Seminal and key studies on how organisational functioning affects screening and brief intervention. Highlights a striking illustration of the importance of organisational context emerging from the unprecedented implementation drive at the US health care system for ex-military personnel. 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 London emergency department pioneers systematic screening and brief advice (1996). Innovations which led to a 10-fold increase in the proportion of patients advised about their drinking included developing and making available a very quick screening test tailored to the setting, allowing doctors to screen only categories of patients most likely to be drinking heavily, and employing a specialist alcohol worker to shoulder most of the intervention load. For discussion and scroll down to highlighted heading.
K Different contexts require different approaches in England (2016). To download scroll down the linked page to the fourth publication listed. Discussions with housing, probation and social work practitioners indicated that standard screening methods and structured brief advice are unlikely to be implemented in many non-health settings, due largely to the practice context demanding more flexible and ‘conversational’ approaches. For discussion and scroll down to highlighted heading.
K Standard screening and brief intervention inappropriate in range of working contexts in Scotland (2015). Move beyond the usual medical settings for screening and brief intervention and in Scotland, even after training seven out of nine practitioners did not engage in these activities at all because they felt them inappropriate for their clients. Message is that implementation efforts must accept the possible downsides of more judgement-driven, non-standardised approaches, including possible loss of efficacy. For discussion and scroll down to highlighted heading.
K Abandon researched packages in favour of a ‘conversation’ about drinking, say midwives in Scotland (2019). Fitting screening and brief interventions into the working context of midwives in Scotland meant abandoning the standard, scripted approaches tested in research in favour of a more individualised and nuanced conversation – and then not initially about current drinking, but the less threatening topic of pre-pregnancy consumption. For discussion and scroll down to highlighted heading.
K ‘Advise this patient’ reminders little use without organisational backing (2010). Contrasting this study with another from the same US primary care system for ex-military personnel shows that screening/intervention rates can hinge on how the organisation handles implementation procedures in general. For discussions click and and scroll down to highlighted headings.
K Implementation at US primary care clinics ranges from zero to near universal (2005). Implementation rates at clinics offered training and support depended on complex provider and organisational characteristics. These differed for the screening and intervention phases of the programme and contributed to a 0–95% range in the proportion of risky drinkers offered a brief intervention. For discussions click and scroll down to highlighted heading, and , scroll down to highlighted heading and unfold the supplementary text.
K Positive organisational climate fosters widespread screening and brief intervention (2013). From Brazilian primary care clinics comes a rare confirmation that a positive organisational climate is associated with overcoming barriers to widely implementing screening and brief intervention programmes. For discussion , scroll down to highlighted heading and unfold the supplementary text.
R Strategies to implement alcohol screening and brief intervention in primary care (2011). Provides a useful map of a large and complex territory enabling you to identify which implementation levers you are already pulling and which you might also turn your hand to. For discussions click , and , and scroll down to highlighted headings.
R Barriers and facilitators to implementing alcohol screening and brief intervention (2011). UK-focused review for Britain’s National Institute for Health and Care Excellence.
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 aims 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 score in at-risk zones 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 for more on typically studied screening and brief intervention activities.
This cell is not, however, about the content of these interventions (for which see cell A1), but whether implementation and impact are affected by the manner and the degree to which the organisation responsible for the programme supports managers, supervisors and staff, and offers an environment conducive to screening and brief intervention.
Though not commonly researched, such issues are crucial. Screening and brief intervention are often implemented by services where harm arising from ‘normal’, non-dependent drinking, and making public health gains, are not naturally seen as ‘our business’. Without a conducive organisational context which makes screening and brief intervention part of the business, they are likely to be sidelined and/or of poor quality. The not inconsiderable task is to make these activities happen not as discretionary add-ons, but as a routine way to find and respond to risky drinking – even when there are no obvious signs of excessive drinking, even if ‘risky’ means low-level risk, and even if alcohol is not why the drinker made contact nor the primary mission of the service.
Where should I start? A map of a large and complex territory showing its extent and identifying continents and sub-continents is always helpful. That’s what is provided by a comprehensive review listed above of strategies to foster implementation. It ranges from the micro-level of designing an effective and suitable intervention, out to adapting the economic, political, and social environment. In the middle, and from US experience (see section below) very important, are the so-called “Inner setting” influences – features of the implementing organisation including the degree to which its structures, communication mechanisms, resources, leadership, and culture facilitate the adoption of innovations, and the degree to which the innovation ‘fits’ its needs and circumstances.
You could use the map provide by this review to locate the focus of your own implementation efforts, as a checklist of the factors which might be obstructing or promoting those efforts, or to identify where else levers might be pulled to aid implementation. In the remainder of this commentary we return to this map several times, an indicator of its utility.
Highlighted study More than any other health provider yet documented, the US Veterans Affairs or ‘VA’ health care system for ex-military personnel has made a determined effort to implement routine brief alcohol counselling in its primary care clinics. Its efforts were found uniquely successful by our starting point review listed above, partly because the VA marshalled the organisational influences described under the previous heading.
Within this uniquely successful context, listed above our highlighted study reveals a telling failure. It showed that electronic reminders to advise risky drinkers were usually ignored at a clinic where there were no active implementation efforts, little leadership encouragement to use reminders of any kind, no culture of routinely responding to clinical reminders, and no incentives for their use or for brief alcohol interventions. The reminders were used for just 15% of patients who screened positive for risky drinking and just 6% were offered brief counselling. Instead, staff gravitated towards advising abstinence to a few very heavy drinkers – not the public health role envisaged for brief interventions.
At other VA clinics, the story was very different. Though again the reminder system was implemented without provider training or support, at these clinics the culture was that reminders were to be responded to and their use was routine. When the alcohol counselling reminder was inserted into this culture, its use also became routine, being documented for 71% of positive-screen patients, and there was no indication that intervention was reserved for heavier drinkers.
Taking these two studies together, the authors’ warning that results from the more successful implementation might not be replicated “where clinical reminder use is not routine” seems well founded – a striking illustration of the difference organisational context can make. For confirmation from elsewhere, take a look at these non-VA studies.
One very wet dampener must be thrown over the VA’s implementation successes: there is very little evidence that these helped curb drinking – perhaps a reflection on the quality of screening and intervention. But of course, unless they are implemented, there is no chance that interventions of this kind can work, and no chance that they can be made to work by improving quality.
Are there ‘good’ reasons for ‘bad’ implementation? It has become accepted that implementation – how well interventions root when transplanted from research to routine-practice – is the weak link in screening and brief intervention. But in any such situation, there are at least two ways to interpret a low level of implementation. First, that this truly is a ‘failure’ – that even the organisation concerned will acknowledge that it should have implemented better, but some deficit prevented this happening.
Then there are those times when what looks like a ‘failure’ from the outside is not seen that way from within, but rather as a justifiable de-emphasis on an intervention which does not meet the needs of clients, or not as well as other interventions which could be mounted with the same resources.
In this context it is relevant to return to the caustic observations of the Lancet medical journal cited in cell C1’s bite. Their editorial complained that “lecturing” patients takes up time in the short primary care consultation which could have been used to more cost-effectively improve health than a detour to drinking or other lifestyle issues.
Note also that the review listed above and discussed in the “Where should I start?” section, says one implementation determinant is whether the innovation ‘fits’ the organisation’s needs and circumstances. Perhaps this is why an inner-London emergency department led the way (listed above) in Britain, while in the USA, it was (studies listed above) the health care system for mainly male former military personnel. But what if there is no ‘fit’? Could it be that sometimes an appreciation of the needs of your population, good leadership, the maintenance of a trusting relationship with clients, overriding priorities, will actually mean rejecting routine alcohol screening and brief intervention programmes, and for very good reasons indicative of organisational quality rather than the lack of it? Can you conceive of any such circumstances in medical, welfare and offender supervision services of the kind targeted for such programmes? For possible examples turn to the section below headed and unfold the supplementary text. But before accepting these reasons for rejecting standardised screening and brief intervention, play devil’s advocate and consider whether really they are less reasons and more excuses for not moving beyond the practitioner’s and the organisation’s comfort zones.
Is what it takes to successfully implement screening the same for brief advice? We talk of ‘screening and brief intervention’ as a package – and justifiably so, because often they are done seamlessly by the same person. Even so, could the organisational influences which affect screening differ from those which affect brief interventions? In primary care settings, the review (listed above) we started with found evidence this was the case. It could make a stab at identifying the characteristics shared by services which achieved widespread screening. Among these were requiring documentation of whether screening had been done, and making the doing of it matter to the organisation and to the staff – in the examples it cited, through a mandatory entry in electronic medical records used to hold organisation and staff accountable. In contrast, what promoted widespread brief intervention was unclear. The reviewers reasoned this was because screening can be a quasi-‘mechanical’ procedure completed by patients filling in a questionnaire, or by non-medical or junior staff following a set script, while whether then to offer advice and of what kind involves judgement and individualised assessment. (In passing it should be pointed out that while this may be the case in the primary care settings on which the review focused, screening can be a highly sensitive issue requiring tact and judgement in other settings; for examples turn to the and unfold the supplementary text.)
The review was unable to incorporate the findings of the major test of screening and brief intervention implementation strategies in Europe – the five-country ODHIN trial. A striking finding was how much the different components of the strategies affected the proportion of patients screened, and how relatively little the proportion then advised about their drinking. Once faced with evidence of risky drinking, even before the implementation strategies were applied, generally the GPs, nurses, and other clinicians in the study felt they should respond and did so; to do otherwise would be contrary to their duties of care and place them in the position of having ignored health-threatening behaviour. But very rarely did they actively and systematically go looking for evidence of risky drinking to respond to. Implementation strategies had more scope to raise the screening rate, and most successful among these was paying for each patient screened.
Another way to approach this question is to examine organisational dimensions associated with screening versus brief intervention rates. If what promotes screening differs from what promotes brief interventions, we would expect to see organisational dimensions associated with one but not the other, or at least, not so strongly – just what was found in a US study listed above of the impacts of training and continuing support in primary care clinics.
For our purposes the key measures were the proportion of patients screened (screening rate), and the proportion who received advice after screening positive for risky drinking (intervention rate). There were factors strongly associated with both, such as time pressures on practitioners and support from the top, but also notable differences. Four particularly salient organisational factors were significantly associated with either screening or intervention rates, but not both. Of these, three were significantly related to the intervention rate, but not to screening: a negative link with competing organisational priorities, and positive links with assistance from the research team’s expert, and with the degree to which clinic staff were able to change relevant operating procedures. Only one – involvement of clinic staff in planning – was related (positively) to screening but not to intervention.
Compared to screening, it seemed that achieving high rates of intervention was more dependent on technical and managerial support and on the scope afforded to staff to change how the organisation works. Only a randomised study or the equivalent could test whether these truly were causal factors which differentially affected screening and intervention – but deliberately creating poorly functioning health services in order to test this proposition is unlikely to be acceptable.
From this US study as we have told the story so far there is of course an important missing chapter – the one recounting whether implementation efforts actually reduced drinking. To finish the story, unfold the supplementary text.
Horses for courses? Though treatment is essentially individualised, the fundamentals of the encounter are set – generally a patient who wants help in the form of treatment seeking that help from a service geared to providing it. From the start it is accepted that the patient has a substance use problem and both see engaging in treatment as their business and what they are there for; at this deep level, the ‘rules of the game’ are clear and mutually agreed.
In contrast, in research studies screening and brief interventions are often highly de-individualised, yet the contexts in which these activities take place lack the shared foundations found in treatment. Instead, someone is intercepted in the course of engaging with a service for a reason nothing to do with drinking, and the service itself has a different primary agenda. Different working contexts demand different approaches, partly because neither party wants to undermine their primary objectives by baldly raising drinking, spending time on this issue when others may have priority, or admitting to risky consumption if the consequences of doing so might threaten other aims. The organisational context demands adjustments, meaning ‘ideal’ interventions which may have proved themselves in research studies are often substantially abandoned.
A trio of UK studies listed above testify to this conundrum. To get addressing alcohol into the service mix requires departing from the research script, meaning the ‘intervention’ that’s left has at best not yet been shown to be effective, and at worst, may be ineffective or counterproductive. Based on interviews and discussions with practitioners and managers, the results of all three studies are freely available and accessibly written (1 listed above; 2 listed above; 3 listed above). In general the findings reveal that outside primary care in housing, probation, social work, antenatal care, and sexual health clinics, standard brief interventions are felt inappropriate to the caseload and standardised screening approaches a threat to the client-worker relationship unlikely to elicit honest answers. In practice, what looks like a simple set of questions followed if the scoring indicates by brief advice which could be rolled out by any modestly trained practitioner, becomes in these settings far more complicated. Though such adaptations risk compromising the effectiveness of research-validated screening and intervention packages, commonly there was a strong feeling that less formal and more flexible approaches to drinking are likely to work best and fit best with current practice.
Points relevant to this cell from these and other studies are detailed in the supplementary text. For more see Alcohol Change UK’s web page dedicated to reports it funded on broadening the base for alcohol screening and brief intervention out from health to non-health contexts such as housing, probation and social work.
This draft entry is currently subject to consultation and correction.