By Dr Jeremy Dixon and Megan Robb, University of Bath, UK
A few years ago one of us worked as a senior practitioner in a medium secure unit for offenders with mental disorders. A key social work task was agreeing discharge arrangements.
The manager was keen to ensure referrals were made to Multi Agency Public Protection Arrangement (MAPPA) meetings – at the time these were relatively new.
As most readers will know, these meetings are chaired by the police or probation with the aim of managing an offender safely in the community. Different professionals are invited and are encouraged to share information. The result is a risk management plan.
Although workers at the secure unit were used to assessing and managing risk themselves, many felt distinctly uneasy about working with the police.
Why? Some social workers worried police concerns about public protection would limit the ability to re-integrate offenders into the community. Psychiatrists, by contrast, felt sharing information with the police challenged duties of confidentiality.
Nowadays multi-agency risk meetings are far more common. MAPPAs have continued and Multi-Agency Risk Assessment Conferences (MARACs) encourage workers to share information to reduce harm to people experiencing domestic abuse.
Tensions with social work values
Yet the unease some social workers feel about working with the police remains. This seems to stem from a belief social work values are different to those held by the police and a worry that joint working may lead to these values being watered-down.
In a recent paper, we argued that this view is too pessimistic. Changes in police practice have actually led to some positive developments in the way that victims are identified and treated.
Social workers can learn from some of these practices but need to approach them with caution. To do so, it’s important to understand how policies aimed at social workers and the police shape the way that both services think about risk.
How we understand risk
Risk is often presented as a matter of common sense. Yet our understandings of risks are actually shaped by a number of things. These include our upbringing, our professional training and our work culture.
Research shows different professions have preferences for assessing risk in particular ways. Even within professions, team culture may alter the way people see risk.
Our paper looked at the way adult social workers and the police have come to assess risk in one area – cases involving women with learning disabilities who have experienced domestic abuse.
We were aware of the social work literature on risk. This often argued that risk assessments have come to shape social work practice. However, when we examined policies around risk for people with a learning disability in the UK we were surprised at how little risk guidance existed.
Lack of resources for social workers
We found plenty of risk assessment tools for offenders with a learning disability but very few guidelines focusing on the risks to people with a learning disability. There were even fewer tools for social workers that focused specifically on domestic violence among learning disabled women.
In practice, this means social workers usually draw on their own knowledge of the service user to predict risk. The police, on the other hand, now routinely draw on a series of risk assessment tools.
These tools are aimed at predicting future harm. They include the Domestic Abuse, Stalking and Honour Based Violence (DASH ) risk assessment.
The DASH was developed to identify factors that would indicate a high risk of injury or death in the future. These include the women’s own view of the situation, a history of violence between the perpetrator and victim in the past, and previous child contact difficulties.
The DASH assessments rely on statistical information to predict how likely it is that harm will occur in the future. This then informs the advice that the police may give.
The benefits
So should social workers be worried about being involved in these types of assessment?
We think the tools used by the police offer some benefits to workers. While some authors have worried that using these types of assessments discourage workers from drawing on their values, we are mindful that research indicates workers underestimate domestic abuse across all areas of practice.
We believe it’s useful to be informed by data from risk assessments. However, workers need to be aware that risk assessments cannot predict the future but only provide probabilities about the likelihood of future harm.
There are also some down-sides to risk assessments. The use of such assessments tends to place an emphasis on making victims responsible for planning their own safety. In the case of women with a learning disability, such an approach may not take into account the needs that these women may have.
Social workers working with the police should therefore aim to help women with learning disabilities to understand what coercive behaviour may look like, but also enable them to make choices about how they want to manage their lives in the future where they have capacity to do so.
In doing so, social workers can demonstrate the ability to be informed by risk assessments whilst also promoting the autonomy of women with learning disabilities.
Dr Jeremy Dixon is a social work lecturer at the University of Bath. Megan Robb is a teaching fellow in social work at the university.
very interesting and timely article; the research is also fascinating and well worth the read. I have an interest in this as I am a social work lecturer and have had a long term and continuing involvement in domestic abuse work. I would quibble with the depiction of DASH as an actuarial tool
“The DASH assessments rely on statistical information to predict how likely it is that harm will occur in the future. This then informs the advice that the police may give”
I am a accredited trainer for DASH and have been training people in its use for some time. When we train people in the use of the DASH we are very clear that although it is based on the best available national and international evidence of risk factors it is not an actuarial tool. Actuarial evidence does inform DASH but you cannot add up the number of ticks and come to a summary risk “number”. Having said that it s clear to me that a lot of MARACs and other multi-agency groups do actually do that, against the advice of the designers of the tool.
The Police do use versions of DASH and this is at least partly dictated by the nature of their roles in domestic abuse cases. Subjective judgements in these cases often lead to poor outcomes; those very factor of personal history, understanding of the dynamics in abusive relationships, team and professional culture and so on, often produce bias in decision making. DASH helps to counter some of that. It is also the case that aide memoires help us to account for factors that we might otherwise ignore or simply forget.
The real experts in the use of DASH though are likely to be IDVAs; independent domestic abuse advocates or their probation counterparts, Women’s Safety Workers. I have also read a great many DASH assessments as part of S.20 reports. The tool is only ever as good as the person using it. The items on the list are only as good as the individuals understanding of how they might operate in the real world, and critically, what they mean in in an individual woman or child’s life. DASH enables structured professional judgement. Used well by a competent professional it can take the needs of women with learning disabilities into account; IDVAs will know that in this context learning disability can be a risk and vulnerability factor.
It is true that tools of this nature are limited in their predictive value for individual people. DASH can tell us something about what cases and what risk factors we might need to prioritise. CAADA have been successful in securing safety for the victims of domestic abuse because they use DASH as part of a multi-agency effort to do that. By itself they DASH is of limited value, but as a part of inter-professional practice it has value. When the inter-professional network does not have a common understanding of what it is and what it can be used for, its efficacy may be limited.