This article is written by Jonathan Shepherd, surgeon, based at the Cardiff University Crime and Security Research Institute, and founder and former chair of Cardiff’s violence reduction unit, where he developed the Cardiff Model.

In the UK half of all violent incidents that lead to emergency hospital treatment are not reported to the police.

This is a serious problem because effective prevention depends on knowing where and when violence takes place. Research shows that police knowledge of violence relies on injured people reporting to the police. But many do not report, for fear of reprisals, because they don't want their own behaviour scrutinised and because they don’t know who injured them and therefore don't think the police can help.

People who carry or use weapons, are involved in the drug trade, are regularly intoxicated with alcohol, and who are gang members, aren’t going to report violence to the police; yet such people are those who are most likely to be injured in violence.

The Cardiff Model is a multi-agency approach to violence prevention which in the UK is built on collaboration between police, the NHS and local government. The model combines data collected in emergency departments with information known to the police. This facilitates much better identification and targeting of violence hotspots and weapons at the right times.

Evaluations show that this approach is effective and highly cost beneficial.

The model has four components: first, the collection in emergency departments (EDs) of specific information key to violence prevention: violence locations, times and dates, weapons used and numbers of attackers (more than one signals probable gang violence)

Next, this information is anonymised and shared, after which this intelligence is combined with information available to the police.  Lastly — and crucially — this comprehensive information is translated into practical violence prevention initiatives by a violence reduction unit.

The Cardiff Model has been implemented in many cities in the UK, including in Cardiff and Swansea in Wales, Glasgow and Dundee in Scotland and in London. In a few places, in the Wirral and Cambridge, for example, it has been implemented for more than a decade. Published evaluations demonstrate that implementation has reduced serious violence by over 40% in Cardiff relative to cities where it was not implemented. According to the 2019 government impact assessment annual savings amount to £2.7 million ($3.5 million) for each community safety partnership which implements the Model.

Lessons learned

Implementing this approach requires senior leadership to be on board. Only senior leaders have the authority and are around long enough to build VRUs, establish and maintain the flow of high quality data, hire analysts, and keep the focus of public sector bodies on ensuring the safety of the people they serve.

At the local and city level, this leadership needs to come from NHS Hospital Trust, police and local authority chief executives with the authority to establish and maintain data flow, analysis and the city’s violence reduction unit. At national level, Cabinet Office coordination is needed so that government departments responsible for the NHS, local government and the police work together.

In England, unfortunately, the Cardiff Model was translated as Information Sharing to Tackle Violence (ISTV), as if information sharing on its own prevents violence, which of course it doesn’t

The Prime Minister’s 2019 serious violence summit in April brought together the relevant cabinet ministers for this purpose for the first time — a hugely significant step in the fight against knife violence and homicide.

Realising the violence prevention benefits of the Cardiff Model depends on implementing it faithfully according to its evidence-based recipe. Changing this can erode the effectiveness of the Model to zero – in the same way that tinkering with the recipe for a therapeutic drug can negate its effect.

In England, unfortunately, the Cardiff Model was translated as Information Sharing to Tackle Violence (ISTV), as if information sharing on its own prevents violence, which of course it doesn’t. Highly responsive, multi-agency Violence Reduction Units (VRUs) — a core component of the Model — are also needed to turn the information into practical prevention. Every city and large town needs one.

Spoiling the recipe

Another example of spoiling the recipe is the addition, on a whim, of extra data items. Helpfully though, the Cardiff Model data set has been carefully formulated based on a series of rigorous trials, published by NHS Digital, and is compliant with General Data Protection Regulations and approved by the Information Commissioner.

The prevention of serious violence on a cross-sectoral basis, together with its regulation, needs specific legislative underpinning. This is promised in the 2019 Queen’s Speech and it’s crucial for the safety of UK citizens and their families that this legislation happens.

In the early and mid-2000s there were many early adopters of the Cardiff Model, in Wales, Merseyside, Scotland and the then South East Government Region for instance. But local authority and police cuts, especially following the financial crisis, meant that many Community Safety Partnerships, which were the vital mechanism for multi-agency crime reduction, established in the 1998 Crime and Disorder Act, were effectively disbanded.

Beware of loud advocates for their interventions and approaches; always scrutinise the evidence and ask to see reports of published controlled trials

Local authority and police analysts were laid off. As a surgeon myself, this seems akin to disinvesting in hospital X-ray departments, radiologists and NHS multidisciplinary teams. How effective patient care would suffer!

The new funding for VRUs and police forces is part of the solution, but without the new statutory duty and regulation of this specific violence prevention approach, the problem of patchy, temporary implementation and realisation of the benefits will remain.

Understanding what works and what doesn’t

Strengthening the data analysis workforce and giving these specialists greater prominence is also needed.

Data analysts who present the latest findings on violence locations and weapon use, for example, are crucially important they are front room practitioners in every Violence Reduction Unit and Safety Partnership.

Without their expertise to inform collaborative decision making, the benefits of substantial increases in police numbers and investment in multi-agency prevention and VRUs will not be realised. Staying with the healthcare analogy, without hospital radiologists who produce and interpret high quality imaging of patient’s bodies, appointing thousands more GPs wouldn’t achieve very much.

Violence Reduction Units and Community Safety Partnerships need to be highly selective in choosing the agencies they recruit and the interventions they implement. They need to scour the websites of authoritative bodies, like the What Works Centre for Crime Reduction – part of the College of Policing – and the Early Intervention Foundation, to find out what works to prevent violence, and then recruit public and third sector executives who can commission and deliver these interventions and services.

Beware of loud advocates for their interventions and approaches; always scrutinise the evidence and ask to see reports of published controlled trials. For interventions which sound promising, but which haven’t been thoroughly tested, recruit a reputable research agency, a local research-intensive university for example, to evaluate them.

A problem locally is that police chief constables have no authority, on their own, to introduce the collection of data in NHS emergency departments. By the same token, NHS chief executives have no authority to ensure that the data they supply will be used by police to target their violence prevention activity so that the considerable burdens which violence makes on emergency departments are reduced. The same issue is evident at national level.

The Home Secretary has no authority to ensure that Cardiff Model data are collected in hospital emergency departments and shared by NHS Trusts. Hence the importance of joint action locally, and co-ordination provided by the Cabinet Office and the Prime Minister at national level.

Where do we go from here?

The tragic spike in knife violence and homicide in London, as well as a number of other localities in the UK, has led to a renewed focus on multi-agency collaboration and the attention on prevention is higher than ever before. As in any campaign, this is prompting new collaborative efforts, tactical and strategic innovations, and progress which needs to be maintained and embedded in business as usual. UK public servants are key to this.

The Cardiff Model has proven effective in ways unimagined when it was first developed, for example in identifying and tackling gang violence and crack houses. It is also a UK export, including to Australia and to the United States where the federal Centres for Disease Control and Prevention (CDC) has published recommendations and a toolkit for implementation there.

Overall, the last decade of cuts has taught us that tackling serious violence depends on a continuous stream of high quality data, purposeful data analysis, and sustained prevention jointly led and carried out by local authorities, the NHS and the police.

Professor Jonathan Shepherd is a surgeon, now based at the Cardiff University’s Crime and Security Research Institute. He convened and for 20 years chaired Cardiff’s violence reduction unit and developed the rest of the Cardiff Model with his local government, NHS and police colleagues. He is a member of the Home Office Science Council and the What Works Council at the Cabinet Office.  — Jonathan Shepherd

(Picture credit: Unsplash)

Make sure to share your own thoughts with the author by leaving a comment below