Despite progress in the care and treatment of mental health problems, violence directed at self or others remains high in many parts of the world. Subsequently, there is increasing attention to risk assessment in mental health. But it this doing more harm than good?
The continuing focus on risk, well-intentioned as it is in reducing harm and increasing people’s safety, has a stigmatising, and, in some cases, traumatic effect on people using mental health services. It reinforces the myth that people who are mentally unwell are an inevitable risk to society, and that through risk assessment we can minimise or even eliminate this threat. It is the often unquestioned acceptance of the effectiveness of risk assessment, and the unconscious bias that emerges from this narrative that poses the biggest risk.
Why do we need risk assessment?
Risk assessment seeks to identify the likelihood of harm to self or others with a view to preventing or minimising such harm. National crime statistics and suicide data show that three times as many people with mental health problems will take their own life, than will take the life of others.
Risk algorithms (based on asking specific questions, and generating a subsequent ‘score’), allow for a calculation of clinical utility similar to ‘numbers needed to treat’ (NNT). For example, before using a particular intervention to reduce harm to others – it would be useful to know how many people would need to be treated using the intervention in order to reduce harm by at least one incident. An NNT of ten would mean that ten people would need to be treated to reduce harm by at least one incident. Thus, the intervention might be seen to have value. An NNT of 100 would mean that 100 people would need to be treated to reduce harm by at least one incident. In this case, the intervention might be seen to have less value.
In a similar vein to such discussions, is the calculation of ‘numbers needed to detain’ (NND). Requirements for legal detention in most jurisdictions are that a person, who is judged to have a mental disorder of a nature or severity that threatens the safety of the person or others, may be detained involuntarily. The issue of risk is incredibly pertinent here. As long as risk is linked to decisions around legal detention, mental health practitioners with the powers to detain people involuntarily would benefit from access to the best available risk assessment methods.
The unseen risks in ‘risk’
In his 2010 Gresham College lecture, George Szmukler, Professor of Psychiatry at the Institute of Psychiatry, Psychology and Neuroscience, illustrated how the issue of risk discriminates against people thought to be mentally ill. A person deemed to be mentally ill who is thought to be at risk of harm to self or others can be detained even if they have never committed a violent act, have insight into their risk, and have plans to manage it, i.e. they have mental capacity. Conversely, a person with a life-threatening physical illness who refuses treatment that can save their life and who is deemed to have mental capacity can refuse treatment without fear of involuntary detention.
The number of people in the general population who may be deemed risky to others, and who are unlikely to have a mental disorder, greatly exceeds those with a mental disorder deemed risky. Yet these people are not eligible for involuntary detention under mental health law. Therefore, detention on the grounds of risk discriminates against people who have a mental disorder, and this situation should sit uneasily in democratic societies with strong civil libertarian principles enshrined in law.
If researchers, clinicians, managers, and policymakers are serious about reducing or preventing harm, they will need to identify the likely causes, develop interventions to address these causes, and evaluate these interventions in well-designed studies. We could take steps to reduce harm with more confidence if we knew that A (e.g. ‘unsafe’ staffing levels) causes B (harm) and taking steps to reduce A leads to a reduction in B. Therefore, only a thorough understanding of what causes harm will allow us to discover causal risk predictors that we can then address to increase safety.
Harm in the mental health context is linked to environmental factors such as the use of physical restraint and seclusion, over occupancy, staff-patient ratios, ward rules, and staff characteristics. Yet, risk measures seldom include such factors; they are almost without exception, individual centric. This person-centric approach helps explain why many service users do not engage with risk assessment or consider the process stigmatising, disempowering, and even traumatic. A consequence of this approach is that risk assessment methods lack credibility in the eyes of many service users for whom risk is something that is done to them. It also violates the principle of ‘no decision about me without me’.
From risk to safety: an idea whose time has come
Risk assessments are seldom linked to improved therapeutic outcomes, and often further marginalise disenfranchised groups by labelling people rather than understanding and helping to resolve their difficulties. Moving forwards, risk assessment needs to be focussed on safety issues – secured by a desire to improve, reintegrate, retrain, and foster recovery. Marginalised groups such as those living with mental health issues are further burdened as they often live in communities associated with recurrent harm and crime. The label of ‘risk’ promotes stigma by classifying individuals as unsafe, thereby giving society’s prejudices and fear the stamp of scientific approval. Clinicians using risk assessment measures usually have benevolent intentions, but history has shown that such intentions do not always lead to benevolent outcomes.
A simple way to address the pitfalls in risk assessment is to focus on the issue of safety. The language of risk punishes and stigmatizes, in some cases it may traumatize. The language of safety nourishes and protects. A collaborative approach to safety assessment – in partnership with service users, is recommended. Given that discussing safety issues is a sensitive topic, this discussion should be fully integrated throughout an assessment interview – not added on as a parting shot ending a clinical encounter. By placing ‘safety’ at the heart of our work around risk; acting with both compassion and clinical-knowledge, we can ensure better outcomes for all involved.
Featured Image Credit: ‘Rime Risk Ryze Pysa Saber Askew Augor Zes MSK LTS WCA TMD AWR SeventhLetter LosAngeles Graffiti Art’ by A Syn, CC BY-SA 2.0 via Flickr.