Catalyst for Harm: Risk Aversion in Mental Health Settings

Sarah Markham discusses harm caused by disproportionate risk aversion in mental health settings, and how this might be remedied

In secure and forensic psychiatric settings, a tension is perceived to exist between the promotion of responsibility and autonomy on one hand, and the need to protect patients from posing a risk to themselves and others, by curtailing their human rights through restriction and confinement. The Independent Review of the Mental Health Act (MHA) 2018 expressed clear concerns about the disproportionately risk averse nature of psychiatric practice, citing clinicians’ subjective, self-protective anxieties as a significant cause. In this article I seek to briefly highlight the nature of the harm caused by disproportionate risk aversion in mental health settings, and how this might be remedied.

Perception of risk and current management strategies

The nature and extent of risk aversion varies between inpatient mental health settings. In secure settings, where risk aversion often dominates clinical thinking and treatment provision, blanket rules, policies and procedures act to stifle proportionate personalised care and opportunities for meaningful recovery. It may also contribute to unnecessarily long lengths of stay.

Psychiatric practitioners can be prone to conceptualising risk in very narrow terms; as risk to others, risk to self and risk arising from vulnerability. Patients are perceived and treated primarily as risk entities, rather than as human beings who are in need of compassionate care and treatment and professionals place far too little gravity on removing an individual’s liberty [3]. The Review also acknowledged it ‘can be very difficult for a patient to demonstrate they are no longer a risk’ [3]. Risks to patients arising from the stigma and social exclusion from being detained under the MHA, as well as the iatrogenic effects of psychiatric treatment, tend to be disregarded.

Traditionally risk assessment and management in secure and forensic units is the purlieu of the multidisciplinary team with the patient’s or carers’ views being either unsolicited or discounted. Recent prompts for more collaborative and recovery model based approaches have originated from both academia and national and local policy makers. However the organisational norm of credibility within mental healthcare services is biased against patients. This means that regardless of the conscientiousness of individual practitioners in giving due regard to their patients’ views and requests, the wider services may function to suppress patients’ input in decision-making regarding their care and increase the epistemic marginalization of the patient body. Epistemic disregard and contempt can be creeping contagions; if one person in authority is seen to silence or disregard a patient’s testimony or self-report, this may encourage others to follow suit. Yet patients are often the best predictors of their own risk [4].

Harm to patients

I argue that risk averse management strategies may in themselves paradoxically act to increase the likelihood of violence. Patients may be reluctant to seek support or disclose violent ideation for fear of being detained under the MHA or prolonging current lengths of stay.

Restrictive cultures can place abnormal expectations on patients in terms of emotional self-regulation. They may not feel allowed to have and express emotions and variations in mood as other people normally do. Patients subjected to disproportionate levels of restriction experience difficulties in (re)developing and expressing a sense of self, leading to institutionalisation boredom, hopelessness, loneliness, or lethargy.

In secure settings the imperative to minimise, if not eradicate the potential for patients to cause harm, conflicts with recovery needs such as self-responsibility, meaningful activity and the opportunity for individuals to learn naturally from their own mistakes [2]. A risk averse culture which emphasises and priorities risk avoidance above all clinical and therapeutic goals may invariably lead to excessive restriction and the compromise of individual patient’s human rights.

Secure and forensic psychiatric settings are ostensibly therapeutic driven, yet designed in accordance with both protective and preventative models of care, are overtly risk-averse and dominated by security measures. They have been framed as unethical manifestations of state control in which behavioural change programmes deny patients any meaningful autonomy [5]. According to the MHA (1983) an individual can be detained where it is ‘necessary for’ or ‘justified in the interests of’ the patient’s health or safety or for the protection of others’. This is vague and potentially sets the threshold for detention at too low a level to be aligned with human rights legislation. To quote the final report this ‘may have allowed professionals to become increasingly risk averse; to become too quick to use ‘risk’ as a catchall justification when they are afraid of consequences that may never happen, indeed probably won’t happen’ [3]. However the European Court of Human Rights (ECHR) has made it that disproportionate risk aversion is potentially at odds with human rights law [1].

Looking ahead

Perceptions of restriction appear to depend on a number of factors, many of which are associated with the quality of the relationship between the patient and the service provider. This strengthens the argument for secure and forensic services to invest in relational as opposed to physical and procedural security. However, I argue that increased opportunities for autonomy to risk averse convictions is that increased opportunities for autonomy, involvement, and developing therapeutic alliances have the potential to reduce risk by improving self-esteem and encouraging patients to become more responsible for their behaviour.

For the practice of proportionate relational security to become embedded in secure services would require a consistent and pervasive commitment made by all staff to disregarding their subjective biases regarding patients and the risk they may or may not pose. Such improvement of the therapeutic nature of secure and forensic services requires a systematic shift away from an anxious and rigid culture of risk aversion to a more fluid and responsive culture with increased emphasis upon relational safety and epistemic regard for patient self-insight and testimony. This would not be merely good practice; it is essential if clinical practice is to be aligned to national and international human rights standards.

A careful and considered balance needs to be struck in the context of collaborative risk assessment and management in order to support patients to recover in addition to preventing harm. If mental health services and clinicians stop prioritising supposed risk above clinical need, and associated disproportionately restrictive practices, they will be able to redirect resources to improving the mental health of the many and away from supposed preventative low volume, high cost secure care. More humane and person-centred care approaches to mental health care are in alignment with the NHS Long Term Plan. Increased autonomy need not necessarily mean increased risk; the positive emotion people may experience on regaining autonomy and greater social status can mitigate risk, whereas losing autonomy and status is associated with heightened risk. It is important to remember that patients, even when experiencing the disabling effects of a mental health condition, have knowledge and agency essential to their recovery.


[1] ECHR. Case of Hiller v. Austria. (Application no. 1967/14). Judgment Strasbourg, 22 November 2016.

[2] Mann, Bradley, Matias, Elizabeth, and Allen, Jo. “Recovery in forensic services: Facing the challenge.” Advances in Psychiatric Treatment, (2014):20(2), 125–131.

[3] Department of Health and Social Care (2018). Modernising the Mental Health Act – final report from the independent review.  Department of Health and Social Care Publications: London, UK, 2018.

[4] Lockertsen, Øyvind. “Screening for risk of violence using service users’ self-perceptions: A prospective study from an acute mental health unit”. International Journal of Mental Health Nursing, (2017).

[5] Holmes, David, and Murray, Stuart. (2011). “Civilizing the ‘Barbarian’: A critical analysis of behaviour modification programmes in forensic psychiatry settings.” Journal of Nursing Management, (2011):19(3), 293–301.


Sarah Markham (@DrSMarkham) is a Visiting Researcher at the Institute of Psychiatry, Psychology and Neuroscience, King’s College. She is a staunch advocate for mental health with especial interest in risk assessment and management.


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.