In 1993 Richard Bentall went a bit mad.
He voluntarily took an antipsychotic drug and at first thought he'd get through unscathed.
"For the first hour I didn't feel too bad. I thought maybe this is okay. I can get away with this. I felt a bit light-headed."
Then somebody asked him to fill in a form. "I looked at this test and I couldn't have filled it in to save my life. It would have been easier to climb Mt Everest."
That was the least of his troubles. Bentall, an expert on psychosis from the University of Bangor in Wales who is in New Zealand under the University of Auckland Hood Fellowship programme, developed akathisia - unpleasant sensations of inner restlessness and an inability to sit still.
"It was accompanied by a feeling that I couldn't do anything, which is really distressing. I felt profoundly depressed. They tried to persuade me to do these cognitive tests on the computer and I just started crying."
Bentall had volunteered to be in a study run by Irish psychiatrist Dr David Healy. Volunteers were given either 5mg of the antipsychotic droperidol, 1mg of lorazepam, a type of tranquillizer, or a placebo.
"The experiment completely failed," says Bentall. "Because first, it's absolutely mind-bogglingly obvious to anybody after an hour whether or not they are taking an antipsychotic or a placebo - the side effects are so marked. There is no such thing as a placebo antipsychotic in that sense."
But it was the fact that most of the healthy volunteers who took the antipsychotic became so unwell, let alone do the cognitive tests, that meant the study couldn't continue. One psychiatrist became suicidal and had to be put under observation.
In his controversial book Let Them Eat Prozac Healy wrote about what the volunteers experienced. "It was not like anything that had happened to them before... Highly personal memories of previous unhappy times - broken relationships or loneliness - seemed to be flooding back. And if they previously held themselves responsible for these unhappy times, they seemed to hold themselves responsible for feeling the way they did now as well."
The antipsychotic experiment, which gave him a hangover for a week, typifies Bentall's approach to mental illness - rigorous scientific research coupled with a clinical psychologist's perspective.
He has a doctorate in experimental psychology. "Most of my arguments are research-based," says Bentall. "I'm just interested in what the evidence says about the nature of mental illness and how best to treat it. I'm a scientist at heart."
Attention to experimental detail and analysing what the science says is also the focus of Bentall's Madness Explained: Psychosis and Human Nature published in 2003. His thesis is that we have for too long focused on diseases like schizophrenia and manic-depressive illness and neglected symptoms such as delusions and hallucinations.
Bentall dislikes categorical psychiatric diagnoses which he says are born out of, and largely unchanged from, the Victorian era: "The assumption is that there is a borderline between mental illness and normal function - that is, you have a mental illness or you don't."
Bentall notes studies that show the majority of people who are diagnosed as depressed are also anxious and vice-versa: "It becomes somewhat arbitrary whether you say someone has depression or anxiety."
Blanket diagnoses such as schizophrenia are similarly not meaningful - especially when two people diagnosed that way can each have completely different manifestations.
Bentall prefers to focus on symptoms, usually by getting the patient to list and identify their problems. "A patient may say: 'I do hear voices, but actually it's never really bothered me, but I've got this terrible relationship with my husband' or whatever."
The idea that psychosis is more understandable than is commonly thought builds on the work of Dutch social psychiatry professor Marius Romme, a founder of the Hearing Voices movement. The central premise is that hearing voices - usually a key symptom in the diagnosis of schizophrenia - is not in itself a sign of mental illness; that many people hear voices with no ill effect and that if hearing voices causes distress, people can learn strategies to cope with the experience.
Bentall, who spoke at a meeting organised by the New Zealand arm of the movement (www.inter voiceonline.org), is not saying antipsychotics should never be used, but that they should be used with great caution.
He says though there is considerable doubt about how effective antidepressants such as prozac are compared to placebo, there's actually no doubt that antipsychotics do, in quite a lot of patients, reduce the severity of their psychotic symptoms - particularly paranoid delusions and hearing voices in the early stages of treatment.
"They appear to work on those two symptoms of psychosis, but not on others," says Bentall. "In fact, some symptoms they may even make worse - especially negative symptoms, such as loss of motivation and apathy."
But they're also seriously dangerous drugs. "There is no street market for antipsychotics," he says. "No one is trying chlorpromazine [the oldest in the antipsychotic family of drugs] on the sly."
As well as mood-altering side effects, variously referred to as "neuroleptic dysphoria, (literally taking hold of the nerves and the opposite of euphoria), neuroleptics, or antipsychotics, can also have life-threatening consequences including diabetes and increased risk of heart attack. "Studies show life expectancy is reduced by them."
What worries Bentall is how many mental health services seem to ignore what the research says and when an antipsychotic medicine doesn't work, simply up the dose.
Once again Bentall refers to the science - that about a third of recipients don't get any benefit whatsoever from the drugs. And research that shows if patients don't respond at a relatively low dose, they're not going to respond to a high dose. And are very likely not going to respond to any other anti-psychotic.
The optimum dose of antipsychotics is about 350mg per day (measured as chlorpromazine equivalents). Yet a recent study in the north of England found the median dose of antipsychotic drugs was about 600mg and about a quarter of those reviewed were on a gram or more a day.
"The average dose was about twice the optimum. How does that happen? It doesn't make any sense." Bentall suggests the reason such "unethical doses" occur is because mental health services have come to rely on these drugs as if they are the only treatment available. "When a patient doesn't respond, they just up the dose in some magical belief that hopefully something will happen."
But while promoting alternatives like cognitive behavioral therapy - the Government-sanctioned treatment of choice for depression and anxiety disorders in England - Bentall also points to research that shows all psychotherapies work, and that no type is more effective than any other. It's a finding that surprised many, including Bentall.
Closer analysis highlights a common theme. "The quality of the relationship between therapist and patient explained most of the result."
It seems blindingly obvious that having a good quality, empathetic therapist is likely to get good results, so why doesn't it happen? "Establishing good relationships with patients shouldn't be that difficult, but most psychiatric services seem to find it very difficult indeed," says Bentall.
He says many services operate from a coercive model: "We know best. We've got the treatment. Better take these no matter what the side effects. Do what we say and if you don't, we'll put you on a community treatment order and you'll be legally obliged to do what we say."
As he points out, it's not, and never was, a good way to try to help anybody. "The evidence that these coercive methods make the public safer is just not there. What they do is turn patients off psychiatric services. It seems completely wrong-headed."
Bentall is a strong advocate of patients being given choices - especially the choice to have drug treatment. "I'm not against people being given antidepressants or antipsychotics, but the evidence is increasingly showing that having services which rely on drugs is a very bad idea."
In the realm of low-end anxiety and depression type illnesses, he says it's quite clear that psychological treatments are more effective than drug treatments. And at the severe end of illnesses where people are delusional, he says choice is very important too.
"The drugs do help some people, but some they harm more than they help. The only person that can tell that is the patient themselves and it's important to provide people with alternatives."
Bentall's book, Madness Explained, has drawn comparisons with iconoclastic psychiatrist R.D. Laing and his books, especially The Divided Self published in 1960. Reviewers point to the similarity with Laing's view that the expressed feelings of the patient are valid descriptions of lived experience rather than simply symptoms of some separate or underlying disorder.
How does Bentall feel about the comparison? "I don't know whether to be flattered or terrified. I have mixed views about Laing. He was a talented observer of patients and had a magnificent ability to understand people. I think he was an incredibly courageous individual, but one who was terribly flawed in many ways. He couldn't recognise when two ideas were contradictory and had that legendary problem with alcohol."
What about the anti-psychiatry label that Laing was branded with? "I'm anti bad psychiatric theories but not anti-psychiatrist," says Bentall. "I think we need more and better psychiatrists. I'd like them to have a very different kind of training than they have today."
Not surprisingly he argues for more emphasis on psychological approaches. And he would like the rivalry between and psychology and psychiatry to disappear. "My objections to psychiatry are science-based, evidenced-based. They are against theory and practices which are harmful to patients. I want psychiatrists to use more rational and more humane methods of treatment."
NEW ZEALAND'S $60M DRUG BILL
There were about 400,000 antipsychotic prescriptions in New Zealand in 2008 at a cost of $60.5 million. Drug funding agency Pharmac began funding for the antipsychotic medicine ziprasidone (Zeldox) as a second-line treatment for people with schizophrenia and related psychoses from August 2007.
Pharmac says ziprasidone, one of the newer "atypical" antipsychotic range of medicines, became funded for people who have tried others (clozapine, risperidone, quetiapine and olanzapine) but stopped using them because of side effects or inadequate response.
In October 2008, after concerns raised in both the United States and Britain about high rates of use of antipsychotics in elderly people, Pharmac and the College of Psychiatrists launched new guidelines for use in that age group.
Concerns included potential health risks (possible increased risk of stroke and higher death rates), and that over-prescription of antipsychotics to people with dementia in care does significant harm.
Pharmac's prescribing data showed there was comparatively high use of antipsychotics among older people.
In 2006/07 about 35,000 prescriptions were recorded for people aged 80-90.