Nine RCTs (n=251) and seven observational studies (n=approximately 228,670, but this included patients who received antiepileptic drugs and non-antiepileptic drugs and may have included duplicate patients) were included. Most studies were funded by pharmaceutical companies.
RCTs: Eight of the nine RCTs were cross-over studies, four were double-blind and three were single-blind. Jadad scores ranged from 2 to 4 out of 5. Sample size ranged from 10 to 64. Treatment duration ranged from four to 28 weeks.
There was no significant difference between brand-name antiepileptic drugs and generic versions in the control of seizures, OR 1.1 (95% CI 0.9 to 1.2; seven RCTs, n=204). No significant statistical heterogeneity was found.
Observational studies: Newcastle-Ottawa scores ranged from 5 to 9 out of 9.
Six of the seven studies attributed the reported increase in seizures (three studies) or increased healthcare utilisation (three studies) with switching from brand-name to generic drugs. The seventh study reported no difference in seizure rates attributed to switching.