The test uses a plastic strip impregnated with a chemical derivative of the B vitamin niacin. This causes a reddening of the skin in normal people which is significantly reduced in schizophrenics. The test should make it possible to distinguish schizophrenia from other conditions at an early stage, so the right treatment can be given.

As one of the research team, Pauline Ward, explains, the test was developed as a result of observations made over 20 years ago: "The observations were made when doctors were attempting to treat schizophrenia with large doses of vitamins. It was noted that, whereas when you give normal people large doses of the vitamin niacin they tend to flush quite considerably, on their bodies and faces, the schizophrenics didn't do that.

"It was suggested then that this might be an indication of abnormal biochemistry in schizophrenia. But it has only been recently that we set out to discover what the actual mechanism is."

It became clear that the reaction seen more strongly in non-schizophrenics is a reaction by cells, macrophages in the skin, which release the locally acting hormone prostaglandin D2 on exposure to niacin. This causes nearby blood vessels to dilate, causing reddening of the skin.

The theory developed to explain the lessening of this effect in schizophrenics is that, as other research has shown, their blood cells produce less arachidonic acid, which is required for the synthesis of prostaglandins.

Pauline Ward and her colleagues decided to develop a simple skin test measuring this effect, which could easily be used to see how significant were differences between schizophrenics and non-schizophrenics, and which could be used clinically if the differences proved significant.

Trials of the skin test began about two years ago. The test used a plastic strip with absorbent paper patches on it. The patches contained four different dilutions of aqueous methyl nicotinate, a derivative of niacin. The strips were placed on the skins of test subjects for five minutes, and the colour change reaction measured over the following 15 minutes.

The tests were carried out on 38 schizophrenic patients and 22 normal controls. Pauline Ward said: "We found highly significant differences between the schizophrenics and the normal controls at all four concentrations of methyl nicotinate. All the schizophrenics were less reactive than the controls."

More extensive trials will be required before the test can be used routinely in clinical medicine. Research is needed to discover at exactly what stage in the onset of schizophrenia the test becomes valid, and what results it gives in other affective disorders (serious mental illnesses.)

The Highland Psychiatric Research Group, that has carried out the work, also intends in future studies to stimulate the skins of schizophrenic patients and controls directly with prostaglandin D, to see if the blockage is at the niacin-macrophage level or at another stage of the pathway.

The test will undoubtedly be valuable in defining a biochemically distinct group, of people with low levels of arachidonic acid, probably as a result of abnormal activity of the enzyme phospholipase A2, increased activity of which has been reported in platelets and serum from schizophrenics, and which may be responsible for increased breakdown of phospholipids in the cell membranes of brain cells in the cerebral cortex of schizophrenics, causing abnormalities in neurotransmitter receptors.

When it has been refined Pauline Ward and her colleagues hope the test will be valuable clinically. "I see it being used initially as an aid to diagnosis, because often when it first presents in a young person it's hard to distinguish schizophrenia from the effects of drugs or straightforward depression," she added.

"It should be useful there. It may also be useful in monitoring the effects of treatment, especially new medications which use diet to attempt to correct abnormalities in brain cell membranes."