Mattson 1992

Methods	Double‐blind, multicentre trial across 13 Veteran’s Affairs medical centres (USA) 2 treatment arms: CBZ and VPS
Participants	Adults (18‐70 years) with previously untreated or under‐treated complex partial seizures, secondarily generalised tonic‐clonic seizures Number randomised: CBZ = 236, VPS = 244 445 male participants (93%) 100% of participants had partial epilepsy. Mean age (range): 47 (18‐83) years
Interventions	Monotherapy with CBZ or VPS Mean daily dose achieved by month 12 CBZ = 722+‐ 230 mg/d, VPS = 2099 +‐824 mg/d Range of follow‐up: 0‐73 months
Outcomes	Total number of seizures (of each type) during 12 months Number of seizures per month Percentage of participants with seizures completely controlled Time to first seizure Seizure rating score (severity of seizures) at 12 and 24 months Composite score (combined score for the control of seizures and incidence of adverse events) Incidence of systemic and neurologic adverse events (and severity) Time to treatment failure
Notes	IPD provided for all outcomes of this review by the Department of Veterans Affairs
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomised to treatment using random permuted blocks with a different randomisation scheme for two seizure groups (complex partial and secondarily generalised tonic clonic)
Allocation concealment (selection bias)	Low risk	Treatment allocation was concealed via sealed, opaque envelopes
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double blind (participants and personnel) achieved with additional matching placebo tablets
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Unclear if outcome assessment was blinded, no information provided
Incomplete outcome data (attrition bias) All outcomes	Low risk	Attrition rates reported, ITT approach, all randomised participants analysed from IPD provided (see footnote 2)
Selective reporting (reporting bias)	Low risk	Protocol provided. All outcomes reported or calculated with IPD provided (see footnote 2)
Other bias	Low risk	None identified