Policing is undergoing substantial mission change. Where acquisitive crime rates and safety on the streets once set the agenda, the police increasingly find themselves dealing with risks to the vulnerable, threats posed by the dangerous, and trying to reduce the harm caused by the latter to the former.

It is work that requires an extended reach, beyond streets and town centres into less visible private and virtual spaces where established methods such as hotspot policing, situational crime reduction and problem orientated policing seem to have less to offer.

Here it is both ironic and problematic that as interest in using the evidence-base’ has grown, attention has shifted to crime problems where research coverage is thin. For example, much less is known about preventing child sexual exploitation than reducing burglary. Where evidence is lacking, orthodoxy quickly fills the gap.

The orthodoxy has two main characteristics. First, the business of threat, risk, harm and vulnerability is increasingly being managed at the level of cases; that is individuals or occasionally, families, couples or small groups with complex personal case-histories that have come to the attention of service providers and have apparently risky’ futures. And second, the mechanisms and arrangements through which these cases are addressed are routinely constructed on a multi-agency basis, bringing the police together with other local services to organise and co-ordinate case activity, share information and expertise, assign tasks and hold each other to account’.

This was the context that confronted the Police Foundation research team when we tried to find ways to improve the response to violence in one English town, as part of our Police Effectiveness in a Changing World project.

Although we mapped and analysed police, ambulance, A&E and other datasets, interviewed offenders and victims and consulted local practitioners, violence in the town resisted formulation into a unified, coherent problem’ and appeared poorly suited to the types of response shown to work in other circumstances. Hotspots were weak, much of the violence (including about 40 per cent non-domestic’ violence) took place behind closed doors, and where crime-drivers could be identified they were distant and difficult to address through practical interventions.

What we also found, however, was a substantial level of recurrence within the crime data (the same individuals were coming to attention time and again) and a committed set of practitioners, from a broad range of agencies and organisations, who were concerned about violence and jointly saw it as their business’.

In this context it is easy to see why multi-agency case management is appealing and frankly there seemed to be few other options available.

Evidence and orthodoxy

As any local superintendent’s wall-planner will testify, multi-agency case management forums (MAPPA, MARAC, MASH, IOM, Troubled Families, and a range of local iterations such as the Violence Multi-Agency Panel we helped develop in this project) are a growing part of police business and increasingly place demand on police resources, as well as those of other agencies. Despite this, it is important to recognise that this approach remains largely unevidenced.

The process and impact assessments we conducted in our project fit the picture emerging from other evaluated practice. On the one hand the scheme succeeded in bringing together an experienced and committed group of practitioners, systematised the way they identified cases, improved information sharing, broke down barriers, and was viewed very positively by those involved in delivery. On the other hand, the outcomes they were working towards in this case a reduction in the rate at which recurrent victims and offenders were involved in further violence did not materialise.

Of course the lack of impact identified, in our study and others, does not necessarily mean that the paradigm is flawed. What it demonstrates is that the orthodoxy alone is not enough to guarantee results in all the circumstances to which it is applied: just because everyone involved likes the working arrangement, it doesn’t automatically follow that positive outcomes will result.

Next steps

So where do we go from here? There is intuitive sense in the approach and perhaps worryingly few other big ideas; pulling back from multi-agency case management does not seem to be an option, but a number of pointers emerge from our research.

First, there is an urgent need to redress the balance between evidence and orthodoxy that informs current practice. We need to know much more about what works in multi-agency case management and in what circumstances multi-agency case management works.

Second, there is a tendency for formal and informal evaluations of such schemes to focus on internal, organisational process improvements, such as better working relationships and improved information sharing. More attention needs to be given to assessing what additional outputs such arrangements generate in terms of the quantity and quality of engagement and intervention work with cases.

Third, multi-agency arrangements seem to contain inherent weaknesses in relation to tasking and compliance monitoring. Keeping partners on board’ while challenging their performance requires a perhaps impossibly finely balanced management style. Here closer integration of multiple specialisms into more vertical delivery structures may be required.

Finally, there remains a question of purpose. Our initiative was explicitly premised on reducing recorded crime, yet the collaborating practitioners often seemed to slip into an admirable yet less tightly defined habit of shared difference making’. More thought needs to go into clarifying the objectives of the partnership project and the role each agency should play within it.

Multi-agency case management: evidence and orthodoxy the third thematic paper in the Police Foundation’s Police Effectiveness in a Changing World series, is published today.