Falls in the days before a first seizure: a possible missed sign

Design

Retrospective case-crossover study of state inpatient and emergency databases

Population

57,547 adults with first-time seizure; 7,307 with a prior fall visit in case or control periods

Primary outcome

Odds of an emergency visit for a fall in windows before first seizure vs control windows a year earlier

Effect

OR 2.6 (95% CI 2.2 to 3.2) in the 7 days before; OR 4.5 (2.5 to 9.0) before first-time status epilepticus

This retrospective case-crossover study used inpatient and emergency databases from 11 US states between 2016 and 2021. It identified 57,547 adults with a first-time seizure, 7,307 of whom had at least one emergency visit for a fall in the case or control periods. Each patient served as their own control: fall visits in windows before the seizure were compared with matching windows a year earlier.

Fall visits were more frequent in the pre-seizure windows, with the highest odds in the 7 days before the seizure (OR 2.6, 95% CI 2.2 to 3.2). The association persisted up to 22 weeks before and attenuated with distance. It held across age and sex strata and after excluding traumatic brain injury. For first-time status epilepticus, the odds in the final week were higher (OR 4.5, 2.5 to 9.0).

The design cannot show that the falls were unrecognised seizures. Falls could also reflect a shared cause, such as intoxication, and administrative coding limits detail.

• Consider a seizure when assessing an unexplained fall, particularly with confusion, tongue biting or incontinence.
• Ask witnesses about stiffening, jerking or unresponsiveness during recent falls.
• Review recent falls in people presenting with a first seizure.
• Do not investigate every fall for epilepsy; the study shows association, not predictive value.

Why it matters

It points to a possible earlier moment to recognise epilepsy in routine emergency care.

Don't overread it

A case-crossover analysis of coded data shows temporal association, not that earlier falls were seizures.

The statistics, in plain English

An odds ratio of 2.6 compares the odds of a fall visit in the week before a seizure with the same week a year earlier. Most people with falls never have a seizure, so this does not make a fall a good predictor on its own.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.