Mental Health special section

A cereal offender

Danish researchers give the 40-year-old link between celiac/schizophrenia the once over. On the meds and off the porridge

By Graham Furness

Schizophrenia often runs in families, and one of the areas of DNA which appears to be implicated in the disease, the dysbindin locus, happens to be located very close to the most important genetic marker for another inheritable condition, celiac disease.

It was proposed 40 years ago that gluten in the diet could trigger schizophrenia in a genetically predisposed person. There was some epidemiological evidence to back this up, and research in the journal Science back in 1976 suggested that a cereal-free diet could improve remission of symptoms of schizophrenia. It has also been suggested, based on anecdotal evidence, that gluten could trigger schizophrenia — but this has never been proven with epidemiological data.

A group of Danish researchers decided to take the 1976 research a step further and set out to examine the epidemiological relationship between celiac disease and schizophrenia — the first time this link has been tested. The researchers retrospectively examined the medical records of 7,997 people diagnosed with schizophrenia. Their research appeared in the February 21 issue of the British Medical Journal (BMJ).

Each of the 7,997 schizophrenic cases was matched with 25 controls of the same age and gender. Patients with dermatitis herpetiformis were grouped with those who had celiac disease because the two conditions are closely related. The analysis controlled for known schizophrenia risk factors, namely socioeconomic position, urban residence and family history of the disease.

They found that patients with celiac disease were 3.2 times more likely (P<0.0001) to have been diagnosed with schizophrenia than those without. Patients with dermatitis herpetiformis had a relative incidence of schizophrenia 3.1 times the average. By contrast, Crohn's disease and ulcerative colitis only appeared to increase risk of schizophrenia by 1.4 times.

The risk relation between schizophrenia and celiac disease "is strong," say the authors, "but reflects a small proportion of cases of either disorder, since both disorders are rare." On the other hand, a recent BMJ study in Britain found that no less than 1% of the population carried celiac disease antibodies, though six out of seven had no symptoms. The Danish study only considered symptomatic celiac patients, most of whom would be on gluten-free diets.

"An important question," conclude the Danish researchers, "is the degree to which removal of gluten from the diet will alleviate symptoms in the small proportion of people with schizophrenia who screen positively for celiac disease but do not show its classical symptoms."