Trying to distil a riddle of mind

A riddle wrapped in a mystery inside an enigma" - Winston Churchill's description of Soviet policy - would not be inappropriately applied to schizophrenia. This devastating illness, with a lifetime risk of nearly 1 per cent for the world's population, is both ubiquitous and elusive. There is no test to identify it, the clinical boundaries are uncertain, its genetics are so far impenetrable, and findings of biochemical or neuroimaging abnormalities sometimes end in blind alleys. As with all highly complex subjects in science, its literature is immense.

Eve Johnstone, professor of psychiatry at Edinburgh University, with four colleagues, has sought to distil the core of current knowledge on the subject - including its practical management - into a modest 0 pages.

Psychiatrists have long been taunted with the accusation that what they treated as a disease had no tangible basis, but the scientific findings summarised in this book should remove any doubt that schizophrenia is essentially a brain disease. Both postmortem and imaging studies show that the whole brain and specific areas of it are smaller than in non-schizophrenic samples. Correspondingly, the ventricular system and volume of cerebro-spinal fluid are enlarged. Since schizophrenia usually emerges clinically during late adolescence, it would be logical to assume that the brain abnormalities also date from that period. It is now clear, though, that they are not degenerative, as that would imply, but developmental, having been present since early in the formation of the central nervous system.

Why these abnormalities occur is still uncertain, though it is probably through the interaction of genetic vulnerability with other risk factors. Complications at birth have been a prime candidate for the environmental risk, though sometimes it might be an already abnormal foetus itself that upsets the birth process. Wartime experience in Holland showed that starvation in pregnancy can have this effect, with connections between different brain areas being disrupted. One consequence of this may be that the brain can no longer distinguish between internal and external speech, resulting in auditory hallucinations.

But if the underlying abnormality is there for so long, why has this fact not become evident earlier? The answer seems to be that no one has looked for it. Years before their psychiatric symptoms appear, children who later become schizophrenic tend to show a fairly distinct profile of cognitive deficits affecting attention, memory and "executive function". Milder forms of the same deficits can be found in unaffected relatives.

Since almost half of identical twins will show concordance for schizophrenia, the genetic contribution to it is undeniable, yet more than 60 per cent of patients have no positive family history. It remains unknown whether there is a single gene of major effect, several of moderate effect, or many of small effect. There might even be sub-groups of the disorder, which have a particular form of inheritance. One fact that strongly supports genetic determination is that the frequency of schizophrenia varies little throughout the world - much less than multifactorial bodily disorders such as diabetes or heart disease.

This psychosis was first identified - as dementia praecox - on the basis that it had a poor prognosis, yet the outcome of any individual case is very difficult to predict. Anti-psychotic drugs were introduced nearly 50 years ago and were later found to block the neutrotransmitter dopamine. Yet, overall, dopamine function does not seem to be abnormal in schizophrenia. More recently, "atypical" drugs are known to act on serotonin, though serotonin is also not dysfunctional overall. In the view of Johnstone et al , the lack of a single neurochemical mechanism may mean that the clinical manifestations of schizophrenia represent a final common pathway from different disturbances.

The fact that schizophrenia disables people early on but does not kill them, except by suicide, is fundamental to the problem of care. Reduced "executive function", or coping with the business of life, means many affected people have to be cared for over long periods. It is rightly pointed out here that "the swing of the pendulum between institutional and community care cannot be shown to be based on any clear rationale". Both the multiplicity of problems that are produced by this disorder and its mostly chronic or relapsing nature call for well-organised and comprehensive services. But these are not widespread in Britain, and even rarer in most other countries.

Family intervention to reduce hostile feelings, cognitive techniques and social skills training can all be helpful, though the precise elements of these that are effective still need to be isolated. On the other hand, the authors regard psychotherapy, non-specific social treatment and "case management" as having no proven value. This view will upset some enthusiasts, but the spread of evidence-based practice is likely to support Johnstone et al, who say their book is designed for use by "psychiatrists and other health professionals involved in the management of schizophrenia".

Its clarity of style and freedom from jargon, though, should make it accessible to readers who lack the technical background. In fact, I would like to lock up every mental health service manager with a copy, and not let him or her out until a satisfactory knowledge of the book had been shown.

Hugh Freeman is former editor, British Journal of Psychiatry, and honorary writing fellow, Green College, Oxford.