Post-seizure psychosis linked to later psychotic-disorder diagnoses in record study

The researchers used de-identified records from the TriNetX database, covering 2006 through 2026. They studied adults ages 18 to 65 with epilepsy-related diagnoses.

They identified people with records coded as probable postictal psychosis, or as a brief psychotic disorder diagnosed within seven days of an epilepsy diagnosis. Postictal means occurring after a seizure. Psychosis describes symptoms such as hallucinations, delusions, or severely disrupted thinking.

The team excluded people who already had a psychotic disorder diagnosis or delirium, a sudden state of confusion. They then matched each person in the probable postictal psychosis group with one person without postictal psychosis. Matching accounted for age, sex, race, ethnicity, and recorded psychiatric and chronic physical health conditions.

A total of 1,275 people were included after matching. The researchers followed the records for up to five years to see who received any psychotic-disorder diagnosis.

People in the probable postictal psychosis group had a higher recorded risk of a later psychotic disorder. The reported hazard ratio was 10.78, with a 95% confidence interval from 6.21 to 18.70. A hazard ratio compares how quickly an outcome occurs in two groups over time; here, the higher figure indicates a much higher rate of later diagnoses in the probable postictal psychosis group.

The study also reported five-year survival probabilities of 85.79% for the probable postictal psychosis group and 98.21% for controls. In this type of analysis, “survival” refers to remaining free of the outcome being tracked, not to death.

These results are important because postictal psychosis is often temporary. The study raises the possibility that, for some people, it may be associated with future psychotic disorders, including forms that occur between seizures.

Still, the study cannot establish cause and effect. It relied on diagnostic codes in existing medical records, and the researchers described the condition as “probable” postictal psychosis. Records may not fully capture symptoms, their timing, or other factors that could help explain the association.

The findings also do not show whether closer mental health monitoring changes outcomes. Future research would need to confirm the results in other groups, examine why the link occurs, and test whether particular follow-up approaches are helpful.