Psychologist logo
Clinical, Legal, criminological and forensic

Prisoner suicide

Graham Towl and Tammi Walker consider public management, punitiveness and professionalism (an 'Online first' publication).

06 July 2015

Prisoner suicide is of particular concern for three reasons. First, there is the loss of life, by way of a potentially avoidable death. Second, prisoners tend to come from socially and economically disadvantaged groups with high levels of need and comparatively little access to support services. Third, they are under the care of the state. Psychologists have key roles to play in contributing to reducing the rates and risk of suicide amongst prisoners.

In this article we argue that if psychologists are to play a full role in addressing the unacceptably high levels of suicide in prisons, then there is a need to recognise that there may well, at times, be tensions between such clinical roles and forensic functions or work roles. Indeed, the introduction in 2003 of the first strategic framework for psychological services in prisons and probation brought these issues into sharp focus (HM Prison Service and National Probation Service, 2003). The strategy was pivotal in the development of psychological services and in particular in warmly welcoming a range of applied psychology specialisms to work in prisons and probation services. Of course, there is much overlap across the practitioner psychologist specialisms, and there can be much variation in job role within the discipline of, for example, forensic psychology. From a statutory regulator perspective the concern is chiefly that all those on the register work within the bounds of their competence.

On the broader canvas of public service developments there is benefit in considering the advent of New Public Management in the late 1980s and the neoliberal agenda of a new punitiveness. If we are to better understand and respond to rates of suicide in prisons there is a need to revisit the psychological maxim that all behaviour takes place in a context. With suicide the evidence seems to indicate that the context is critical. Just by way of broad illustration, the rates of suicide in prisons have been internationally and historically reported as being significantly higher than those outside prison. And in prisoner suicide assessment we know that the earlier days of imprisonment are those with the highest degree of risk, by a significant margin, for most prisoners.

New Public Management and ‘new punitiveness’

The term New Public Management (NPM) broadly refers to an ideology that the business disciplines of the private sector may be routinely transferred to deliver more effective public services. This has contributed to a marketisation of public services including prisons (and probation services).

Since around the 1990s there have been impacts of what is sometimes referred to as the ‘new punitiveness’ in criminal justice (Garland, 2001; Pratt et al., 2005). The approach is perhaps most starkly reflected with the doubling of the prison population in England and Wales over the past 20 years or so (Berman & Dar, 2013; Ministry of Justice, 2013). This new punitiveness has also impacted directly upon the work of psychologists in prisons, where the prisoner may be seen as having been ‘dealt with’ rather than ‘worked with’. The shift in the work of many psychologists in prisons to focus upon reducing the risk of reoffending, rather than on areas such as seeking improvements in everyday functioning, may well have detrimentally impacted upon the relationships between psychologists and prisoners. The net result is that instead of being seen fundamentally as health and care professionals, such chiefly forensic psychologists may simply be seen as agents of the state. An example of this may be illustrated by the accredited ‘programmes’ that have been rolled out in the UK since 1996 within prisons and probation with their emphasis on procedure rather than broad psychological knowledge or process (see McGuire, 2001). This has led to psychological therapy with prisoners being deconstructed into a manual with set procedures. It has been argued that as a result of this there is no need to understand the individual to whom these procedures are applied; they just need to understand their part in the process (Thomas-Peter, 2006). As a consequence of this situation, psychologists have been known to use psychometric tests without the intent of changing their intervention according to the results. They simply forward the results of the psychometric instruments to programme managers as if the data is irrelevant to their own clinical work. In this capacity, some believe, psychologists have become ‘intellectually dependent programme drones’ that run the ‘risk of never developing the competence… to derive unique psychological solutions for unique psychological problems’ (Thomas-Peter, 2006, p.32).

The development of such structured groupwork interventions then had three key antecedents; NPM, new punitiveness and some initially promising empirical findings. But one unintended consequence of such an approach may well be an increased challenge in being able to engage most effectively with prisoners, for our purposes here in working with the suicidal. Thus in view of such differences in the purposes of such interventions it presents professional challenges – for the individual and also the recipient of the service – in switching such roles.

New professionalism

If as psychologists we are to have a full role in saving lives, then there is a need for change. There are some signs of such changes. As mentioned earlier, the 2003 strategic framework for psychological services included recognition of the need for a range of applied psychology specialisms working in prisons and probation instead of the traditional more or less closed shop of forensic psychological practitioners (HM Prison Service and National Probation Service, 2003). The framework provided a strategic recognition of the need to view prisoners as citizens with a range of needs and wants, and not just to be seen in terms of what had become somewhat crassly known as ‘criminogenic factors’. There were good ethical grounds to broaden the range of practitioner psychologists with the richer range of approaches whereby all could learn from each other.

In 2009, practitioner psychologists achieved statutory regulation with the then Health Professions Council (HPC). This afforded the opportunity to not only learn from the range of applied psychology specialisms but also a range of health professionals, in view of the HPC being a multi healthcare profession regulator. Thus the combination of the strategic framework and the advent of statutory regulation provided a professional context whereby there was scope for a more rounded understanding of prisoners and their needs as members of the public. It also provided a firmer foundation from which to further develop ethical policies and practices, with the potential for a greater diversity of contributions.

Since the report of the public inquiry into the Mid Staffordshire NHS Foundation Trust (Francis, 2013) there have been further opportunities for us as psychologists to reflect more upon our policies and practices. For example, one impact that we may all take away is a greater openness to being challenged or questioned about our practices from others. Such shared learning is important if we are to do our best for service users. Here we use the terms service user very broadly to include both the prisoner and wider public.

Helping suicidal prisoners

We have outlined above the central importance of having an appropriate professional relationship and standing with the prisoner. Crucially, this is by no means solely a product of the skills base of the individual practitioner. As we have seen, there are wider forces at work in terms of the politics of the day. Without a basic trust it is unlikely that the prisoner will come forward and disclose their feelings. But even if they have, sometimes to another member of staff or prisoner, in working with prisoners for therapeutic goals there is a need to have a suitable rapport, which includes the development and maintenance of a relationship of trust. When work is focused chiefly on protecting those members of the public who are not currently prisoners this makes work with prisoners to reduce the risk of suicide even more difficult than it intrinsically is. So, there is perhaps a discussion and debate to be helpfully had around which psychologists are best placed to provide such services.

Additionally it may be helpful to develop policy and practice further amongst staff chiefly undertaking assessments of risk of harm to the public. One potential benefit is that such psychologists tend to have a good understanding of the concept of risk assessment and management – the logic is the same in terms of risk assessments whether the ‘risk’ being assessed is harm to self or others. The process involves the identification of those factors likely to be associated with an increased risk of suicide and those factors likely to be associated with a decreased level of risk (Towl & Forbes, 2002; Towl & Hudson, 1997; Walker, 2015).

One problem that has bedevilled those seeking to produce ‘suicide risk screening tools’ in prisons is that, against a backdrop of most prisoners having factors associated with a higher lifetime risk of suicide (Hawton et al., 2002; Shaw et al., 2004; Towl & Forbes, 2002; Towl & Hudson, 1997; Walker, 2015), these are of little utility and sometimes have significant expense. A problem with the snapshot design associated with such screening is that suicidal ideation may significantly fluctuate over time, so the chances of missing many through false negatives may be relatively high. Screening for suicide in prisoners is further compounded as the classification of individuals also uses a two-by-two table identifying those who truly are at risk (i.e. the sensitivity of the instrument), and those who truly are not at risk (i.e. the specificity of the instrument). Therefore the evaluation of any screening tool for suicide will include a trade-off between the sensitivity and specificity of the tool by manipulating the cut-off scores used to identify a case. Perry et al. (2010) highlighted that adverse effects from screening can occur from the misclassification of individuals in this way. In addition, to these considerations, Perry et al. (2010) also highlighted that there are issues with the transferability of existing scales as many are originally based on psychiatric populations and there is a lack of an apparent gold standard test in this area (Correia, 2000; Mills & Kroner, 2000). Overall, they concluded that there are concerns about the sensitivity and specificity of such instruments and that more research is needed to assess the predictive validity of such tools for offender populations in the identification of those at risk.

So where does this leave us? There are a number of levels at which psychologists may be usefully able to contribute.

First, at the organisational level by influencing policies and practices. Ensuring that policies are enacted and challenging colleagues of whatever discipline if this is needed. Contributing to structuring the environment whereby there is more likely to be trust than distrust, where prisoners feel safe to disclose is in and of itself a crucial contribution. Professionally it can be difficult to challenge those in authority, but sometimes speaking truth to power is a key role of the practitioner psychologist. This is consistent with what would be entirely expected post-Francis, and indeed from a health regulatory perspective too.

Second, there needs to be a leadership contribution from Heads of Psychology Units. In particular this involves role-modelling and ensuring that psychological staff recognise the importance of being therapeutic in all direct work with prisoners. This necessarily involves having a more rounded view of the prisoner, not just seeing the prisoner as criminal, whatever the balance of work of the individual psychologist.

Third, and this is, in large part predicated upon the two above conditions being met, there is a need to make more widely available psychological work with prisoners to reduce the risk of suicide. In other words what we are suggesting is a public health–based model whereby all prisoners experience a positive environment and those groups identified at a higher risk (e.g. all prisoners in their first week of imprisonment) have targeted interventions, as do those who have been identified as being at a specific inflated risk. The field is replete with resources to inform any such assessments or interventions. The problem is not one of a lack of knowledge; we do not need any more research to make very significant progress. What is needed perhaps most of all is for it to be recognised that the life of a prisoner is worth as much as the life of any other citizen. If we accept that then there is real potential to provide the right services, to the right prisoners, in the right place at the right time.

- Graham Towl is Professor of Forensic Psychology and Pro Vice Chancellor, Durham University [email protected]

- Tammi Walker is a Chartered Psychologist at the University of Manchester tammi.walker@

BOX: Learning lessons about suicides in prison
Nigel Newcomen, Prisons and Probation Ombudsman

Tragically, the number of suicides in prison increased sharply in 2013/14, reversing a welcome decline in recent years. My office independently investigates all deaths in custody and, in 2013/14, we investigated 64 per cent more suicides than the year before, as well as 7 per cent more deaths from natural causes (In 2013/14, we investigated 130 deaths from natural causes, 90 suicides, 4 homicides and 16 ‘other deaths’). While the increase in deaths from natural causes was largely due to an ageing prison population, explaining the increase in suicides is more difficult.

Some have argued that the rise is related to austerity, which has reduced staffing and protective factors, such as time out of cell and interaction with others. While plausible, the evidence for this is limited, and suicides also increased in prisons that had not suffered cutbacks.

While the increase may be hard to explain, a higher rate of suicide in custody than in the community is common to most jurisdictions, not least because of scale of mental ill-health among prisoners (Preventing Suicide in Prisons, Department of Mental Health and Substance Abuse, WHO). Sometimes the cases I investigate illustrate just how difficult it is for prisons to deliver their duty of care to the most vulnerable. For example, last year a prisoner on constant watch in a healthcare unit killed himself by jumping from a height before staff could stop him. The mental ill-heath and abject despair evidenced by such cases is shocking.

Despite this depressing picture, we must recognise the positive efforts of those working and living in prison. Every day prison staff and prisoner peer supporters save prisoners from harming themselves, an achievement that goes largely unreported and without which the statistics would be much worse.

But more must be done. So to signpost the way forward, I recently published two thematic reviews of the lessons to be learned from investigations into suicides since 2007. The first examined how well prisons identify and assess risk of self-harm or suicide. The second examined the quality of support to prisoners identified as at risk. Both reviews provide case studies to illustrate their findings.

Worryingly, I found recurring weaknesses in suicide and self-harm prevention procedures in prison (known as ACCT – Assessment Care in Custody and Teamwork). When assessing risk, prison staff often placed too much weight on how the prisoner presented, rather than known risks, such as previous self-harm. The professional judgement of staff is crucial, but known risks should not be ignored as they can be predictive of future action. Similarly, for those identified as at risk, too often the support was not good enough.

In many ways, ACCT procedures are impressive, and I know of few better approaches in other jurisdictions. However, the real test is implementation, and this can be poor. The reviews highlight some lessons that prisons need to learn.

First, there is a need for better risk-assessment training for prison staff, which ensures that all known risk factors are considered.

Second, once ACCT documents are opened, they need to be adjusted to new events affecting the prisoner, with regular multidisciplinary reviews of progress. Monitoring needs to be supportive, engage those at risk and, where possible, involve their families. Records need to be comprehensive and regularly tested for quality assurance.

These are important lessons that clearly still need to be learned. Given the unacceptable rise in suicide in prison, the urgency of the situation is obvious. That is also why, in my annual report for 2013/14, I called for a review of the implementation of ACCT, which was designed for a prison service with fewer prisoners and more staff. The review is under way – we must hope it succeeds.

BOX: John Podmore, Prison Governor

The first dead body I ever saw was as a trainee Assistant Governor in Kent wing of Maidstone Prison in 1985. He was a young man who had died by hanging. I was summoned to the scene as the duty governor of the day. The staff that found him were too shocked to cut him down and the task was left to me. In the next 25 years I saw too many others. In a joint question and answer session with another governor some years later we were asked what our main priority was as a Governor. My colleague replied ‘to prevent escapes’ I replied ‘to keep people alive’. I failed more often than I was comfortable with. Such deaths always had a significant impact on staff and prisoners alike. I saw sadness, regret, guilt and remorse. There was also disdain, disrespect and indifference, such is the nature of the difficult, damaged and complex environment that is a prison but it rarely came to the fore.

It is prison officers who shoulder the greatest burden in such circumstances. I saw as many officers in tears as those attempting unprotected resuscitation. People cared. There was compassion and a desire to learn from mistakes. To help those delivering care we developed processes to measure its effectiveness or defend its failure. There were assessments aimed at the recognition of the vulnerable. There was suicide awareness training but no mental health awareness training. The Prison and Probation Ombudsman and Inquests sought answers except that they came years later peppered with ‘urgent recommendations’ which may or may not have had implications for the subsequent suicide still to be reported on. Reports often contained, and rightly so, findings that related to a lack of care, of people not doing their job properly, or at all. The culture of a jail and the nature of staff-prisoner relationships however, were concepts that too often escaped the coroner and the Ombudsman.

It is the prison officer who is at the heart of a prison and on whom the safety, security and decency of a jail depends. But they do not operate in isolation. They need to be led and motivated by someone who has a vision for the establishment. As individuals with limited skills and training they need support from specialists: psychologists, probation officers, medical practitioners. The fact that so many staff across the specialisms are now working under separate contracts has had the effect of isolating prison staff rather than bringing everyone together as a team working for the overall benefit of prisoners. Psychologists in particular are perceived by many, not least prisoners themselves (see any edition of Inside Time, the national newspaper for prisoners) to have become pre-occupied with screening tools and programmes, untroubled by evidence . Probation has all but been abandoned to the ravages of the market. How health and social care will emerge from The Lansley reforms remains to be seen. All the while the Governor tries to act as ringmaster with the danger that he or she may simply withdraw to the role of contract manager. Prisons are communities, albeit highly complex ones and need to be treated accordingly. For them to be reduced to a series of complex commercial contracts will do little to protect the public. 

Further reading

See also the news item on the Harris Review into self-inflicted deaths in custody.


Berman, G. & Dar, A. (2013). Prison population statistics. House of Commons Briefing Paper SN04334.

Correia, K.M. (2000). Suicide assessment in a prison environment: A proposed protocol. Criminal Justice and Behaviour, 27, 581–599.

Francis, R. (2013). Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry. London: The Stationery Office. Garland, D. (2001) The culture of control: Crime and social order in contemporary society. Oxford: Oxford University Press.

Hawton, K., Rodham, K. & Evans, E. (2002). Deliberate self-harm in adolescents: Self report survey in schools in England. British Medical Journal, 325, 1207–1211. HM Prison Service and National Probation Service (2003). Driving delivery: A strategic framework for psychological services in prisons and probation. London: Author.

McGuire, J. (2001). What works in correctional intervention? Evidence and practical implications. In Bernfeld, G., Farrington, D. & Leschelid, A. (Eds.) Offender rehabilitation in practice: Implementing and evaluating effective programs. Chichester: WiIey & Sons.

Mills, J.F. & Kroner, D.G. (2000). Depression, Hopelessness and Suicide Screening Form (DHS): User guide. Unpublished manuscript. Ministry of Justice (2013). Story of the prison population: 1993–2012 England and Wales. London: Author.

Perry, A., Marandos, R., Coulton S. & Johnson, M. (2010). Screening tools assessing risk of suicide and self-harm in adult offenders: A systematic review. International Journal of Offender Therapy Comparative Criminology, 54(5), 803–828

Pratt, J., Brown, D., Brown, M. et al. (2005). The new punitiveness: Trends, theories, perspectives. London: Routledge.

Shaw, J., Baker, D., Hunt, I.M. et al. (2004) Suicide by prisoners: National clinical survey. British Journal of Psychiatry, 184, 263–267.

Thomas-Peter, B. (2006). The modern context of psychology in corrections: Influences, limitations and values of ‘what works’. In G. Towl (Ed.) Psychological research in prisons. London: Wiley.

Towl, G. & Forbes, D. (2002). Working with suicidal prisoners. In G. Towl, L. Snow & M. McHugh (Eds.) Suicide in prison (pp.93–101). Oxford: BPS Blackwell.

Towl, G. & Hudson, D. (1997). Risk assessment and management. In G. Towl (Ed.) Suicide and self-injury in prisons: Issues in Criminological and Legal Psychology, 28. Leicester: British Psychological Society.

Walker, T. (2015). Self-injury and suicide in prisoners. In G. Towl and D. Crighton (Eds.) Forensic psychology. London: Wiley-Blackwell.