Alcohol Treatment Matrix cell A1: Interventions; Screening and brief intervention

2020/21 update funded by

Alcohol Change UK

Previously also funded by

Society for the Study of Addiction

Developed with

Alcohol Treatment Matrix

Includes brief interventions

Drug Treatment Matrix

Includes harm reduction

About the matrices

Articles, slides, and video

Comment/query

Matrix cell

Alcohol Treatment Matrix cell A1

Interventions; Screening and brief intervention

Seminal and key research on the effectiveness of screening for risky drinking followed by brief interventions for people screening positive. Discusses the implications of the major UK trials intended to inform government policy, questions how ‘real world’ trials have been, and how strong the evidence is for the UK. See the rest of row 1 of the matrix for more on screening and brief interventions.

S Seminal studies K Key studies R Reviews G Guidance more Search for more studies

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text

S First primary care study harbinger of later ‘no benefit’ findings (1987). Conducted in Dundee in 1985 and led by a researcher later involved in the crucial SIPS study listed below, this first primary care trial found similar drinking reductions whether or not screening to identify risky drinking was supplemented by a typical warning from the doctor, or a brief intervention featuring feedback on an assessment of the patient’s drinking, a self-help booklet, and further consultations – results strikingly similar to those from SIPS. For discussion click and scroll down to highlighted heading.

S Pioneering British studies question need for extended treatment (1999). Three studies from the 1970s and ’80s which showed alcohol problems could be reduced by brief interventions in alcohol clinics, hospitals and GPs’ surgeries, findings which challenged the need for the extended treatments of the time. For discussion click and scroll down to highlighted heading.

S Assessment plus feedback enough to reduce drinking (1988). Assessment (usually a screening questionnaire), feedback on assessment, and a motivational interviewing counselling style – these core components of a typical brief intervention originated in a trial of the offer of a check-up to people wondering if their drinking might be harmful.

K Whether conducted in GPs’ surgeries (2013), emergency departments (2014) or probation offices (2014), results from the SIPS trials were the same: two scientifically developed brief interventions were not shown to have been more effective than a much less expensive terse warning plus a leaflet, intended as a ‘control’ against which the brief interventions could shine – a surprise which prompted a ‘less is more’ interpretation of the findings. For discussions click here and here and scroll down to highlighted headings.

% drinking excessively in intervention and control groups

K Violent injury plus brief advice prompts young men to moderate drinking (2003). Relative to usual care, in Cardiff young men facially injured in a drunken altercation substantially cut their drinking after 20 minutes’ alcohol advice from a face-clinic nurse chart.

K Referral for counselling works for heavy-drinking emergency patients (2004; free source for original article at time of writing). Typically very heavy drinkers who saw their emergency as alcohol-related drank less after referral for brief counselling. Conducted in London, the first study to record benefits from an almost entirely routine procedure, including 30% fewer return visits over the following year. Similar results seen in France among another set of patients (all drunk on admission) patently in need of moderating their drinking. For discussion click and scroll down to highlighted heading.

K Booster phone call needed to make brief counselling effective in US trauma centres (2014). Serious alcohol-related injuries are common at US trauma centres, which compared to usual emergency departments offer a conducive environment for addressing patients’ drinking. Nevertheless, only if reinforced with a follow-up phone call was brief motivational counselling found significantly more effective than minimal advice.

K Brief counselling for hospital inpatients not found more effective than handing over a booklet (2007). Compared to usual care, in Scotland handing heavy-drinking patients a guide to sensible drinking reduced consumption about as much as roughly 20 minutes of FRAMES-based advice, seemingly demonstrating the impact of simply being identified by a clinician as a risky drinker who should consider cutting back. For related discussions click here and here, and scroll down to highlighted headings.

K Student population responds minimally to routine screening/intervention (2013). Compared to offering no screening and no intervention at all, this rare ‘real-world’ trial of a routine programme found that offering web-based screening for risky drinking plus a computerised brief intervention feeding back screening results led to slightly fewer (about 45% v. 48%) university students in Sweden scoring as risky drinkers. However, on no measure did supplementing screening with the offer of the brief intervention appreciably or significantly improve outcomes. For related discussion click and scroll down to highlighted heading.

K Widespread routine implementation at best marginally effective (2010). The US service for former military personnel is one of the few large health systems to have got close to universal screening and brief intervention, but this pilot study found only minor drinking reductions and others (1 2) could not demonstrate any reductions at all. Screening too missed most risky-drinking patients. For discussions click here and here and scroll down to highlighted headings.

K Nurse-led brief intervention not found more effective than standard care in ‘real-world’ English trial (2006). Rated as the most ‘real-world’ of the primary care trials included in an influential review listed below. Did not find that practice nurses trained in a structured brief intervention reduced drinking significantly more than those instructed to offer a leaflet plus standard advice to cut down. For discussion click and scroll down to highlighted heading.

R Relatively real-world brief primary care interventions just as effective (2018). Influential synthesis of research findings concluded that brief advice in GPs’ surgeries and emergency departments reduced risky drinking even in trials assessed as closest to routine practice – but how real-world were any of these trials, and why did the most recent studies indicate zero effect? Similar results from an earlier version were considered to strengthen NICE guidance (listed below) in favour of implementing brief alcohol interventions in primary care. For discussions click here and here and scroll down to highlighted headings.

R Face-to-face brief interventions modestly effective in a range of settings (2016). Randomised trials with results published in English record modest but statistically significant drinking reductions from brief interventions in primary care, universities, and (though halved in size) emergency departments; evidence was insufficient for inpatient wards and non-clinical settings. For discussion click and scroll down to highlighted heading.

R Brief interventions patchily effective among hospital inpatients (2011). Synthesis of international studies found some significant impacts but these were inconsistent, perhaps because merely being identified as a heavy drinker has an impact which brief interventions find hard to better. Our analysis found this patchy record applied also to UK studies. A later review (2013) found single sessions unconvincing but brief interventions with booster sessions more effective than usual care/no intervention. For discussion click and scroll down to highlighted heading.

R Advanced analysis casts doubt on effectiveness of brief interventions for college students (2015; free source at time of writing). A new way to synthesise findings from studies of brief motivational interventions for college students did not find that overall they affected the probability of drinking or the amount drunk when drinking occurred. Suggests previous syntheses found such effects partly because they did not adjust for the large number of non-drinkers and for differences between individuals.

R How GPs can identify risky drinkers (2014). In the British context studies suggests the best combination of accuracy and brevity is achieved by asking patients just two questions and further confirmatory questions only if patients screen positive initially.

G NICE calls for UK to invest in screening and brief intervention (National Institute for Health and Care Excellence, 2010). UK’s official health interventions advisory body recommends investing in widespread screening and brief intervention using the FRAMES approach as part of a public health ‘invest to save’ strategy. For related discussion click and scroll down to highlighted heading.

G Practical advice and information for health professionals in England on implementing national guidance on drinking (Public Health England, 2019). National health body responsible for supporting substance use work offers advice on responding to alcohol problems, including screening and brief interventions. The advice is tailored for practitioners, service managers, and for senior or strategic leaders involved in assessing local needs and commissioning services.

G UK screening and brief intervention implementation aids and guidance (accessed July 2019). Web site offers discussion, news and a portal to screening instruments and guides on how to advise patients.

more Search for all relevant Effectiveness Bank analyses or for sub-topics go to the subject search page. See also hot topics on brief interventions and computerising therapy and advice.

Close Matrix Bite guide and commentary Close Matrix Bite

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text

What is this cell about? In contrast to treatment, screening and brief interventions are usually seen as public health measures. Rather than narrowing in on dependent individuals or just those seeking help, the aim is to reduce alcohol-related harm across a whole population including those unaware of or unconcerned about their risky drinking. Screening programmes aim to spot drinkers at risk of or already experiencing alcohol-related harm while for some other purpose they come in contact with services whose primary remit is not substance use. In studies the typical response to those who seem at risk is from five minutes to half an hour of advice, counselling and/or information aiming to moderate their drinking or its consequences, delivered not by alcohol specialists, but by the worker the drinker came into contact with – the ‘brief intervention’. Click here Unfold supplementary text for more on what screening and brief interventions typically consist of.

The thinking is that by reaching many millions of low-risk drinkers, small risk reductions achieved by broadly implemented, resource-light interventions could contribute at least as much to improving public health as tackling the greater risks faced by the far fewer drinkers with more extreme and obvious problems.

For a population-level impact, a high proportion of risky-drinkers must be reached. In the UK, GPs’ surgeries are the principal venue, but programmes are also mounted in other medical settings such as emergency departments and sexual health clinics, on inpatient wards, at antenatal clinics, as well as in non-medical settings such as criminal justice, social care, community and housing services.

The effectiveness of structured, research-developed brief interventions has become accepted to the degree that they are now embedded in UK policy and even if imperfectly and patchily, also in practice. This cell redresses the balance by giving you reasons to reconsider that acceptance, whilst acknowledging that there have been positive findings and that studies of the kind available to us can only raise doubts over effectiveness, not definitively prove ineffectiveness. Even if they could, their findings are not an argument for failing to check on patients’ drinking (especially when this is potentially relevant to their health) and responding to risky drinking with usual medical care including individualised advice, information and follow-up, and where this seems warranted, with a more structured and extended response.

Where should I start? A good place is our hot topic on brief alcohol interventions listed above. Research findings, policy trends and practicality issues are drawn on to address the key question: whether screening and brief intervention really can be sufficiently effective and widely implemented to appreciably improve health across an entire population.

Among the studies considered are the English SIPS trials (listed above and discussed in the “Highlighted study” section below) intended to inform national policy, and evaluations listed above of the programme mounted by the US medical service for former military personnel – important as a rare example of a large health system achieving widespread routine screening and intervention. These US studies are also instanced below (click to highlight the relevant text) under the heading, “How real world were the ‘real-world’ trials?”

In the hot topic you will also learn about the scaling back of UK policy from universal to ‘targeted’ screening of new patients and/or those possibly at risk, and about the UK’s attempts to make screening and intervention actually happen on the intended scale.

Highlighted study For Britain the highlight has to be the SIPS study, reports from which are listed above. Funded by the Department of Health in 2006 to help determine national policy, SIPS mounted three trials of screening/brief intervention: in 29 general practices; in nine emergency departments; and in 20 probation offices. Importantly, the trials were intended to test real-world interventions delivered by the staff who would conduct them in routine practice. The results would help answer criticisms (see “How real world were the ‘real-world’ trials?” below) that positive results relied on to promote widespread screening and brief intervention programmes emerged from trials divorced from how these programmes would actually be implemented.

All three SIPS trials tested three interventions, each building on the other. Considered further below, most basic was a terse 30-second warning that the participant was risking their health through excessive drinking, plus an alcohol information and advice leaflet – a package which according to Britain’s National Institute for Health and Care Excellence would not count as a brief intervention at all. It was intended as a ‘control’ comparator against which to benchmark the benefits of adding more sophisticated and longer brief interventions. One of these interventions was a (very) brief intervention lasting just five minutes, which included comparing the participant’s drinking to typical drinking levels across the population. Best outcomes were expected from supplementing these shorter and less sophisticated interventions with 20 minutes of individualised ‘lifestyle counselling’ from staff equipped with motivational interviewing and health behaviour-change skills.

That expectation was comprehensively contradicted, a lesson in how the randomised trial can defy what seems obvious on common sense and scientific grounds. In the event, the simple control comparator turned out to be the best option, gaining what clinical benefits there were at the lowest direct cost. There was little or no indication that adding brief interventions gained anything additional. The implications of this surprising result are explored below.

Those were the critical findings, questioning national policy and accepted wisdom about the value of brief interventions, but the study also generated other important findings. Among these were that implementation often required specialist support and patient/offender throughput was low, suggesting there will be difficulty in reaching a large proportion of the population. Though these findings were never formally published, it seems there were no great differences in how well the three different screening methods tested in the trial identified risky drinkers (1 2).

In addition, the primary care trial compared ‘targeted screening’ only of patients whose medical complaints suggest excessive drinking (currently favoured in England) against trying to screen everyone. Though general practices allocated to the targeting strategy started with fewer patients eligible for screening (1274 v. 1717), they ended up netting more patients who screened positive (461 v. 439) because targeting was more likely to reserve screening for patients who actually were risky drinkers. However, over a quarter of the patients who did not meet the targeting criteria turned out to score as risky drinkers; in a targeted programme, their drinking would remain unaddressed. A more intuitively expected outcome was seen in Sweden, where some primary care practices were randomly allocated to screen all adult attendees for risky drinking (the results of which were passed to their clinician before the consultation), while in other practices clinicians were encouraged to ask about drinking whenever they found this relevant due to possibly alcohol-related medical or social problems – similar to the targeted screening in the English SIPS study. Universal screening led to twice as many risky drinkers being talked to about their drinking; without this, the issue tended to be addressed only among heavier drinkers. For the authors it seemed likely that ‘if relevant’ screening misses risky drinkers “because the signs are either difficult to recognize during the consultation or turn up late in the course of developing problems”.

Issues to consider and discuss

Do results from the SIPS trials mean do just the minimum? From these important English trials (described in the “Highlighted study” section above), that seems likely to be the message grasped by austerity-hit commissioners and hard-pressed practitioners faced with the finding that the simple warning of the control script was just as effective as a fully-fledged brief intervention. In this they would be encouraged by the “less is more” take on the findings from the Department of Health’s Director of Health and Wellbeing.

However understandably, this would be to over-read the study’s implications. Whatever benefits there were came after patients or offenders had been quizzed by research staff about their drinking and related problems and their readiness to do something about these, possibly thought-provoking interventions in themselves. And while what the interventions were intended to be is clear, what was actually done is not. Faced with someone for whom they had responsibility who they suspected was risking their health, did nurses, doctors and probation officers allocated to the control script really confine themselves to five sentences and a leaflet, and did the recipients accept this without probing further?

The minimal control intervention in the SIPS trials

Experts have stressed that the findings do not mean handing over an alcohol advice leaflet is all it takes. Screening plus the script ( panel) of SIPS’s control intervention incorporated assessment, strong feedback on that assessment, an implicit call to action to stop “excessive” drinking above “safe”, “recommended” and “sensible” levels, as well as a reminder in the form of the leaflet. There is little evidence that anything longer or more sophisticated than this (such as motivational interviewing) has much if any greater effects (1 2 3 4 5 6 7), and some from a study listed above that among hospital patients, basic intervention along these lines can be better than screening only.

If that tempts you to campaign for SIPS’s most basic option, remember that without a no-advice comparator, there is no way of knowing whether any of the SIPS interventions had any effects at all, including the control script (1 2). Neither can it be said that the trials proved longer and more sophisticated interventions worthless, only that they were not proved more worthy than the basic option, perhaps because a fluke of sampling meant their advantages did not emerge in these particular trials. The unsatisfactory upshot is that SIPS did not show any of the interventions were worth doing, nor that any were not worth doing. Usefully, what it did do was throw a cat among the pigeons, upsetting expectations and causing us to think again – to develop new expectations and ways of testing them.

In the end these niceties have seemed to matter little. Some will be convinced the trials have shown that for brief interventions, very small and very basic really is beautiful, while UK guidance and policy (described in the hot topic listed above) and expert opinion continue to advocate the more extended interventions the SIPS trials failed to vindicate. Consider which side of this fence you are on and why. Are research findings the main reason, or gut feelings that ‘brushing off’ a heavy drinker with a few sentences just can’t be right, or perhaps ‘common sense’ considerations, experience, caution about abandoning what may yet prove to be valuable interventions, cost considerations … or something else? How big is the role played by research in your decision?

How real world were the ‘real-world’ trials? Here we address what the British researcher seen as “a key architect of contemporary thinking on brief interventions” has called “the most pressing challenge” facing the sector – whether research findings will survive transplantation of programmes from the unrealistically controlled context of a clinical trial to more ‘messy’ and less controlled routine practice.

For screening and brief intervention this issue is particularly pertinent because their essence is that they are implemented in services and by staff who see their ‘real’ business as something else, must balance competing demands, and who know the patient is not there to talk about their drinking. Many such staff and services will simply not join studies, leaving a few to imperfectly represent the rest. These few may benefit from unrealistically high quality and extensive training, monitoring and supervision, and be aided by research staff who do the screening which informs the feedback element of brief interventions. Sometimes research staff also deliver interventions and follow-up patients for booster sessions. Even if evaluation results are promising, they may relate to a programme which will never be implemented beyond the context of a clinical trial.

First published in 2007, an attempt to address this issue divided primary care and emergency department trials in to those which more versus less closely approximated how brief interventions would be conducted in practice. Finding no outcome differences between the two sets of trials, the analysts argued this meant the findings were “relevant to the real world of clinical practice” rather than an artefact of the more rarefied context of a randomised trial. Conducted for the renowned Cochrane Collaboration, the review had considerable authority, and its findings were considered to strengthen NICE guidance (listed above) in favour of implementing brief alcohol interventions in primary care.

Eleven years later and again for the Cochrane Collaboration, the review was updated (the version listed above) and the attempt to assess real-world applicability repeated. Overall the estimate of the effectiveness of brief interventions in reducing drinking had nearly halved but remained statistically significant, and again the classification of trials as more versus less ‘real-world’ suggested the findings would apply to routine practice. But around the same time another comprehensive review of primary care and emergency department brief interventions covering essentially the same studies saw things differently: “… it seems that some of the benefits of [brief interventions] could be lost when translated from a special research condition to the natural conditions of typical clinical practice. Indeed, a common criticism of [brief intervention] trials is that they are efficacy studies (optimizing internal validity) rather than pragmatic trials.”

How can we reconcile these differing interpretations? The answer is that while the Cochrane reviews divided trials into more versus less real world, the whole set of trials were clustered towards the non-real world end of the spectrum. For the Cochrane reviews, ‘real-worldness’ applied mainly to the brief intervention phase of the trial, the phase they were seeking to evaluate. But before this came the selection of services, of the sometimes very few patients at those services screened and willing to participate, and the crucial screening process without which interventions could not have targeted appropriate patients, which often supplied data for use in the interventions, and which was often done not by the service’s usual staff, but by research assistants. A close look at the trials included in the earlier review reveals that few if any categorised as relatively real-world tested anything close to what a widely and routinely implemented programme would look like, a gap which carried over to the updated review; unfold Unfold supplementary text the supplementary text for examples and further methodological considerations.

Close supplementary text

Conducted in the UK, what according to the Cochrane reviews was the most real-world of its set of trials (it is listed above) recruited only a quarter of the GP practices it approached (many said they had no time) and just over 1 in 10 contributed data to the analysis. This selectivity means results cannot be assumed to represent what would happen in a normal GP practice less motivated, or less well placed, to join and complete a brief intervention trial, or in a programme which did successfully go beyond enthusiast practices to achieve the widespread implementation needed to realise public health gains. In the event, the trial did not find that a structured brief intervention was more effective than standard advice.

Closely behind in the real-world ratings was an Australian study of screening and brief intervention in an emergency department. Its real-worldness included a modest training programme for department staff, who conducted both screening and brief intervention, and the criteria for excluding patients from the trial, which were simply those expected in routine implementation. The trial was also real-world in its inability to persuade staff to screen more than a minority of patients despite “sustained effort over more than 1 year to encourage universal screening”. There were large differences in screening rates – one clinician screening 700 attendees, others, none – perhaps affected by the research requirement to obtain consent from patients to participate in the trial if they screened as risky drinkers. What this means is that even if the brief intervention had been found to work, we could not be confident that a successful effort to widely implement screening and brief intervention in emergency departments would have similar effects. Like the trial described in the previous paragraph, in the event the trial did not show that a brief intervention was more effective than usual care.

Among other features distancing studies from widespread, routine practice are that often research staff screen patients, providing information which then feeds into the intervention. In their absence, this load would have to be taken up by probably less well prepared and less motivated practice staff. Also, the latest Cochrane review notes that of the 69 trials it uncovered, 56 were or might have been at high risk of bias due to loss of patients to follow-up. As well as possibly biasing outcomes within the patients in the trial, this means results cannot be assumed to apply across an entire population of risky drinkers screened and advised by clinicians, but only to those for some reason relatively easy for researchers to follow up. Perhaps most seriously impeding generalisation of results to a whole population is the fact that often very few of the possibly eligible participants are screened and join the studies. Not only does the process of sifting potential participants impede generalisation, but it can undermine the reliability of the findings even for those who did join the trials and were followed up.

Close supplementary text

On this issue of real-worldness the original Cochrane review became the site of a head-to-head between the leading UK analyst referred to above – one of the authors of the review – and an uncompromising US critic. Read their freely available argument and counter-argument. The critic argued that the review’s criteria for a study close to routine practice were like saying a typical car is one which has wheels and an engine and can be steered by any qualified driver. On this basis, all but a few cars up to and including a Formula 1 racer would be ‘real world’, because this is the essence of what a car is. We would need to go further to be able to realise that the performance of a racing car driven by a professional racing driver is not really a guide to the typical performance of more prosaic vehicles and drivers. So too, he argued, the essence of a brief intervention is such that the review’s criteria would judge even an unrealistically tightly controlled trial to be relatively typical of routine practice. Studies truly close to real-world conditions have, he says, failed to find brief interventions work – witness the English SIPS trials discussed above in the “Highlighted study” section.

His parting shot was that “If the current literature were truly replete with real-world … effectiveness studies, then we would likely see [alcohol screening and brief intervention] being widely disseminated (it is not), and implemented successfully. Yet in one of the only large health systems with widespread … implementation, the benefit … is undetectable.” He was referring to the US medical service for former military personnel, evaluations of which (listed above) have found little or no significant drinking reductions from routine brief interventions. A similar result was found when a national brief intervention programme for university students was evaluated (study listed above) in Sweden.

Do these findings mean real-world brief intervention programmes are inherently ineffective and not worth further study? Or is it just that they are difficult to do well, but with further study and refining of implementation and intervention, might yet be found worthwhile? That possibility cannot be denied (after all, the findings from the US service for former military personnel have been described as “early” results from a programme finding its feet) and the potential prize in terms of inexpensively reducing alcohol-related harm without having to impose availability restrictions is huge. Worth pursuing – or should we cut our losses and stop chasing a losing strategy?

Even if they were effective, are they now? Yet more fundamental issues are raised by the review discussed above – not just the real-worldness of primary care trials, but whether real-world or not, more recent trials have demonstrated an effect on drinking, and whether any such effects translate into reduced alcohol-related harm. On these issues the evidence is either weak or fading as it accumulates.

Differences in the way harm was measured in the 20 studies which reported on alcohol-related harm meant the review was unable to amalgamate the findings. In turn this meant it could not say whether small drinking reductions associated with brief interventions have any clinical significance in terms of preventing harm, ultimately the outcome targeted by these programmes. Instead the review tallied results from the individual trials: 16 of the 20 did not register significant effects on alcohol-related problems/harms, and two others found only fleeting reductions – not definitively negative, but not reassuring either.

However, there is a reasonable expectation that less drinking will mean less harm across an entire population, even if small-scale, short-term studies are unable to detect this to a statistically significant degree. But the latest Cochrane review cast doubt on drinking reductions themselves. With its finding of an overall 38 g a week reduction in alcohol consumption, the earlier version had provided an estimate of the effectiveness of brief interventions which fed into cost-effectiveness calculations influencing national policy. Collecting its evidence 11 years later, by the time of the latest review (listed above) that estimate had nearly halved to 20 g a week.

Over the years findings of the impact of brief interventions on drinking a year later steadily diminished until by 2014 they averaged near zero

Part of the reason was a significant tendency for studies published most recently to record lesser effects. In fact, by 2014 the ‘best fit’ graph of this tendency suggested studies were on average finding zero effect chart. A number of credible explanations were given for this tendency, but it does mean that contemporary practitioners should have much less confidence than their predecessors that diverting the scarce resource of time to formal brief intervention programmes in primary care settings is worthwhile – and primary care (in this review taken to include emergency departments) is the pre-eminent setting for these programmes.

In 2017 such considerations plus doubt over the applicability of trials to routine practice led a US and UK expert to see (free source at time of writing) “good reason to question whether brief interventions work in routine [general] practice”. They queried on what basis practitioners faced with patients who typically have several other problems part from heavy drinking would choose to focus the few minutes available in a consultation to focus on drinking rather than, say, diet, exercise, or smoking – or, we might add, the reasons why the patient came to the surgery in the first place.

It is important to understand what these findings may mean – that among the general run of primary care patients, formal, structured brief interventions of the kind devised for research studies are now on average possibly no more effective after screening than usual or more minimal responses. They are not an argument for failing to check on patients’ drinking (especially when this is potentially relevant to their health) and responding to risky drinking with usual medical care including individualised advice, information and follow-up, and where this seems warranted, with a more structured and extended response. And just as studies can trend towards an ‘ineffective’ verdict, in future that trend may turn the opposite way.

What about the UK? Two of this cell’s seminal studies (listed above: 1 2) show Britain has a strong claim to being the home of brief alcohol interventions and of their evaluation. What does the record show about effectiveness in the UK context?

Conducted in Dundee, the first primary care trial (listed above) was unable to demonstrate that what seems to have been a well thought-out and extended brief intervention had any greater effect than screening and research assessments only, or these plus a simple warning to cut back. Bringing us more or less up to date, the results were reminiscent of the 2009 SIPS trials in English primary care, emergency and probation settings. Discussed above in the “Highlighted study” section, these offered no reassurance that structured brief interventions improved on a simple warning. What of the intervening studies, especially those which (unlike SIPS) could tell whether intervening at all was worthwhile because they featured a no-intervention control group?

For primary care trials look at the background notes to our analysis of a Danish real-world trial which did not find significant drinking reductions. The most convincing British study (signposted “Wallace 1988” on page five of the notes) dated from the 1980s and tested its intervention in relatively ideal conditions with highly (self-)selected patients. Of 4,454 whose drinking had been excessive or caused them concern, just 748 ended up supplying data for the 12-month follow-up. This possibly particularly intervention-friendly minority did reduce their drinking more if they had been advised by their GPs and offered a follow-up session than if left to decide for themselves whether to seek advice. Screening was done by research staff, and the practices in the study may not have been typical, all having opted to join an existing research network. Another UK trial (Anderson 1992, p. 6–7) which found drinking reductions was similarly limited in its applicability to usual practice. Other British trials were either not reflective of primary care, or inconclusive about the benefits of intervention.

Next take a look at our detailed account of the British hospital inpatient studies featured in a review listed above. Again, most studies did not find no significant extra drinking reductions attributable to brief interventions, amongst which was one of this cell’s key studies (listed above). Two studies did find significant drinking reductions, but in one it was unclear whether the brief intervention was the cause, and in the other, nearly half the excessive drinkers identified by research staff refused to join the study, a third who did join could not be followed up, and on most measures no significant impacts were recorded. After this review another UK study conducted in London also found no evidence that a brief intervention had reduced drinking or related problems among patients admitted to general hospital wards. A randomised trial, it was a relatively rigorous test which (as some studies have not) included female as well as male patients, and excluded the alcohol-dependent patients thought not to be susceptible to brief advice.

Lastly, for emergency departments there is simply not enough quality evidence. Apart from the negative SIPS trial listed above, just one further UK trial qualified for inclusion in a recent review listed above. Also listed above, it did find drinking reductions due to routine referral for brief alcohol counselling, but involved a department in London with an unusually strong commitment to addressing drinking and with established procedures to train and motivate staff, and just a third of patients identified as potentially risky drinkers contributed data to the 12-month follow-up.

Do these studies convince you that routine implementation of screening and brief intervention in the UK is likely to appreciably curtail risky drinking and improve population health? That rather non-routine implementations can modestly reduce drinking is not in question. In the UK as elsewhere, it comes down to that “pressing challenge” for the sector discussed above – whether the most promising of research findings will survive transplantation of screening and brief intervention programmes from clinical trials to routine practice.

Thanks for their comments on this entry in draft to Nick Heather, Emeritus Professor of Alcohol and Other Drug Studies at Northumbria University in England. Commentators bear no responsibility for the text including the interpretations and any remaining errors.

Close Matrix Bite guide and commentary Close Matrix Bite