McGovern R., Newbury-Birch D., Deluca P. et al.
Institute of Psychiatry, King's College London, 2012.
Unable to obtain a copy by clicking title? Try this alternative source.
The probation arm of the largest alcohol screening and brief intervention study yet conducted in Britain found that the proportion of offenders drinking at risky levels fell just as much after the most minimal of screening and intervention methods as after more sophisticated and longer (but still brief) alternatives.
Summary This account has been superseded by an account based primarily on a later formally published report. It is retained here for archival purposes only. Please go to this analysis for the latest report.
This account is based on preliminary findings released by the SIPS project in the form of factsheets and conference presentations rather than peer-reviewed publication in academic journals. Later more detailed and scientifically formal accounts of these and other findings (such as what was actually done by the interventionists and what patients thought of it) will be incorporated as they emerge, and mailing list subscribers will be alerted to any alterations in the findings or their implications. Some of the documents to which links are provided may no longer be available on the SIPS project web site.
The project was funded by the UK Department of Health in 2006 to evaluate the effectiveness and cost effectiveness of different ways of identifying risky drinkers through routine screening, and different forms of brief advice to help them cut back. Other aims were to assess the feasibility of implementing such procedures in typical practice settings, and to discover what made these more or less likely to succeed.
Conducted in three English regions (London, the South East, and the North East), the project took the form of three randomised controlled trials in different types of settings: nine emergency departments; 29 GP surgeries; and 20 probation offices. After summarising common features across the three trials, this account focuses on the probation study, relying largely on a factsheet produced before formal publication of the findings.
All three trials involved random allocation of practices, departments or offender managers to different variants of screening and intervention. Staff seeing adult patients or offenders for usual purposes in these settings asked them to consent to screening and basic data collection. Those who screened positive were further asked to join the study of the interventions, usually The exception was the most intensive of the interventions in emergency departments and probation offices, intended to be reserved for specialist alcohol workers. to be delivered by the same staff after training by the study. To assess changes in their drinking and related issues, patients and offenders who were eligible for Across all three trials patients or clients already in treatment for drink problems were excluded as were those severely injured, suffering with a serious mental health problem, or who were grossly intoxicated or homeless. and agreed to participate in the intervention study were followed up six and 12 months later.
Three quick ways to identify risky drinkers were tested for feasibility and accuracy, the latter defined by how well they duplicated corresponding results from the AUDIT screening questionnaire, widely used to determine whether someone is probably drinking at hazardous, Scoring at least 8, the cut-off used by the study. harmful or possibly dependent levels.
Single question: The simplest and quickest method was to ask, "How often do you have eight (or for women, six) or more
Each drink is roughly a UK unit of 8gm alcohol.
on one occasion?" Monthly or more was considered a positive screen, meaning the respondent would be offered a brief intervention to help them cut back.
FAST Alcohol Screening Test: As used in the study, this begins with the question above and registers a positive screen if the response is weekly or more often. Otherwise three further questions About how often in the last six months the respondent has been unable to remember what happened during the previous night's drinking, failed due to drink to do what was normally expected of them, or experienced concern over their drinking from a relative, friend, or health professional. are asked. Scores in response to the four questions are summed So that, for example, people who have (even if in all cases less than monthly) drunk excessively, and forgotten what happened and failed to meet obligations would screen positive, as would someone who said they had never drunk excessively yet had either forgotten or failed to meet obligations at least weekly, or experienced concern on more than one occasion. to determine whether to proceed with intervention.
Paddington Alcohol Test (PAT): Used only in the emergency department study.
Patients or offenders identified as risky drinkers by these methods were all offered advice of some kind, so the study could not assess the absolute impact of this advice, only how the impacts of one variant differed from those of another. The main yardsticks These figures were adjusted for any significant differences between patients or offenders on characteristics the study found related to how likely they were to end up drinking safely. were the proportions of patients or offenders who six and 12 months later did not score as hazardous (or worse) drinkers on the AUDIT questionnaire, which assesses alcohol intake and other indicators of harmful or dependent drinking. Other assessments included drink-related problems, quality of life, and use of services. Crime and health service costs before the study and over the 12-month follow-up were used to assess cost effectiveness in terms of gains in quality-adjusted years of life per £ change in costs to society.
All the patients and offenders in the intervention trial were given a standard The UK Department of Health's How much is too much? booklet, was the standard official public information available at the beginning the SIPS project. alcohol information and advice booklet, supplemented by a sticker with contact information for local alcohol treatment services. At issue was whether also offering different types and degrees of advice would make a difference to later drinking.Brief feedback: At its most basic, the booklet was accompanied only by very brief feedback from the health care or criminal justice practitioner who did the screening that the screening test had indicated the patient or offender was drinking "above safe levels, which may be harmful to you".
In 2007 in the UK there were about 950,000 incidents of alcohol-related violence, and drink is a factor in nearly half of all violent crimes. In Britain hazardous and harmful drinkers outnumber dependent drinkers 7:1. The greatest population-wide impact on alcohol-related problems can be made by identifying and intervening with these drinkers, even before they are aware of any problems or seek help. One proven way to do this is to screen for risky drinking and then offer brief advice to people who screen positive. However, questions remain about the best screening strategy, whether longer or more sophisticated interventions work better, impacts in normal practice, and cost effectiveness, all questions addressed by the overall design of the SIPs trial. One further gap is the lack of evidence on the impact of screening and brief interventions within the criminal justice system, a gap specifically addressed by the probation arm of the SIPS study.
Probation was chosen as the criminal justice setting on the basis of a pilot study of this setting plus prison and police stations. It found that offenders seeing probation officers were most likely to agree to join such a study, and that this setting offered the highest study recruitment rate. This pilot also found that offenders in these settings were three times as likely as the general population to be problem drinkers.
Across the North East, London and South East regions, 20 probation offices Selected to have no existing programme of alcohol screening and brief intervention. and 197 offender managers working in those offices agreed to join the study. Within each region, offender managers were randomly allocated to one of the six possible combinations of two screening methods (the single question and the FAST Alcohol Screening Test) to be applied to all eligible adult offenders, and the three interventions for those screening positive.
A slight methodological variation was that staff were first to give offenders the alcohol advice leaflet to read. At the next appointment, consent was sought for screening, screening was completed, and, for positive-screen offenders only, consent sought for the intervention phases of the study, brief feedback or advice given, and for offenders allocated to this, an appointment made to see the alcohol health worker for counselling.
Despite staff enthusiasm, barriers to implementation cited by staff included workload pressures, lack of knowledge, and lack of follow-up treatment services. Compared to staff in the other two settings (primary care and emergency departments), screening and brief intervention was felt to meld more naturally with routine probation work, but staff were less convinced these procedures would be useful and tended to feel they were best reserved for offenders with obvious drinking problems. Of the 197 staff in the trial, 44 did not recruit any offenders to the study, and just 45 were able to implement screening and brief intervention as intended without extra help from researchers and the specialist alcohol workers. Implementation was more successful where research staff were able to engage with and provide ongoing support to individual staff and where they and the alcohol health workers were more often 'on site'.
In the end, over 16 months, 976 offenders were approached about the study of whom 860 were eligible to participate. Of these, 574 screened positive and 525 agreed to join the intervention study. Typically white men, they averaged 31 years of age and an AUDIT score at the threshold for a high severity Scores of 16 or more. Scores of 8–15 indicate medium severity and lower scores, non-hazardous drinking. of drinking problems, a range which accounted for 43%. Around two thirds were followed up six months later and 59% at 12 months.
In terms of identifying people who screened positive for risky drinking on the AUDIT, the FAST Alcohol Screening Test was preferable (92% were identified) to the single question (81%), and significantly better at identifying people whose AUDIT scores indicated a high severity of alcohol problems. The results confirm the (not statistically significant) trends in favour of FAST in the preceding pilot study in prisons and police stations as well as probation.
Positive screen offenders were then allocated to different forms of intervention. Virtually all allocated to brief feedback or advice received this plus the alcohol advice booklet, the full intended interventions. This was not the case for those allocated to lifestyle counselling; though nearly all received the five-minute brief advice and booklet delivered immediately after screening, only 41% attended a later appointment for more extended counselling.
Six and 12 months later the proportions of offenders scoring as at least hazardous drinkers on the AUDIT questionnaire (initially 82–90%) had fallen overall by 11% and nearly 16% respectively, but neither on this measure nor on alcohol-related problems or health-related quality of life had there been significantly greater changes after one type of intervention than another, and the offenders were equally satisfied with all the options. The expected extra impacts of more extensive advice and counselling had not materialised; at 12 months the reduction in the proportion of risky drinkers was 19% after the briefest option, 4–5% more than after the other two. However, at the six-month follow-up only, the high severity drinkers among the sample were significantly more likely (according to their AUDIT scores) to be drinking safely if they had been allocated to counselling rather than just brief feedback. Across the entire sample, police records also revealed that brief feedback offenders were significantly more likely to be reconvicted (50% v. 36–38%) than offenders offered either of the more extended interventions.
Another difference was in costs, averaging £1.04 per offender for the brief feedback option, £8.55 for brief advice, and £32.45 for lifestyle counselling. These costs were however overshadowed by the costs of the patients' health service use and crime over the 12 months of the follow-up. Changes in these costs from the period immediately before the offender joined the study meant that the most extensive option (the offer of counselling) saved society £2760 more than brief feedback. At the 12-month follow-up, all three groups registered slight declines in their average quality of life. Valuing each of these years at £20,000, there was a 98% probability that the briefest intervention was more cost effective than brief advice and 78% that it bettered the offer of more extensive counselling.
Though possible, due to workload pressures, implementing alcohol screening and brief intervention in probation will be difficult. Successful implementation is associated with promotion by local managerial champions. Even then, successful and sustained implementation requires sustained and significant support from specialist alcohol workers. Particularly difficult to implement was the lifestyle counselling intervention, which required appointments to be made and kept rather than integrating screening and briefer interventions in to an existing probation supervision meeting.
When it came to how to respond to these risky drinkers, the more extensive interventions offered no significant clinical benefits, except temporarily at the six-month follow-up, when particularly high severity drinkers had responded better to counselling than brief feedback only. On average all the options were followed by reductions in the severity of drinking, results which may have been due to the interventions, but may instead have been due to natural changes in relatively extreme behaviour, or to the impact of being repeatedly assessed for drinking and recruited to a trial of drinking interventions.
commentary See these Findings analyses for the sister studies conducted in emergency departments and GP surgeries. The following commentary explores common themes across these settings and any differences, and supplements these with comments focused on the featured setting, probation. The general picture was that implementation often required specialist support, and there were no great differences between how well the screening methods identified patients and no significant differences between how well the interventions helped them reduce the severity of their drinking. What was intended to be a 'control' condition against which scientifically developed and longer interventions could shine, turned out instead to be the better option, reaping what clinical benefits there were at the lowest direct cost in money and time.
Seeing the effectiveness of brief interventions as established in principle, the studies aimed to assess whether they would also work in normal practice. First issue was the feasibility of implementing such programmes with training, support and incentives of the kind that might routinely be available. In each setting, the intention was that usual staff would undertake screening and intervention, except for the longest intervention of the three, lifestyle counselling. In probation and emergency departments, this was delegated to a specialist alcohol worker provided by the SIPS project, an extra resource which mirrors how such programmes would probably be (and in emergency departments, commonly have been) implemented in routine practice. The project also undertook training, though for the briefer interventions this was minimal. Apart from research tasks, enough to enable screening involving an understanding of the 'standard drink' used by the study to assess alcohol consumption, and an hour on how to deliver the brief advice option. For these interventions too, no structured ongoing support and supervision is mentioned, except "Research staff and trainers will maintain regular contact with practices throughout the study period, including site visits and telephone support." http://dx.doi.org/10.1186/1471-2458-9-287 for the primary care study, though researchers and alcohol health workers may have been available to offer ad hoc support.
One possibly important way the studies departed from normal practice was that usual staff also undertook the research tasks involved in recruiting patients to the trial and collecting baseline information. Compared to brief screening and intervention, this must have been a relatively substantial extra burden, one which may have suppressed the numbers screened Little pre-screening data collection was required, but staff would have known that a positive screen would demand more explanation of the intervention study, obtaining consent and data collection. and offered intervention.
Broadly, each study found that while implementing the tested programmes was possible, at many sites researchers and specialist alcohol workers who had trained the staff had to help with screening and intervention. Workload pressures, lack of knowledge, and feeling there were insufficient back-up alcohol services, were common themes. In emergency departments and in probation, inability to implement was the norm. Incentivized with per patient payments, most primary care practices managed to implement fully, but still 4 in 10 were unable to do so. While the denominators in terms of overall patient and offender throughput are unknown, the numbers screened seem to have been small, equivalent to about 12 per emergency department per week, less than two per GP practice per week, and one or two a fortnight in each probation office – and this despite the intention that half or more of the sites would screen nearly all the adults they saw who were capable of participating in the trials.
These findings have two possible implications. The first is to cast doubt over the potential for screening and intervention in these settings – as implemented and resourced in the trials – to make a significant contribution not just to the welfare of the individuals actually screened, but to the nation's health; numbers reached may simply be too small. Reinforcing this doubt was the uncertainty over the resultant impacts on those who were screened and advised (of which more below). Second is the possibility that those recruited to the trials and screened were not representative of all who might have been, and therefore too the possibility that how they reacted would not be duplicated in a national programme with the leverage to ensure widespread implementation.
In probation in particular, perhaps because risky drinking was so common (two thirds of offenders screened positive compared to a third of primary care patients), and because their interest was not in chronic long-term disease but in intoxication-related crime, probation officers were interested mainly in the most problematic drinkers, and it was these they tended to identify more than in the other settings. For more on screening, brief intervention and other alcohol-related work by probation in England and Wales see this Findings analysis, suggesting that while in theory widespread, screening is not the norm, and that on-site specialist alcohol workers are an important resource.
In relation to screening, results from the trials have been amalgamated in a conference presentation. Of the three methods tested, the FAST Alcohol Screening Test had the broadest applicability, in all three settings virtually equalling or bettering the alternatives in terms of its ability to identify risky drinkers. Generally only the first (about frequency of excessive drinking) of the four questions had to be asked, and the test picked up 8 in 10 of the risky drinkers who would have been picked up by the longer AUDIT questionnaire.
In probation the advantage of the FAST test was clearer than in the other settings, perhaps because it asked questions about symptoms of the kind associated with intoxication-related impairment and lack of responsibility, but FAST may not meet the perceived need to identify high severity drinkers, for which AUDIT (the most commonly used screening tool in probation) may still be preferred. Whether screening is best implemented universally or targeted at certain offenders was not tested in the probation setting, but the prevalence of risky drinking was such that universal screening seems the most sensible option.
The final link examined by the studies was how best to advise risky drinkers identified through screening. Once patients and offenders had been sorted in to risky drinkers who had agreed to join the intervention study, there was a remarkable uniformity in trends in their drinking. Six months later the proportions scoring as risky drinkers had fallen by 11%, 12 months later, by 16–17%. With one exception, on this, the primary yardstick used by the studies, an alcohol advice booklet plus a sentence or two of feedback alerting someone to their risky drinking was not improved on by adding more extended and individualised interventions.
The exception was in the featured probation study. At six months and among particularly heavy drinking offenders offered counselling, this study recorded a fleeting extra reduction in the proportion still drinking at risky levels. Given the many tests of significance made in the studies, this single finding may have breached the threshold of statistical significance purely by chance, but it seems possible that the on average higher severity of drink problems among the offenders meant that enough to register a statistically significant effect responded better to extended advice than to minimal feedback, feeding in to fewer reconvictions, and possibly too accounting for the relative reduction in health and crime costs associated with counselling. Even if this was the case among offenders, findings among the patients gave no grounds for triaging heavier drinkers in emergency or primary care settings to more extended brief advice.
As the researchers acknowledged, the general findings do not mean the interventions were equally effective; they may have been equally ineffective. Without a no-intervention comparator, there is no way of knowing whether the interventions played any hand in the outcomes. Even before the interventions, 15–20% The figure is not available for probation. of emergency patients and nearly a third in primary care said they were trying to reduce their drinking. Apart from the possible reasons for the drinking reductions mentioned by the researchers, this in itself could account for the findings.
It cannot even be said that screening plus a sentence of feedback is all it takes to get whatever benefits are available. These came after patients and offenders were quizzed about their drinking and related problems and their readiness to do something about these, possibly thought-provoking interventions in themselves. Also, while what was intended in the interventions is clear, what was actually done is not. In particular, it seems reasonable to question whether brief feedback interactions really ended abruptly after a doctor, nurse or probation officer, had warned the person for whom they had welfare responsibilities that their drinking risked harm – that the recipients of this news did not respond and staff in turn respond back, in what could have become an interchange rivalling in length and perhaps exceeding in individualisation the brief advice option.
Reinforcing doubts over the impact of the interventions is the general finding that control A group of people, households, organisations, communities or other units who do not participate in the intervention(s) being evaluated. Instead, they receive no intervention or none relevant to the outcomes being assessed, carry on as usual, or receive an alternative intervention (for the latter the term comparison group may be preferable). Outcome measures taken from the controls form the benchmark against which changes in the intervention group(s) are compared to determine whether the intervention had an impact and whether this is statistically significant. Comparability between control and intervention groups is essential. Normally this is best achieved by randomly allocating research participants to the different groups. Alternatives include sequentially selecting participants for one then the other group(s), or deliberately selecting similar set of participants for each group. groups in alcohol brief intervention studies who received no or minimal intervention on average reduced their drinking by amounts comparable to those seen in the SIPs trials. Though the review which collated these findings did not single these out, the studies which offered only usual care to control groups often also registered such reductions.
The clearest difference between the interventions was in cost, likely to be persuasive given equivocal or no evidence that spending more gained more. Not only did this directly cost least, but on the health service's primary yardstick – quality adjusted life years – in both probation and primary care, the briefest intervention gained most years for each £ of social costs incurred by the drinkers. Only in emergency departments did the longest intervention have the edge, but this was minimal, and may have been partly due to these patients starting the study with the lowest quality of the three intervention groups and catching up somewhat in a natural levelling up.
However, the featured probation study gives greater grounds in this setting for offering extended advice, especially among higher severity drinkers. Here the crime yardstick is more salient and though fleeting, the extra reduction in reconvictions at six months and the (possibly associated) lower social costs among high-severity offenders offered counselling may tip the balance. There is probably also a greater social expectation on staff working with convicted offenders to deliver more than a one-sentence warning, when failure to address drinking appropriately could result in serious crime.
All quality of life calculations are partly dependent on how quality is measured. SIPS used a health-related measure, ill equipped to capture losses or gains in the quality of social and leisure life, major domains within which drinking plays a role and is seen by consumers to have value (for which they are prepared to pay), just as excessive drinking can cause damage. Discounting such possible benefits of substance use as judged by consumers also makes a substantial difference to cost-benefit calculations.
The UK alcohol strategy published in 2012 said government was awaiting the results of the SIPS project before deciding whether to incorporate alcohol screening and brief intervention in to the national quality framework for primary care, a major national driver of primary care practice. Already, however, brief alcohol interventions are among the practices commissioners can incentivize through cash rewards, and from April 2013 this work will be incorporated in the NHS Health Check for older adults. The strategy also encouraged accident and emergency departments and hospitals in general to check for and offer brief advice about hazardous drinking, in the case of hospitals by employing alcohol liaison nurses.
In emergency departments and GP surgeries, the SIPS findings appear to contradict guidance recommending the equivalent of the mid-level intervention, or for heavier drinkers the most extended option. If anything, the findings offer most support for screening and brief feedback. Similar guidance has been disseminated to criminal justice services. However, in these settings, given the mixed findings from the SIPS probation trial on the relative impacts on drinking, quality of life, and crime and social costs, at the moment the findings seem to offer no convincing rationale for altering this advice in favour of a yet briefer option.
Last revised 19 May 2012. First uploaded