Monday, October 01, 2001

Psychiatry Needs Brain Science To Shine

The Unbalanced Mind

By: Jonathan A. Leonard

Are we entering an era of reform in American psychiatry directed against overreliance on psychiatric drugs? There are signs that we are and, if so, that is a good thing.

Reform is needed because psychiatrists entering practice today are often shoved into pill-pushing posts by managed care organizations and HMOs that give them little time (commonly 15 or 20 minutes) to diagnose each case and prescribe a drug. It is needed because few U.S. medical school graduates want to practice psychiatry this way, and so more than half of today’s psychiatric interns are recruited from abroad.  It is needed because most mental hospitals have closed their doors. It is needed because time spent in psychotherapy is often minimal or is eliminated altogether. It is needed because, even when therapy is provided, the psychiatrist’s coordination with the psychologists, therapists, and counselors providing therapy, as well as with people providing assorted social services, is often weak. Is is needed, in the final analysis, because this pattern of ill-coordinated, marginal treatment causes many of the mentally ill to get disconnected from health care, thereby feeding a vast torrent of human misery that embraces— among others—roughly 200,000 mentally ill homeless people, 280,000 mentally ill prison inmates, and more than a million other Americans with severe mental disorders like schizophrenia, bipolar disorder, and major depression who receive no health care whatsoever.

This pattern of ill-coordinated, marginal treatment causes many of the mentally ill to get disconnected from health care, thereby feeding a vast torrent of human misery.

Because therapy is on the neglected side of all this, new books effectively espousing therapy are welcome. Julian Leff’s The Unbalanced Mind is such a book. 


Leff is an English psychiatrist, a professor at the Institute of Psychiatry of London with a distinguished career in social and cultural psychiatry. As befits a leader in this field, he has an ax to grind: He wants everyone to comprehend the great impact that social and cultural factors can have on mental illnesses, and how investigation of those factors can lead to improved therapy. He devotes much of his book to this task, directing attention to his work and the work of others on the role of caregivers in managing schizophrenia and depression. 

In discussing schizophrenia, Leff reviews work begun by George Brown and others in the 1960s. Brown’s team studied relatives caring for schizophrenics to determine whether emotional criticisms from these relatives had any influence on the course of the disease. The researchers did this by interviewing the relatives, using a standardized technique, and finding the number of emotionally critical comments made by each relative about the schizophrenic patient during the interview. On this basis, the researchers divided the family members into two groups, those with relatively high and those with relatively low levels of expressed emotion directed at their charges. They found that within nine months of leaving the hospital, most (58 percent) of the schizophrenics being cared for by the more critical relatives had a relapse, compared with only 16 percent of those less criticized. 

Leff relates how he and his team replicated Brown’s work in the 1970s and ’80s and supported other work indicating that if the schizophrenic adopted a strategy of keeping out of the critical caregiver’s way, this was roughly as effective as the antipsychotic medications of the day in preventing a relapse. Leff’s team then cobbled together a technique for working with the more critical caregivers to try to take the emotional sting out of their attitudes. They found that this technique succeeded in most cases and that, where successful, it drastically reduced the rate of schizophrenic relapse in the first nine months following discharge from the hospital. Similar trials in Great Britain, the United States, and other countries supported the essential validity of these results. 


The Unbalanced Mind also recounts how Leff and others applied this approach to depression. Early work had indicated that, in contrast to schizophrenics, many depressed people lived with partners, and that depressed persons tended to be even more sensitive than schizophrenics to an emotionally critical attitude by the partner. Studies in the 1980s and ’90s supported these results. 

Taking this as his departure point, Leff recently devised a trial to see whether couples therapy could improve the relationship between partners and whether that improvement would be an effective treatment for the depression. Patients with depression were randomly assigned to couples therapy or antidepressant medication. Those getting the medication received tricyclics for six weeks and were then given Prozac if they did not respond to the tricyclics. The couples therapy was provided for nine months; the antidepressants were provided for a year and were then tapered off. Leff found that for two years after the start of the trial, the group of depressed patients in couples therapy did better than the group on antidepressants. According to Leff, “this result gives support to the theory that depression is maintained by a problematic relationship between the two partners and that working to improve the relationship gives more lasting benefit than taking an antidepressant drug.” 

Beside this ringing conclusion, unfortunately, Leff’s description seems vague. He does not tell us basic things such as how depression was defined, how often the couples met with the therapist, how long the therapy sessions lasted, whether the doses of medication were tailored to individual patients, what those doses were, and how the course of depression in the two sets of patients was assessed. It seems reasonable to assume that this information has been provided elsewhere, but even assuming such information, the conclusion I just quoted seems overstated. At least in theory, one can compare a specific form of drug treatment to a specific form of therapy in dealing with a certain kind of depression.

But assessing the general benefits of drugs versus therapy in depression is much harder; for the benefits obtained in any individual case depend on hordes of variables, some obscure, that would seem difficult or impossible to accommodate in one study.


Despite such imprecision, most of the findings cited by the author from the schizophrenia and depression studies are reasonably consistent with our current knowledge. Consider Leff’s work on schizophrenia as it relates to what we have learned about psychosis. The schizophrenic relapses that Leff studied involved psychosis (severely disordered thinking and withdrawal from reality). In the case of schizophrenia, such psychosis is commonly associated with renewed or intensified “positive” symptoms such as hallucinations (usually hearing voices), delusions (commonly of being watched or followed), disorganized or catatonic behavior, and disorganized speech. 

Of course, psychosis can take many forms, and many things besides schizophrenia can cause it. Mania can cause it. So can depression, alcohol abuse, or use of psychoactive drugs such as LSD; and so can REM sleep, a brain state involving active dreaming that happens to be psychotic, which is something that we all pass through several times a night. 

While the psychoses produced by these various causes have widely varying symptoms, they share certain things in common. For instance, they are all associated with brain chemistry—specifically with enhanced activity by a couple of neuromodulators (acetylcholine and/or dopamine) relative to two others (serotonin and/or norepinephrine). Dopamine appears to be involved in the specific case of schizophrenia, although here we are likely to be dealing with increased neural sensitivity to dopamine rather than with increased levels of dopamine inside the brain. In a general way, these relationships among the four major neuromodulators in psychosis help to explain why certain drugs that tend to reverse the situation by favoring the work of serotonin/norepinephrine over that of acetylcholine/dopamine (as all our leading antipsychotic medications do in one way or another) can help to relieve psychosis. 

It is also relevant that all the psychoses seem to share certain elements with REM sleep, in which parts of the prefrontal cortex serving self-reflective awareness, insight, judgment, and executive guidance are selectively deactivated while visual perceptions and emotions are intensified—because all the psychoses appear to tilt the brain’s balance against thought in favor of internal perceptions and emotion. 

That being so, we should not be surprised to find that something such as frequent criticisms that evoke emotional responses can promote psychotic relapses in schizophrenics, and can also break through defenses erected by preventive doses of antipsychotic medication. Conversely, it seems reasonable that a style of care directed at reducing emotional tension and calming the patient by providing reassurance and support should tend to prevent relapses. All of this underlines something obvious: Well-coordinated therapy and effective management of the patient’s social setting play crucial roles in the sound treatment of schizophrenia.    

Turning to the Leff team’s intervention in depression, in a general way the results reported seem consistent with what we know: For instance, prolonged stress can cause a deep cerebral emotion center (the amygdala) to become chronically overactive, thereby promoting the chronically high levels of adrenal stress hormones found in many depressed patients. This is not the only mechanism involved in depression (the full story is complex and only partly understood), but it seems clear that both prolonged stress and abnormally high stress hormone levels have close ties to depression. 

Something interesting that Leff’s team found was that about one-third of the depressed patients’ partners were also depressed. This suggest that many of these partners may have been exposed to prolonged stress as a result of the patient’s illness, problems in the relationship, factors outside the relationship, or some combination of these. In any case, it seems likely that therapy directed at improving the partners’ relationship, and thus reducing friction, would reduce the stress felt by the patients. It would not be surprising to see such stress reduction countering their depression at least as well as a standardized course of antidepressant medication. 

So far so good. The author then branches out into snapshot accounts of various other subjects: treatment of mental illnesses in different cultures; a shift in diagnosis away from anxiety and toward depression that relates to the emergence of popular antidepressant drugs; a purported preference by the British public for therapy over drugs; a brief history of the asylums and their shutting down; and public misconceptions about levels of violence arising from mental illnesses. All this is a bit disjointed. Because the snapshots are brief, they require generalizations so broad that few specialists are likely to agree with everything Leff says. Even so, this part of the book is thought-provoking. 


The last two chapters, dealing with causes of schizophrenia, are more problematic. Studies of identical twins show that if one twin develops schizophrenia, the other twin has about a 50-50 chance of getting the disease. That means genes are important. What is more, prenatal and delivery complications tend to amplify these genetic factors, because they have a marked tendency to prompt eventual development of schizophrenia among children who are already genetically inclined in that direction. Together, these causes appear to account for almost all schizophrenia cases. 

Leff is determined to test the mettle of this view. He wisely chooses not to question the idea that genetic factors are involved. Instead, he reviews studies, including some of his own, suggesting that social factors—including poverty, urban stress, lack of cultural integration, and racial discrimination—could play a role. These studies typically depend on comparisons of different locales, often different countries, as well as different socioeconomic, cultural, and racial groups. 

All this looks rather questionable. The conclusions from studies of identical twins and of prenatal and delivery complications are well established. But comparisons that cross geographic, socioeconomic, cultural, and racial lines are hard to make because clouds of rebellious variables obscure the view. Separating cause and effect, not to mention separating nature from nurture in a social setting, makes disentangling the Gordian knot look like child’s play. 

Leff knows this. He admits that the sorts of studies he performs include extensive gray areas. But something in the hearts of most good researchers makes them want to assault the windmills of orthodoxy. The success of science depends on it. Moreover, while Leff has yet to prove his theory that urban settings cause schizophrenia, and while his chances of doing so seem slim, he definitely has some interesting data and he could well make worthwhile discoveries in other areas. Thus, there is reason to think that his quest should be encouraged. 


The big weakness of this book lies elsewhere, in Leff’s conviction that biological science is responsible for today’s dominance of biomedical psychiatry—and that biological science should be punished. To press this point, he picks out two dramatic strands of bioscience, brain imaging and genetics, and makes them into straw men. “My vision for the future of psychiatry,” he asserts, “is one that depends not on technical advances in making images of the brain or replacing bad genes with good ones, but on increasing our understanding of relationships between people....Progress will come from achieving the right balance between biological and social research into the complexities of the human mind.” Here he seems to direct his ire not only at the unjustified neglect of therapy but also at biological research and brain science. 

Many readers, even strong supporters of brain science, will sympathize with Leff’s frustration. They will see that his results are worthwhile. They will know that there has been enormous public and institutional interest in human genetics, brain imaging, and other brain research—and that this has been a magnet for research funds and attention, draining support away from those seeking to pin down social factors related to mental illnesses. So now that humanity’s overreliance on biomedicine is becoming evident, and psychotherapy’s star is showing signs of rising, it is understandable that Leff should single out brain science as his target. 

But it is the wrong target. Brain science gives no ringing endorsement of biomedicine’s victory over psychotherapy. Its discoveries are highly compatible with both. Scarcely anyone familiar with today’s neuroscience imagines that spectacular achievements such as unraveling the human genome or improving brain imaging are likely—at least in the short run—to have much impact on mental illnesses. Nor do brain scientists regard today’s psychiatric drugs as any cure-all, because they know that those drugs treat only symptoms. Since brain science affirms that many, even most, mental illnesses tend to be chronic (as suggested by an array of anatomic and other findings) but also tend to have biochemical ups and downs that can be influenced by a wide range of external conditions, there is every reason to consider therapy, monitoring, and ongoing assessment of the patient’s social circumstances to be key elements of treatment. 

Thus, the correct target is not brain science. The correct target is the split in psychiatry’s personality that has been worsened by the pharmaceutical companies, health insurers, and HMOs, as demonstrated by the current neglect of psychotherapy and the common failure to provide well-coordinated care. 

A frequent public misconception, at least on this side of the Atlantic, is that brain science’s progress justifies our overdependence on drugs to treat mental illness. That is tragically wrong.

A frequent public misconception, at least on this side of the Atlantic, is that brain science’s progress justifies our overdependence on drugs to treat mental illness. That is tragically wrong; but America’s mid-century experience with Freudian psychoanalysis suggests that single-minded pursuit of psychotherapy is no better. Instead, what we need is a strong foundation of knowledge that will restore public confidence, encourage funding, empower psychiatry, help the profession heal its split personality, and promote a sound blending of psychiatric medicine with psychotherapy. 

Right now, no established humanistic or social psychology can provide this foundation, because none can provide sufficiently compelling knowledge. There is another reason, however. Over the past few decades, hard science has been learning a lot about how the brain works. This bright light now being shed upon the mind and the nature of mental illness by brain science is so exciting that no theoretical psychology stands much chance of competing. So those of us seeking to reform psychiatry should not try to punish or outshine this beacon. Rather, we should seek to harness the source of its light to our purpose. 

This will happen someday, as the store of knowledge grows and it becomes increasingly evident that brain science can provide the foundation for a unified psychiatry. However, that day could be a long time coming if no one leads. Therefore, it is urgent that the leaders of psychiatry, including Julian Leff and many others on both sides of the issue, come to see the true nature of the problem. We all need to recognize that it is not just a matter of swinging the pendulum from drugs to therapy; that the foundation being built by brain science is compatible with social research and friendly to both sides; that this foundation can be used to heal psychiatry’s medical-therapeutic rift; and that such healing is vital, because the ultimate health and vitality of psychiatry, and of psychiatry’s neglected mentally ill multitudes, depends on it. 


From The Unbalanced Mind by Julian Leff. © 2001 by Julian Leff. Reprinted with permission of Weidenfeld and Nicolson. 

MEASUREMENT OF EXPRESSED EMOTION WAS THE BRAINCHILD of George Brown and Michael Rutter....It took them several years to finalize the technique, which centers on an interview with a relative who is caring for the sick person. Initially, in the 1960s, the research concerned people with schizophrenia, but it has now been successfully applied to a wide variety of psychiatric and physical illnesses. The interviewer asks the carer about the patient’s symptoms and behavior in the previous three months, a subject most carers will talk about eagerly. The interview, which occupies up to one and a half hours, is audiotaped. No ratings are attempted during the interview, but are made when the interviewer can listen to the recording at leisure. This entails going over crucial sections several times in order to be confident of the assessment, so that the rating can take up to twice as long as giving the interview and is obviously a painstaking procedure. The rater listens carefully not only to what the relative says, but also to the way in which the words are spoken. The rater treats the relative’s voice as a channel of information about the relative’s emotional attitude to the patient. To make this clear I am going to ask you to conduct a little experiment. Try saying as calmly as possible, “My son lies in bed til the afternoon, and when he gets up he doesn’t wash himself.” Now repeat exactly the same words, but convey as much anger with your voice as you can. If you think about how you changed the emotional tone of the sentence, you will find that the second time, you probably spoke louder and faster, and there were more ups and downs in the pitch of your voice. These clues from the voice, along with the content of the speech, are used to identify the emotions expressed by the relative. In this example, what is rated is termed a “critical comment” and the number of such comments made in the course of the interview is counted. 

The principle of the technique is that the rater’s own emotional response to the relative’s utterances is used to identify the emotion the relative is expressing. There is an obvious parallel with the psychoanalyst’s use of countertransference, except that analysts do not discuss with each other the clues they use to identify particular emotions in their patients; this ability is taken for granted as common to all human beings. But in a research project it is essential that the experimenters are equally sensitive to these clues. Therefore they have to attend a two-week training course in interviewing and rating, during which the techniques are explicitly discussed and a number of master tapes are rated. Trainees have to reach a high level of agreement on rating the master tapes before they gain approval to do their own studies.

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Helen Mayberg, M.D., Icahn School of Medicine at Mount Sinai 
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine
Charles Zorumski, M.D., Washington University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.