People with schizophrenia are six times more likely to develop epilepsy, says a study which finds a strong relationship between the two diseases.
Writing in Epilepsia, researchers in Taiwan say this could be due to genetic, neurobiological or environmental factors.
The study followed around 16,000 patients with epilepsy and schizophrenia between 1999 and 2008.
An epilepsy expert says it is an interesting and convincing study.
The study used data from the Taiwan National Health Insurance database and was led by researchers from the China Medical University Hospital in Taichung.
They identified 5,195 patients with schizophrenia and 11,527 patients with epilepsy who were diagnosed during the nine years period.
These groups of patients were compared to groups of the same sex and age who did not have either epilepsy or schizophrenia.
The findings show that the incidence of epilepsy was 6.99 per 1,000 person-years in the schizophrenia patient group compared to 1.19 in the non-schizophrenia group.
The incidence of schizophrenia was 3.53 per 1,000 person-years for patients with epilepsy compared to 0.46 in the non-epilepsy group.
Previous studies had suggested a prevalence of psychosis among epilepsy patients.
Researchers in this study also found that schizophrenia levels were slightly higher in men with epilepsy than in women with the disease.
Dr I-Ching Chou, associate professor with the China Medical University in Taichung inTaiwan, said:
"Our research results show a strong bidirectional relation between schizophrenia and epilepsy.
"This relationship may be due to common pathogenesis in these diseases such as genetic susceptibility and environmental factors, but further investigation of the pathological mechanisms are needed."
Dr Manny Bagary, consultant neuropsychiatrist in Birmingham, said it was a "very interesting" study.
"We have been aware that epilepsy sufferers seem to be have an increased risk of psychosis but this is the first convincing study to suggest that people with schizophrenia could also be at risk of developing epilepsy, suggesting a bidirectional relationship has been found between depression and epilepsy".
"The association may be due to a common environmental factors such as traumatic brain injury or brain haemorrhage in utero. Alternatively, a genetic association may be relevant such as LGI1 or CNTNAP2 genes which have been associated with seizures and psychosis."
"However, there may be some questions to ask about the reliability of the schizophrenia and epilepsy diagnoses in the study because it is a retrospective health register study and both conditions can be difficult to diagnose.
"Nevertheless, this study will serve to guide further research into the relationship between epilepsy and psychosis."