ObjectivesWe evaluated the on-scene time of emergency medical services (EMS) for cases where discrimination between acute stroke and epileptic seizures at the initial examination was difficult and identified factors linked to delays in such scenarios. Materials and MethodsA retrospective review of cases with suspected seizure using the EMS database of fire departments across six Japanese cities between 2016 and 2021 was conducted. Patient classification was based on transport codes. We defined cases with stroke-suspected seizure as those in whom epileptic seizure was difficult to differentiate from stroke and evaluated their EMS on-scene time compared to those with epileptic seizures. ResultsAmong 30,439 cases with any seizures, 292 cases of stroke-suspected seizure and 8,737 cases of epileptic seizure were included. EMS on-scene time in cases of stroke-suspected seizure was shorter than in those with epileptic seizure after propensity score matching (15.1±7.2 min vs. 17.0±9.0 min; p = 0.007). Factors associated with delays included transport during nighttime (odds ratio [OR], 1.73, 95 % confidence interval [CI] 1.02–2.93, p = 0.041) and transport during the 2020–2021 pandemic (OR, 1.77, 95 % CI 1.08–2.90, p = 0.022). ConclusionThis study highlighted the difference between the characteristics in EMS for stroke and epileptic seizure by evaluating the response to cases with stroke-suspected seizure. Facilitating prompt and smooth transfers of such cases to an appropriate medical facility after admission could optimize the operation of specialized medical resources.