Blakeborough L., Richardson A.
[UK] Home Office, 2012.
UK government-funded pilot schemes found no crime reduction benefits from brief alcohol counselling for arrestees under the influence of drink, disappointing hopes that arrest referral would help quell late-night alcohol-related disorder. The schemes did however uncover many dependent drinkers.
Summary From 2007 alcohol arrest referral was piloted in England by the Home Office UK government department responsible for maintaining law and order within the UK. to help tackle alcohol-related offending, in particular in the night-time economy. Schemes involved an alcohol counsellor offering brief counselling to adult arrestees identified by police as being under the influence of alcohol. In most schemes arrestees could refuse the offer without legal consequences; in a minority of cases, accepting counselling was a mandatory condition of being bailed (avoiding a spell in custody awaiting trial) or cautioned after admitting the offence as opposed to being prosecuted in court. Initial sessions (follow-up sessions were offered where appropriate) for voluntary clients generally took place in the police station; those for offenders who accepted counselling in relation to bail or cautioning generally took place elsewhere.
Schemes were piloted at first in four police areas and then in another eight. The project ended in September 2010. This account is based on a summary of findings from both schemes, but also draws on the two more detailed reports on phases one and two.
In both phases, the brief AUDIT questionnaire was used to assess the severity of the arrestee's drinking, and they were offered information on the risks of drinking (including links to offending) plus practical advice and techniques for reducing their consumption. Among other documents, counsellors were instructed to complete an Alcohol Intervention Record form for each client to document their intervention. Clients for whom these forms were completed constitute the samples of arrestees subject to the arrest referral process.
Evaluation involved interviewing staff and clients, observing counselling, analysing documentation (including 6916 Alcohol Intervention Record forms), and comparing arrest rates from police records for the six months before and after the intervention against those of similar offenders in the same police areas before the schemes had been implemented. Systematic records of whether the offender was under the influence of drink were not available for the comparison groups, so instead they were selected on the basis of their similarity in terms of sex, age, offence, and the fact that this took place at night. Changes in drinking were assessed by recontacting arrest-referral arrestees, but of 6916 just 829 – about 1 in 8 – both agreed to be followed up and reported on their drinking.
Over three quarters of the interventions were delivered on a voluntary basis. The mandatory routes were found by workers to be cumbersome and difficult to enforce. These did however help ensure arrestees at least attended an initial session, though just 34% attended further sessions away from the police station compared to 62% of voluntary clients. Alcohol counsellors found early and ongoing engagement with police essential to the development and running of the schemes, facilitating the vital role played by custody officers in identifying and referring clients.
Over 12 months the 12 schemes documented 6916 interventions involving mainly young white men aged 18 to 24, just over a third of whom were suspected of violent offences and nearly 16%–17% of being drunk and disorderly. According to their AUDIT scores, over a third (37%–38%) were dependent drinkers and nearly half (46%–49%) were not dependent but drinking at hazardous/harmful levels. For around 60% this was their sole arrest in 12 months (six months before and six after arrest referral).
The key issue was whether offending fell more after arrest referral than among comparison arrestees not subject to this procedure. Across the four phase one schemes (and within each one) it did, but only very slightly and not to a statistically significant degree. Results from the eight phase two schemes were statistically significant, but in the 'wrong' direction. The raw figures were that from six months before to six months after the focal arrest, arrests rose by 5% for arrest-referral arrestees but fell by 6% for comparison offenders. After adjusting for some of the differences Age, sex, month of focal offence, focal offence type, and number of 'pre-intervention' arrests between the arrest referral and comparison samples, the greater reduction among comparison offenders remained statistically significant, and none of the eight schemes registered a significant positive difference while three registered a negative one. Neither could arrest referral be shown to be associated with re-arrest reductions for particular types of offenders.
Without comparison group data it was not possible to infer that any changes in drinking were due to arrest referral as opposed to (for example) the passage of time or the experience of being arrested, but at least on this measure statistically significant post-referral reductions were found among the minority of arrest-referral clients who could be followed up.
A small and possibly unrepresentative minority of clients were interviewed and expressed mixed views on the value of their arrest referral experiences. Some said these had prompted reflections on their drinking and they had made changes such as reducing the speed at which they drank or avoiding certain people or places. However, many did not attribute these changes wholly to the counselling, and memories of the sessions were often vague. In phase two pilot schemes the overall impression given by clients was that interventions were able to identify and make use of motivational levers. Details about alcohol units and how long alcohol stayed in the system were frequently recalled. Clients who had already identified a need to reduce their drinking found the sessions useful, but benefits were less obvious for those who did not believe they had a problem.
The average cost per intervention ranged from £62 to £826. Since no benefits could were proven to occur, it was not possible to set the value of these against the costs of the intervention. It was however possible to establish how much of a reduction in offending (had there been one) would have been needed to cover costs. Phase one schemes would have needed to produce a reduction of between 0.6 and six arrests per 100 interventions, phase two around a 5% reduction in re-offending. On this basis, most schemes did not break even, though possible health and other benefits were not accounted for.
Alcohol arrest referral schemes proved feasible in several police force areas. Both to refer and to treat individuals, schemes depended on effective working relationships between custody staff and alcohol agency workers. Doing the counselling while the arrestee was detained in the police station helped ensure at least an initial session. For procedural reasons, mandatory routes for referral were more difficult to implement than voluntary, particularly in respect of conditional bail, but did lead to better attendance at the first sessions. However, subsequent sessions were less well attended, suggesting that mandatory routes are useful in securing initial attendance, but thereafter motivation and engagement with the intervention is more influential.
Before the schemes were established it was thought that most clients would be non-dependent risky drinkers. In fact these were slightly in the minority, and there were more dependent drinkers than expected. Furthermore, as reflected in arrest records, arrestees were generally not prolific offenders, making it difficult for arrest referral to further reduce their known or suspected offending. More generally, this finding raises questions about any offender-centred approach to tackling crime and disorder related to the night-time economy.
commentary Evaluations of alcohol arrest referral schemes in Worcester and Dudley before the Home Office pilots had found enough in the feedback from offenders and staff to recommend commissioning of similar schemes elsewhere. Before the findings of the featured report were known, the Home Office had issued guidance on establishing alcohol arrest referral schemes. But in the context of disappointing findings from the Home Office pilots, not surprisingly the UK government alcohol strategy for England and Wales released in 2012 gave no encouragement to expanding alcohol arrest referral, merely commenting that such schemes can operationally and in funding terms be integrated with similar schemes for users of illegal drugs.
With phase one results already known, the report on the phase two schemes concluded that "the overall direction of the evidence does not support the continuation of the [alcohol arrest referral] process in its current form or for the current outcome measures". The key limitation to demonstrating crime-reduction benefits was that, to judge by arrests, there was little crime to reduce, a contrast to the often extensive criminal histories of arrestees whose offending is assessed as related to illegal drug use. Related to this, given that for most this was their first recent arrest, it seems possible that the impact of this experience would overshadow the impact of brief alcohol counselling.
Instead the authors of phase one and two reports suggested that arrest seems to provide a valuable opportunity to identify dependent drinkers who are more motivated than 'binge' drinkers to do something about their drinking, and to direct them to more intensive intervention (such as specialist treatment for alcohol dependence), potentially justified on health and broader social as well as crime reduction grounds. Such a change would, as suggested by the featured report, distance the schemes from the original intention to address public nuisance and disorder related to late-night drinking, and reposition them as a treatment entry route for individuals with serious drink problems.
Thanks for their comments on this entry in draft to Anna Richardson of the Home Office Drug and Alcohol Research Team. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 19 April 2012. First uploaded