Fişgin 2002.

Methods	Prospective quasi‐randomised study (odd and even days randomisation of the 2 drugs) over 15 months
Participants	45 children of both sexes and aged 1 month to 13 years presenting to the emergency room with a seizure lasting at least 5 minutes No exclusion criteria stated
Interventions	intranasal midazolam versus rectal diazepam
Outcomes	Stopping of seizure within 10 minutes Time to cessation of seizure Efficacy of anticonvulsant effect Need for a second drug to control seizures Presence of complications
Notes	Some methodology described unclear, particularly relating to seizure type and aetiology of included children. It is therefore unclear if the population of this study is generalisable
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: "Diazepam was given on odd days of the month and midazolam on the even days"
Allocation concealment (selection bias)	High risk	See above; no concealment of allocation
Blinding (performance bias and detection bias) All outcomes	Low risk	Study unblinded, but blinding would not have been possible, due to the different routes of administration of the 2 study drugs, but this is not likely to have affected outcome
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants were included in the analysis
Selective reporting (reporting bias)	Unclear risk	Quote: "information about previous convulsions and history of antiepileptic medication was obtained.." Comment: this information was not reported in the Results section but as this is not one of the primary outcome measures it is not likely to be significant
Other bias	Unclear risk	Unclear description of the seizure type and aetiology of included children, so it is unclear if the population of this study is generalisable