Skip to main content

19-03-2014 | Mental health | Article

Researchers estimate psychosis rates in epilepsy


Free full text

medwireNews: Patients with epilepsy are nearly eight times more likely than people without the condition to experience psychosis, show results of a systematic review and meta-analysis.

The study, conducted by David Cotter (Royal College of Surgeons in Ireland, Dublin) and colleagues, also estimated that 5.6% of all patients with epilepsy and 7.0% of those with temporal lobe epilepsy have comorbid psychosis, which is largely consistent with the findings of previous qualitative reviews.

As reported in BMC Psychiatry, the researchers analysed data from 58 studies investigating the prevalence of psychosis in patients with epilepsy. The majority (n=48) of studies were cross-sectional, nine were cohort studies and there was one case–control study.

The number of participants ranged from 50 to 34,494 and the prevalence rates of psychosis varied from 0.02% to 27%.

The pooled prevalence of interictal psychosis was slightly lower than the pooled prevalence of all psychosis, at 5.2%, while the rate of postictal psychosis (PIP) was 2.0%.

This figure is much lower than the previously reported rates of around 6%, but may be attributed to the transitory nature of PIP or the fact that a consensus on the classification of psychotic symptoms associated with epilepsy is lacking, say the researchers.

They explain that “the psychopathology of patients with epilepsy can be atypical and does not readily conform” to the diagnostic manuals; therefore, “[i]mproved diagnostic classification would allow for better characterisation of prevalence rates”.

Just four studies looked at psychosis in a control group, Cotter and team used the pooled data from these studies to show that patients with epilepsy have a 7.8-fold increased risk of psychosis compared with those without it.

Overall, there was substantial heterogeneity (I2>70%) in the prevalence estimates of psychosis among the studies. A model incorporating study setting, the assessment instrument used, study year, family history of psychosis and type of epilepsy explained the largest amount of variance observed, at 10%.

Cotter and co-authors conclude that “adequate recognition and treatment of psychosis in epilepsy is essential for patient management because of their considerable burden in morbidity and quality of life.”

They recommend that future studies should be conducted to define psychotic presentations in patients with epilepsy.

medwireNews ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2014

By Laura Cowen, medwireNews Reporter