Applying an Alcohol Brief Intervention (ABI) model to the community justice setting: Learning from a pilot project

Abstract

A pilot study was implemented in three local authority areas in Scotland to test the feasibility and acceptability of delivering Alcohol Brief Interventions (ABIs) as part of routine practice in community justice. Screening for Alcohol Use Disorders (AUDs) was undertaken with offenders given Probation or Community Service Orders, with randomization of participants into control (information only) and experimental (ABI) groups. A total of 419 offenders were assessed; 195 meeting inclusion criteria, consenting to take part and for whom a report was returned. Of these, 42 per cent (n = 82) fell within the harmful/hazardous drinking ranges (eligible for an ABI) and just over half (52%; n = 43) were randomized to receive an ABI. Despite limited follow-up data to measure potential effectiveness of ABIs, the study showed that alcohol screening tools can be used successfully in this context, and that screening and ABIs can be accommodated into routine community justice practice with reasonable levels of acceptability.

Keywords

Alcohol brief interventions community justice offenders probation screening

Introduction

The rationale for exploring the application of an ABI model in offender populations is based on the strong links that have been evidenced between alcohol and offending. Information from the Scottish Prison Service (2009) Prisoner Survey 2009 shows that 50 per cent of those in receipt of a custodial sentence or on remand in a Scottish Prison reported being drunk at the time of their offence. Alcohol problems are also more prevalent among prisoners than they are in the general population with almost half (46%) of prisoners maybe having harmful alcohol use or potential dependency (Scottish Prison Service, 2009) compared to 14 per cent of the adult male general population (Corbett et al., 2009). Research has also shown that there may be a direct association between alcohol misuse and increased risks of violent offending among young people in particular (Fergusson et al., 1996; McKinlay et al., 2009). Indeed, more than three-quarters of young offenders (77%) report being drunk at the time of their offence (Scottish Prison Service, 2010). Alcohol consumption has also been shown to be linked to risk of victimization (World Health Organization, 2010). Data from the Scottish Crime and Justice Survey (SCJS) 2010/2011 showed that around two-thirds (63%) of victims of violent crime perceived the offender to be under the influence of alcohol at the time of the offence (Scottish Government, 2011). Again, evidence shows that young people in particular may be at risk of alcohol-related victimization (Galloway et al., 2007) and women have also been found to be at risk of alcohol-related partner abuse (Hamlyn and Brown, 2007; World Health Organization, 2010). Interventions which reduce alcohol consumption in offenders have the potential to reduce not only the prevalence of alcohol problems in offenders but may also reduce re-offending.

Despite the strong links between alcohol and crime, the relationship between the two is not definitively causal. Alcohol-related crime can be described in three broad categories (Deehan, 1999):

direct causal relationship: alcohol-specific offences such as drink-driving and drunkenness;

contributory factor: alcohol as a trigger or facilitator to offend (e.g. assaults, antisocial behaviour); and

co-existent relationship: crimes unrelated to perpetrators’ alcohol consumption.

For instance, with regard to violence, alcohol is recognized as both a causal (Boden et al., 2012) as well as a contributory factor (World Health Organization, 2005). Not all alcohol consumption, however, leads to violence and not all violence is due to alcohol. There is a complex interplay between the quantity of alcohol consumed, drinking patterns and individual and contextual factors (Graham et al., 1997).

For individuals drinking at hazardous or harmful levels, Alcohol Brief Interventions (ABIs) are among the most extensively evaluated approaches (Raistrick et al., 2006). In Scotland, Alcohol Brief Interventions have been defined as: ‘A short, evidence-based, structured conversation with a client/patient that seeks in a non-confrontational way, to motivate and support the client/patient to think about and/or plan behaviour change’ (Scottish Government, 2009a: 49). The evidence base for the effectiveness of ABIs to reduce alcohol consumption when delivered in the primary health care setting is considerable (Kaner et al., 2007; Raistrick et al., 2006; Scottish Intercollegiate Guidelines Network, 2004). Indeed, in response to increases in alcohol-related harm in Scotland, the Scottish Government in 2008 set a national ‘HEAT’ (Health Improvement, Efficiency, Access and Treatment) target for the National Health Service (NHS) for the delivery of ABIs (149,449 in primary care, accident and emergency (A & E) and antenatal care by 2011) supported by a considerable increase in funding for services providing alcohol treatment and support. This target was subsequently met (Information Services Division, 2011) and extended for a further year to help local Health Boards embed screening and ABIs into routine practice, and support their development in other settings including criminal justice. Indeed it has been hypothesized that ABIs may have potential not only to reduce alcohol consumption in offenders, but also to contribute to a reduction in re-offending as well (Newbury-Birch et al., 2009a).

ABIs in isolation would have only a limited effect in reducing population alcohol consumption and health inequalities. To address these outcomes, effective alcohol policy is one that encompasses a range of interventions (including regulatory measures, support and treatment interventions and changes in culture and attitudes) delivered via a comprehensive approach aimed at the whole population with particular targeting for high risk groups, including offenders. Scotland is currently implementing such an approach which includes provisions for minimum unit pricing for alcohol, licensing reform and increased investment in treatment and care services (Scottish Government, 2009b).

Community justice alcohol studies

Evidence for the effectiveness of ABIs in other settings, including community justice settings, is only beginning to emerge. In 2007, a pilot study was undertaken to evaluate the most feasible criminal justice setting in which to conduct a randomized control trial of ABIs (Coulton et al., 2012). A survey of adults being held in three police custody suites, three prisons and three probation offices was used to compare scores on three different screening measures (Fast Alcohol Screening Test (FAST), Modified-Single Alcohol Screening Question (M-SASQ) and Alcohol Use Disorders Identification Test (AUDIT)). Of the three, probation was found to be the most suitable for screening with the greatest levels of consent for study participation in this group.

Following on from this work, in England and Wales a national Screening and Intervention Programme for Sensible Drinking (SIPS) was implemented which sought to provide evidence on the delivery, effectiveness and cost effectiveness of a range of screening and ABI approaches across settings and regions in England (Cochrane, 2010). The programme comprised three linked trials: one in the primary health setting, one in the A & E setting and one in the criminal justice setting. The criminal justice arm of the trial operated in 20 probation offices in three geographical regions of England, with trained offender managers testing two screening tools (FAST and M-SASQ) and three forms of intervention to explore the effectiveness of each for screening and brief intervention for alcohol use disorders (AUDs) in probation. It also explored staff and client views about the appropriateness and acceptability of screening and intervention in the probation context and factors associated with successful implementation. Follow-up data at both six months and 12 months showed a significant decrease in the proportion of participants who scored 8 or more on the AUDIT screening tool, with reductions being even greater at 12 months than six months. This was true for all clusters (McGovern et al., 2012). Despite this shown effectiveness at reducing alcohol consumption, however, the study encountered some challenges with implementation. Workload pressures and high staff turnover of criminal justice staff, as well as a greater than anticipated need for support in delivering the scheme suggested that significant managerial support and local champions of screening and brief intervention were necessary in the probation setting to ensure successful implementation. Concerns were also raised about confidentiality and resistance to questions about drinking, as well as concerns about the length of screening and intervention being too long. In addition, offender managers reported feeling overworked by the scheme. Training was also a barrier with one-to-one training having to take place due to lack of group availability. Staff also challenged the appropriateness of the scheme given their work was primarily with heavy drinkers (Cochrane, 2010).

The limited evidence base, to date, for the feasibility and effectiveness of screening and ABIs in community justice is bolstered by earlier work carried out in the wider criminal justice context, mostly in police custody and prison settings.

Police custody alcohol studies

An arrest referral scheme set up in Nottingham, England in 2006, which used the AUDIT tool to screen over 2000 arrestees, showed that the mean score for this population was 12 (in the hazardous drinking range) (Hopkins and Sparrow, 2006). Similarly, in 2008, a study of screening and ABIs with detainees held in police custody in Plymouth, England in which 3900 arrestees took part (83% of those contacted), showed that the proportions of those screened who scored in each of the three AUDIT ranges were 36 per cent low risk, 43 per cent hazardous/harmful and 21 per cent dependent respectively (Barton and Squire, 2008). A feasibility study for the delivery of screening and ABI with arrestees detained specifically for offences linked to drinking behaviour (Brown et al., 2010) showed similar patterns to other studies, with a mean score of 15.3 and with 50 per cent hazardous drinkers, 15 per cent harmful and 35 per cent showing signs of dependence. More recently, in 2010, researchers working alongside the police in Aberdeen (Scotland) made contact with 207 detainees, of whom 16 per cent scored in the low risk category, compared to 39 per cent hazardous/harmful and 45 per cent dependent (Gibbons-Wood et al., 2010). The overall proportions of those in the hazardous/harmful and possibly dependent clusters in the Scottish-based study were greater than the comparable English studies, perhaps indicating higher levels of problematic consumption in the Scottish context.

Prison alcohol studies

Scottish research by Graham and colleagues (2012) showed approximately 30 per cent of remand and short-term prisoners to be drinking at hazardous and harmful levels, and just under 50 per cent at possibly dependent levels. A study of alcohol use in four prisons and three probation offices in the north of England also showed that AUDs are highly prevalent among these populations (Newbury-Birch et al., 2009b). More recently, data from a Prison Health Needs Assessment for Alcohol Problems (Parkes et al., 2011), showed that, from a total of 259 screenings, 73 per cent of prisoners had scores in the hazardous/harmful range (8+ AUDIT score), including 36 per cent possibly dependent (20+ AUDIT score). Those in the highest risk range were at opposite ends of the age spectrum (18–24 and 40–64 age groups), and higher AUDIT scores were present among those with shorter sentences (less than six months). The research showed that prevalence of alcohol problems in the Scottish prison population was markedly higher compared to the general population, at all ages, and for both men and women. The research also showed that there was no formal alcohol screening using a validated instrument in the Scottish prison estate and, as a result, many prisoners with alcohol problems were going undetected. Further, there was limited access to alcohol-specific interventions with most prisoners instead accessing more general substance misuse interventions. The research also revealed self-reported associations between drinking alcohol and index crimes among two-fifths of participants, increasing to half of those involved in violent crimes, as well as links to recidivism (MacAskill et al., 2011).

In addition to screening studies within prison settings, other recent research has tested the effectiveness of brief intervention/motivational interviewing approaches. This research has cut across a wide range of prisoner sub-groups, including adult males (Forsberg et al., 2011), women (Begun et al., 2011; Clarke et al., 2011; Stein et al., 2010) and juvenile offenders (D’Amico et al., 2010; Stein et al., 2011). As well as variation in target groups, the length of intervention delivered varies considerably and results have been mixed. For women offenders, the studies which report positive results only do so at short-term follow-up (less than three months). For juvenile offenders, studies show that ABIs have potential in both individual and group settings, but more research is required. In the one study on adult male prisoners, reductions in alcohol and drug use were observed in both control and intervention groups.

Learning from the previous research

The evidence shows that there is a high prevalence of drinking in the offender population and also that the criminal justice setting is one which affords a potential opportunity to engage with clients, who would not necessarily engage elsewhere, to address their alcohol problems. The criminal justice setting therefore presents a potentially ‘teachable moment’ (Havinghurst, 1952: 7), when offenders may be reflecting on the reasons why they have become involved in criminal activity. There are numerous alcohol screening tools and interventions available, many of which have been tested rigorously and shown to work in applied settings, predominantly health. That said, there is little evidence of tools designed specifically for criminal justice populations and only a few studies which test standard tools in this setting. Indeed, an independent review of procedures adopted by the National Probation Service (NPS) for identifying and intervening with offenders who have alcohol problems has shown that there is currently limited scope for developing empirically informed guidance to instruct staff about the effective targeting of interventions within a criminal justice context, or to identify which ones are likely to be most effective for whom (McSweeney et al., 2009).

It is noted that most screening tools that have been used in the criminal justice setting rely on self-report and researchers have noted that offenders are likely to under-report their alcohol-related problems, because they do not wish to be labelled as having alcohol problems (Lapham, 2004/2005). It is also well evidenced that alcohol problems among offenders often exist alongside other problems such as drugs misuse and mental-health issues which means that screening for alcohol in this group can be problematic since self-report estimates may be confounded (Lapham et al., 2001).

From the existing literature, it seems that evaluation of screening and interventions is difficult with this population and the limited evidence that is available suggests that this is mainly due to lack of follow-up and outcome data and being able to track clients, as well as issues with staff training and workloads which make it difficult to collect reliable and representative data (Barton and Squire, 2008; Brown et al., 2010; Cochrane, 2010; Hopkins and Sparrow, 2006). Sample sizes are generally too small to say anything conclusive about actual impacts or outcomes of ABIs, although the SIPS study in England and Wales has provided some evidence of effectiveness at six and 12 months after initial testing (Cochrane, 2010). Research has also shown that, even when some arrestees acknowledge that their drinking may be problematic, they feel they either could not, or would not, engage with services of their own volition (Barton, 2011). Perhaps the main outcome of such research to date, therefore, is that it provides intelligence of the nature and scale of AUDs within the detainee and offender population, but limited knowledge on effective interventions to address these behaviours.

This article presents screening and intervention findings from a pilot project which was established to test the feasibility and acceptability of applying an ABI model in the community justice setting in Scotland, which was operational between January 2010 and April 2011. The article also includes qualitative insights from staff participating in the pilot. This formed part of a wider evaluation (Skellington Orr et al., 2011), designed directly to inform Scottish policy and practice developments to understand better the extent and nature of alcohol problems in offenders and which effective interventions can address them, recognizing that the criminal justice setting is an opportunity to detect and intervene in an often ‘hard to reach’ population.

Methods

The pilot was designed to be delivered as part of routine practice for staff working in community justice services, and was to be delivered to clients in receipt of the community-based sentences of Probation Orders (POs) or Community Service Orders (CSOs) after Orders had been issued as part of the initial appointment with the designated social worker/community service officer. This stage of the community justice process was chosen for the study instead of at an earlier stage so as not to interfere with the sentencing process. It was also felt that if screening and ABIs were carried out at an earlier stage there would be higher levels of attrition among participants, as not all those recruited would go on to be sentenced (and therefore have onward contact with community justice staff). Carrying out the screening and ABI at the initial appointment stage also meant that the same staff member would be engaged with the client over time and so they could carry out both baseline and follow-up data collection, thus limiting any potential bias in the sample.

Staff in all areas received standardized training which comprised pre-event online training as well as a one-day face-to-face session delivered by National Health Service training staff with ABI expertise. Staff in one area also requested a follow-up training session that was delivered by a local co-ordinator recruited to the area. A total of 121 staff took part in the training sessions, which included front-line and managerial staff.

A baseline questionnaire was developed to assess clients for eligibility and to act as a consent form. Only those providing verbal consent were considered for inclusion in the study. Individuals had to satisfy a number of inclusion criteria before being eligible to participate. These included ensuring alcohol treatment or education was not a condition of their current Order, and screening and an ABI had not been carried out elsewhere in the preceding 12 months. This was because it was assumed that such individuals were already likely to be engaged with alcohol treatment or support providers, or to have done so recently, and so would be unlikely to benefit from an ABI. Other exclusion criteria were based on demographic and health factors, including age (the pilot was only open to those aged over 18) and physical/mental fitness to take part.

For screening purposes, the AUDIT questionnaire was selected as this represents the ‘gold standard’ test for identifying hazardous, harmful and dependent drinkers (Babor et al., 2001) and has also previously been shown to work effectively in offender populations (Parkes et al., 2011). Developed by the World Health Organization (WHO), the AUDIT is a self-report measure that comprises 10 pre-coded questions which produce a score in the range of 0 to 40. AUDIT scores can be clustered to four levels. Those scoring 8 or below are considered low risk, and those scoring 8 or above can be classified as having an AUD. The four clusters, as described in the National Institute for Health and Clinical Excellence Public Health Guidance 24 (NICE, 2010) are:

1 to 7: low risk drinking.

8 to 15: hazardous drinking.

16 to 19: harmful drinking.

20+: possible dependence.

The AUDIT has been internationally validated, is quick and simple to deliver and has been proven to integrate well into standard practice in numerous applied health and social care settings and across a range of drinking cultures (Reinert and Allen, 2007).

The pilot adopted a randomized control design in which staff were randomly allocated to either a control group (screening and delivery of a booklet) or an intervention group (screening and delivery of ABI if appropriate) using a statistical software package (SPSS). This approach was taken, instead of randomizing clients themselves, so that there was less chance of the sample being contaminated by staff bias which might have occurred if decisions were made on a client-by-client basis. It was also an approach that seemed most practical in terms of staff training and delivery. The researchers were blinded to this randomization for the duration of the fieldwork, and data were only allocated to the two groups at the start of the data analysis.

Data were intended to be collected at three stages – at initial appointments and at three- and six-month follow-ups as appropriate, to explore changes in AUDIT scores over time. Probation Order clients already had an existing review structure every three months which the pilot model mapped onto. For Community Service Order clients, the pilot required that staff would undertake a three-month face-to-face ‘mini-review’ and a six-month review over the telephone or face-to-face, as appropriate. Data were collected from three separate pilot sites, but were merged for analysis purposes. The three separate pilot sites are identified here as pilot areas A, B and C to retain anonymity.

Attitudinal and perception data were collected from participating staff via online questionnaire surveys, training questionnaires and one-to-one qualitative face-to-face interviews.

Data were analysed descriptively using SPSS v15.0 (SPSS Inc. 2007).

Results

A total of 419 offenders were assessed using the baseline questionnaire to ascertain eligibility to take part. Figure 1 shows the percentage breakdown of those who met the criteria and who provided consent to take part. A total of 195 offenders participated in the study, representing 94 per cent of those who were eligible and provided consent. AUDIT forms were not returned for the remaining 6 per cent of offenders who had agreed to take part.

Figure 1.

Eligibility, consent and participation rates.

Table 1 shows that most offenders who did not meet the criteria did so on the basis that they were already subject to alcohol treatment/education as a condition of their Order. This was the case for over a third (39%) of clients. A further 35 (29%) clients were already in treatment for their alcohol problems (not attached to an Order), and so were not included in the pilot on that basis. Seven (6%) clients had already received screening or an ABI elsewhere in the previous 12 months through the national ABI programme, and so were not eligible to take part for that reason. This meant that just under three-quarters of clients who were excluded from the pilot could potentially be classified as having an AUD and would already have had an opportunity to receive alcohol advice and support from another source.

Table 1.

Number and percentage of clients who did not meet eligibility criteria.

Reason for ineligibility	N	%
Alcohol treatment/education included as a condition of Order	47	39
Currently in treatment for alcohol problems	35	29
Under 18	15	13
Received screening and/or ABI elsewhere in previous 12 months	7	6
Intoxicated at time of interview/not alert and orientated	6	5
Suffering serious mental health problem	6	5
Cannot read and write English sufficiently well to take part	4	3
Total	120	100

Of the 295 clients (70% of the overall sample) who met the eligibility criteria, 207 (70%) consented to take part. Participation in the pilot was entirely voluntary and, at 70 per cent, was lower than other similar studies. This percentage was skewed, however, by a low consent rate in Area A (47%), as in Areas B and C it was over 80 per cent. There was also a higher ‘refusal to consent’ rate among community service clients. Only just over half (51%) of community service clients who were eligible to take part provided verbal consent, compared to 93 per cent of probation clients.

Figure 2 details the AUDIT scores and randomization that occurred for the 195 participating offenders. It shows that 42 per cent of all those who completed an AUDIT form fell within the hazardous/ harmful drinking range, and were allocated to one of the ABI experimental groups.

Figure 2.

AUDIT scores and randomization for participating offenders.

From the 195 clients where gender data were recorded, 85 per cent were male and 15 per cent were female. This gender split was exactly the same as the proportionate split for all those who completed the baseline questionnaire. The gender ratio in the different settings showed that there was a slightly higher presence of females in the probation setting (80% male compared to 20% female) than the community service setting (89% male compared to 11% female).

These figures compare closely with the gender split for CSOs and POs for the selected case study areas combined which, in 2009/2010 were 89 per cent male, 11 per cent female and 83 per cent male, 17 per cent female respectively (Scottish Government, 2010). Essentially, therefore, the participating population was not different from the total population assessed for eligibility or from the wider CSO/PO population, in terms of gender.

The average (mean) age of participating clients was 31, ranging from 18 to 65. The average age of participants was the same as the average for all those assessed for eligibility. Although the average (mean) age of those receiving CSOs and POs nationally and locally was not known, national statistics show that those aged 31–40 typically make up around 21 per cent of all those in receipt of Orders (Scottish Government, 2010). While this age band has the greatest prevalence among PO recipients, it is second to 21–25-year-olds for CSOs.

In 97 per cent of cases, clients were classified as being ‘White’ and in 2 per cent of cases their ethnicity was described as ‘Other’. Ethnicity data were missing in 1 per cent of cases. Again, this did not differ from the total population assessed for eligibility or the national data for POs and CSOs (Scottish Government, 2010).

There was some variation between probation and community service clients in terms of the overall distribution of scores, with 38 per cent and 45 per cent respectively scoring in the 0–7 range, 40 per cent and 46 per cent respectively scoring in the 8–19 range and 22 per cent and 9 per cent respectively scoring in the >19 range. This shows that, although the proportion of clients from each designation who were eligible for an ABI was broadly similar, a higher proportion of community service clients fell in the low risk category compared those in receipt of POs. Conversely, a higher proportion of probation clients were in the possibly dependent category compared to community service clients.

For males, the mean AUDIT score was 12.2 compared to 9.3 for women. There were also differences in the distribution of scores among males and females, with 46 per cent of males compared to 24 per cent of females falling in the intervention AUDIT range (8–19). Far fewer males fell in the low risk score range (36% of males compared to 59% of females). The proportion of males and females in the high risk category was 17 per cent in each case, showing that the difference between the genders was all located in the lower two categories.

Table 2 shows AUDIT scores by age. It shows that those aged 18–24 were more likely to have an AUDIT score in the hazardous/harmful range (55%) compared to those aged 25–29 (38%) or 30–39 (40%) or older (21%). Indeed, 18–24-year-olds accounted for just over half (51%) of all those falling into the ABI range despite only accounting for 40 per cent of the sample overall.

Table 2.

Age band and AUDIT scores.

	18–24	25–29	30–39	40+
	(n=76) (%)	(n=40) (%)	(n=40) (%)	(n=34) (%)
Score 0–7	30	42	42	59
Score 8–19	55	38	40	21
Score >19	15	20	18	20

Follow-up data

From the 82 clients who scored between 8 and 19 in the initial AUDIT assessment, only 18 follow-up forms were returned. This included 11 clients followed up at the three-month stage, and seven clients followed up at the six-month stage. Of these, only two clients had both three- and six-month follow-up data (both probation clients) and so the 18 forms received represented 16 individual clients, 22 per cent of the sample. This included three community service clients and 13 probation clients. It also means that in four cases, initial and six-month data were available, but there were no accompanying three-month data.

Of those with no follow-up data, in 21 cases Orders were described as ‘successfully completed’ and so contact with the client had been terminated before it was possible to carry out a three-month follow-up. In 10 cases the Order had been breached or a breach had been submitted. In one case an ABI had been done in another setting which the worker felt made it unsuitable to carry out follow-up work. At six months, one person who was due to be followed up had been referred to a specialist agency. A further participant was described as being ‘currently suspended’ at the time that follow-up data would have been captured. Information relating to all other clients who were not followed up was not available from the pilot areas. Table 3 below summarizes the follow-up and outcome data for each of the three pilot areas, and also shows the number of cases that were lost to follow-up (n = 30; 37%).

Table 3.

Follow-up and outcome data by area.

	Area A	Area B	Area C	Total
Total participants (scoring 8–19 at initial assessment)	20	22	40	82
Total followed up (3 months)	5	4	2	11
Total followed up (6 months)	5	–	2	7 ^a
Total where other outcome was known (of which:)	8	5	21	34
Order completed	3	3	15	21
Order breached	3	1	3	7
Order still active – breach submitted	–	–	3	3
Already had ABI screening in another context	1	–	–	1
Referred to specialist service at 3 months	–	1	–	1
Currently suspended	1	–	–	1
Lost to follow-up	2	13	15	30

Note: ^atwo people were followed up at both three and six months, so the 18 follow-up forms relate to 16 individual clients.

Staff perceptions

Views expressed by staff were generally supportive that screening and ABIs could be useful for encouraging clients to think more readily and realistically about their drinking behaviours:

It’s a reminder and a useful tool to start a discussion about alcohol issues, particularly where the person has initially said there are no alcohol problems.

People just didn’t think drinking was a problem.

Overall, however, although the AUDIT tool and the ABI were viewed as easy to administer, and were seen as useful tools in themselves, some negative perceptions of the appropriateness and likely success of screening and ABIs in this environment were expressed by some participating staff. In particular, some staff felt that harmful/hazardous drinking was not a key priority for this client group relative to their other criminogenic needs:

It’s, kind of, targeted at more middle of the road clients, and I don’t have that, I work with the extremes … they are either alcoholics or they are drug users.

Comments were also made about the timing of screening and ABIs for this group, and there was a strong view that they may capture more people and be of greater use in determining sentencing outcomes if undertaken at an earlier stage in the community justice process.

Some staff also felt ill-equipped to deliver screening and ABIs, and this was more prevalent among community service staff:

I do not feel that I have the detailed, specialist knowledge or skills to undertake such work.

This contrasted quite strongly with views from criminal justice social workers, for whom the training, in particular, was seen as being pitched too low:

I mean, alcohol interventions are kind of our bread and butter, you know, we do this a lot so it was pitched at the wrong level for us, definitely.

Even where the training was viewed more positively, some staff still had reservations about the suitability of ABIs in the community justice context and the strict eligibility criteria were also seen as compounding the challenge of finding willing and eligible clients to take part:

Appropriate training provided made it easy enough to deliver the ABI pilot in the area. There have been issues, however, as to when it is most appropriate to use this, and to find people who are suitable to use the information with.

One of the biggest gaps in the data collected from offenders was the follow-up data, discussed earlier, and interviews with staff suggest that this may have been largely due to a lack of understanding on their part about the need to collect such data:

Reminders for follow-ups are really helpful because when there are months between doing one [client interview] and doing the next one, you do forget about it … you do lose track and something like ABI can get lost if you don’t get a reminder.

Overall, feedback indicated that appropriately targeted training for staff, bolstered by on-going administrative and co-ordinative support and a broadening of eligibility criteria would maximize the potential effectiveness of screening and ABIs in this population.

Discussion

The pilot sought to explore the feasibility and potential effectiveness of using ABIs in the community justice setting. As one of the first controlled studies of its kind in this setting, the pilot was successful at identifying AUDs within the community justice setting and, as a screening exercise, provided valuable evidence of the levels of problem drinking in the community justice population which was otherwise not known.

The screening showed that around 59 per cent (n = 116) of offenders in receipt of CSOs or POs in the pilot areas, and who fulfilled the eligibility criteria and provided consent, were categorized as having an AUD. Of these, 42 per cent fell into the hazardous/harmful category and could potentially have benefited from receipt of an ABI. A further one in five offenders screened demonstrated high risk drinking behaviour that might indicate possible dependence. This may be an underestimate of the true prevalence of AUDs in this population, however, since an additional 21 per cent of those screened for eligibility (n = 89) were excluded from the pilot as they were already in contact with alcohol services. Table 4 compares the AUDIT scores from the current study with previous studies in the criminal justice field. It shows that the proportion of clients falling into the hazardous and harmful category were broadly in line with other studies, although there were slightly more in the low risk category compared to other research, and fewer in the high risk cluster. This was perhaps to be expected given that this pilot excluded those who were already in treatment for alcohol-related problems, or who had alcohol education/treatment as a condition of their Order. Thus, the low proportion of dependent drinkers in this sample is artificially low if used as an indicator of likely prevalence in the wider community justice population. Accepting that the sample was small, and that there may therefore have been some inherent bias, these findings nonetheless indicate that screening for AUDs in this population could potentially be used to identify effectively and intervene with a large number of people with alcohol problems who might otherwise not be identified as being in need.

Table 4.

ABI pilot scores and other studies (%).

	CJ ABI pilot (current study)	Barton and Squire (2008)	Newbury-Birch et al. (2009b)	Gibbons-Wood et al. (2010)	MacAskill et al. (2011)
Setting	Probation/community service	Police custody	Prison/probation	Police custody	Prison
Score 0–7	41	36	38	16	27
Score 8–19	42	43	27	39	36
Score >19	17	21	35	45	36

The main strengths of the study included the randomization process, the use of validated tools and standardized training delivered to all participating staff. Perhaps the main weakness was the failure to capture sufficient follow-up data and, as a result, the research was not able to say anything conclusive about the impact of ABIs on those who took part. This occurred largely as a result of lack of engagement from offenders with the offer to take part in follow-up qualitative interviews with researchers, as well as failure by staff to collect follow-up data, compounded by failure to attend meetings, breached and revoked Orders. Future research in this area should be mindful of these challenges as a lack of evidence on ABI impact in this setting will act as a barrier to future implementation.

Willingness to take part in initial alcohol screening among the community justice population was, however, around 70 per cent, and this is encouraging as an indicator for future likely engagement should the model be rolled-out elsewhere. It is important to note, however, that there was a higher refusal to consent rate among community service clients. Only just over half (51%) of community service clients who were eligible to take part provided verbal consent, compared to 93 per cent of probation clients. This supports observations of better willingness to participate among those on probation compared to other criminal justice populations, as found in England and Wales (Newbury-Birch et al., 2009b).

When examining all those who completed the baseline questionnaire, fewer of those in the community service group were excluded because they were already accessing alcohol treatment/education elsewhere, or to have it as a condition of their Order, compared to probation clients. This could be interpreted that, as a group, community service clients are less likely to be already receiving alcohol education, support, advice or intervention and so might perhaps have benefited most from the opportunity presented by the pilot. Their decision to decline participation may, however, also indicate that they were less likely to consider themselves to be in need of alcohol education, support, advice or intervention. Another possible explanation for the lower participation rate among community service clients may have been staff influence over decisions to take part. The training analysis and subsequent interviews with staff showed that community service staff felt less comfortable with the skills required to administer the AUDIT screening and ABI, possibly because of lack of previous motivational interviewing training which would have been received routinely by their qualified social work staff counterparts. Ensuring self-efficacy among staff to deliver such work is as important as client self-efficacy to take part.

Feedback from staff also revealed some negative perceptions about the suitability of screening and ABIs for this client group, as found in previous work (Brown et al., 2010). This may have resulted in reluctance to deliver the pilot, and also the limited follow-up data, although it is recognized that data collection problems may also have occurred due to other factors beyond staff control. In particular, staff expressed views that alcohol problems were of less immediate concern in light of their other issues for clients and this is perhaps one of the strongest themes emergent from the interviews carried out with staff. Again, this is not unique to this research and the large scale SIPS trial in England and Wales also reported reluctance from staff engaged in delivering screening and ABIs where they considered their clients to be ‘heavy drinkers’ and thus beyond the reach of such initiatives as ABI (Cochrane, 2010). From a process view, this maybe suggests a need to adapt training materials for this client group, although it is possible that this still would not overcome some staff reluctance to deliver ABIs if they perceive that they are not appropriate for their client group. From a research perspective, this highlights a requirement for future studies to explore criminal justice staff attitudes to alcohol in more depth, in particular the other concerns they prioritize ahead of drinking when dealing with clients with identified alcohol problems and their perceptions on the complex causal links between alcohol and offending. Although alcohol is cited as a key factor in a number of offences, it is likely that it is interacting with other social, environmental and individual factors. Rational choice theory (Cornish and Clarke, 1986) and the Situational Crime Prevention literature (Clarke, 1997) provide more rigorous analyses of the interplay between these factors. The ABI approach bears some comparison to the latter; while explaining offending with reference to an individual’s disposition or social and physical circumstances may be inherently problematic, manipulating situational or cognitive factors (in this case, encouraging individual reflection by discussing alcohol consumption) may encourage behavioural change and help prevent offending.

Staff perceptions of competing priorities in their clients’ lives also support the observations of MacAskill et al. (2011) who suggest that the high prevalence of socio-demographic indicators of disadvantage has implications for both successful desistence and rehabilitation. It is clear that alcohol problems are co-existent with and cannot be entirely disentangled from wider social, personal and behaviour problems often seen in the offender populations. Holistic approaches are therefore required which tackle progressively the various competing challenges facing offenders in their day-to-day lives.

Importantly, the current work also provided indicative evidence of the importance of readiness to change. A lack of willingness to engage in follow-up work with researchers and staff, as well as some of the attrition brought about by breaches of Orders, suggests that some of those targeted by the study simply were not interested in changing or reducing either their drinking or offending behaviour. Similar previous work has also shown that, even when drinking problems are understood, some offenders are unlikely to seek help or support proactively (Barton, 2011), possibly because they do not wish to. For others, however, lack of awareness of how to reach such support and other social and practical issues will arguably be a barrier. Different strategies may be required for those who are unwilling to engage with service providers because they are suspicious, over-targeted or disaffected compared to those who are service resistant for other reasons. Achieving a better understanding of the cultural bias in the way that patterns of drinking are understood for offender groups (Bernburg and Thorlindsson, 1999; Galloway et al., 2007), as well as understanding the positive sub-cultural norms about hazardous use and alcohol consumption and group identity that may exist for this group may enable better targeting of interventions.

Conclusion

In conclusion, this pilot study provides important new knowledge on alcohol problems within the community justice population. Despite challenges to determining impact, the pilot has shown that, on an administrative and resource level, there should be few barriers to introducing screening and ABIs into community justice settings. However, other issues require further consideration and exploration, in particular staff attitudes towards the complex causal relationship between alcohol and crime and the importance of the priority they afford client alcohol problems within a wider spectrum of needs. In addition, demonstrable impact of ABI effectiveness, not only on reduction of alcohol consumption but potentially on other outcomes such as a reduction on re-offending, within future studies is necessary before more widespread delivery can be achieved.

Overall, however, the pilot highlighted that the community justice setting does afford an opportunity to reach some of those most at risk of alcohol-related harm. This should be viewed within a wider national alcohol policy context which adopts a comprehensive approach aimed at the whole population, with particular targeting for high risk groups, including offenders.

Footnotes

Acknowledgements

Iain MacAllister and Sharon Grant (Scottish Government) were active members of the Project Advisory Group and provided detailed commentary on the initial report. In addition, the co-operation of criminal justice staff was essential to the smooth running of the study and their supportive participation was much appreciated. We would also like to thank all the criminal justice clients for their participation in the study.

Declaration of conflicting interest

The authors report no conflict of interest. The opinions expressed in this publication are those of the authors and are not necessarily those of NHS Health Scotland or the Scottish Government.

Funding

This work was supported by funding from the Scottish Government. The study was commissioned by NHS Health Scotland on behalf of the Scottish Government.

Author biographies

Kate Skellington Orr is Director of KSO Research. She was Principal Investigator on the Alcohol Brief Interventions Community Justice Pilot (2009–2011), and is a member of the British Psychological Society, British Society of Criminology and European Association of Psychology and Law.

Andrew McAuley is Public Health Adviser for Substance Misuse. He co-led the Alcohol and Offenders Criminal Justice Research Programme (2009–2011) funded by the Scottish Government and has been working on substance misuse research within the NHS for the past eight years.

Lesley Graham is an Associate Specialist in Public Health. She co-led the Alcohol and Offenders Research Programme, funded by Scottish Government. She has worked in the alcohol and offender health field for many years, including for the Scottish Prison Services, Scottish Government and the World Health Organization.

Shirley McCoard is a Social Researcher who works for the Scottish Prison Service. She led on both quantitative and qualitative aspects of the Alcohol Brief Interventions Community Justice Pilot (2009–2011) and has recently completed an MSc in Criminology and Criminal Justice.

References

Babor

Higgins-Biddle

Saunders

(2001) AUDIT: The Alcohol Use Disorders Identification Test. Geneva: World Health Organization.

Barton

(2011) Screening and brief intervention of detainees for alcohol use: A social crime prevention approach to combating alcohol-related crime? Howard Journal of Criminal Justice 50(1): 62–74.

Barton

Squire

(2008) Screening and Brief Intervention in the Charles Cross Custody Suite – final report. University of Plymouth, Social and Public Policy Research Group.

Begun

Rose

Lebel

(2011) Intervening with women in jail around alcohol and substance abuse during preparation for community re-entry. Alcoholism Treatment Quarterly 29(4): 453–478.

Bernburg

Thorlindsson

(1999) Adolescent violence, social control, and the subculture of delinquency: Factors related to violent behavior and nonviolent delinquency. Youth & Society 30(4): 445–460.

Boden

Fergusson

Horwood

(2012) Alcohol misuse and violent behaviour: Findings from a 30-year longitudinal study. Drug and Alcohol Dependence 122: 135–141.

Brown

Newbury-Birch

McGovern

Phinn

Kaner

(2010) Alcohol screening and brief intervention in a policing context: A mixed methods feasibility study. Drug and Alcohol Review 29: 647–654.

Clarke

Anderson

Stein

(2011) Hazardously drinking women leaving jail: Time to first drink. Journal of Correctional Health Care 17(1): 61–68.

Clarke

(ed.) (1997) Situational Crime Prevention: Successful Case Studies. New York: Harrow and Heston.

10.

Cochrane

(2010) Implementation of screening and brief intervention in Probation. Presentation to OHRN, PETaR Network event, ‘Alcohol and offenders – Challenges and new developments’, Holloway Prison, July.

11.

Corbett

Given

Gray

. (2009) Scottish Health Survey 2008. Edinburgh: Scottish Government.

12.

Cornish

Clarke

(1986) Introduction. In: Cornish

Clarke

(eds) The Reasoning Criminal. New York: Springer-Verlag, pp. 1–16.

13.

Coulton

Newbury-Birch

Cassidy

. (2012) Screening for alcohol use in criminal justice settings: An exploratory study. Alcohol and Alcoholism. Epub ahead of print 2 May 2012. DOI: 10.1093/alcalc/ags048.

14.

D’Amico

Osilla

Hunter

(2010) Developing a group motivational interviewing intervention for first-time adolescent offenders at-risk for an alcohol or drug use disorder. Alcoholism Treatment Quarterly 28(4): 417–436.

15.

Deehan

(1999) Alcohol and Crime: Taking Stock. London: Home Office.

16.

Fergusson

Lynskey

Horwood

(1996) Alcohol misuse and juvenile offending in adolescence. Addiction 91: 483–494.

17.

Forsberg

Ernst

Sundqvist

Farbring

(2011) Motivational interviewing delivered by existing prison staff: A randomized controlled study of effectiveness on substance use after release. Substance Use and Misuse 46(12): 1477–1485.

18.

Galloway

Forsyth

Shewan

(2007) Young People’s Street Drinking Behaviour: Investigating the Influence of Marketing and Subculture. Glasgow: Glasgow Centre for the Study of Violence.

19.

Gibbons-Wood

Tait

Morrison

(2010) An Analysis of Alcohol Use and Disorders amongst Those Detained in Custody in Aberdeen. Aberdeen: Robert Gordon University.

20.

Graham

Wells

West

(1997) A framework for applying explanations of alcohol-related aggression to naturally occurring aggressive behaviour. Contemporary Drug Problems 24(4): 625–666.

21.

Graham

Heller-Murphy

Aitken

McAuley

(2012) Screening of short-term/remand prisoners using the AUDIT screening tool. International Journal of Prisoner Healthcare 8(2): 51–59.

22.

Hamlyn

Brown

(2007) Partner Abuse in Scotland: Findings from the 2006 Scottish Crime and Victimisation Survey. Edinburgh: Scottish Government.

23.

Havinghurst

(1952) Human Development and Education. New York: Longmans, Green.

24.

Hopkins

Sparrow

(2006) Sobering up: Arrest referral and brief intervention for alcohol users in the custody suite. Criminology and Criminal Justice 6(4): 389–410.

25.

Information Services Division (2011) Alcohol Brief Interventions 2010/11. Edinburgh: NHS National Services Scotland.

26.

Kaner

Dickinson

Beyer

. (2007) Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews 2: Art. No: CD004184.

27.

Lapham

(2004/2005) Screening and brief intervention in the criminal justice system. Alcohol Research and Health 28(2): 85–93.

28.

Lapham

Smith

C’De Baca

(2001) Prevalence of psychiatric disorders among persons convicted of driving while impaired. Archives of General Psychiatry 58: 943–949.

29.

MacAskill

Parkes

Brooks

Graham

McAuley

Brown

(2011) Assessment of alcohol problems using AUDIT in a prison setting: More than an ‘aye or no’ question. BMC Public Health 1: 865.

30.

McGovern

Newbury-Birch

Deluca

Drummond

(2012) Alcohol Screening and Brief Intervention in Probation: SIPS CJS Factsheet. London: King’s College.

31.

McKinlay

Forsyth

Khan

(2009) Alcohol and Violence among Young Male Offenders in Scotland (1979–2009). Edinburgh: Scottish Prison Service.

32.

McSweeney

Webster

Turnbull

Duffy

(2009) Evidence-Based Practice? The National Probation Service’s Work with Alcohol-Misusing Offenders. London: Ministry of Justice Research Series 13/09.

33.

National Institute for Health and Clinical Excellence (NICE) (2010) Alcohol-Use Disorders: Preventing the Development of Hazardous and Harmful Drinking, NICE Public Health Guidance 24. London: NICE.

34.

Newbury-Birch

Bland

Cassidy

. (2009a) Screening and brief interventions for hazardous and harmful alcohol use in probation services: A cluster randomised controlled trial protocol. BMC Public Health 9: 418.

35.

Newbury-Birch

Harrison

Brown

Kaner

(2009b) Sloshed and sentenced: A prevalence study of alcohol use disorders among offenders in the north east of England. International Journal of Prisoner Health 5(4): 201–211.

36.

Parkes

MacAskill

Brooks

. (2011) Prison Health Needs Assessment for Alcohol Problems. Edinburgh: NHS Health Scotland.

37.

Raistrick

Heather

Godfrey

(2006) Review of the Effectiveness of Treatment for Alcohol Problems. London: National Treatment Agency for Substance Misuse.

38.

Reinert

Allen

(2007) The alcohol use disorders identification test: An update of research findings. Alcoholism: Clinical and Experimental Research 31: 185–199.

39.

Scottish Government (2009a) Changing Scotland’s Relationship with Alcohol: A Discussion Paper on Our Strategic Approach. Edinburgh: Scottish Government’

40.

Scottish Government (2009b) Changing Scotland’s Relationship with Alcohol: A Framework for Action. Edinburgh: Scottish Government.

41.

Scottish Government (2010) Criminal Justice Social Work Statistics. Edinburgh: Scottish Government.

42.

Scottish Government (2011) 2010/11 Scottish Crime and Justice Survey: Main Findings. Edinburgh: Scottish Government.

43.

Scottish Intercollegiate Guidelines Network (2004) The Management of Harmful Drinking and Alcohol Dependence in Primary Care: A National Clinical Guideline (Updated). Edinburgh: Scottish Intercollegiate Guidelines Network.

44.

Scottish Prison Service (2009) Prisoner Survey 2009. Edinburgh: Scottish Prison Service.

45.

Scottish Prison Service (2010) Young Offender Prisoner Survey 2010. Edinburgh: Scottish Prison Service.

46.

Skellington Orr

McCoard

Canning

McCartney

Williams

(2011) Delivering Alcohol Brief Interventions in the Community Justice Setting: Evaluation of a Pilot Project. Edinburgh: NHS Health Scotland.

47.

Stein

Lebeau

Colby

Barnett

Golembeske

Monti

(2011) Motivational interviewing for incarcerated adolescents: Effects of depressive symptoms on reducing alcohol and marijuana use after release. Journal of Studies on Alcohol and Drugs 72(3): 497–506.

48.

Stein

Caviness

Anderson

Hebert

Clarke

(2010) A brief alcohol intervention for hazardously drinking incarcerated women. Addiction 105(3): 466–475.

49.

World Health Organization (WHO) (2005) Alcohol and Interpersonal Violence Policy Briefing. Copenhagen: World Health Organization.

50.

World Health Organization (WHO) (2010) Preventing Intimate Partner and Sexual Violence against Women: Taking Action and Generating Evidence. Geneva: World Health Organization.