Brodie 2002

Methods	Randomised, multicentre, double‐blind trial conducted in 41 centres in Europe and Australia 2 treatment arms: GBP and LTG
Participants	Participants > 16 years with at least 2 partial seizures with or without secondary generalisation or primary generalised tonic clonic seizures in the last 12 months. All participants were untreated in the previous 6 months or AED naive Number randomised: GBP = 158, LTG = 151. Evaluable population (exclusions due to protocol violations): GBP = 148, LTG = 143 152 male participants (52%) out of evaluable population 233 participants with partial epilepsy (80%) out of evaluable population Mean age of evaluable population (SD, range): GBP: 35.8 years (16.4, 13‐78), LTG: 37.9 (16.7, 16‐78)
Interventions	Monotherapy with GBP or LTG Titration of 2 weeks for GBP to a dose range of 1200 mg/d‐3600 mg/d and titration of 6 weeks for LTG to a dose range of 100 mg/d‐300 mg/d Titration period followed by 24‐week maintenance period. Range of follow‐up not stated
Outcomes	Time to exit Percentage of completers/time to withdrawal for any reason Time to first seizure Percentage who remained seizure‐free during the final 12 weeks of the 30‐week evaluation period Withdrawal rate due to adverse events
Notes	IPD requested from trial sponsor Pfizer but data could not be provided due to time elapsed since the trial was completed. Additional information provided in a clinical study report. Aggregate data extracted for time to exit from the trial and time to first seizure extracted from the publication
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed with permuted blocks, stratified within each centre by seizure type.
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Masking was achieved by double‐dummy dosing. A dose range was permitted within the trial to maintain the blind of two drugs with different titration rates (2 weeks for GBP and 6 weeks for LTG)
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No information provided
Incomplete outcome data (attrition bias) All outcomes	Low risk	Attrition rates reported, all randomised participants included in an ITT analysis (even though demographics presented for 'evaluable population' only)
Selective reporting (reporting bias)	Low risk	All efficacy and tolerability outcomes specified in the methods sections were reported well in the results section. No protocol was available.
Other bias	Low risk	None identified