Carsten Dernedde

Mental health is an area of medicine which is uniquely shaped by the law. The development of the law relating to mental health is in turn uniquely driven by incidents involving mentally ill people, and the way the public perceives the role of the mental health services in these incidents.

The key word here is "risk". In this essay, I will try to trace how the recognition and prevention of risk has both positively informed day-to-day psychiatric practice with the evolution of the Care Programme Approach, and created a defensive atmosphere in real-life psychiatric practice in which the priority given to dealing with "risk" diverts attention and resources away from addressing "need".

For the purposes of this essay, a simple definition of risk will suffice, rather than an overview over the vast literature that tries to grapple with the conundrum of how to identify and predict risks. It does not require specialist knowledge to understand that mental illnesses can induce sufferers to commit suicide or become violent and harm others. The problem is knowing who will and who will not: contrary to what the public seems to demand of psychiatric services, the only instrument at hand to answer this question is refined common sense, rather than any kind of exact science.

Consequently, short of keeping everyone who suffers from mental illness locked up and under supervision, suicides and murders will happen.

1-2% of global mortality is due to suicide. The lifetime risk of suicide in schizophrenia is 4-13% and affective disorders increase the suicide risk 15-20-fold (Lonnqvist, 2000). Murder or injury caused by mentally deranged people is relatively rare. Only 1% of schizophrenics commit violent offences, and only 5% of murderers are mentally deranged (Chiswick, 2000).

The MHA 1983 provides the only legal basis in medicine on which patients can be treated by doctors against their express wishes (apart from the compulsory treatment of certain highly contagious diseases). It cannot be invoked for treatment for conditions other than mental disorders even in mentally disordered patients who refuse consent. It also cannot be invoked once it is established that a patient who has been detained under the act suffers from a mental disorder which is untreatable, even if he or she poses an undisputed risk. (I will reflect on this problem below).

The justification of the act hinges on risk. Although the Human Rights Act 1998 permits detention of the unsound of mind as an exception to Article 5, the Right to Liberty and Freedom, without further qualification other than recourse to the relevant legislation (i.e. the MHA), the HRA itself permits only such exceptions as are in the overriding public or individual interest. It is a condition that patients to whom the MHA is applied must be reasonably deemed to pose a threat to themselves or others. The MHA also puts in place a system of checks and balances to ensure that the question of whether such a threat exists is not merely in the hands of all-powerful professionals who might otherwise be inclined, for their convenience, to cast the net rather too widely. The Mental Health Review Tribunals (S.65 MHA 1983), whose task it is to exert independent control, have become reluctant to lean too far out the window against the advice of the professionals administering the act after the case of Maria Caseiro, who in 1995 stabbed a GP following her discharge by her Mental Health Review Tribunal against the advice of the Resident Medical Officer (Nicholson et al.1995).

Detention and treatment under S.3 MHA must stand on the dual pillars of "treatable mental illness" and "risk".

Hence, if a person who does not satisfy the first condition (and this has perhaps been established under S.2 MHA during a 28 day assessment order) declares an intent to kill himself or herself, it would be unlawful to prevent the suicide by applying the MHA. Wheat (2000) argues that although suicide itself is not unlawful, prevention of suicide is in the interest of the state. Therefore if one were to forcibly, temporarily prevent the suicide of someone who is not mentally ill, the courts would tend to take the view that it is proper to "keep the last door open", thus giving that person time to explore all avenues. Prisons and mental hospitals do not have the option to leave the choice to their inmates, as they are bound by their duty of care.

In analogy to the problem of suicide, dangerousness alone cannot sanction any form of detention unless an offence has been committed.

This bears out the liberal stance on mental illness taken by the current MHA, which puts great emphasis on the threat to civil liberties which could emanate from the authority in the hands of a profession to detain people perhaps merely on the grounds of being "different" or "difficult".

The media, while ready to shout "Human Rights Act" in support of the most unlikely causes, do not at present like to stress the essentially liberal attitudes that came to bear in the Mental Health Act, designed to protect vulnerable people from avoidable interference in their personal freedom. Laing (2000) warns of a shift back to the traditional legalism of the Victorian era, and of the government’s obsession with public protection, risk and control. The government’s obsession with avoiding bad press is well documented elsewhere and there certainly is more than enough bad press when it comes to mental health: Condren and Byrne (2000) find that media reporting about mental health focuses almost exclusively on dangerousness, and on perceived holes in the net of mental health care provision. We are spoon-fed a vision of murderous lunatics on the loose amongst us, despite empirical evidence of a decline in homicides by mentally ill people over the last 40 years (Cutcliffe et al. 2001). It is no wonder, then, that the government finds itself lobbied to strengthen the law to pre-empt the possible wrongdoings of the dangerous, and it would be naïve to assume that it is exclusively the well-being of the mentally disordered that motivates change in the legal framework of mental health.

However, it would be equally unfair to say that nothing to benefit patients has come out of the ongoing debate about risk and dangerousness. Indeed, to put it pointedly, it could be said that the problem of dangerousness is the strongest ally of the psychiatric population in driving change in practice, and securing funding for mental health. The perceived danger to the man in the street loosens the purse strings more than the silent suffering of the "others" (the mentally ill).

The development of care in the community would not have been possible without the revolution of the first effective treatments which only arrived in the second quarter of the last century, namely Chlorpromazine for Psychoses, Electroconvulsive Therapy (ECT) and Tricyclic Antidepressants. But the good intentions of care in the community might well have ended up as an excuse to shut down facilities and dump patients in the street, to become all but invisible, had it not been for spectacular failures of the system. Subsequent media "feeding frenzies" provided the necessary impetus to push through into law and statutory guidance recommendations for change which were made at, since 1985, 140 statutory public inquiries into potentially avoidable murders and other fatal incidents. (Sheppard, ongoing publication; also Biennial Reports of the MHA Commission).

The MHA 1983 stipulates that for those detained under the act, the local Health Authority and other relevant agencies have a duty to, prior to discharge, arrange appropriate aftercare (Section 117).

The act does not detail how Health Authorities should monitor whether the planned aftercare actually comes about. Consequently, patients were easily lost to follow up when they moved out of the area where they had been detained. Since 1991, Mental Health Services are obliged to follow the Care Programme Approach (CPA). The Department of Health has over time laid down rules in the form of statutory guidance detailing who this should apply to, how often reviews of the care plan should be carried out, how the individual care plan should be communicated, and who in relation to an individual patient should overlook this process. Prior to the Care Programme Approach, each discipline involved kept individual files and communicated relevant information, e.g. about emerging risk factors, to the other professionals as and when necessary. Inevitably, this led to lapses in communication with sometimes catastrophic consequences, which have been heavily criticised in subsequent public inquiries and publicity campaigns. The prime example is the case of Christopher Clunis, who had been lost to follow-up repeatedly in the time leading up to his stabbing of Jonathan Zito (Howlett and Zito, 2002). The Zito Trust was established as a consequence to campaign for tightening of the law, and prescribing rules for effective communication. Their campaign (and others) have not gone unheard: With the 1999 National Service Framework for Mental Health (DoH 2000), targets have been set to fully implement the Care Programme Approach across the NHS. The approach now covers not only patients who have been subject to the Mental Health Act, but all patients who receive ongoing specialist follow-up, as well as their carers (who in theory should receive their own regularly updated care plan).

I would argue that risk containment (through improved communication) rather than improving provision for patient’s needs is the vested primary objective of introducing CPA. It is modelled upon S.117 MHA aftercare planning and covers a wide range of issues other than risk, such as housing, transport, education, childcare or whatever else might impact directly or indirectly on the service user’s mental health. Ongoing assessment of these issues of need, however, is long established normal practice of the various professionals involved (Nurses, Doctors, Social Workers, etc.). In effect, CPA constitutes a set of formal rules requiring all this information to be regularly collated and updated in the presence of the patient and his/her carer in a multidisciplinary meeting, under the responsibility of an allocated Key Worker. The only real innovation, then, is that risks are assessed, discussed openly and documented in "real time" among all involved, avoiding delay and confusion when action needs to be taken.

The full local implementation of CPA is only now happening across the NHS according to targets published in the National Service Framework.

Although in theory, CPA is essentially normal practice applied and communicated in a formalised way, with the well known workload pressure in the NHS it comes as no surprise that before CPA, only the exceptionally difficult to manage patients or those who fell within the statutory care planning under S.117 MHA enjoyed the benefit of regular multidisciplinary planning meetings. CPA is not a system of unified note-keeping – individual disciplines looking after a patient keep separate files, as they often work in geographically distant locations (Community, Mental Hospital, Social Services…). It follows that there is some duplication involved when individual pieces of information are brought together. Planning and carrying out multidisciplinary meetings probably also tends to increase staff time spent (e.g. travelling to the venue where the meeting is convened), although this is hard to quantify as all those involved spend time with the patient whether or not they meet among themselves. I would hazard a guess that CPA does cost money. At least, I would argue (although no figures have been collected to prove my point), it costs time. If there is no additional staff time provided to run CPA – then does it not follow logically that, once they have properly identified and documented a risk, staff have less time to do something about it?

In theory however, CPA is supposedly quality controlled normal practice, and should have been implemented 10 years ago when NHS trusts first received directives to do so. This, the DH takes to mean that there is no need to allocate additional funding and consequently, the implementation has to come out of existing Mental Health Care NHS Trust budgets.

The NSF 1999 promises an increase in NHS spending on Mental Health towards achieving its main targets (such as 24 hr access to services) of £700 million over 3 years. Taking the example of the Budget of the local Norfolk Mental Health Care NHS Trust (NMHCT) for the year 2001/2002, this works out as minus £1.4 million which have had to be freed with cuts to existing services in order to be redirected to the struggling local Acute Trust (figure taken from the minutiae of a trust board meeting circulated within the trust).

I do not have sufficient understanding of the financial affairs of this Trust to comment how it fares with its financial management compared to other trusts, but the figure does give a flavour: that the blanket of Mental Health Service provision is very thinly stretched indeed. In these circumstances, the primat of risk containment over provision for need is bound to impact on the care given to service users.

Considering the unique position that the sharp end of the law is never far from the patient-doctor relationship in psychiatry, it seems at first surprising that there is not a great deal of case law involving negligence in psychiatry. I will examine why this might be so and review some relevant cases.

In applying the mental health act, professionals discharge a statutory duty. This is different to the legal standing of the voluntary patient-doctor relationship, which imposes on the doctor a duty of care in common law. The psychiatrist acts as a statutory agency when he applies the Mental Health Act. Doing so does not automatically confer a duty of care in common law, an issue that was raised in Clunis v Camden and Islington HA. The decision to apply the act cannot be challenged as negligent: when the doctor’s obligations to the patient flow from statutes and the patient doesn’t agree that the Mental Health Act justly applies to him (which is in fact one of the conditions for its lawful application!) the proper way to challenge this is by Judicial Review, normally after going through the statutory process of appeal to the Mental Health Review Panel (S.65 MHA). Judicial Review is part of public law (unlike negligence, which is private law) and its remit is solely to examine whether existing law has been correctly applied, not whether the law itself is correct. I will therefore now concentrate on negligence litigation. The principles by which the courts determine whether a defendant has been negligent towards a plaintiff were established in 1932 (Donoghue v Stevenson). The defendant has to have a duty of care towards the plaintiff, it has to be shown that there was a breach of this duty, and this breach of duty has to have caused actual damage. Duty of care (proximity) was the moot point in Palmer v. Tees HA. A psychiatric out-patient, who was known to be dangerous, murdered a four year old child. The claim was that the Health Authority was negligent not to disclose the known risk. However, the court dismissed the claim on the grounds that there was no duty of care towards the child, as any child, at any time, was in the same danger. Furthermore, as the patient did not suffer from a treatable mental illness, there was no legal instrument that could have been employed to prevent the patient from putting his fantasies into action. The court acknowledged, in other words, that the Health Authority had both hands tied behind its back in terms of predicting and preventing this crime.

A similar case, where a duty of care did exist, was Osman v Ferguson. The defendant here was a police officer but the decision could conceivably have been similar if a psychiatrist had been involved. Here, the identity of the child who was injured was known, as well as the possible intentions of the paedophile stalker. The claim was dismissed, as it would be unreasonable to make the police liable for damage caused by criminals.

The action in the case of Clunis v Camden and Islington HA was brought not on behalf of the victim, but by the offender after he had been convicted of manslaughter. He sued the psychiatrist for not pursuing statutory follow-up (S.117 MHA 1983) proactively enough to prevent the killing (and thereby his own detention in a secure hospital). Had the psychiatrist undertaken stronger efforts to find C after he failed to attend follow-up appointments, he might have found him so deranged as to warrant detention in hospital. The case fell through in the court of appeal for two main reasons: firstly, it was unacceptable, even if negligence were undisputed, that C should benefit from a claim that was based on his own criminal act. Secondly, while a statutory duty of aftercare did exist, it was not in the spirit of the Mental Health Act that the Health Authority should take on the same level of responsibility for the patient’s movements and actions in the community as in the hospital.

In other areas of medicine, the issue of who owed a duty of care to whom is rarely contentious. Violent incidents in psychiatry are more difficult to unravel, because the victim is a third party, and the damage is inflicted by the patient, not the doctor. Only if the patient was unable, due to the nature of his mental illness, to refrain from committing the act or to understand its nature, can the doctor be held responsible (provided he gave substandard care).

Actions for negligence in connection with suicidality have more chance of success. In Drake v Pontefract HA, a patient who suffered from agitated depression was admitted to hospital by a junior doctor and, two days later, absconded and injured herself badly by jumping from a bridge. The Health Authority was found to have been negligent for not supervising the junior doctor’s assessment and treatment of the patient, and for not instating close nursing observation and/or sedative medication to prevent the suicide attempt. It would have been interesting to see if the stamp of authority conferred by a consultant psychiatrist’s assessment of the patient would have kept the Health Authority out of trouble. There are many vagaries in assessing suicide risk, and there is a broad margin of discretion as to what degree of intrusive observation should be imposed on a patient at risk of suicide. In other words, provided notes are kept detailing why a particular decision was taken, it is unlikely in connection with the assessment of suicidality that the courts would find that no responsible body of medical opinion could have reached the same decision. This so-called "Bolam Test" (Bolam v Friern Hospital Management Committee) has come under criticism more recently for leaving the question of standard of care to the medical profession to decide. Lord Woolf (2001) mockingly calls the test "any responsible group of doctors knows best" and Brazier and Miola (2000) would like to see it supplanted by the notion that judges should be convinced of the inherent logic of the medical opinion they are presented with, before passing it as a sound opinion (cf. Bolitho v. City and Hackney HA). It remains to be seen whether doctors in general will face a tidal wave of successful negligence action after Bolitho, and whether the muddy waters of opinion surrounding the question of prediction and prevention of suicide or homicide will become any clearer in our courts of law. Adhering to professional guidance such as the rules on CPA laid down by DoH may well have a protective effect (Hurwitz 1998) in terms of negligence litigation, though whether it will improve suicide statistics is less clear.

Although Negligence strikes fear into the clinician’s heart, I would argue that the danger of actually becoming embroiled in litigation over negligence is not so great in psychiatry. "Defensive medicine" (the term is usually understood to mean over-cautious practice which does not primarily serve the interest of the patient) in psychiatry centres on risk. Professionals are anxious to be seen to take risks seriously at all cost, to avoid the finger of blame being pointed should it come to an inquiry into an untoward incident. This is a weakness which is exploited time and again when patients use claims of being at risk as a lever to get into the system in pursuit of some secondary benefit unrelated, or even detrimental, to their mental health. Examples of such secondary benefit would be avoiding homelessness or court appearances by being admitted to hospital "acutely suicidal". (Of course, a person who finds himself in such a predicament may well be genuinely suicidal, rather than just manipulative). Inherent in defensiveness is an element of collusion between professionals and service users – both have their reasons to fudge the issue if a clinical decision has been taken without real conviction, "just to be on the safe side".

The problem is that defensive practice, such as filling in endless risk assessment forms or admitting people to hospital who are unhappy rather than ill, uses up limited funds. Using a local example again, the waiting list for Psychological Therapies (which are of proven effectiveness for non-psychotic anxiety and depression) within the NMHCT stands at around twelve months – after the most critical screening to pick out the most "promising" and most unwell patients. Neurotic disorders for which such therapy may be the only treatment of substantial benefit have a prevalence of 16% in the UK population, and they are thought to account for 1 in 4 visits to the GP (quoted in Bhuj et al, 1997). Would we allow this proportion of the population to go largely without effective treatment if they were dangerous?

The government has presented a White Paper in 1998, which proposes to extend the Mental Health Act in two significant ways: Community treatment orders, to enforce effective treatment (mainly depot injections) on non-compliant high-risk patients in the community, and detention of untreatable people with personality disorders who are considered dangerous. At present, in the absence of a custodial sentence, and without hope at least of benefit for the patient, it is against the law to detain personality disordered people.

My scepticism with regard to the detention of psychopaths is that expecting psychiatrists to give a watertight prediction of who will commit serious crime on the basis of their personality is a ludicrous proposition, not far off the discredited "science" of phrenology a hundred years ago, which claimed to make such predictions on the basis of measurements of skull shape. Supposing this obstacle can be overcome, the next rock this boat will hit is the European Court of Human Rights which may well want to be shown evidence that detention is of benefit to the "patient" (i.e. the dangerous man).

I would hope that the vessel would finally flounder when the medical profession remembers their own dignity and refuses to lock up people whom they cannot help, on the grounds that society doesn’t like them. In a society which values personal freedom, psychiatry can only do so much towards reducing risk, and psychiatrists should not be afraid to say so. Psychiatry should fight for the legal freedom and the necessary resources to concentrate on treating mental illness.

Sheppard, D: Mental Health Inquiries published since 1985. Ongoing internet publication of summaries, www.davesheppard.co.uk/inqs.htm