Key studies on the impact of the practitioner in brief interventions. Highlights Swiss studies which dissected how these work and helps develop evidence-informed understanding of four issues: Why does the practitioner matter? Are some naturally effective? Does getting it wrong matter more than getting it right? What do we know about non-motivational interventions? 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 Confrontation provokes resistance (1993). US study which seemed to validate motivational interviewing’s empathic, client-centred style rather than the more confrontational approach typical of the era. When therapists fed back the results of an alcohol-harm check-up in a motivational style, heavy drinkers who had responded to ads were less resistant to change and later made greater cuts in their drinking. For discussion click and scroll down to highlighted heading.
K Best to stick to the script? (2012; free source at time of writing). US researchers developed a scale to measure fidelity to a commonly researched form of brief intervention. It worked well at assessing practitioners’ competence in an emergency department study – but scores on the scale were not found to be related to how much patients drank at follow-up, showing that ‘competence’ of this kind does not always make a brief intervention more effective. For discussion click and scroll down to highlighted heading.
K Nurses or doctors: who is most effective? (1997). Relatively ‘pure’ test from Australia of whether impacts are greater when a GP does the brief intervention as opposed to a practice nurse. No significant differences, but the nurses had the edge. However, neither’s two-session intervention was found more effective than much briefer and simpler advice. Related reviews below: 1 2. For related discussion click and scroll down to highlighted heading.
K Emergency patients respond to motivational style (2009). Rather than topping the league in making motivational interviewing’s recommended types of comments, practitioners’ impacts on risky drinkers at emergency departments in Switzerland depended on their ability to embody the client-centred spirit of the approach and to avoid comments inconsistent with this ethos. For discussions click here, here and here, and scroll down to highlighted headings.
K Reflective listening key to provoking intention to change (2010). Micro-analysis of interactions between heavy-drinking Swiss army conscripts and counsellors found reflective listening (when the counsellor signified their attention and understanding by selectively echoing back the young men’s feelings and comments) the key to provoking signs of an intention to reduce drinking. Counsellor comments inconsistent with motivational interviewing provoked the opposite indications. Similar findings below suggest drinking will actually have been reduced. For discussion click and scroll down to highlighted heading.
K Reflective listening related to later drinking (2014). Study of Swiss army conscripts (also participants in study above) exposed the influence of the counsellor’s proficiency in motivational interviewing on subsequent drinking reductions, especially the degree to which they reflected back the client’s remarks enriched by an additional or deeper meaning. In contrast, piling on lots of the other types of recommended comments was not associated with reductions, and even a single interjection inconsistent with the approach seemed to nullify the session’s impact. Experience seemed to count, but only if it led to proficiency, while whether the counsellor was a doctor or a psychologist was not associated with later drinking. For discussions click here and here and scroll down to highlighted headings.
K “Even one behavior inconsistent with [motivational interviewing] can be particularly damaging” (2014; free source at time of writing). Giving the examples of confrontation or unsolicited advice, that was the conclusion of a study which re-analysed session recordings from three trials (1, free source at time of writing; 2; 3, free source at time of writing) of brief interventions based on motivational interviewing. In these studies where counsellors were generally proficient in the approach, another “surprise” (free source at time of writing) was how little being yet more proficient, qualities such as acceptance and empathy, or the strength of their relationship with the client, affected subsequent drinking. Two of the trials involved Swiss army conscripts also involved in studies above (1 2). For discussion click and scroll down to highlighted heading.
R Doctors or nurses? (2016). Found that brief alcohol interventions in primary care are not significantly less effective if conducted by nurses or other primary care staff instead of doctors, and using non-physician staff means a given amount of funding would be able to pay for more interventions for more patients. Related study above and review below. For related discussion click and scroll down to highlighted heading.
R Nurses may be slightly more effective than doctors (2016). Review not confined to primary care generally did not find that impacts on drinking of brief alcohol interventions differed when conducted by interventionists from different professions, but nurses had the edge over doctors and other categories. Related study and review above. For related discussion click and scroll down to highlighted heading.
R Barriers and facilitators to implementing screening and brief intervention (2011). UK-focused review for Britain’s National Institute for Health and Care Excellence concentrating on the views of professionals and patients and the implementation process; includes the influence of staff characteristics and attitudes on whether they implement screening and brief interventions. For related discussion click and scroll down to highlighted heading.
R Implementation strategies in primary care (2011). Investigates the international literature on implementing alcohol screening and brief interventions in primary care, including the influence of staff characteristics.
more Search for all relevant Effectiveness Bank analyses or for sub-topics go to the subject search page.
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 however not about the content of the intervention (for which see cell A1), but about whether its impact depends on the interpersonal style and other features of the person doing the advising – much less commonly researched. In fully-fledged psychotherapy it is well established (see cell B4) that the approach of the person doing the therapy and their ability to forge therapeutic relationships are important influences on outcomes – but is the same true in what are often the fleeting, one-off encounters of brief interventions?
There are reasons to believe this might be the case. Compared to treatment which people often patently need and have sought or at least accepted, in brief interventions there is an added dimension. For a population-level impact, a high proportion of risky-drinkers must be reached. Since by definition the impetus to engage in screening and intervention does not come from the risky drinker, it must come from the practitioner, who needs to have the opportunity, rationale and motivation to prioritise screening and to carve out the time to advise risky drinkers – even if this is not their main role or their or the drinker’s priority. A briefing based on a conference held in England in 2015 needed no convincing: “ultimately quality [alcohol identification and brief advice] comes down to the beliefs, motivations and skills of the practitioner”. They were not alone: in 2016 south London’s Health Innovation Network asserted that, “Central to quality assurance and improvement of the [alcohol identification and brief advice] pathway are the frontline staff delivering the service,” and in particular “the relationship between the individual and the staff member delivering [identification and brief advice]”.
Where should I start? No better place than our analysis of a the seminal study listed above led by William Miller, originator of motivational interviewing – an approach first tested as a brief intervention and still the dominant inspiration behind these when they extend beyond simple advice. Importantly, motivational interviewing is an interactional style, not a programme, one some people are more adept at learning and implementing than others (for which see cell B4). Rather than confronting directly, it nudges/leads the drinker to themselves find reasons to cut back – an approach likely to be more acceptable to people not actually seeking help over their drinking.
The results of Miller’s study turned the spotlight on how practitioners actually behave, not how they are meant to. Compared to a confrontational style typical of the day, patients allocated to what was meant to be a motivational style did better, but only marginally – perhaps because despite the researchers’ intentions, in practice the differences in the counselling styles (the same practitioners delivered both) were slight. When drinking was related to how practitioners actually behaved, the results were more clear-cut: the more they had confronted, the more the client drank a year later.
Read our analysis and you will see that while the link between confrontation and drinking was clear, the study could not determine what caused what. Instead of confrontation causing more drinking, perhaps patients who drank more led exasperated practitioners to become more confrontational. But within the context of other research, it seems probable that a real effect had emerged – a harbinger of later findings ( below) on the negative impacts of even infrequent confrontation or other counsellor behaviours which fail to concede to the client the right to decide whether, what, and how they want to change.
Despite its limitations, this study highlighted the importance of how the practitioner relates to the client. What would now be considered one of those limitations may have been critical to the emergence of this finding. Unlike many later studies, the practitioners had no detailed, manualised script to follow and (as they would in normal practice) their approaches varied considerably. Though they were trained to a level of competence in the intended approaches, no mention is made of the monitoring and supervision later studies used to ensure therapists stay on track. In contrast, modern studies want to know exactly what they are studying, so commonly tightly control the intervention, minimising what would normally be an important influence – the characteristics of the counsellor. Published over 20 years later and conducted on a different continent, in a different setting, and with a different caseload, a Swiss study (highlighted below) which shared this methodological ‘weakness’ seemed to confirm these early findings.
Highlighted study Capitalising on the fact that at age 20 Swiss men are assessed for fitness to be conscripted into the army, a series of studies has uniquely dissected how alcohol brief interventions work with a representative set of young European men whose screening results indicated risky drinking. One of the studies (listed above) set out to reveal the impact of the counsellors rather than the intervention by deliberately recruiting 18 counsellors who differed widely in professional status, clinical experience, and experience of the motivational interviewing approach used in the brief intervention. In other respects it duplicated the conditions of the seminal study discussed above rather than more tightly controlled studies which seek to eliminate variation due to the practitioners. Bringing it closer to normal practice, after initial training counsellors conducted the 20–30-minute session without manuals or further guidance or supervision. Ratings of audio-recorded sessions showed that left to their own widely differing devices, counsellors delivered interventions which also varied widely, though on average of good quality and modestly effective in reducing drinking.
When drinking outcomes were linked to several indicators of the experience and confidence of the counsellors or the quality of the interventions, three months later only conscripts who had seen counsellors or experienced sessions in the top half of the indicators’ ranges drank less relative to a control group offered screening and assessment only. Best results came from the more experienced counsellors, those who were more confident of their motivational interviewing abilities or of the efficacy of the approach, or who were rated as especially proficient. On average, it could not be shown that counsellors or sessions scored in the bottom half of these dimensions were any better at reducing drinking risky than no counselling at all.
That much is perhaps expected, but the details in this study were more thought-provoking. Experience was important, but only because it was associated with better motivational interviewing skills, an amalgam of demonstrating acceptance of and empathy with the client and embodying the collaborative spirit of the approach. Simply reflecting back the back the client’s feelings or comments seemed less important than so-called ‘complex reflections’ – the times when the counsellor reflected back, but with a spin which extended or deepened meaning beyond that expressed by the client. When these formed a relatively small portion of all reflections, the brief intervention made no difference to drinking; when a larger portion, drinking was reduced – the kind of finding which in 2014 led reviewers to highlight “the key importance of competent reflective listening skills (ie, the use of more complex reflections)”.
Complex reflections are an advanced competence, involving an ability to delve underneath a comment or behaviour to a deeper meaning, one which the client might see as a persuasive reason to moderate drinking. In surprising contrast, simply accreting more of the other responses considered compatible with motivational interviewing actually seemed counterproductive. These sorts of comments were common, occurring on average once or twice every minute of the sessions. When sessions were evenly divided into those in which they were more or less common, only sessions in the bottom half were significantly associated with reduced drinking – sessions supposedly less characteristic of proficient counselling. Sessions in the top half – packed with more of what was supposed to make motivational counselling work – were followed by reductions no greater than no intervention at all.
The other side of the equation was counsellor comments incompatible with motivational interviewing. These were very uncommon – usually one or none per session – but when they happened, that session was no more effective at moderating drinking than no counselling at all.
What was happening in these and similar studies and what are the practice implications? Below under the heading, “Is it what you don’t do that matters?”, you are invited reflect on these unexpected findings.
Are you surprised that in such brief encounters the practitioner matters? Or would you expect sensitivity to be critical when someone unexpectedly starts talking about your drinking when you came to them for another reason entirely? Turn to our analysis of the Swiss emergency department study listed above. There we point out that the dynamics of interacting with someone intercepted through screening are likely to be very different from those involving patients at substance use treatment clinics, who have usually already acknowledged their problems and resolved to do something about them. Brief interventionists have to generate this resolve in people who were not even thinking about their drinking, are suffering no noticeable consequences, and do not see it as something they need to change; doing that well seems in some ways a trickier skill than engaging treatment clients.
In particular, in brief interventions based on motivational interviewing it seems critical (see section below) to avoid confrontation, direction and those other behaviours which treat the patient as less than equal. In contrast, when it comes to initiating treatment, being told by an expert what is wrong with them and how they can get better may be just what a patient is looking for. Given this context, the kinds of behaviours proscribed in motivational interviewing are not necessarily destructive, but appropriate to a relationship implicitly invited by the patient. But when drinking is raised ‘out of the blue’ in the form of screening and brief intervention, on this health issue there is no implicit permission for the practitioner to ‘play doctor’, and the risky drinker has not implicitly adopted the patient role.
Check for yourself: is the evidence for the impact of the practitioner as strong for fully-fledged treatment (compare this cell with cell B4) as it is for brief interventions?
Are some practitioners naturally effective? Look at our analysis of the Swiss emergency department study listed above. Qualifications, experience and training were equalised, yet still counsellors varied widely in effectiveness. A counsellor who often made responses and comments of the type recommended in motivational interviewing ended up with the worst drinking outcomes – an average 18 UK units more per week – while another who emitted these responses and comments less often averaged a nine-unit reduction. The more effective counsellor’s advantage seemed to arise from better embodying motivational interviewing’s accepting, non-judgemental spirit, and more completely avoiding comments antithetical to this ethos. Check the background notes to that analysis and you will see that variability in the outcomes achieved by brief interventionists is not uncommon. What makes the difference is, it seems, not necessarily the approach itself, nor the practitioner’s strengths and weaknesses, but the combination of the two; some practitioners who do well with a more directive approach do badly when they attempt the client-centred style of motivational interviewing.
What accounts for this variation in practitioners’ effectiveness? Our highlighted study discussed above teaches us that in itself experience may not be important – what matters are the skills and confidence which can (but not always) come with it. Years of experience was associated with better motivational interviewing skills, but sessions during which counsellors had been rated as relatively skilled were followed by reduced drinking, regardless of the counsellor’s experience. Presumably, less experienced counsellors could nevertheless either be ‘naturally’ skilled in motivational interviewing, or far enough along that road to have developed these skills without the need for extensive experience.
From this study, from another listed above with a different set of professions, and from reviews of brief alcohol interventions in general (listed above) and in primary care (listed above), we can gather that the impacts of brief interventions do not depend on the seniority of the interventionist in their profession nor on whether they are a doctor, nurse or psychologist. (Though in European primary care, non-physician staff like nurses conduct screening seven or eight times more often than GPs – quality and impact may not differ, but quantity certainly does.) Nor it seems (document listed above) does sticking to the intended brief intervention programme necessarily maximise effectiveness. Personal qualities matter more – and that may include basics like sharing the the patient’s native language and culture. The message about the primacy of personal qualities extends also to treatment based on motivational interviewing, the basis for many brief interventions; cell C2’s bite cites studies which suggest that recruiting the ‘right’ clinicians who have not been trained in motivational interviewing would be better than choosing the ‘wrong’ ones who have been.
Is it what you don’t do that matters? Here we address a possibly dispiriting finding: that doing lots of the ‘right’ things in a brief motivational intervention matters little, while just one lapse to the ‘wrong’ sort of comment can be destructive.
Run your eye down the list of seminal and key studies above and you will see that many testify to the negative power of doing (in client-centred counselling terms) the ‘wrong’ thing. We saw it right from the start in the seminal study’s findings (discussed above) on confrontation – those times when the counsellor imposes their perspective or will on the client. According to a manual for assessing motivational interviewing skills, “These are the expert-like responses that have a particular negative-parent quality, an uneven power relationship accompanied by disapproval, disagreement, or negativity. There is a sense of ‘expert override’ of what the client says.”
Confrontation is one of those counsellor behaviours considered incompatible with the client-centred core of motivational interviewing. What they share is the non-collaborative stance of someone who knows best and is therefore in a position to confront, warn, direct, or advise the drinker. Across the studies listed in this cell, these behaviours were consistently related to weaker intentions to change drinking and lesser reductions in drinking itself.
Of most relevance to real-world practice is the latest of the studies of Swiss army conscripts discussed in the “Highlighted study” section. The authors warned that “even one behavior inconsistent with [motivational interviewing], such as unsolicited advising or confrontation, can be particularly damaging”, while “the quality and the exact combination of skills matters more than the quantity ... using a high number of open questions and simple reflections without eventually showing support or in-depth understanding might not be sufficient”. In this study it seemed that frequent interjections by the counsellor which conveyed support, affirmation, reflected back the client’s comments, and so on, were fine, but when these became very, very frequent, something was happening to make the session ineffective.
Look at the other studies. Findings were very similar in a Swiss study listed above of emergency patients. Move down the list to another study of Swiss army conscripts, and we gain insight into the mechanisms which generate desired outcomes. Reminiscent of the findings of our highlighted study discussed above, ‘complex reflections’ adding a deeper or alternative meaning to the client’s comments topped the list of counsellor behaviours which provoked ‘change talk’ indicative of an intention to drink less. In contrast, more of the other motivational-type comments such as open questions, simple reflections, and affirming and supporting, were like the proverbial water off a duck’s back, not moving the conversation in any particular direction. That was a surprise, but another finding should by now come as no surprise – that non-motivational comments like confrontation and directing or advising the client, counterproductively prompted comments indicative of an intention not to change one’s drinking. Finally, a joint re-analysis listed above of two Swiss army conscript trials and a US trial on medical wards confirmed the negative impact of these kinds of comments.
What do you make of these findings? Can they be dismissed as not cause and effect or even a reverse causality, the more ‘difficult’ clients who were always going to carry on drinking eliciting non-motivational responses from exasperated practitioners? What was going on in sessions when practitioners made very, very frequent – but not deeply probing – motivational-type responses? Why did these fail to generate change and sometimes seem to obstruct it? Were the counsellors energetically doing what their training had mandated, without really paying attention to and trying to understand the individual before them? Thoughtful, attentive attempts at understanding are signified above all by complex reflections, which most consistently among motivational interviewing’s repertoire were associated with positive outcomes.
Here we seem to have a picture forming of any number of perhaps superficially positive or affirming remarks (the equivalent of ‘Have a nice day’ wishes) failing to connect with clients, while just a few which showed the counsellor really is trying to understand make a positive impact, and just a few which demonstrated the counsellor was not on and by the client’s side, but pursuing their own agenda from a superior position, scuppered the session.
While their physiological responses were measured, 90 participants carried out a series of arithmetical tasks designed to be stressful. They were randomly assigned to receive validating, invalidating, or no feedback from the experimenter during these tasks. For example, if the participant said they found the task stressful, the experimenter might say, “I don’t understand why you found it stressful – it’s just adding and subtracting numbers” (an invalidating response), or “That’s understandable – lots of people have said they found this task stressful” (a validating response). Compared to no such feedback, invalidation had a detrimental impact, provoking “significantly higher psychological arousal, significantly lower perceived safety ratings, higher ratings of negative mood, and less willingness to take part in the study again”. In contrast, validating responses made little difference.
Bear in mind that this would be no peculiarity of brief interventions, but probably (research is lacking) a feature of medical practice in general. In 2015 a duo of UK-based authors writing in the American Journal of Medicine advanced the argument that in interactions between patients and clinicians, “bad is more powerful than good”. Specifically, their thesis was that communications which ‘invalidate’ the patient by failing to accept or understand their perspective damage the patient–clinician relationship and lead to feelings of hopelessness, anger and being dismissed and disbelieved (these had actually been recorded in real consultations at a pain management clinic), which in turn can be detrimental to the patient’s treatment and to their health. One of the authors had conducted a more definitive study (but not in a medical context) which seemed to prove the impact of invalidation and the relative non-impact of validation panel.
Even an apparently empathic response can be invalidating if it fails to chime with the patient’s perspective, as might be expected from seminal therapist Carl Rogers’ stress on accurate empathy. However, accuracy can’t be guaranteed, and authoritative advice is at least not to impose what may be a damaging understanding. Instead, “Empathy should always be offered with humility and held lightly, ready to be corrected.” Though particularly sharp for people in distress, concerned over their health, and with a right to expect validation, such responses are not limited to medical practice: “It is part of being human to want to feel understood by others, and our general life experience tells us that harsh words and criticism can hurt and have lasting negative effects.”
Are we now in a better position to understand why in brief interventions sometimes just one remark confronting or directing the client, or issuing uncalled-for warning or advice, has been associated with nil or counterproductive impacts on drinking? Does that make sense to you from your work or personal life?
Non-motivational interventions – the elephant in the room Nearly everything known about how brief interventions work relates to those based on motivational interviewing. Yet this is not typical of the briefer interventions which usually include feedback on how the patient’s drinking compares to the norm, advice to cut down, encouragement to set a date to do so, and perhaps a few simple hints on how this might be achieved, often informed by the FRAMES principles. Direct advice of the kind avoided in motivational interviewing has been identified as the essence of alcohol brief interventions, yet “the lack of study of the effects of direct advice is striking”. That such interventions are worth investigating is suggested by the findings of the English SIPS study detailed in cell A1’s bite. In this key study, longer interventions which drew on motivational interviewing were not found more effective than relatively unsophisticated and brief advice in the form of the study’s ‘control’ script panel right.
Little is known of how these kinds of interventions work, or whether the interventionist and how they behave makes a difference. Perhaps variation in outcomes due to the interventionist applies only to interventions which are harder to acquire and practice. Or perhaps the interventionist is just as important when the intervention is simple advice, but in different ways.
One thing we do know – that the staff concerned are a strong influence (see review listed above and these notes) on whether screening and brief interventions get done at all. Though partly determined by the organisational and policy contexts within which they work, their views on the legitimacy of this activity and its effectiveness and acceptability to patients, and the priority they give to it, all seem to matter. Perhaps these factors also influence their effectiveness via the impressions given of the importance of the issue and their advice, their belief in what they are doing, authenticity, and persistence in the face of an unenthusiastic reception from the patient, all conveyed not just in words but perhaps also in tone of voice; unfold the supplementary text for an early example.