Stalled on the brink of insight

Schizophrenia's secrets are still waiting to be unlocked, Hugh Freeman writes

The affliction of human minds that is schizophrenia has not yielded many of its secrets to researchers of the past 50 years. As methods of investigation have come into use that were no more than science fiction not so long ago, knowledge of the brain's functioning has expanded enormously.

Yet, in practical terms of understanding and treatment, it has made surprisingly little difference.

Certainly though, if more effective use were made of treatment methods already available, things would be a lot better for the mentally ill. But even then, existing treatments would seem more like tinkering at the edges of the problem than being able to confront it comprehensively.

One major pitfall in the way of progress is society's profound ambivalence towards this disorder, related to its stigma. It is low on most research agendas, while those responsible for services for the people afflicted remain unwilling to accept the implications of chronic, disabling illness.

For politicians, there is little to be gained from intervening, even if outcomes are positive. What is much more likely is that some of the distaste that surrounds the subject will stick to those who become involved. Furthermore, no question about schizophrenia has a simple answer.

The Epidemiology of Schizophrenia is the most comprehensive assembly of existing knowledge to date. Its five editors are all distinguished scientists, and the contributors include many of those most active in the field, with the surprising omission of Assen Jablensky. Why any book needs five editors is not clear; possibly it has something to do with professional politics.

There is no doubt, though, that the epidemiological approach is fundamental to schizophrenia, whether in terms of scientific inquiry or the development of services for sufferers of the disorder. Schizophrenia represents a major public health problem, largely through its frequent chronicity. It usually appears by a person's mid-20s and may then be present for 50 years or more; closing mental hospitals does not reduce the size of this problem, but simply means that it has to be managed in more complex ways. The lifetime risk, which does not vary much between different societies, is about 1 per cent of the general population. That represents a substantial number of people to be monitored, even though the appearance of new cases (incidence) is relatively rare. There is evidence at present that this incidence is declining; if it is, social factors are likely to be responsible, but what these factors are is unknown.

As with so many of the issues covered in this work, the answer to the question, "Is schizophrenia common?", depends on how you look at it. A good illustration of this methodological issue is the question of violence, described here by Elizabeth Walsh and Alec Buchanan. Epidemiological studies have shown that the likelihood of committing violent acts is several times higher for people with schizophrenia. Yet, in any given year, 99.97 per cent of those diagnosed with schizophrenia will not be convicted of serious violence. A complicating factor, though, is substance abuse, which does greatly add to the risk. Yet the contribution of mental disorder to violence is modest compared with a combination of male gender, youth and low socioeconomic status. So, although people with schizophrenia are significantly more likely than the population average to be violent, the proportion of total societal violence that can be attributed to them is small. At the population level, eliminating schizophrenia would make relatively little difference to violence, though each case is, of course, a personal tragedy.

Suicide, discussed by Hannele Heila and Jouko Lonnqvist, is the most common cause of premature death in schizophrenia, accounting for 28 per cent of the excess mortality rates noted among sufferers. There are indications that this suicide rate is increasing, which emphasises the need to look carefully at social policies such as community care. Since suicide victims have often been inadequately treated, the advice that "it is important to offer hope by optimising both psycho-social and pharmacological treatment" is timely. Epidemiology thereby illustrates again the interaction of clinical and social processes.

One of the main functions of epidemiology is to identify risk factors, though this does not always lead to immediate progress. For example, it has been known for 80 years that children born in the winter months have a 10 per cent greater risk of later developing schizophrenia (more so in the northern hemisphere). But no attempt to explain this has stood the test of further research. Similarly, both in Britain and Holland, second-generation immigrants from the Caribbean or Africa have shown a greatly increased rate of schizophrenia, and, once again, no explanation up till now fits all the facts. Racism, however defined, certainly does not.

Epidemiology developed mainly out of the study of infectious diseases, where diagnosis is rarely a problem. With psychiatric disorders, though, diagnosis is rarely straightforward, an issue that emerges many times in this book. Is psychosis an all-or-nothing phenomenon, or is it the greater presence of features that exist at lower levels in the non-psychotic population? For the clinician having to decide whether to intervene or not, it is a categorical answer of yes or no. But using epidemiology to throw light on a particular disorder, it is essential that all individuals in a sample should have the same kind of morbidity.

In the largest section of the book, Jim Van Os and Hel ne Verdoux deal with this issue. Concluding that it is unwise to rely on traditional diagnostic categories alone, they suggest that there "may be advantages in using a combination of polydiagnostic categorical and multidimensional representations of psychosis". But this sounds like a progress from diagnostic oversimplification to total confusion. More usefully, they point out that two individuals who have identical clinical states of psychosis may yet differ greatly in their degree of impairment and need for care. Presumably, genetic risk, personality and social factors may account for this difference.

An epidemiological issue that has become increasingly prominent is that of urbanisation and migration. In the 1930s, a classic study in Chicago showed that first-admission rates were highest in the centre of the city and diminished towards the periphery. The explanation then was that social isolation provoked the illness - the "breeder" hypothesis. Later studies suggested that local migration of affected individuals caused the excess.

(A lengthy reference list surprisingly omits the important 1974 paper by Levy and Rowitz.) The breeder view is back in favour, though urban birth is seen as more important than urban residence; why this should be so remains unexplained. With a few exceptions, all studies have been done in western cities, yet the largest urban populations are now in the third world, so that, theoretically, there should be an epidemic of schizophrenia there. So far, there is no evidence that this is happening, although the question has yet to be seriously studied.

The editors say that they have aimed to capture "the new and exciting themes that have been emerging over recent years". In particular, a science of "genetic epidemiology" is said to be emerging. Though genetic and environmental factors have often been seen as discrete, there is much discussion here of the ways in which the two can interact. For instance, a foetal brain that was genetically vulnerable might be further damaged by birth complications. It has been known for nearly a century that there is a genetic contribution to schizophrenia, but Stanley Zammit et al conclude that "its aetiology still remains something of a mystery - a complex, multifactorial condition that results from the influence of both genetic and environmental factors". That statement could have been written a long time ago.

This is a volume packed with information, but much of it consists of rather dense summaries of published work; not all the chapters have the kind of concluding overview that would be welcome. The question of schizophrenia in prisons seems to have been overlooked. One also wishes that the five editors had intervened at times to make the text more reader-friendly. But this is not to question the value of a substantial achievement in the compilation of scientific knowledge on this subject. Although most contributors are tentative in their conclusions and very conscious of the limitations of existing research methods, the general feeling is perhaps one of standing at the threshold of insights that new techniques and the better coordination of existing disciplines should be able to provide before too long.

Hugh Freeman is honorary visiting fellow, Green College, Oxford.