Suicide, national inquiries and professional judgement

Prevention of suicide is one of the cornerstones of the current government’s public health agenda. In Our Healthier Nation (Department of Health, 1998) and later in Saving Lives: Our Healthier Nation (Department of Health, 1999b ), the government identified mental health as a priority area alongside heart disease and stroke, accidents, and cancer. The aim was to reduce deaths by suicide and undetermined injury by at least a further sixth (17%) by 2010 from a baseline at 1996. It was estimated that if this were achieved, 800 lives would have been saved in 1996 and 4000 lives in total between 1997 and 2010. In this context, the National Confidential Inquiry into Suicide and Homicide by People with Mental Illness took on additional importance. In all, there have now been five reports, including the preliminary report published in 1994 (Steering Committee of the Confidential Inquiry into Homicides and Suicides by Mentally Ill People, 1994) and the progress report published in 1997 (Department of Health, 1997). Undoubtedly, these reports have been influential in determining government policy, in raising the profile of mental health services and in keeping the services in the eye of ministers, even if, as a result, clinicians often feel more under scrutiny than is comfortable.

The principal findings are now well established. For example, approximately 25% of all people who die by suicide have been in contact with mental health services in the year before death; half of these have been in contact with mental health services in the week before death; suicides cluster in the first year after onset of illness; and 63% of those who die by suicide have a history of self-harm. In addition, 16% of cases are of psychiatric in-patients and in those suicides occurring on the wards, most are by hanging. A substantial proportion (24%) of suicides occur within 3 months of discharge, and post-discharge suicides are at a peak in the first week after leaving hospital. Furthermore, 41% of post-discharge suicides occur before the first follow-up appointment (Department of Health, 1999a ). These findings have now been further confirmed in a more recent report (Department of Health, 2001).

Report recommendations

Safer Services

In Safer Services (Department of Health, 1999a ), 31 recommendations were made, and it could be argued that many of these went beyond the evidence and also that fewer recommendations might have concentrated all minds on a choice list of feasible and important objectives. For example, the first recommendation was on training:

‘All staff in contact with patients at risk of suicide should receive training in the recognition, assessment and management of risk, of both suicide and violence, at intervals of no more than three years’ (p. 10).

On the face of it, this recommendation was reasonable. However, there was nothing in the Confidential Inquiry itself that indicated that suicides had occurred either because members of staff were inadequately trained or had attended refresher courses infrequently. There was certainly no evidence to indicate that 3-yearly refresher courses, rather than 1-yearly or 5-yearly, were necessary in order to keep staff skills up to date. The impression given to independent observers was that the Inquiry had become an excuse to make national mental health policy. However, some of the recommendations that were pertinent were lost in the long list. For example, the recommendations about in-patient suicides, particularly those concerned with changes to the physical structure of wards, were important, as were those regarding post-discharge suicides. In both cases, these recommendations became part of national mental health policy and as such were part of the targets on which managers were judged.

Safety First

The 2001 report Safety First had fewer recommendations: 18 in all (Department of Health, 2001). This is still a considerable number and, I believe, still open to the charge that the recommendations go beyond the evidence. There is an emphasis on procedure, the usual resort of bureaucrats far from the field of action, whose aim is to control outcomes for which they are responsible but which they cannot directly influence. Recommendations covered suicide prevention strategy, in-patient care and post-discharge follow-up, the care programme approach (CPA), training, substance misuse, ethnic minorities, the criminal justice system and stigma.

Specifically, it was recommended that the CPA must be overhauled such that national criteria for enhanced CPA be established and all care plans for enhanced CPA include explicit plans for responding to non-compliance and missed contact. Enhanced CPA should normally apply to patients with schizophrenia.

As regards training, it was recommended that National Health Service (NHS) and social care organisations should have a system for approving training courses, and staff should attend only approved training. Approval should be based on the evidence that training leads to benefits.

Implications

The recommendation on the CPA and schizophrenia ignores what is priceless about professional judgement, the ability to make clinical decisions that are unfettered by administrative interference, yet are conscientious and competent. It demonstrably treats all patients with schizophrenia as a homogeneous group in need of enhanced CPA. Clinical judgement and individual difference are apt to suffer in such a bureaucratic environment.

The strictures on training are unlikely to encourage the development of training courses; rather, they will end up strangling courses. I admit that training has to be functional and utilitarian, that is, that it should at least adequately equip a trainee with knowledge and skills for a purpose. But the stricture that only training that leads to benefits should be approved raises questions about what one means by ‘benefit’ and how ‘benefit’ will be measured. And this is in the context of the other mandatory training that is now required of nurses, in particular, and also doctors. One is almost driven to ask ‘will there be time to do the actual work?’

Structured professional judgement

In this issue of APT, Bouch & Marshall (2005) describe a structured professional judgement approach to clinical risk assessment and management, focusing on suicide. They argue, on the basis of epidemiological data, that the assessment and management of suicide risk is of public health importance. This line of reasoning is in keeping with UK policy as laid out in Our Healthier Nation (Department of Health, 1998). Furthermore, they argue that informal, i.e. unstructured, clinical and actuarial approaches respectively are inadequate for the task. The informal clinical approach is described as ‘subjective, intuitive and based on experience’. The implication is that it is inaccurate and prone to error, for the subjective and intuitive are almost by definition whimsical and idiosyncratic. This is, of course, a straw man against which to contrast a structured professional judgement approach.

What passes for an informal clinical approach is a highly systematised but automatic algorithm based on the formalised principles that Bouch & Marshall explicate but which, like all over-learnt procedures, are hidden from view. Walking is an example of an over-learnt procedure. Thankfully, one no longer needs to perform a ‘structured professional judgement’ before embarking on a walk. But if one did, it might be based on consideration of the static risk factors of the road and terrain, the dynamic risk factors determined by the speed and number of vehicles, including the dynamic road conditions, and future risk factors depending on the strength and agility of one’s lower limbs. In general, when setting out on a walk, subjective and intuitive judgements seem to be good enough. So it is too with the informal clinical assessment of risk.

There is no doubt that the structured professional judgement allows the clinician to make explicit the factors that determine the nature of risk that a patient poses to self or others. The structure that Bouch & Marshall propose is a good one. Their clinical case illustration of Peter demonstrates how complex the task is. A chronically elevated risk is distinct from an imminent and foreseeable risk. This structured approach is admittedly a systematic and explicit method. However, once the approach is accepted and widely taught it is liable to become implicit and automatic, and it will be the better for that reason.

The danger of all approaches, including structured professional judgement, is that clinicians can be lulled into thinking that risk, once assessed, is static. The reality is that risk needs continuous reassessment. This means that measures determined by questionnaires or assessment tools are out of date as soon as completed. A rapid, human approach, relying on conscientious and competent clinicians, built on a structured but implicit system is the only approach fit for purpose. It ought to provide a system that allows for repeated assessments of risk, unobtrusively carried out, efficient and as reliable as possible in the circumstances. The false security engendered by ticked boxes and completed forms is like fool’s gold.

Conclusions

The Confidential Inquiry into Suicide and Homicide by People with Mental Illness, like the government’s suicide prevention policy, has focused our attention on suicide. In general this is a good thing. Mental health services at last have an index that the government scrutinises, and mental illness is alongside heart disease, accidents and cancer as national priority areas. However, the subtle difference between suicide as a mortality measure in mental illness and death from coronary artery disease should not be ignored. The patient has a direct relationship with his own death in suicide, whereas the patient’s own action is only distally related to his death from coronary artery disease. Thus, the degree to which clinicians or services are responsible for a suicide death is far less than for coronary artery disease death. Yet, in the current environment and sustained by documents such as the Confidential Inquiry reports and An Organisation with a Memory (Department of Health, 2000), the clinician’s role and room to influence the outcome is quite often exaggerated. Bouch & Marshall’s article is a welcome addition to the clinician’s toolbox in this highly complex and important area of practice.