2 or more seizures in the previous year with at least 1 seizure in the last 6 months Number randomised: LTG = 102, CBZ = 48 83 males (55%) 105 with focal seizures (70%) Not stated if any participants had received previous AED treatment Mean age (range): 77 (65 to 94) years
Interventions	Monotherapy with LTG or CBZ for 24 weeks 4‐week escalation phase leading to LTG = 100 mg/day, CBZ = 400 mg/day Range of follow‐up = 0 to 280 days
Outcomes	Time to first seizure after 6 weeks of treatment Time to treatment withdrawal Percentage of patients reporting an adverse event Proportion of patients who were both seizure‐free in the last 16 weeks of the trial and did not discontinue treatment
Notes	IPD provided by trial sponsor GlaxoSmithKline for time to treatment failure and time to first seizure (plus seizure freedom rates at 24 weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random sequence (information provided by drug manufacturer). Participants randomised in a 2:1 ratio (LTG:CBZ)
Allocation concealment (selection bias)	Low risk	Allocation concealed with pharmacy‐dispensed treatment packs labelled with participant's trial number
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind achieved using LTG tablets formulated to be identical in appearance to CBZ tablets
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Trial investigator blinded, not stated if other outcome assessors were blinded
Incomplete outcome data (attrition bias) All outcomes	Low risk	Attrition rates reported; all randomised participants analysed from IPD provided (see footnote 2)
Selective reporting (reporting bias)	Low risk	All outcomes reported or calculated with IPD provided (see footnote 2)
Other bias	Low risk	None identified