PRACTICAL LESSONS FROM HOMICIDES AND OTHER DISASTERS
On Sunday 28th June 1914 Archduke Franz Ferdinand’s chauffeur, Leopold Loyka, took a wrong turning on the way to visit victims of a failed assassination attempt in Sarajevo Hospital. Gavrilo Princip, one of a group of Yugoslav nationalists, was in a café having a snack and happened to catch sight of the Archduke’s car. Seeing Princip approach with a pistol, Loyka tried to reverse the car but it stalled, enabling the Yugoslav to fire just two shots, which killed both the Archduke and his wife, Sophie. The repercussions of this event were so momentous that within less than 10 weeks the whole of Europe was at war. The Archduke’s incompetent chauffeur was thereby responsible for World War 1.
Or so runs the logic or, more important, the instincts of many employers when patients commit homicide or children die in the hands of parents or others. A junior official, with recent involvement, is nominated as being responsible (whatever that might mean) for the actions of others. Partly to counteract this response, employers and investigators were instructed instead to look for root causes, an approach which is no less misguided.
A preoccupation with causality (and a particular kind of linear causality) not only carries with it a seed of blame, however much employers and investigators protest otherwise, but also takes attention away from the single factor which will reduce the chances of serious incidents taking place: that is, sound, professional practice.
In fact, all the lessons for practice which needed to be learned by staff from the death of Maria Colwell in 1974 to Baby P in 2008, and from the killing of Isabel Schwarz by Sharon Campbell in 1988 to the death of Dennis Finnegan in Richmond Park in 200? at the hands of John Barrett are already known. The only productive question for inquiry panels to ask is: how did it happen that staff weren’t aware of these lessons or, if they were aware, what were the factors which led them to ignore them?
A stark example of how institutional, professional and cultural waters close as quickly as is decently possible over a serious event can be seen in the responses to the killing of Jenny Morrison, an experienced social worker, by Anthony Joseph, a young man with active paranoid schizophrenia, in November 1998. For the next few years, a very dark cloud hung over her colleagues at the hospital where she worked and he had been a patient. An explosion of activity led to Action Plans, Internal and External Inquiries, followed by a blizzard of initiatives, policies and procedures. Staff went through the motions of hoisting these in but instinctively sensed that the factors and failings which led to a canny professional being on her own in a room with a person with a florid psychotic illness and intense delusional beliefs about her persecution of him had very little to do with essentially administrative instructions contained in large ledgers back in the office.
Within perhaps three years, most of her immediate colleagues had moved on and further homicides and suicides had blotted out what remained of the collective memory of the terrible incident and what it taught staff about how to work more safely with people when they are very disturbed. Subsequent inquiries, instead of gathering up again the lessons from her death and perhaps distinguishing some of them from those of their particular case, recommended fresh measures – that is, clinical, professional, organisational and procedural ones – in the essentially superstitious hope that future disasters could or would be prevented.
In practice, such inquiries either have no effect – because staff ignore them in the magical hope that lightning is unlikely to strike again in a similar place; or their unintended effect is to increase anxiety, defensive practice and workload, and impair therapeutic confidence and effectiveness. The reasons for this are complicated but the primary function of this kind of inquiry is to demonstrate active concern on the part of government and senior management rather than to educate staff in a benign and helpful way about how they can work more effectively, safely and (because they are connected) rewardingly.
Because of this primary function, inquiry panels are generally chaired by people with high status but little or no experience of direct work in mental health or childcare and the majority of panel members tend to be senior staff who left practice some time ago. Administratively minded investigators will inevitably produce administrative recommendations.
More important, their fading familiarity with day-to-day practice lures them into making a fairly minor logical error which tends to have major ramifications. In examination of practice in a particular case they come across failings and inefficiencies which they would find if they were to look at every case or any case at random – that is, sketchy record-keeping, sclerotic systems, interpersonal conflict or rivalry, misinterpretation of information and so on. However, they treat these factors as though they were both peculiar to the case before them and causative, following this up with wide-ranging and highly demanding recommendations for changes in practice and procedure which are impossible to implement. Not only does this have a negative effect on staff morale and confidence, it also misses the chance to educate staff about how to distinguish between routine cases where risk is low and the very few exceptional cases where risk is high.
It is difficult – perhaps because it requires a lot of confidence derived from years of practice – to say that something is relatively easy or straightforward. However, a further task of inquiries should be to distil the huge amount of information they inevitably collate into simple, soft principles that staff can internalise so that they become an automatic part of their practice.
Each homicide or death of a child throws up one or perhaps two crystal-clear lessons for practice. ShrinkTank provides some general principles about safe and effective practice, and some crisper lessons for minimising risk in day-to-day work. They come from 60 or so homicide or childcare reports over the past 25 years, starting with Louis Blom-Cooper’s seminal account of the death of Jasmine Beckford in 1984 (A Child in Trust), which to some extent he reprised in The Falling Shadow in 1995, the inquiry into the killing of a member of staff by an in-patient, Andrew Robinson. The recommendation of all subsequent inquiries need only be:
read these reports.
GENERAL PRINCIPLES
On 14th April 1986 a senior social worker, MR, went to the home of Kimberley Carlile, a 4-year-old girl about whom there were very serious concerns. She was to die less than two months later from a traumatic subdural haemorrhage consistent with having been hit or kicked. The post-mortem revealed multiple injuries which had been inflicted over the past weeks and months.
Part of the concern was that she had had no recent medical examination, she hadn’t been seen by any agency for a month and she had not taken up her place at a day nursery. MR was let into the house by Nigel Hall, the boyfriend of Kimberley’s mother, Pauline. Mr Hall was adamant that MR could not see Kimberley and said that the family just wanted to be left alone by the authorities. In the end MR was allowed to look through the glass panel at the top of a door to an upstairs bedroom where there were two small children asleep on the floor with their backs to the door. All that was visible were the tops of their heads.
MR nevertheless felt he had seen enough and later put a note on the file saying “no further action at present”, relying on other agencies to contact him if they had concerns. He explained to the inquiry panel which met the following year that he was “positively reassured by the fact that he was allowed to look through the glass panel by Mr Hall who could not have known how much of Kimberley would have been seen”.
What we see is not the same as what is there (insider blindness)
When vulnerable people die or others die at the hands of vulnerable people the reaction among outsiders – that is, the general public and media commentators – is usually one of anger but also of puzzlement at a baffling paradox: that professionals who are trained to observe and assess often seem to suffer from a curious kind of insider blindness.
In the days or weeks leading up to a serious incident there is, in a sense, one certainty – which is that the patient was on the way to killing someone or there was a child who was soon to die at the hands of a care-giver. The primary or basic task of professionals responsible for protecting vulnerable people from themselves or others is to see what there is to see: that is, a patient whose state of mind is such that he is on that path towards committing homicide or a child in immediate danger.
However, a number of filters get between the professional and the object of their work. Some of these are a function of human nature: that is
- we tend to see what we want to see
- we tend not to see what we don’t want to see
This may partly explain MR’s curious behaviour on his fateful visit to the Carlile-Hall household in April 1986. He did not want to see an abused child – perhaps because he feared taking action against Nigel Hall or because earlier failings on his part would have come to light – so he instead saw a child who was peacefully sleeping. Conversely, the infamous social work excesses of Cleveland, the Orkneys and Rochdale happened partly because social workers – and doctors – saw what they wanted to see: that is, abused children, who were in fact not abused at all.
There are other filters which cloud the lens through which professionals view their patients or clients, for example:
stereotyping
Christopher Clunis, who killed Jonathan Zito in a random attack, tended to be seen as a nomadic black man who took drugs rather than a person with a serious, chronic and lethal mental illness. Michael Buchanan, whose random killing of stranger in an underground car-park received virtually no publicity, was similarly seen as a young black man who mugged in order to fund his drug habit rather than as a patient with untreated paranoid schizophrenia.
Conversely, Andrew Robinson’s dangerousness was fatally diluted because he was seen as an intelligent, charming young man from a good middle-class family. The management and supervision of Richard Allott, who killed two pedestrians in Leamington Spa while on a manic spree, was clearly compromised because he had a good job and a psychologist girlfriend.
Daniel Joseph, who killed an older woman who had offered him sanctuary, was a young black man with a serious manic illness but in a sense was the victim of a different kind of stereotyping. His primary diagnosis, and the aspect of his functioning which determined both the services he was given and the ways in which staff responded to him, was one of profound deafness. He therefore tended to be seen as a victim of misfortune and circumstances rather than as someone who had the potential to create victims himself.
over-identification
Professionals who identify too strongly with people who pose a risk thereby reduce the room for manoeuvre and objectivity. In other words, their psychological closeness to the patient or client cramps the space between them which should be used for reflection and weighing up the costs and benefits of intervention. This is more likely to happen if the two parties share key characteristics such as class and ethnicity but can also happen if there are other strong echoes such as family history or dynamics.
The most obvious dangers of over-identification lie in child protection work where staff may unconsciously form an alliance with the adult-parent, with the result that potential for change or rehabilitation is idealised and pathology is minimised. This gives rise to what Louis Blom-Cooper called in A Child in Trust the rule of optimism. Since levels of identification are always likely to trump procedural obedience (partly because procedure is often necessarily vague and cannot cater for or home in on the precise facts of individual cases), it is clear that this issue is something for a particular kind of supervision rather than the crude dragnet of procedure.
under-identification
Protectiveness involves empathy, a derivative of identification. Part of the reason that a disproportionate number of black patients end up in secure or long-term care, many having committed serious offences while unwell, is that staff from different ethnic backgrounds fail to identify with them, a dangerous consequence of which is a lack of interest in their well-being and welfare.
Staff from the same or a similar background may well also fall short of the degree of identification necessary to act in an empathic and protective way towards black patients. A common motive for working in the field of mental health is that it gives professionals a continuing opportunity to underline the difference between them and those who use their services: that is, the existence of people who are demonstrably unwell seems to corroborate the sanity or health of those who are looking after them. An unconscious need to preserve this vital distinction may even at times work against patients being given really effective treatment. To put it another way, a professional who is comfortable with his or her own psychopathology is likely to be more clinically successful than one who relies psychologically on maintaining difference.
In childcare, a lack of protective interest ( that is, a personal-psychological one rather than one which is administratively driven) in the parent and/or child leaves staff at sea: when they should be caring about, they are only caring for. When the employing organisation also lacks a real interest in staff, as it almost always will, these difficulties will be compounded.
So-called passive surveillance, a common hallmark of professional or formal systems of protection and a theme which runs through the majority of cases which become the subject of inquiry reports, is a symptom either of a fear of intervention stemming from over-identification or of a lack of motivation and impetus resulting from under-identification. Again, converting observation into well-timed intervention is a function of professional skill, aided by sound supervision, and is something which is beyond the reach of administrative solutions.
who or what the patient represents
In only one of the 60 or so reports is there a suggestion that transference – in this case that of father-daughter – may have blinded a key clinician to the risk a patient posed and therefore muddied his management of her care. Persistent omission in inquiry reports of soft psychological factors and a preoccupation with hard, concrete ones is a symptom of the institutional wish to rinse humanity and, with it, human failings out of the care and treatment of those with mental health problems. This gives rise to the fantasy that mental health care can first be homogenised and secondly digitised on the road to the eventual elimination of all risk.
Transference also often plays a part in the exclusion or marginalisation of relatives, especially parents, which again partly explains why policy and procedure have so little effect on the amount and quality of contact that professionals have with key relatives. Intelligent and sensitive supervision can reach parts of the professional-patient relationship that guidelines never will.
poor inter-professional relationships
Difficulties between professionals and between professions may also cloud the view that individual staff have of patients or clients. These may derive from long-standing frictions between, say, hospital or residential staff and community staff, or generic professionals (for example, GP’s) and specialist ones; or they may have more to do with personal opinion or animosity. Underlying the former there is usually either rivalry or envy connected mainly with status or hierarchical position, or professional stereotyping.
These dynamics impede the free flow of information necessary to sharpen professional focus on events and actors. They are also difficult to talk about and are rarely addressed openly in inquiry reports. One explanation for hostel staff not telling Jenny Morrison that Anthony Joseph had made direct threats against her and not passing on other clinical information about him was that the relationship, in general at any rate, between residential and community mental health staff was an uneasy one. Further along the chain, other hostel residents had not passed on to staff that he had described some extremely disturbing psychotic experiences to them, which may have contributed to staff seeing his threats as a facet of his heavy drinking rather than of his paranoid illness. What fellow-patients could see clearly, Jenny Morrison could not.
Ideas can be dangerous as well as people
The belief that the primary aim in mental health and childcare is to prevent the need for admission into various forms of care is powerful and influential. It is both unconscious (because it is something staff would want for themselves) and conscious – that is, it is institutionalised in policy and in the reduction of hospital beds along with the creation of teams with the explicit brief of maintaining people at home. In both fields voluntary care is almost always felt to be better than compulsory care.
Because these attitudes coincide most of the time with what government, senior managers and the general public appear to want, they feel sound and uncontroversial. However, there are times when the primary aim should be to ensure that a person is admitted into care and, on occasions, to ensure that admission is enforced. Professional skill lies in working out which kind of person and which kind of situation is being dealt with.
Be clear about the objective task and how it may change
The psychoanalytic writer, Otto Fenichel, wrote that those who misunderstand the objective task are doomed to inefficiency and disappointment. In mental health and childcare the two dominant positions tend to be those of omnipotence – I can or should be able to predict or foresee disaster – and helplessness – disaster could strike at any time. They are in fact two sides of the same coin, as a consequence of which both individuals and organisations can switch rapidly between the two.
The professional task, as distinct from the idealised wishes of the public and politicians, is to find the middle course: that is, simply to make defensible judgements and to follow them through. Inquiry reports, on the other hand, tend to push senior management into the first position and more junior staff into the second, creating a destructive dynamic of blame and anxiety with a common element of inevitable failure.
But the objective task may change, sometimes in an instant. For example, circumstances may demand that the task of maintaining an alliance with a patient (which many staff feel to be an end in itself rather than a means to an end) should be superseded by a custodial or even quasi-judicial function. Again, understanding and being comfortable about moving between these two countervailing activities is a skill which comes with experience. Being open with patients about the possibility of the relationship changing in this way will in fact strengthen the alliance and make custodial action less likely. Formally revisiting this pivotal area in the professional-patient exchange from time to time will remind both parties that their relationship is necessarily both formal and fluid.
Anxiety can be dangerous too
Omnipotence and helplessness are essentially manifestations of a kind of therapeutic narcissism: that is, a feeling the professional has that he or she is somehow at the centre of things and that his or her capacity to succeed or to make mistakes may be vast. The anxiety that derives from the onus of this basic position will take its toll internally on the individual – perhaps in the form of burnout or problems with mood.
However, it also has more far-reaching consequences because it inevitably leads to very large numbers of children being placed on child protection registers (with high turnover as a result) and, in mental health, to every case being subjected to very complicated electronic systems of record-keeping and highly detailed risk assessments. This means that, instead of staff being encouraged, trained and supported to distinguish the very few high-risk cases from the very many low-risk ones, every case looks much the same and must therefore, it follows, have the same amount of energy and time expended on it.
When things go wrong, there has always been a failure in the therapeutic relationship
Medication may be the message but the medium is the relationship
In the days before the birth and subsequent dominance of root cause analysis, inquiry panels would often track back through the whole of the patient’s career and discover, to the surprise of those responsible for his current care, that he had a longish phase of stability in which symptoms and risk were minimal. Andrew Robinson, for example, was settled and accepted medication for two years while under guardianship, a legal arrangement under which staff have no power to compel a patient to accept medication. The panel found that he agreed to take his medication “because of the relationship he had with members of the team”. This may not have been a very advanced or mature once since his motive was not to “upset them by refusing medication” but it was nevertheless a platform for further work.
In fact guardianship was eventually not renewed for reasons which show that the subtleties and limitations of the relationship were not understood, namely: “the new found trust which he had developed would have been irretrievably damaged and the benefits of guardianship would have been limited at best”. The relationship was idealised with the consequence that the patient’s capacity to work without the authority of the law in the background was also idealised.
Many, if not most, inquiry reports involving both adults and children reveal
the absence of a functioning therapeutic relationship, which is not same as therapeutic contact (see LWL, RV, RS, CC, MB, MM, JB and others). To develop a truly protective and progressive relationship the aim should be to maximise the possibility of openness, a precursor of which is the development of trust – in particular, in the capacity of the professional to exercise her judgement, authority and powers in the best long-term interests of the patient. This is not a fanciful aim but it does mean that patients who sense that professionals are simply enacting rigid procedural guidelines or algorithms, or are likely, out of anxiety, to reach for the most restrictive and therefore safest option, are much less likely to disclose, for example, command hallucinations, or a wish to commit serious self-harm.
The sum of pathologies in a team should add up to something less than pathology
Therapeutic teams exist, in part, to iron out the idiosyncrasies of individual members and distil them into sound and effective interventions. For example, individuals are likely to be:
- over-anxious or under-anxious
- over-decisive or under-decisive
- over-responsible or under-responsible
- tending to see low risk or tending to see high risk
- passive or interventionist
- inclined to over-treat or inclined to under-treat
The healthy team under healthy leadership will understand and manage the failings and blind-spots of individuals. It will also be very suspicious of unanimity since this is likely to lead to excesses such as those in the Cleveland and Orkneys childcare disasters. Instead, it will understand the significance, say, of low-risk individuals seeing high risk in a particular case or of under-decisive non-interventionists wanting to intervene. The difference between unanimity and consensus needs to be understood.
Teams and individuals should also be aware of the ways in which their systems or practice may be dangerous which, from time to time, they inescapably will be. For example, there might be a culture of individual staff operating in isolation, perhaps borne out of competiveness or an unwillingness to ask for help or a lack of confidence in the competence of colleagues; or of pivotal decisions, perhaps about discharge, being taken when senior staff are present but more junior staff, who have had more contact with the patient, are absent.
Informal team audit of near disasters which fall short of serious incidents should be take place promptly before waters can close over them and useful lessons for practice are lost.
Continuity is all-important but can be dangerous too
Services should be organised in such a way that continuity, which helps professionals and sufferers develop lasting and protective relationships, is maximised. Allocating different functions – intervening in crisis, treating at home, assertive practice and so on – to different teams, instead of providing them in a single generic team of specialists, interferes with continuity. Allocating different aspects of a patient’s inner life and external functioning to different professionals has a similar effect. It may also mirror or maintain the internal fragmentation which the patient is struggling with, thereby missing an opportunity to help him work on and process factors which lead him to pose a risk to himself or others.
Continuity, of itself, is not necessarily beneficial. It may mask complacency, collusion, redundant patterns of working together and other factors which tend to increase risk. However, a personal historical knowledge of the patient is generally more informative than a written or electronic history.
Procedures should not be net consumers of energy and thinking
Working with people who suffer from mental illness is often stretching, though when it is also effective it is rewarding in equal measure. It requires a range of skills – that is, cognitive, personal, interpersonal, therapeutic and practical ones; a bundle of qualities – tolerance, patience, a sense of humour, a capacity for empathy and what is sometimes called non-possessive warmth, an ability to be both consistent and flexible, openness, a willingness to put personal experience to good use and to learn from patients and colleagues, and many more; and a certain amount of intellectual sophistication, which may be needed to apply trial and error, to balance competing or conflicting rights, to live with or resolve ethical dilemmas, to work from different perspectives simultaneously, to interpret or reframe what patients say or report, to use training to test intuition and vice versa, and so on.
Procedures and policies, properly conceived and drafted, should guide activity so that its effectiveness is maximised. They should be the director, not the show itself. They should channel energy and promote or facilitate thinking. What tends to happen, because the purpose of policy and procedure is misunderstood, is that obedience to them becomes the primary goal, with any benefit to the patient (and the professional, for that matter) being secondary or incidental. This absorbs huge amounts of energy and thinking which, more important, it also dulls or blunts.
In a sense, the bureaucratic machine becomes the patient who must be satisfied or appeased. Unable to be truly and fully available to the real patient, a prerequisite of effective and safe practice, the professional is like the bigamist anxiously commuting between rival partners.
TWENTY SOFT LESSONS FROM SIXTY HOMICIDE REPORTS
1. If in doubt, go out (DJ)
2. But make sure it is the right people going out (PH)
It is almost always more informative to see a patient than to speculate about risk before he is seen. Prolonged arguments about responsibility for patients are likely to increase risk. Having the right people present – for example, a GP who knows the patient well – is likely to reduce risk.
3. Relatives (almost always) know best (AJ, AR, CC, MD ++)
Relatives do not fall victim to insider blindness and, in assessing risk, professionals should always seek out relatives who are in contact with patients.
4. The knowledge that professionals have of patients is always superficial (AJ, JB, NB et al)
It is very easy to assume that knowing little means there is little to know. Professionals are always on the periphery of patients’ lives but tend to overestimate the significance or importance of the information they have, while underestimating the knowledge that others in the network hold.
5. Bias is permissible: allowing it to cloud one’s judgement isn’t (LL, MD)
Having negative views of relatives or colleagues is unavoidable but this shouldn’t be allowed to affect one’s judgement of their opinions or information they may have. Disliking or being afraid of patients are both normal and common but should be ring-fenced by supervision and team discussion so that they don’t influence decision-making.
6. Spending time with relatives will always save time in the long run (CC, AJ, TW, AM, DW ++)
7. Don’t use confidentiality as an excuse for passivity, inertia, or avoidance (SA, RV)
The institutionalised and instinctive tendency to marginalise relatives has to be actively resisted. Taking time to track down and visit relatives will improve decision-making and is likely in the long run to create time. Confidentiality is often wrongly used to legitimise a failure to consult relatives.
8. Be able to distinguish between procrastination and good practice (CC, DJ)
Handling cases through duty systems increases the risk of procrastination, which may be masked by staff stressing the importance of gathering information or disputes about who is responsible for the patient. Duty systems have a built-in incentive to delay intervention so that it falls to a colleague on another day.
9. In psychiatry there is no such thing as an idle threat (MM, AJ, JR, MD)
Threats to staff or others should be treated as a declaration of intention. Professionals who believe they must use the same burden of proof as the police or courts increase risk.
10. Beware of assumptions about patients becoming truths (RS)
Views that inevitably get passed around about patients – for example, “he is becoming dependent on the hospital” (see RS) – must not be allowed to move from opinion to objective fact. In any event, dependency is the springboard of autonomy and should sometimes be an initial therapeutic goal.
11. Always inquire further about violent thoughts (RS, MM, GN, RV++)
Many patients who go on to commit serious events have already told staff about having violent thoughts towards their victim. They may well feel ambivalent about this, because anxiety about what action might be taken may be stronger than their wish to protect the person concerned, and as a result may mention these thoughts obliquely or in passing. Staff should always inquire further about them in a thorough but relaxed, low-key way, bearing in mind that violent thoughts do not generally lead to violent actions. They should gauge, together with the patient, the extent to which they are likely to act on them and try to agree on what steps to take.
Equally, some patients will make no mention of violent thoughts which they are likely to act on but their silence does not necessarily mean that they are indifferent to the possibility of others coming to harm. Again, if staff understand their ambivalence about disclosing violent thoughts and the internal conflict which this sets up, they will find it easier (and therefore more productive) to ask about them in a benign and open way.
12. Inability to cooperate with services is often a symptom to be managed or understood, and not something patients should be excluded or punished for (SB, MB)
All behaviour has meaning and motive (SA)
Co-operation with professionals requires a certain amount of volition, capacity, organisation, socialisation, an ability to understand the likely benefits and to tolerate the possible drawbacks, and many other strengths.
When there is a history of not co-operating with professionals which is not understood or at least interpreted, responses to the patient are more likely to be polarised: that is, they will oscillate between compulsion (via the Mental Health Act) and rejection (in the form of discharge).
This dynamic tends to be established when the implicit or unconscious aim on the part of the professional is that the patient should conform or comply, thereby acknowledging her greater status or learning: not co-operating may then be experienced as disparagement or an attack which is met by covert retaliation. Thinking about the meaning of behaviour and the management of problems which follow from it will ensure that intervention is less reactive and more the result of considered judgement. Because certain patients very quickly get used to institutionalised, quasi-parental disapproval of their behaviour or attitudes, it will also contain an element of surprise which is a useful therapeutic tool.
13. The knowledge that professionals have of patients is always superficial (AJ, TW, SA ++)
Before making an assessment of risk or a decision about intervention, professionals must understand that the information which they hold, even if it has been diligently gathered from a number of different agencies, is necessarily superficial and therefore, to use Virginia Woolf’s expression, “like rain on the surface of the sea”.
Failure to understand this is the result of a perspective error. In planning or coordinating intervention staff feel themselves to be at the centre, pulling in information from the margins: whereas seeing themselves as one of many points on the circumference of the patient’s network circle is likely to give them a more comprehensive view of his functioning and the risk he may pose.
14. Take serious assaults, even if the outcome is trivial, seriously (CC, SC, AR, GN, SB, SA)
The idea that assaults in a sense have not happened if they did not lead to a conviction is a pervasive and persuasive one. Facts which are not corroborated by conviction are felt to be a kind of fiction and close investigation of facts is felt to be a matter for the police rather than mental health professionals.
However, preventive detention, based on a prediction of future risk, is sanctioned by the Mental Health Act. Staff should always make the time to find out and accurately record the details of assaults, remembering that with violent behaviour intent is as significant as outcome in the calculation of future risk. In fact, if a patient commits assaults which are not met with the deterrent effect of conviction and sentence, he may be more likely to commit further assaults in the future and therefore pose a higher risk.
A systemic reluctance to charge mentally ill people who have committed assaults or other dangerous acts means that there will be many patients on the caseloads of ordinary generic mental health teams who, apart from the fact that they have not been arrested or convicted for them, are no different clinically from their counterparts who have been convicted and are looked after by forensic services.
The man who, having lit the curtains in his room with a candle before walking out of his hostel, is admitted voluntarily to hospital for a couple of months treatment because he is well-known and everyone agrees this would be more humane than arresting and charging him, may pose exactly the same risk and require the same amount of clinical rigour as someone who happened to be convicted of arson and detained under a restriction order of the Mental Health Act.
15. Beware of concrete solutions to complex problems (RV)
Things go wrong when timing goes wrong
Humane and effective mental health practice depends on good timing. That is, when to take risks and when not to; when to expend energy on understanding or gathering more information and when to channel it into action; when to complicate and when to simplify; when to be permissive and when to be interventionist; when the rights of one person should be set aside in favour of the rights of another, and so on.
Concrete thinking, which is a common symptom of psychotic illness, may also be a facet of the professional approach to families where a member has mental health problems or child abuse is a feature. A perceived solution – often rehousing, a move into a hostel or supported housing, or daycare of some kind – gathers legitimacy and may ease protection out as the primary goal. Factors which increase risk – active symptoms of paranoia, interpersonal tensions or conflict, an inability to process anger, ambivalence, dependence and so on – are then seen as the secondary fall-out to an essentially practical problem.
16. Don’t let the tail of bureaucracy wag the dog of sound practice (NB)
The primacy of the bureaucratic imperative over clinical or therapeutic ones is a hallmark of the dangerous organisation or service. Bureaucracy – form-filling, procedures, CPA requirements, meetings and so on – should facilitate clinical practice rather than impede, delay or fragment it.
Fulfilling bureaucratic requirements tends imperceptibly to become the goal of practice rather than the means to the end of safe and therapeutic practice. This is partly because it is much harder to demonstrate that safe practice has been achieved than it is to show that bureaucratic requirements have been met – and easy to claim or assume that the latter will necessarily lead to the former.
17. Beware of the car mechanic syndrome (A da C/DP, AJ)
Just as the new car mechanic – or dentist, for that matter – tends to be critical of the work of his predecessor, so teams will often instinctively, but generally covertly, question the approach or diagnosis of the previous team or service of a patient whose care they have taken over. This is usually a function of unspoken rivalry or competition between Trusts or between teams within Trusts, combined with an endemic tendency to denigrate the work of others.
It is aided by a symptomatic reluctance on the part of staff in the receiving service to scroll thoroughly through old notes and past history, and to find time to visit the referring service (which is both symbolically preferable and generally more informative than the receiving service visiting the accepting one).
This potentially lethal failure to invest time at pivotal points in patients’ travels round the mental health system is partly caused by the primacy of so-called throughput over output. To put it another way, the immediate rewards of moving patients on outweigh the more distant ones of therapeutic effectiveness – which in fact is the best insurance against risk.
18. Diagnosis is generally not that important (MB, SA, DJ, NB, A da C/DP, PS, SB, JM, SC)
The significance of diagnosis depends on perspective and position. It will generally be of little concern to relatives, the public, the police and certainly the media when a patient commits a serious offence. On the other hand it may be a central issue in deciding whether a homicide was murder or manslaughter. It is also important, for example, for trainee psychiatrists learning their trade and in distinguishing between mental disorders which have a direct physical cause and those which do not.
For well-established patients, however, the primary function of diagnosis is often to legitimise internal transfer from one service to another. This may be important for individual teams or clinical staff, because it shifts responsibility elsewhere (a tendency which is encouraged by the growth of specialist services based on diagnosis), but is of much less importance for their organisation as a whole which remains responsible wherever the patient happens to be in the system.
In practice, however, most patients who go on to commit homicide have a variety of interrelated problems to do with their personality, mental state, use of alcohol or drugs, history of abuse, trauma or loss, difficulties in relationships and sexuality, and so on. Concern about precise diagnosis is likely to take energy, thinking and focus away from management of risk and the pragmatic control of symptoms.
19. Actions, just as much as words, may be an expression of a wish to be helped (LWL, AJ, A da C/DP, JB, PS +)
Many patients who are normally resistant to help or treatment turn up unannounced at hospital, their GP’s surgery or somewhere similar in the days before they commit a serious offence. Because it is often assumed that patients who are experienced as difficult, uncooperative or avoidant are wholly unaware that they may be ill or pose a serious risk to others, the potential significance of this behaviour tends to be overlooked.
20. When no-one is in charge, events will dictate action (DJ, MM, CC, A da C/DP, MD)
Homicide inquiries often reveal that the patient had quite a long phase of stability some years before the offence was committed. Invariably this was when he was looked after by one or two professionals who developed an understanding both of him and his potential for violence, felt a sense of commitment to him and worked to a simple and realistic plan.
When this phase comes to an end, which it inevitably does when staff move on or the patient is transferred to another service, its significance and the reasons for its success quickly get lost. Responsibility for the patient tends to become increasingly diffuse and, as instability begins to build, more teams, agencies and even panels become involved in the mistaken belief that more means better or safer. The endgame is characterised by a contradictory mixture of passive surveillance and islands of reactive intervention spread across several agencies.