Police custody risks being as much a revolving door as prison. The same vulnerable people are picked up time after time, seldom getting to court, let alone being sentenced. Police cells are disproportionately occupied by people who have mental health problems, alcohol and drug addictions and/or are homeless – some of the most vulnerable in society.

Police custody staff have long realised that police action will not stop this door revolving. So they’ve turned to health practitioners to support suspects’ legal rights and to help keep them out of the criminal justice system altogether. Custody suites now host liaison and diversion (L&D) teams. L&D practitioners provide a great service to suspects in custody, assessing their needs and vulnerabilities and referring them to services where relevant. They were trialled many years ago, and set up nationwide in the wake of the 2009 Bradley report on people with mental health issues and learning difficulties in the criminal justice system.

A major evaluation of L&D scheme outcomes has just been published, but it begs more questions than it answers. And suggests that L&D schemes can only be effective at keeping vulnerable people out of the criminal justice system if health and other services pick up the pieces before and after they are arrested.

The good news is that those who have been referred to L&D seem to be diverted in greater numbers from prison sentences, but we don’t know if vulnerable people were diverted in other ways. Diversion usually refers to diversion from court, diversion from the criminal justice system altogether or diversion from police custody. Rand, the researchers, didn’t have the data to assess the effect of L&D on these outcomes (or weren’t asked to look at it). Nor does the research analyse the other CJS outcomes L&D suspects receive – both court and out of court disposals, NFA, and whether they are more or less likely to get remanded.

The bad news is that the report does measure reoffending, and the findings are pretty negative. The researchers used official reconviction data and also asked those who had been referred to L&D whether they had committed crime again. Both sources suggested that these vulnerable people were more likely to commit crime after referral to L&D than before. On the self reporting measure, shoplifting went down, but drug selling, other theft and burglary and violence and assault went up. The official reconviction data suggests that the proportion of people committing offences slightly increased post L&D referral though “we cannot draw firm conclusions on the impact of L&D referral on offending behaviour”. Reoffending is not the only measure we should use for the success of the criminal justice system, but it is important – further offending creates more victims, and more stigma and criminal records for those committing crime.

It is not clear why L&D doesn’t have a more positive impact on reoffending, but the rest of the report gives some strong hints. The researchers tracked people’s engagement with A&E and with drug and alcohol services both before and after their detention in police custody and referral to L&D. This suggests that L&D clients are already heavily engaged with services, and that arrest comes at a crisis point. “The 6-12-month prior period to L&D referral is often characterised by a steep increase in contact with Accident & Emergency (A&E) services, specialist mental health services, and declining self-reported health in those attending drug treatment services. However, L&D service users with and without previous criminal justice system contact were as likely to go to A&E after involvement with the L&D service. This suggests that for L&D service users it is acute health vulnerabilities that lead to contacts with the CJS, rather than the other way around”.  L&D referred many people to services but, unfortunately, the people referred often did not attend appointments, particularly at the Improving Access to Psychological Therapies (IAPT) service.

At first glance it may seem that L&D is failing in its purpose – to engage vulnerable people in services which will support them to stop offending. But the findings also suggest that L&D will never make a huge difference while health services are not joined up. We know that those with complex needs are in increasing contact with the health system before they are arrested. Could such contacts trigger a preventative response? Could addiction and mental health services be reconfigured so vulnerable people were more likely to attend? One of the most disappointing findings of this report is that L&D engagement did not seem to influence offending. But if health services engaged more effectively with vulnerable people post arrest and referral by L&D, the offending would surely reduce?

L&D practitioners understand the issues which face their clients and what would help. ‘If someone has got benefit issues and they’re stealing to eat or to survive then obviously if you can break that cycle by helping them in the community to access benefits, then you may stop that offending cycle. Or if someone is offending because of relapse due to mental health issues and they’ve not been to mental health services, then [when appropriate services are provided] hopefully you’ll stop the cycle of reoffending with that person and…before it gets to an even more serious stage.’

But this study reinforces previous research – that the answers to crime lie outside the criminal justice system. Unless and until mainstream services prioritise meeting the needs of this cohort on their terms, they will continue to be frequent visitors to both police custody and prison.

NB we have just published a guide on how to more effectively message about diversion from prosecution.