Abstract
We are very fortunate tonight to have a member of the Society, Dr Andrew Johns come and speak to us. He is also a Guy's man – I put that second after the Society. He trained as a psychiatrist and then he went to St George's and dealt with addictive behaviour as a consultant. He also trained in forensic psychiatry and he is now a forensic psychiatrist in London, but he spreads his wings because he goes abroad: he has been called to Hong Kong and Jamaica on medico-legal work. He is also lead clinician for an 89-bed forensic unit at the Bethlem Royal Hospital. So tonight we welcome Dr Johns and I hope very much you will give him a warm welcome in the usual way. (Applause.) Thank you very much and good evening. Yes, I trained at Guy's, but a lot of us overcome adversity! I am going to talk about “murder, alcohol and madness”, to give it a lay title and give a medical perspective on the overlap between serious violent offending and the law. I will spend at least half the time dealing with medico-legal responses, because these cases are very untheoretical. You generally have a person on the floor, call him, for the benefit of the argument the “deceased”, and a man in jail charged with said offence, and we all appear before our learned friends in court who struggle to understand these issues and to explain them to the jury. There is a risk of considerable confusion when Medicine meets the Law headlong in the courts in front of a jury and tries to explain these issues. So this is what I am going to talk about.
Substance misuse, mental illness and violence
Those of you here who are medical will be aware of the overlap between substance misuse, mental illness and violence and that drugs, alcohol and violence go together, but we need to understand the nature of that connection in order to manage the health and risk of offenders. So I will start with a little epidemiology (a word that normally empties lecture theatres). As many of you will be staying for drinks afterwards though, you are likely to stay in your places. Epidemiology has a lot to tell us about the overlap between drugs, alcohol and crime.
Firstly, I will consider the range of possible offences. If you are a drug- or alcohol-using offender the law gives you a rich scope if you fancy breaking the law. You can commit a violent offence, and generally those offences are committed in an altered mental state. Separate, but not entirely non-overlapping, are acquisitive offences, the so-called “I nicked that because I have a drug habit and need to sustain the cost”. By non-overlapping, I mean that if you are a drunken burglar or shoplifter you will not be a very good burglar, you will get rapidly arrested. So generally these offences are committed in a more or less normal, non-intoxicated state of mind. There are other areas which I don't have time to talk about, for example drugs offences, substance misuse and impaired parenting; I am mainly going to talk about violent offences because it is my area of professional interest and work.
What proportion of crimes do you imagine are committed under the influence of alcohol? Well, it doesn't matter, but it is high as reflected in the police statistics of any major city. In one European survey, it was 30% of the crimes against life, 20% of robbery and assaults, 13% (a conservative number, in my view) of sexual crimes were shown to have been committed under the influence of alcohol, not necessarily outright intoxication, which I will come on to later.
How about drug use and offending? The data on drug use and offending is poor. The biggest study by far is that of the National Treatment Outcome Study (Stewart et al 2000), that looked at several thousand drug users who were in treatment. Now, bear in mind you get into drug treatment if by and large you are an opiate misuser and you are attending a methadone maintenance programme or a detox programme. So, with that bias aside, overwhelmingly these individuals had committed acquisitive offences, but they were not all criminals – so that three-quarters of all the crimes were committed by just 10% of the sample. Robbery is an acquisitive offence with violence or threat of violence and the percentage there is low, it is only 5%, so in general the drug users in this sample were not violent. But this interpretation is flawed, because the study tends to exclude users of drugs other than opiates. In my view it is probable that a high proportion of street robberies are related to stimulant misusers, but as they are not tested for those drugs, thus these offences are not recognised as a drug-related crime. So, drugs and alcohol and crime appear to go together.
What about the effects of mental disorder? Here is a question: “Are the mentally unwell at increased risk of violence?” Well, you might think it is obvious that that is so, but 20 or 30 years ago the main text books of psychiatry said that mental illness was protective and that having such a thing as schizophrenia or manic depression in general meant you were less likely to commit a violent offence. That view is entirely at odds with the public perception of these people, who have passed into the national lexicon as shorthand for violent, mad and dangerous to know; Peter Sutcliffe, the Yorkshire Ripper, Myra Hindley, who has recently died, Ian Brady, her confederate, Michael Stone, who murdered Lynn and Megan Russell; and the public is perplexed by the professional view that these people were ever safe. As ever, the truth lies in between, and here is some data on this.
Let us consider the MacArthur study (Steadman et al 1998) from the East Coast of the United States. This is an important study, because this is difficult epidemiologically to do and it is very expensive – what the MacArthur Violence Risk Foundation funded was a study in which over 1,000 male and female psychiatric inpatients were followed up a year after discharge. The very old and the very young are excluded, and the main criterion is that of needing psychiatric admission, so the cohort included diagnoses such as schizophrenia, manic depression and other conditions. These patients were then followed for a year after discharge and the research question was: have those discharged patients shown an episode of violence and what is their rate of violence compared with that of their host population? So if you are a man discharged from a psychiatric hospital to Baltimore the violence that you may show is compared with that for the male population in Baltimore. Regarding episodes of violence, this study does not draw only on conviction data; the researchers asked your partner, police, probation, anyone who you knew you, and they did so every three months for a year. A violent episode was here defined as “hitting a child; hitting an adult hard enough for the adult to have to go to the emergency room; or raising a gun against anyone”. After one year, 18% of those who had been patients showed major mental illness, and then you add in substance misuse and that tweaks the figures upwards, so that nearly 30% of those with major mental illness also co-morbid for substance misuse had shown an episode of violence. So there is something about the interaction between substance misuse and violence which is dangerous.
Now, interestingly, this substance misuse as shown by these patients was maximal in the three months after discharge. It is as if they've been in hospital for a while and they say what is the first thing you do when you leave hospital? You have a party, you have a drink. So the prevalence of drinking and drug taking problems goes up in those subjects in the first few months after discharge and that is when they show the highest rates of violence, after which their drug or alcohol misuse falls to that of the host population, as does also their predisposition to violence.
A similar study has been done in this country. This is the UK 700 study done by colleagues at the Institute of Psychiatry (Walsh et al 2001) – in this, 700 psychotic patients were followed up over two years and 20% were found to have been violent.
There was a risk ratio of nearly 1.5 for drug misuse. So drug misuse makes things worse. So that is goodish corroborative evidence. This study also showed that personality disorder is an independent variable for violence. Now, this is important, because we run away with the idea that mental illness alone is important. Fully half the patients I treat have a serious mental disorder and a record of serious violence. You give them a drug like Clozapine and do they turn into nice, law-abiding, liberal voting citizens? They do not. Half of them have material personality disorder which also needs treatment, and that is what the UK 700 study refers to. This research shows that mental disorder is a risk factor for violence and that substance misuse increases that risk.
What about possible explanations for this? Now, this is important, because the simple equation of cannabis or alcohol + weapon + victim = homicide is rarely so simple and as doctors, we need a good understanding of these issues before we get into the realm of legal defences. So I want to consider the ways in which drugs or alcohol can interact with mental illness and lead to violence.
Causes of violence in mental disorder and substance misuse
Now, you might think this is obvious but the obvious is rarely straightforward in medicine. Firstly, drinking at moderate levels can increase the risk of violence but the threshold for this varies according to social and psychological factors. If you accidentally jostle an intoxicated football fan in a pub, you are at risk of a more violent response than if you had knocked the elbow of your neighbour at a dinner party who had drunk an equivalent amount of alcohol – the main variable is that social setting and expectation condones or represses the expression of violence.
You don't have to drink much to fall into the risk area for violence. Goddard (1991) surveyed the self-reports of anger and violence in men who were “light-drinkers”, i.e. under 21 units of alcohol a week, compared with “heavy-drinkers”, i.e. more than 50 units a week. Only 1% of light-drinking men said they had been violent, compared with 9% of the heavy-drinking men. This points to the level of consumption as a risk factor for violence.
If you drink more you get into the realm of simple intoxication, which is anything but simple. By “simple” I mean that there are no effects which could not have been anticipated from the dose–response curve. You will bear in mind that depressants cause an initial disinhibition and that stimulants generally lead to arousal and irritability. But think how weird most intoxicated states are. Think of the impaired perception, judgement, impulse control and irritability – all states in which misunderstanding and violence may occur in a way that you will understand.
There is in some text books the notion of pathological intoxication. This arose as a French diagnostic entity. It seems to be raised by the defence where the man whom we will call the accused happens to have killed his wife after a small amount of alcohol and the defence are saying, “Could this be pathological intoxication, doctor? Could this man have drunk a small amount of alcohol but then inexplicably lost control and could that relate to the homicide?” It is of doubtful validity except among lawyers. Most cases, when you look hard at them, seem to have a better explanation in terms of the underlying organic brain damage/head injury. Remember, alcohol lowers blood sugar, so alcohol-related hypoglycaemia can account for odd behaviours. However novel medications can cause unexpected behaviours.
I have dealt with a case where a man was given a novel SSRI, a group of drugs like Seroxat and Paroxetine, and there was an episode of violence within a short time of his taking the medication for the first time, say within two or three weeks. There was a defence psychiatric report which stated that there is a well-known association between Paroxetine and episodes of violence. That is a possible legal defence because the young man would have not taken it before, but the violence would have to be totally unexplainable. It is no good saying, “I had an argument … with so-and-so”, or “I thought she had cheated on me. I then lost control”, and blame the Paroxetine. The violence would have to be almost totally random and directed in an unmotivated way for that to succeed in court. So that has yet to come to court and that is an interesting example.
We shouldn't lose sight of withdrawal effects because they are odd and aggressive. If you and your brain like alcohol or drugs, then you and your brain will miss it and cocaine is a good example. If you take too much cocaine, then you are going to be intoxicated, high, buzzy and impulsive – all risk factors for violence. The effects of cocaine when snorted last for a few hours whereas crack cocaine when smoked has an effect for only a few minutes, such is the short half-life. And so you have a contrast between the best you have ever felt when you have a lot of crack cocaine in you and the contrast a few minutes later when the level falls, and that is quite a destabilising contrast and that is, to my mind, a reasonable explanation for the high rate of impulsive violence shown by cocaine misusers when they are withdrawing from the drug.
I want to say a little about an “altered mental state”, because there are lots of ways in which you can be psychotic on various drugs and yet the more text books you open, the more the nomenclature looks confusing. You will see phrases like “toxic psychosis”, “cannabis psychosis”, “drug-induced psychosis”, all of which can be and have been raised as defences to violent offences. So I want to try and make life simpler by suggesting the following classifications: that a toxic psychosis is having too much of the drug inside you, and LSD is a good example; 25 µg would have an interesting effect on any of you; 50 µg a very psychotic effect; and the higher the dose the more psychotic you become and the psychosis wears off when the drug effect wears off. Cannabis can produce a toxic psychosis, but you have to smoke a fair amount of it.
The best example I can think of is magic mushrooms. I dealt with two young men who had picked magic mushrooms on Richmond Common and dried them in their airing cupboard, which is illegal, and which concentrates the psilocybin, and they then made mushroom soup or the equivalent. They were each psychotic and in a psychiatric hospital for the best part of two months. Now, neither had been psychotic before; there was no family history of schizophrenia. I saw them a few months later and there was no recurrence of psychosis, they didn't have schizophrenia. A month or so of unwellness is far longer than the half-life of psilocybin, so clearly they had a vulnerability, probably genetic, the psychogenic effects of psilocybin had pushed them over the edge. So that is what I call a drug-induced psychosis, and I will come on to that with cannabis in a little more detail.
Withdrawal psychosis: coming off a high consumption of depressants can produce a psychotic state, but it is uncommon. I have only seen one case of DTs due to alcohol and a few cases related to benzodiazepine withdrawal.
Now, what about the person with schizophrenia who says “cannabis is a good drug for me”? Many of my patients believe that to be true. We should not be surprised that patients with major mental illnesses take drugs or alcohol – data from a large three-setting study in this country (Weaver et al 2003) shows that just under 45% of community mental health team patients had past drug or alcohol problems and three-quarters of substance-misuse team patients had problems with mental health issues. So given that we structure our services in a very unitary way, we should not be surprised if patients do not present with a neat single illness classification.
Cannabis and psychosis
I want to talk a little about cannabis and schizophrenia, because there has been some interesting research here. There is some research data which suggests the more cannabis you smoke as a young person the more you are likely to to get schizophrenia as a young adult. But the prevalence of schizophrenia is largely the same in most countries in which it has been studied. The prevalence of cannabis smoking varies, so cannabis as an aetiological factor for schizophrenia can only be relevant in some populations. This was examined in the study (Caspi et al 2005) of Dunedin in New Zealand. A cohort born in 1972 were followed up over 20 years, so much was known of their drug consumption and later risk of psychosis. The key relevant finding is that of an enzyme showing a polymorphism which affects your vulnerability to schizophrenia. This is COMT (catechol-o-methyl transferase) which is an enzyme that breaks down amines, so it is important in the biochemistry of schizophrenia. There are two polymorphic forms – one allele generates a methionine amino acid and the other valine. If you have the methionine allele the rate of psychotic illness in these young people is 3%, and whether you smoke cannabis or not has no impact, but if you have a single based permutation to valine and you don't smoke cannabis your risk is at baseline, i.e. 3%, but the risk goes up of later psychosis to 15% if you smoke cannabis, which is a five-fold multiplier of the risk – but only if you have the underlying genetic abnormality. So why aren't we testing our patients for vulnerability to this particular cannabis psychosis? I don't know. I suspect my patients, if they were better readers of the medical literature, would say, “Will you test me for this valine/methionine/allele, doc, because if I have the methionine source then I can smoke cannabis, can't I?” to which I say, “You don't read the papers thoroughly enough. This only alters your threshold and if you smoke cannabis heavily you will still get psychotic on it”, because cannabis worsens most symptoms of schizophrenia.
A question that is often asked is whether the violence shown by psychotic patients is our fault, and perhaps your fault as well? Is it the fault of the medical profession that more psychiatric patients are not detained in large mental hospitals and that means that there are more of them to commit harm? That is the impression you may get from the tabloid Press, which describes homicides by mentally disordered patients in a lurid way. This was studied by Taylor and Gunn (1999): between 1957 and 1995, i.e. 38 years, homicide convictions in England and Wales increased from 116 to 522 a year. All countries in Western Europe showed that increase. Bear in mind that a homicide is a crime that is thoroughly investigated, so if the defendant has a mental illness it is likely to come out at trial. Mental illness is likely to be raised by the defence, who consider defences such as diminished responsibility and the issue is further tested by the court. So the proportion of homicides that attract a mental health disposal as a finding in court is known; it is the proportion of homicides that get a diminished responsibility and (much less) infanticide or an insanity verdict. So we can look at the percentage of homicides in which there is that mental health disposal, and that has fallen over that time period. It was a third at the start of the survey and it has dropped to 11%, i.e. the contribution of mental illness to homicide has fallen a little each year. What is the reason for this? Well, society has become more dangerous, drug and alcohol misuse has increased. You are far more likely to be mugged at the cash machine by a substance misuser than a person with mental illness. So there is a lower threshold for the expression of violence, a greater use of alcohol and drugs, and that partly accounts for the increase in violence in our society.
Legal aspects of intoxication and offending
Now, enough of medicine; I want to deal with the law and to consider what happens when the courts try to grapple with intoxication and homicide. For example, the law is concerned with locating personal responsibility for actions. With very few exceptions, committing a crime requires a particular mental intent. Among the few offences which do not, is parking your car on double yellow lines. If you leave it on double yellow lines you have committed what is known as a statutory offence and your mental state is irrelevant. But most offences are more complicated and, if you think of the role of drinking, drinking has been variously regarded by the medical profession as voluntary or as a symptom of illness, and that poses problems for the law. For example, if I kill when drunk am I (a) less guilty than killing when sober, because I may not have intended the act, or (b) more culpable than killing when sober, because I am responsible for getting drunk? Either view leads to uncomfortable conclusions. If (a) is taken at face value then you have an absolute defence to drunken homicide, then that excuses most of the Friday night mayhem that plagues hospitals and police stations. On the other hand, if I kill in a drunken state, is that killing as reprehensible as that of a terrorist who kills a hundred innocent people? They are not equivalent and to deal with this, law has evolved some complicated rules.
Generally, the acute effect of voluntary use of drugs or alcohol is not a mitigating factor, and the law is pithy about this: a drunken intent is still an intent. So if you stand up in court charged with an episode of violence and say, “Your Honour, you and I are men of the world; I had eight cans of lager and it was all a bit of a blur, but then I hit him. I didn't mean to”, this is no defence and you will be convicted.
But intoxication is relevant for those offences which require a particular mental element. I want to consider three medico-legal areas in some detail: amnesia, intoxication and capacity to form intent, and diminished responsibility, which applies only to homicide.
Amnesia for serious offending
Amnesia is important because it is commonly raised by defendants, and in one study (Pyszora et al 2003) up to 30% of sentenced lifers claimed a degree of amnesia for the violent offence. Three years later 40% of these were still amnesic. Now, you or I might think that if we do serious mayhem to another person we might remember it, but I have seen a number of defendants who say, “Well, I was sitting on the sofa with my girlfriend toying with my silk scarf and I'd had a lot to drink and it's all a blur after that. The next I knew I was looking down at her dead body”. The courts are often sceptical about this, but it is a fact that 12% of lifers show persistent amnesia. There are a range of possible medical causes, such as intoxication, alcohol-related blackouts or heightened emotional states. From a medico-legal point of view, amnesia does not affect fitness to plead, but it may contribute to a defence if due to a mental disorder such as fits or psychosis, which can form a defence.
Intoxication and capacity to form intent
I now turn to intoxication and intent, if only to show what a muddle the law gets into in this area. As a general rule, a drunken intent is still an intent. But the law recognises that there are offences which require a particular mental intention. The mental element in murder is an intention to kill or to cause really serious harm. If having killed, the jury find you did not have that intention, you are not guilty of murder. Back in the 1920s, an intoxicated man suffocated a girl during an act of rape. The case went all the way to the Lords and it was found (DPP v Beard [1920] AC 479) that where a specific intent is an essential part of an offence, if he was so drunk as to be unable to form that intent, he could not be convicted of a crime. Let me translate: the charge here is murder, defined in common law as an intention to kill or to cause really serious harm. If he was so drunk as to be incapable, unable of forming that particular mental intent, then he cannot be guilty of murder, and that is still the law. So what are the implications? The Lords' ruling went on “drunkenness can reduce the offence from murder to manslaughter”; for example, drunkenness can negate capacity to form a “specific intent”.
So there arises in English case law a group of offences for which the law takes the view that you can be too drunk to be able to form the specific intent required. This is difficult stuff. Firstly, there is nothing specific about this “specific intent”, which differs from offence to offence (and I will come on to that), because case law goes on further to draw a contrast between offences of “specific intent” and offences of “basic intent”, and a fellow called Majewski contributed to our understanding of this. In 1976 he got drunk, he took barbiturates and amphetamines and he assaulted the police. He had no memory of this and he argued in court that he couldn't have been guilty of assault because he had no memory of that. The Lords ruled that for offences of basic intent, self-induced intoxication is no defence (DPP v Majewski [1977] AC 443). If a defendant said in court, “I got drunk and took drugs, but I have no memory of assaulting the policeman and I can't be guilty of assault, can I?”, then the law says otherwise, i.e. allowing oneself to get into that drunken state is recklessness and that constitutes the necessary guilty mind. In summary, if you are so regardless of the consequences as to drink to excess, that is no defence to an offence of basic intent.
So we now have a curious list of offences of “basic intent”. From case law may be derived a set of offences for which drinking to excess is no defence: these are manslaughter, rape, malicious wounding, GBH, going all the way down to arson or criminal damage. These offences of “basic intent” are contrasted with offences of “specific intent”. Remember here “specific” is only defined by the relevant law. So for murder you have to have the intention to kill or to cause really serious harm. The law says that you can be so intoxicated as to be unable to form the necessary specific intent to murder. In practical terms, the forensic psychiatrist may be asked whether the defendant was so intoxicated as to lack the capacity to form the specific intent whereas the jury has to decide whether the specific intention was in fact formed or not.
Let me give an example from the courts. Two young men were charged with burglary of a supermarket. They said in their defence that they had drunk alcohol and taken cannabis all afternoon with their family. The defendants then equipped themselves with items to jemmy a skylight. They got on to the roof of the supermarket, they forced open the skylight and they got in and they stole various items. They were seen by a policeman, they got down from the roof of the supermarket and were arrested. They said in their defence that they were too intoxicated to form a specific intention to steal from the supermarket. I suggested to the jury, that they had the capacity to form an intention to steal as shown by their purposive actions in going equipped and on stealing. They were convicted.
Diminished responsibility
Finally, I wish to consider “diminished responsibility”, which is a particular construct from the Homicide Act of 1957 and a partial defence in that it has the effect of reducing murder to manslaughter.
The Homicide Act 1957, s 2 states:
“Persons suffering from diminished responsibility
For a finding of diminished responsibility, the law requires that when a person kills there has to be an abnormality of mind arising from one of the bracketed causes leading to a substantial impairment in responsibility. A doctor has to say in court if there is an abnormality of mind arising from one of those causes; it is for the jury to decide whether that led to substantial impairment or not. If a person with schizophrenia kills, there is not much medical doubt that schizophrenia is a disease that can lead to an abnormality of mind. But what of the defendant who was in a habitual and drunken state at the time of the killing? Case law has determined that the effects of self-induced intoxication do not constitute a legal abnormality of mind. So simple drunkenness doesn't get you within scope of the Homicide Act. But what of drinking that may be beyond someone's control? Here we turn to the strange case of Mrs Tandy. In 1987 and when drunk, she strangled her 11-year old daughter. At the time of the killing it was accepted by the court that she had drunk most of a bottle of vodka. Her defence was: “I am an alcoholic, I have the disease of alcoholism” and her legal team ran diminished responsibility. So the court heard from various experts. Some said alcoholism is a medical disease, studied by doctors and in medical schools, other experts said alcoholism is a different sort of mental condition to things like schizophrenia, it is entirely excluded from the Mental Health Act. The judge thought about this and he invented his own definition of “alcoholism”, i.e. that if Mrs Tandy was unable to resist the first drink of the day, then the whole of her drinking on that day was involuntary and any resulting abnormality of mind could fall within the partial defence of diminished responsibility. The jury found that she was able to resist the first drink of the day, she was convicted of murder and appealed, and the Court of Appeal upheld the “first drink of the day” test (R v Tandy 87 Cr App R 45, CA).
The difficulty here is that the judge cut through the Gordian knot of tangled medical view and came up with a definition of alcoholism which is unknown to medical science. I can remember the case of a Newcastle man charged with murder. He had been found very drunk beside the body of his drinking friend. It was accepted that the defendant had carried out the homicide yet he had no memory of this or motive for the killing. He relied on the Tandy test and he said, “I'm an alcoholic, and that's ‘diminished’”. I was for the prosecution, and took a careful alcohol history. He said that in the time before the homicide he drank every day on waking except on Sunday when he played football with his friends. I argued that his habitual drinking was under his voluntary control. The jury agreed and he was convicted of murder. Although the Tandy approach has been modified by subsequent case law, this illustrates the difficulty that the law has in working with medical concepts.
So I will finish my brief excursion on drugs, alcohol and the law. This is a fascinating area in which there are complex cases and the need for clear thinking on these medico-legal issues in the courts. Thank you very much. (Applause.)
Discussion
We have about 10–15 minutes for questions. Yes, Dr Roy Palmer. Roy Palmer, Coroner. Has anyone ever defined “mind”? Yes – sort of. I refer you to the case of R v Byrne [1960] 2 QB 396, (1960) 44 Cr App R. This says that “abnormality of mind” covers the mind's activities in all its aspects. It doesn't define “mind”, it says what “mind” encompasses. The legal test here is that of impaired perception, judgement or willpower, so if a person with severe psychosis may hear voices talking about the CIA, he may delusionally know you are in the CIA, and he may be impelled to attack you in consequence – all that is a clear legal abnormality of mind. You didn't mention methadone at any time. Is it going out of fashion? Should it go out of fashion? Well, it's a commonly prescribed opiate substitute. My impression is that the amounts of methadone prescribed each year go up and then the Government can say that there are more drug misusers in treatment. There are various doctors experimenting with alternatives to methadone. There is the Swiss trial of prescribed heroin, which they say is successful, and there is a current trial in London of prescribed heroin. I wonder if many of those doctors tend to forget that we have been here before, in that towards the end of the 1960s you could get prescribed heroin in London and you had “shooting-galleries” in Camberwell where you could go as a heroin user and you would be shown how to inject into your femoral vein by a nurse. That approach fell into disrepute, I think, because the physical health complications among injectors did not reduce; in fact they increased; and the later advent of HIV/AIDS meant that there was a move away from injection per se. My name is Bill Gardiner; I am a lawyer, but this question comes because I am a non-executive Director of a Mental Health Trust and we are now applying the 2007 amendments, and the section 1 definition of “mental abnormality” now refers to it as being something wrong with one's mind or brain. But the main point of my question is that it is a usefully huge definition now, which will permit personality disorder labels, autism labels, Asperger labels. The question is: as a professional giving advice to the court did you find this open door of many, many more labels useful or not? Well, that is an interesting question because the Act has only been in force since the beginning of November – s 1 of the revised Act, which states “mental disorder means any disorder or disability of the mind”. Remember, the 1983 Act had the four categories of mental illness, mental impairment, severe mental impairment and psychopathic disorder. The 1983 Act also excluded promiscuity or other immoral conduct, sexual deviancy or dependence on alcohol or drugs. There is also a treatability test for psychopathic disorder – all these are removed from the 2007 Act. As far as my personal experience goes, this has been tested by Mental Health Review Tribunals, where it is my impression that the broader definition makes it easier for the psychiatrist to detain or re-detain those patients with personality disorders. The test is only that the treatment is appropriate and that it is available. Bearing in mind that I only work with high-risk offenders, I find that the broader wording of the law is useful, because you are spared all the rigmarole about whether a psychopathic disorder is treatable. Richard Slee, surgeon and expert witness. I understand that one of the things that can happen is that a psychiatrist can say, “I can't treat this patient, he has got an incurable illness, and since I can't treat him I am going to discharge him.” Is that correct or not? No. Because the treatability test, so-called, has gone out of the Mental Health Act. It was there in the 1983 Act and is not in the 2007 Act. All the doctor has to say is that the treatment is available and appropriate, not that the patient is benefiting from it or the patient is treatable or untreatable. I think that is fine, because as doctors we don't have a list of untreatable conditions. We don't say that diabetes mellitus is treatable or untreatable, we say that some cases are difficult, some cases are advanced, but in most cases there is an intervention, whether palliative or not, which can make a difference. There is a current scope of thinking within personality disorder that there is hardly anything that is curative, but there are possibly some things that can make a difference and the treatability criterion was misused in the past, in my opinion, as a way of doctors saying “not in my ward” and many patients with personality disorders were excluded from mental health services. Dr Neville Davis. Neville Davis, Forensic Medical Examiner, at least until next month (laughter). FMEs (forensic medical examiners) are often faced with a situation where you have an opiate addict who is screaming blue murder with his withdrawal, “I want this and I want that!”, and the police are saying “we need to interview this guy”. What are your views with regard to whether they should be interviewed prior to being given medication, or should they be interviewed after medication, say with dihydrocodeine? Well, my views are shared with those of Dr Margaret Stark who has written a text book on this very subject (Stark 2005). The current approaches seem to me to protect the doctor and the patient, because what happens down the line is that the intoxicated or drug-using patient often says too much in the police station; disobligingly on video they confess to the burglaries or the assaults. It is later suggested that those admissions cannot be relied on because the defendant was intoxicated or withdrawing. However if the FME has a good record of the mental and physical state of the detainee, and to justify prescribed interventions, then generally, the medical management of the detainee can withstand criticism. Those FME records are frequently so robust that I have been able to say to the defence there is no evidence here of your client withdrawing, they were sleeping soundly for six hours before the interview. The present guidance is that the treatment should be symptomatic, so I think that judicious use of dihydrocodeine is entirely appropriate. Michael Brooks, Circuit Judge, and also I sit on the Restricted Patient Panel. In your slide looking at the evolution of murders over about 20 years, what was the definition of “mental disorder”? I took it to be a Mental Health Act disposal which is a pretty soft thing, isn't it, because on the tribunal you come across lots of people who have been sent to prison and then end up in Broadmoor. For example, the Yorkshire Ripper got life imprisonment and he's in Broadmoor. I think it remains true that, notwithstanding the small number of mentally culpable people who will be missed at trial and end up in prison, when you have numbers like 500 homicides a year, if there is a suggestion of a mental element I think it would be tested by the courts. Whether it's accepted or not I agree is a variable that we can't control for. I have run no end of mental defences which appear to stand little chance yet all are tested by the court, that I submit is robust. Thank you very much for an excellent evening.