Listening to Prozac 
by Peter Kramer.
Fourth Estate, 409 pp., £16.99, April 1994, 1 85702 233 5
Show More
Show More

Of all professionals perhaps doctors are the most gullible; and psychiatrists are perhaps the most gullible of all doctors. Over the last hundred years they have treated mental illness with cold douches, removal of the ovaries or thyroid, castration, hysterectomy, cooling almost to the point of death (and at least once beyond it), extracting the teeth and tonsils, enucleating the cervix, drilling holes in the skull, inducing coma through insulin, using metrazol to cause convulsions and slashing through a large chunk of the frontal lobe. All these savage and useless treatments were praised by their inventors and most were acclaimed by other psychiatrists.

Modern treatments are more effective and much less cruel, but there is still a tendency to hail the latest pill as a wonder drug before the evidence is in. Prozac, a new type of antidepressant, has given rise to extraordinary hype in the United States, where it is known as ‘the happy drug’: physicians are besieged by patients requesting prescriptions for it, even though they are not in the least mentally ill.

Peter Kramer is a bit more cautious, but he does make one seemingly remarkable claim. He asserts that the drug gives people who are unhappy but not depressed (‘dysthymics’ in the esoteric language of psychiatry) great self-confidence and the ability to accept rejection without being upset. He supports this claim with a series of case-histories drawn from his own patients. Unfortunately psychiatric case-histories are a modern form of fairy story, whose fallibility is well documented. First, the expectations of the psychiatrist can distort his judgment. Second, there is massive evidence that people pay more attention to positive instances than to negative ones: the psychiatrist will be struck by cases that support a certain thesis, and is likely to forget those which don’t. Finally, the psychiatrist only sees the cases he treats. He administers a drug and many of the patients (though not all) improve. But it is well known that almost everyone recovers from mild depression and the psychiatrist cannot tell whether there is any difference in recovery rates between those he has not treated and those that he has.

A controlled clinical trial was, of course, carried out with Prozac before it was licensed, and it was indeed found to have anti-depressant effects, though these were no stronger than those of the two types of drug already in use, the mono-amine-oxidase inhibitors (MAOIs) and the tricyclics. It appeared, however, to have some advantages over existing drugs. MAOIs can produce fatally high blood pressure if the patient eats certain foods – for example, cheese – while the tricyclics have unpleasant side-effects, like dryness of the mouth. Although they may possibly be less severe, Prozac, too, has side-effects; to name but a few: rash, nausea, vomiting, diarrhoea, headache, insomnia, anxiety, tremor, dry mouth, dizziness and difficulty in reaching orgasm. There is no such thing as a free lunch when taking medicines. Moreover, as Kramer acknowledges, the tricyclics sometimes work better than Prozac and vice versa – which anti-depressant works best with a given patient is largely a matter of trial and error.

Given all this, it is hard to see why Prozac has been hailed as a wonder drug. The answer is surely that it was the first of its kind (there are now three others) and new drugs are consistently overestimated. The drug companies, abetted by the media, stir up excitement: ‘Prozac works just as well as other anti-depressants’ is hardly news. Doctors and the general public are equally keen to believe that at last the perfect pill has been found. When the benzodiazepines (of which Mogadon and Valium are examples) were introduced, it was widely affirmed by doctors that they were not addictive. It has since been found that they are highly addictive; any benefits of Mogadon, which is prescribed as a sleeping tablet, wear off if it is taken continually, while its side-effects, such as impairment of memory, increase.

To understand how Prozac and other antidepressants work, it is necessary to know something about neurotransmitters, the substances in the brain they are thought to affect. When an impulse arrives at the endings of a nerve cell, it does not (for the most part) stimulate the next cell electrically; instead, it releases a neurotransmitter into the synaptic cleft – the space between two nerve cells. The second cell has receptors to which molecules of the neurotransmitter can attach. When they do so in sufficient quantity, the second cell fires (or in some cases is prevented from firing) through a complex biochemical mechanism that is not yet fully understood. There are at least four ways in which the activity of the neurotransmitter can be increased. A substance needed for its manufacture may be introduced, thus producing more of the neurotransmitter in the first cell; it is possible to increase the amount of neurotransmitter in the synaptic cleft by introducing a substance that destroys an enzyme responsible for breaking it up; the first cell has pumps that take back any neurotransmitter left in the cleft, and this re-uptake can be lessened; finally, the number of receptors on the second cell may be increased, making it more sensitive to the neurotransmitter. There are over forty known neurotransmitters in the brain – nobody has any idea why so many are needed. In addition, there are a large number of ‘neuromodulators’, which cannot produce firing on their own but which alter the action of the neurotransmitters. The picture is extremely complex; none of it is fully understood.

All the anti-depressants were discovered more by luck than judgment. Having accidentally discovered a particular molecule that is effective, the drug companies ring the changes on it in the hope of discovering a drug that will be even more effective or – perhaps of more importance to them – one that will circumvent the patent on the first one. There are now four main varieties of anti-depressant. The first, lithium, which is very effective for bipolar manic-depression, was originally used on guinea-pigs to test its effects on the kidneys. It was noticed that the animals became much calmer and a salt of lithium was subsequently found to have a calming effect on manic patients. The second anti-depressant, Iproniazid, an MAOI, was also discovered by chance: it had been in use to ameliorate tuberculosis and it was noticed that it made tubercular patients more cheerful. Its use as an antidepressant was greeted with the same publicity as Prozac, but it was withdrawn when it was found that it could cause jaundice.

About the same time the tricyclics were developed. Imipramine, the first of them, was a modified neuroleptic, a class of drugs known to be effective for schizophrenia. It was initially tested on schizophrenics, on whom it had no effect, but a psychiatrist persevered by administering it to patients with almost every known form of mental illness. To his surprise, it helped depressed patients; and the tricyclics were born.

Both the MAOIs and the tricyclics affect a class of neurotransmitters called the bioamines, the first by destroying the enzyme responsible for breaking the neurotransmitter down, the second by reducing re-uptake from the synaptic cleft by the first cell. There are several bioamines: the two of most concern here are norepinephrine (a close relation of adrenalin) and serotonin. Underactivity of these neurotransmitters was suspected to be a cause of depression partly because the existing antidepressant drugs increased the activity of both, partly because both were present in the ‘pleasure centre’ of rats. If whenever a rat presses a bar, electric stimulation is given to a particular area of its brain, it comes to press the bar faster and faster: it would appear that electrical stimulation of particular sites in this area is rewarding. Hence, it is possible that too little activity there might produce a lack of pleasure – in other words, depression. Moreover, the area in question, which is subcortical, was already known to be implicated in emotion. Both MAOIs and the tricyclics increase the activity of norepinephrine and serotonin, though they also affect the activity of other neurotransmitters, like acetylcholine, which controls the autonomic system, itself heavily involved in mood. The unpleasant side-effects of antidepressants are probably due at least in part to their effects on the autonomic system.

In the early Seventies, a biochemist began searching for a drug that would increase norepinephrine activity without affecting acetylcholine and which would therefore have reduced side-effects. Eventually, by serendipity, he produced a drug that had little effect on acetylcholine, but which increased the activity of serotonin much more than that of norepinephrine. The trade name given to the drug was Prozac. Kramer tells its story well, though with many omissions. He points out that the bioamine hypothesis of depression is by no means accepted by everyone. Many neuroscientists believe that anti-depressants actually reduce bioamine activity. The additional neurotransmitter in the synaptic cleft caused by an antidepressant may result in a decrease in the number of receptors on the postsynaptic cell to compensate for the excess neurotransmitter. If this process overshoots, the cell will receive less not more neurotransmitter. This would account for the as yet unexplained fact that all anti-depressants take about four weeks to work, although their blood levels peak a few days after one first takes them: some time would be needed to produce a reduction in the number of receptors on the second cell. So it may be either too much or too little activity in bioamines that causes depression; we do not know which. What is clear is that almost any hypothesis about the complex world of neurotransmitters can be turned on its head.

The reason, incidentally, that it is not possible to help depressives by injecting bioamines directly into the blood is the blood-brain barrier, the membrane between the bloodstream and the brain that only allows certain molecules to pass. Serotonin and norepinephrine cannot gain entry: they have to be manufactured inside the brain.

Kramer’s most startling claim is that, unlike other anti-depressants, Prozac increases self-confidence. He believes that this is because it affects mainly serotonin, whereas other antidepressants operate more on norepinephrine. But Prozac is also likely to affect the brain in many other ways of which we have no knowledge. Apart from his case histories, Kramer’s evidence for Prozac’s effect comes from studies of dominance in male vervet monkeys living in small colonies. It was found that the dominant monkey had on average 50 per cent more serotonin in the blood than non-dominant ones; and when the dominance hierarchy changed, the serotonin in the formerly dominant monkey decreased and that in the newly dominant one increased. Finally, if the dominant monkey was removed and one of the remaining monkeys was given Prozac to increase serotonin activity, and another a drug that reduces it, the former monkey always became dominant: by reversing the drugs the dominance, too, could be reversed. These are extraordinary findings, but they must be treated with care. When an originally dominant monkey was returned to the troop after being removed, he regained his dominance, even though his serotonin levels had dropped to those of a non-dominant monkey. Moreover, only 2 per cent of the serotonin in the body is in the brain; it is found in much larger quantities in the stomach and elsewhere. How then can a 50 per cent rise be attributed to increases in its activity in the brain?

It is probable that the formation of hierarchies by monkeys has a biological basis: it appears to be instinctive. And it is likely that the equivalent in people – self-confidence – also has a biological basis. Indeed, studies not mentioned by Kramer suggest that this is so and that personality has a strong genetic component. The real problem, however, with Kramer’s conclusions is that the effects of Prozac are just as readily explained if, like other anti-depressants, it renders some people (and also some monkeys) mildly hypomanic. Whether this condition is induced by a drug or arises spontaneously, anyone who has experienced it will vouch for the fact that it is accompanied by supreme self-confidence. This raises the question whether the effects of Prozac are any different from those of other anti-depressants. We do not know, nor will we know until a large-scale clinical trial on Prozac’s effects on personality is conducted.

Despite these reservations and despite a tendency to exaggerate, Listening to Prozac is an interesting and for the most part well-written book, with quite touching moments of naivety. Kramer is sometimes simplistic in his attribution of ideas: for example, the term ‘functional autonomy’ was not invented recently by D.F. Klein (a psychiatrist) but in the Thirties by a psychologist, Gordon Allport. It applies to a motive which, although initially operating to achieve some other goal, has now become autonomous. Again, Kramer expresses surprise that a drug can have at least as powerful an effect on personality as psychotherapy has. But hormone replacement therapy has beneficial effects on personality, thyroxin can have a dramatic effect on someone with a thyroid deficiency, illicit drugs like marijuana or crack have powerful if temporary effects, and the other anti-depressants can change mood, while lithium can eliminate the violent mood-swings of the manic depressive. Kramer’s surprise that the workings of the brain should be correlated with personality and mood is odd: who has ever doubted – at least in recent years – that there is a complete correlation between the activity of the brain and of the mind? Moreover, despite his touching faith in psychotherapy, based no doubt on his own case studies, the evidence strongly suggests that in most of its many guises it has little if any effect on personality or mood. Nor, pace Kramer, is it surprising that there may be specific centres in the brain controlling personality; we have known since the Thirties that cats can be made extremely aggressive by stimulating their brain in a particular place.

Kramer raises but does not answer the question whether it would be moral to change personality through drugs were it possible to do so. The question of designing personalities has been repeatedly discussed in the context of genetic engineering, but it is unlikely that we will need to provide an answer for some time. The delicate balance of biochemicals in the body must surely have evolved to be nearly optimal, at least in normal people, and drugs that alter it are likely to have unforeseen side-effects. Despite much searching, no one has discovered a drug that will improve memory or intelligence in normal people. It is another matter when something has gone wrong, as in Alzheimer’s disease or manic depression.

Even if it were proved, which it is not, that increasing the activity of serotonin raises confidence, it would not provide any understanding of the effects of drugs on personality. We would need to know which neural pathways bring this transformation about and to understand exactly how they produce the changes in mood and the accompanying changes in behaviour. Many of the biochemists working on neurotransmitters seem to be unaware that finding a correlation between their activity and a personality trait is only a tiny step towards discovering the mechanisms by which the brain determines personality. Despite the fortuitous discovery of useful drugs, the field of neurochemistry is a mess and few of the research workers in it agree with one another. In the meantime, there is no case for putting Prozac in the drinking water, to bring about a general increase in confidence, nor, as an eminent psychiatrist has sensibly suggested, for developing an anti-Prozac drug to reduce the over-confidence that exists in so many people, including some psychiatrists.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 16 No. 15 · 4 August 1994

Stuart Sutherland, in his review of Listening to Prozac by Peter Kramer (LRB, 7 July), disqualifies himself badly by having a go both at psychiatrists and psychotherapists. I expect a psychologist to do the former, but he has no basis for his scornful announcement that ‘the evidence strongly suggests that in most of its many guises’ psychotherapy ‘has little if any effect on personality or mood’. I cannot let this pass in the LRB, which has so many readers who’d just love to believe that Sutherland is right. Writing in the British Medical Journal, a proper authority, Jeremy Holmes, notes from an analysis of the research literature that the therapeutic effect of psychotherapy is greater than that of aspirin.

Sebastian Kraemer
Tavistock Clinic, London NW3

I hope he doesn’t have in mind the therapeutic effect of aspirin on mental distress.

Editor, ‘London Review’

Vol. 16 No. 16 · 18 August 1994

That Peter Kramer has written an almost bad book on Prozac is one thing; that Stuart Sutherland has written a bad review is entirely another (LRB, 7 July). Dr Kramer’s book is bad, not least because he bases his extensive conclusions on a mere 12 cases. His idea of Cosmetic Psychopharmacology misses the point by a clinical mile. Prozac, like another new anti-depressant, moclobemide, has a particularly beneficial effect on social phobic anxiety. The link with improved confidence is self-evident. Psychiatrists are well aware of the benefits and limitations of Prozac: it is no better than any of the other anti-depressants but can be remarkably well tolerated by patients. With depression increasingly being viewed as a chronic relapsing disorder, akin to diabetes, tolerance and compliance are very important in management.

Dr Sutherland, however, is throwing out the patient with the pill. His examples are simplistic, and in blaming Dr Kramer for naive use of facts, he should take more care with his own. Lithium is rarely, if ever, used solely for treatment of depression; only occasionally to boost the effect of other anti-depressants. The statement that ‘it is well known that almost everybody recovers from mild depression’ is meaningless in the clinical sense: what is even better known is that a very high percentage of people do not recover from chronic depression without help.

The evidence that psychiatrists are more gullible than other members of the medical profession is dubious. If the persistence of psychoanalysis is any guide, they tend, if anything, towards the conservative.

Robert Kaplan
The Liaison Clinic,

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences