Key studies on the ‘common factors’ underlying psychosocial therapies for problem drinking and the effectiveness of specific approaches. Explores the famous ‘Dodo bird’ hypothesis that all bona fide therapies are equivalent, examines the legacy of the UK’s most ambitious treatment trial, asks whether therapy can really make things worse, and questions how research amalgamating impacts from many patients can be applied to the treatment of an individual in their individual circumstances.
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). Among heavy drinkers in the USA, motivational interviewing’s non-confrontational counselling style reduced both resistance to change and drinking itself compared to an explicitly challenging approach. See also the first account (1983) of motivational interviewing and a review (free source at time of writing) of the positive role of subtle forms of ‘confrontation’ in the approach. For discussion click and scroll down to highlighted heading.
The member briefly talked to the client about the meeting, its location, provided the client with a ride, met the client before the meeting so that he/she would know someone at the meeting who would answer his/her questions, and reassured the client if he/she felt uncomfortable. [Then] the AA or Al-Anon member made a phone call to the client at his/her home the night of the meeting to remind him/her of the meeting and to inform him/her what time he/she would be by to pick him/her up. The … member then called the therapist the next day to tell him whether the client had made it to the meeting. The client was then encouraged to continue in AA by therapist after the first meeting. The procedure was repeated in the next counseling session.
S Treatment services can radically affect access to mutual aid (1981). Meeting 12-step group members during treatment who encouraged patients to attend 12-step mutual aid meetings and helped them get there ( panel right) meant all attended compared to none just offered the service’s usual information and encouragement. By a toss of a coin, only 20 patients were allocated to these alternative procedures, but this early study convincingly showed that treatment services can radically affect access to mutual aid. Perhaps influential was a lift to the meetings in a “sparsely populated, rural area”.
S Client-centred group therapy works best (1957). For its time methodologically advanced, this US study found a Rogerian, client-centred approach characterised by non-directive, empathic listening generated healthier self-perceptions among alcohol-dependent patients and reduced relapse compared to approaches based on learning theory or psychoanalysis. This approach which underpins motivational interviewing was also tested in the trial listed below. For related discussion click and scroll down to highlighted heading.
K No added value from motivational interviewing (2012). Other than fleetingly and non-significantly, motivational interviewing’s change-prompting strategies did not generate extra drinking reductions among the stable, moderately dependent drinkers recruited to this US trial. If anything, non-directive Rogerian therapy (also tested in study listed above) stripped of motivational interviewing’s ‘active’ ingredients worked best, and a self-change option was almost as effective as both. For discussions click and scroll down to highlighted heading and see “Is Rogerian listening really all that’s needed?” in cell B4; for related discussion click and scroll down to highlighted heading.
K Expected differences between therapies not found in UK (2008). Results of the largest yet UK alcohol treatment trial (dubbed ‘UKATT’) confounded expectations that a motivational approach would best suit unmotivated or hostile clients, and that clients lacking social supports would do best when this was explicitly addressed. Overall too, neither therapy significantly bettered the other. Inspired by US trial listed below. For discussion of this study click here and for related discussion here and scroll down to highlighted headings.
K Definitive US trial confounds ‘matching’ expectations (1999). Project MATCH was a landmark US trial designed as a definitive test of how differences between therapies mean they work best for different types of clients. Instead it confirmed the importance of the ‘common factors’ underlying seemingly distinct approaches; for more see the book (2002) of the project. Inspired the UK trial listed above. For related discussions click here and here and scroll down to highlighted headings.
K In relapse prevention, practice makes (more) perfect (1997 and 2000). In Scotland practising relapse prevention skills rather than just discussing them boosted confidence and helped newly detoxified patients stay sober longer.
K Modest boost to outcomes from adding “state-of the-art” psychological therapy (2008). The large US ‘COMBINE’ trial found that adding psychological therapy to primary care-style medical support plus medications or placebos elevated by 20% the proportion of patients experiencing a “good clinical response” during the year after treatment, though by the end the difference was negligible. Despite over twice the treatment-contact time, psychological therapy plus only basic medical care generated no significant advantages compared to medical support with placebo pills. These findings and similarly good outcomes whatever the therapy have been interpreted (2008, free source at time of writing) as suggesting there is a “robust placebo effect when using medications for alcohol dependence,” or that ‘common factors’ including the healing context, medications, and interactions with medical staff, account for most of the improvements. Further results from the trial in our report (2006) on main findings during and after treatment. For related discussion click and scroll down to highlighted heading.
K 12-step group attendance boosted in London but not abstinence (2012). Tested the ambition to extend recovery beyond formal treatment by systematically linking patients to mutual aid groups. Most patients had just been detoxified from alcohol. Attendance at 12-step groups was substantially boosted but not abstinence from the primary problem drug, a pattern seen in similar US studies. Related review below.
R Active ingredients and common features of four major psychosocial therapies for problem substance use (2007; free source at time of writing). The presumed active ingredients of motivational interviewing and allied therapies, 12-step based treatment, cognitive-behavioural approaches, and contingency management and community reinforcement approaches based on systematic rewards and sanctions usually linked to substance use. Finds that the effective elements they share include: support, structure, and goal direction; rewards for abstinence and rewarding activities that can replace substance use; abstinence-oriented norms and models; and building self-efficacy and coping skills. For related discussion click and scroll down to highlighted heading.
R All bona fide ‘talking therapies’ work equally well (2008). After combining results from relevant alcohol studies, this ingenious analysis found any structured approach grounded in an explicit model as good as any other. We have, it was argued, been looking in the wrong direction for therapy’s active ingredients. Other reviews have delivered similar verdicts on motivational interviewing (Cochrane review, 2011) and cognitive-behavioural therapy (2009). For related discussions click here and here and scroll down to highlighted headings.
R Common core of effective therapy: therapeutic relationships (American Psychological Association, 2018). Introduces and synthesises finding from 16 reviews of the psychotherapy literature based on the understanding that therapeutic change is generated not only by technical interventions, but by the ways clients and therapists relate. Represents the culmination to date of work begun in 1999 which rebalanced the focus on interventions with an appreciation of the power of ‘common factors’ underlying seemingly distinct approaches. From here you can also access the component reviews. For related discussion click and scroll down to highlighted heading.
R Motivational starts to treatment better without the manual (2005). Findings review discovered that motivational interviewing is not always preferable to more directive approaches, and has worked best when the therapist is not constrained to a manual, no matter how expertly drafted – a conclusion confirmed by a synthesis of research findings (2005). For related discussions click here, here and here and scroll down to highlighted headings.
R Cognitive-behavioural therapies for substance use: what they are and the studies which have tested them (2015). In the UK cognitive-behavioural methods are probably the dominant framework for psychosocial therapy for substance use. This freely available review is particularly useful for its accessible account of what the major subtypes consist of and the studies which have tested them.
R Mindfulness meditation takes its place among addiction therapies (2009). Variants of mindfulness meditation are among the ‘third wave’ of behavioural therapies allying Western and Eastern traditions. This first review of their application to addiction finds them equivalent to other structured therapies; similar conclusions from a more recent review (2014). Neither could include a later trial (2014) which found mindfulness more effective than both 12-step based group therapy and (on some measures only) a cognitive-behavioural relapse prevention programme. Many of the patients were drinkers, though for just 14% was this their sole substance use problem. For related discussion click and scroll down to highlighted heading.
R Weak evidence for extinguishing cue-induced urge to drink (2017). ‘Cue exposure’ therapy tries to rob cues which prompt drinking (like seeing the pub or your favourite beverage) of their urge-provoking power by repeatedly pairing them with non-drinking. Especially in respect of heavy drinking (the most clinically meaningful of the measures), evidence was generally lacking, and in the few available studies often but not always negative. Combining cue exposure with cognitive-behavioural urge-control strategies offered the greatest promise. For related discussion click and scroll down to highlighted heading.
R Directiveness is a key difference between therapies (2006). Rather than specific techniques, the interpersonal style (eg, directive v. patient-led) associated with different therapies is why some work better with some clients than others. For discussion here and for related discussion click here and scroll down to highlighted headings.
R Helping each other get better (2009). Monograph from leading authority on peer-based recovery from addiction includes a chapter on the evidence for AA and allied mutual support networks and treatments based on the same principles.
R Contested vindication for 12-step groups and treatments (Cochrane review, 2020). Finds “high quality evidence that manualized AA/TSF interventions [based on the mutual aid groups and 12 steps of Alcoholics Anonymous] are more effective than other established treatments,” and that manualised or not, they “may be at least as effective as other treatments” – conclusions hotly contested and contrary to those of an earlier review (1999) which found such approaches either no better or worse than other treatments. One of the later review’s authors has also reviewed (2004) how treatment services can promote mutual aid. Related study above.
R If patient is in suitable couple, work with both (2013). Problem substance users in a stable relationship sometimes do better (in terms of substance use and family harmony) when the focus is at least partly shifted to working with the couple to promote sobriety-encouraging interactions. In this review the largest number of studies were of a ‘behavioural’ couples therapy applied to drinking problems. Another review (2011) focused on this approach across substance use, coming to similar conclusions. See also the Effectiveness Bank collection of relevant analyses from our database.
R How lasting are the effects of offering prizes for abstinence? (2014; free source at time of writing). Systematically giving substance use patients a chance to win valuable prizes if they test abstinent offers a lower-cost alternative to other ‘contingency management’ systems which provide rewards each time. Research synthesis shows that in the short term it works, but effects soon fade.
R No significant advantage for residential treatment – due to severity limits on participants? (2019). Review of studies published 2013 to 2018 found no evidence from rigorous trials that residential treatment generated better outcomes than non-residential for dependent drinkers. Some more positive but still weak evidence from earlier review (2014) across substance use, but none from yet earlier review (1994) focused on alcohol dependence, summarised (paragraph 1.3) in unpublished notes (2002) from Findings. Some evidence across these reviews that more severe cases differentially benefit from residential care. Research is limited by the exclusion of some such cases to ensure that patients can safely be allocated to non-residential care.
R Some patients get worse (2012). Reminder that after psychosocial therapy for substance use problems up to 15% of clients end up worse than before. The client’s social context and psychology are factors but so too are: a poor client-therapist bond; less “involving, supportive, and expressive” services; failing to assess how clients are doing; lack of goal direction and structure; being confrontational or critical, especially in the context of a lack of empathy; not adapting structure, expectations and degree of challenge to the client; provoking emotional arousal in disturbed individuals. For discussion click and scroll down to highlighted heading.
G NICE guidance on treating problem drinking (National Institute for Health and Care Excellence, 2011). Recommendations from England’s gatekeeper to the public provision of healthcare on overall principles for treatment and particular interventions. For related discussions click here, here and here and scroll down to highlighted headings.
G Principles rather than programmes for how to relate to clients and what to say/do (2006; free source at time of writing). Based on reviews commissioned by the American Psychological Association. Argues that the principles it extracts from research “provide a more research-informed and potentially effective approach to treatment than either the application of a one-size-fits-all standard treatment protocol or the use of idiosyncratically selected interventions”. For discussions click here and here and scroll down to highlighted headings.
more Search for all relevant Effectiveness Bank analyses or search more specifically at the subject search page. See also hot topics on contingency management, residential rehabilitation, motivational interviewing, and 12-step based mutual aid.
Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text
What is this cell about? Every treatment involves direct or indirect human interaction, but this cell is about treatments in which interaction is intended to be the main active ingredient – ‘psychosocial’, or more colloquially, ‘talking’ therapies. They attempt to change how the patient behaves via their beliefs and attitudes, how they relate to others, and how others relate to them, or directly by ‘shaping’ it through rewards and sanctions. For an ‘alcoholic’ or ‘addict’, simply being cared for and about in a healing context focused on your aspirations, needs and welfare, may be a ‘re-moralising’ experience (in seminal therapist Jerome Frank’s formulation), but usually there are also specific techniques or strategies derived from various understandings of how dependence arises and how it can be overcome or ameliorated.
Interventions designed to achieve this vary in type and in intensity and duration. They range from brief advice and counselling to extended therapies (examples listed above) based on psychological theories and all-embracing residential communities (review listed above) where clients stay for several months. Techniques and strategies might include: rewards and punishments contingent on client behaviour, as in contingency management (document listed above); leading the client to see their substance use as contrary to desired self-images or objectives, as in motivational interviewing (document listed above); harnessing social influences, as in group and family (document listed above) therapies and community living arrangements; identifying with the client what triggers their undesired substance use and training them how to manage or avoid those triggers, as in cognitive-behavioural therapies (document listed above); ways to manage thoughts and moods which might otherwise precipitate relapse, as in mindfulness approaches (document listed above); and more practical elements, such as those intended to upgrade the patient’s employability. These and other interventions and services are sometimes orchestrated (especially for patients with multiple difficulties) by someone who performs a case management (2006) function, which may itself involve considerable client contact and therapeutic content.
Whether based on theory, empirical research, religion, morals or experience, belief systems underlie these approaches. Most prominent in the research are the 12 steps (document listed above) of Alcoholics Anonymous, Narcotics Anonymous and allied fellowships, and the understanding that addiction can be learnt and unlearnt, which underpins major psychological therapies, including those recommended (document listed above) by England’s gatekeeper to the public provision of health interventions, the National Institute for Health and Care Excellence (NICE).
Based on these belief systems, all the approaches mentioned above have presumed active ingredients which distinguish them from other approaches, but also share important common features (discussed further in the next section). Listed above and available free of charge is an exemplary exposition of the supposed distinctive active ingredients of four major categories of psychosocial therapies for problem substance, and the elements they share. According to its renowned author, these duplicate the developmental experiences outside the clinic which help prevent most people getting into trouble with substance use. In other words, they are not distinct to therapy, but are the social processes which foster psychologically healthy human development: “These processes involve bonding, goal direction, and monitoring from family, friends, religion, and other aspects of traditional society; participating in rewarding activities that preclude or reduce the likelihood of substance use; selecting and emulating individuals who model conventional behavior and shun substance use; building self-confidence and effective coping skills.” The same observation was also made by Drug and Alcohol Findings in one of our Manners Matter reviews: “the human qualities which cement relationships outside treatment also do so within it”. From this perspective, psychosocial treatment is about encapsulating and importing into the clinic or consulting room nurturing environments and interactions lacking in the past and present lives of many people who develop unhealthy dependencies.
Where should I start? This cell is partly about the relative merits of different psychosocial therapies, but also about the therapeutic properties they share and how such ‘common factors’ and ‘placebo effects’ (related but conceptually distinct categories of influences) can be made more potent. Since these have become seen as the major influences in the treatment of substance use problems, let’s start there, and in particular with the rebalancing focus on common factors implemented by the American Psychological Association’s (APA) appropriately named “interdivisional” task forces. Here we’ll also glance at the roots of a movement which to date has culminated in the APA’s third such task force, one whose iconoclastic importance in psychosocial therapy can hardly be under-estimated.
The first APA task force was commissioned in 1999 to investigate and promote the evidence for the influence of relationships in therapy. It represented a symbolically and practically significant turning point, harnessing the APA’s resources and authority to draw attention to an important common factors domain. A decade later a second task force updated the work; the results were summarised in an article analysed for the Effectiveness Bank, from where you can also find analyses of the component reviews. The findings of a third iteration of the same project were published in 2018. Our starting-point article from the APA’s experts set the context for and interpreted the findings of this new task force’s 16 reviews, representing it was said “the best available research and clinical practices on numerous facets of the therapy relationship”. An “impressively wise overview” was how we described it; read our analysis (listed above) and from there access the original article (click on the blue title) and the component reviews.
Then go back a few years to the guidance (listed above) offered by another APA task force, integrating relationship issues with the content of therapy. In both note the stress on collaborative working, and warnings against being confrontational, hostile, pejorative, critical, rejecting, or blaming. These behaviours can be thought of as at the negative end of common factors dimensions to do with therapeutic relationships.
But note too that the APA’s experts have cautioned that when it comes to relating to clients, there are no universal rules, and to attempt to follow the same strategy for each client would be sub-optimally effective and possibly unethical. The warnings came in a companion set of reviews to the second task force’s work on the generally effective features of therapeutic relationships. This set of reviews instead focused on effective methods of adapting or tailoring treatment to the individual – ‘What works in particular’ rather than ‘What works in general.’ An introductory article stressed the rationale: “Decades of research now scientifically support what psychotherapists have long known: different types of clients require different treatments and relationships … In the tradition of evidence-based practice, psychotherapists can create a new, responsive psychotherapy for each distinctive patient and his or her singular situation, in addition to his or her disorder.” Focusing on the relational dimension of directiveness, it was a proposition we too explored in one of our Manners Matter reviews (document listed above). Unlike public health approaches, treatment is essentially the treatment of an individual, for whom what is generally counterproductive may be productive and vice versa – a maxim expanded on in the final section below.
Common factors including the APA’s relationship dimensions cut across the supposedly specific ways of working of approaches at opposing theoretical poles such as cognitive-behavioural therapy and psychoanalysis, spotlighting instead the non-specific ‘mechanisms’ they share. Explored further in cell A2, thought to be among these common factors are a confiding, emotionally charged relationship with a helping person, a setting in which psychological healing is expected, a plausible explanation for the patient’s symptoms, and a ritual or procedure for resolving them – any ritual will do, as long as in that culture and for that patient it makes sense of their difficulties and signposts a convincing and feasible way out (1; 2, free source at time of writing).
Across psychotherapy shift of attention to what therapies share has partly been prompted by the so-called ‘Dodo bird’ (an Alice-in-Wonderland reference illustration) hypothesis that different therapies have similar impacts because they largely work through similar mechanisms. It was first advanced (free source at time of writing) in this form in 1936 by Saul Rosenzweig and later elaborated (free source for this account of the genesis and impact of Rosenzweig’s theory) by his Harvard university peer Jerome Frank in a book analysed for the Effectiveness Bank, where a collection has gathered together all our relevant analyses. But just in this one cell of the Alcohol Treatment Matrix can repeatedly be found evidence that in substance use as in other sectors, Rosenzweig’s hypothesis retains considerable support: 1, listed above; 2, listed above; 3, listed above; 4, listed above; 5, listed above. Further evidence of the centrality of common factors in substance use treatment has come from findings that therapies stripped down to these factors can work as well as supposedly more active approaches (1, listed above; 2, listed above), and that when therapies do vary in effectiveness, this may not be due (document listed above) to their purportedly distinct active ingredients, but to variations in the degree to which they engage certain common-factor type influences.
When the theoretical pendulum swings to counterbalance a previous imbalance, it sometimes swings a little too far and the evidence demands some pushing back. So too with the swing away from seeing different types of therapies as working in different ways and differentially effective and towards endorsing the Dodo bird hypothesis. Across psychotherapy that hypothesis remains broadly supported, but even when confirming (free source at time of writing) this in 2002, analysts cautioned that the few research syntheses which produced contrary findings cannot safely be dismissed as chance variations. Among the other reasons for not giving a carte blanche to the ‘no difference’ Dodo bird hypothesis was that “Specific outcome measures such as depression and anxiety may tempt us to forget that there may be other differences between treatments. Two patients, for example, after participating in different treatments, may feel better and not currently depressed, but one of them also may have made other important gains.”
That proposition was explored in one of the latest amalgamations (free source at time of writing) of research findings to test whether different psychosocial therapies are on average differentially effective. Published in 2014, its major innovation was to separately answer this question in respect of the primary objectives targeted by the therapy (such as depression in a study of the treatment of depression) versus secondary outcomes also measured but not actually targeted by the therapy (such as general quality of life measures in the same study). On their primary outcomes, therapies did actually significantly if modestly differ in effectiveness both at the end of the treatment and at follow-up, but not on the secondary measures. However, omitting one ‘outlier’ study (of social phobia) with an unusually distinct advantage for one therapy over another rendered the difference in primary outcomes at follow-up non-significant. These analyses largely vindicated the Dodo bird hypothesis with the exception of primary outcomes at treatment end (not follow-up) – an exception largely due to studies involving very specific complaints such as tics or panic attacks rather than more diffuse conditions such as depression. Also the analyses took no account of whether the researchers in the studies were to some degree committed to one of the therapies they were testing. In the 2002 analysis cited in the previous paragraph, adjusting for this potential source of bias substantially eroded differences between the effectiveness of therapies, to the point that it eliminated all the statistically significant differences found across six amalgamations of research.
Highlighted study For Britain the highlight has to be the £1.5 million UKATT trial (listed above), the most ambitious ever in the UK. Implemented in the late ’90s, it was informed by emerging findings from the US Project MATCH trial listed above, which found relatively brief therapy based on motivational interviewing no less effective than longer and more structured approaches. In response, the UKATT team set out to devise a research-based therapy which would better the standard set by MATCH’s motivational intervention. What they came up with was ‘social behaviour and network therapy’. It integrated elements from several approaches and geared them to harnessing the “crucial contribution” of social networks supportive of positive change in the client’s drinking. A scheduled eight sessions of this more extensive, intensive and comprehensive therapy were compared with three sessions of the basic motivational approach. Unexpectedly, there were no significant differences in effectiveness or cost-effectiveness. Neither did (as had been expected) the network option particularly benefit patients lacking anti-drinking social support, or motivational interviewing particularly help angry patients or those lacking motivation, or do more to nudge those merely thinking about changing their drinking at the start of treatment into actively making the attempt.
Unexpectedly, there were no significant differences in effectiveness
Expectations confounded, the researchers accounted for the seeming equivalence of the therapies by the equivalence of the training “in a professional and rigorous atmosphere” undergone by the therapists, their expert and regular supervision, and their guidance by a manual with scientific credibility, generating expectations that the approach they had been steeped in would yield good results: ‘expectations’, ‘optimism’, ‘morale’, ‘structure’, ‘support’, ‘credibility’ – common factors rather than the distinctions in approach and content which it was thought would prove critical.
On this they now agreed with the patients. Asked at follow-up to what they attributed any positive changes, general influences not specific to either of the treatment approaches were cited more often than those through which each approach was supposed to distinctively exert its effects. For the researchers these results bolstered the “contextual model of effectiveness of psychosocial treatments, emphasizing such factors as client commitment, therapist allegiance, and the client-therapist alliance ... as opposed to a medical-technical model, which emphasizes specific factors related to mechanisms of change promoted by specific forms of therapy”. Later the lead author gave his personal interpretation of these and other findings: “Treatment research has been asking the wrong questions in the wrong way … the strange thing is how we have persisted with trials of named psychosocial therapeutic techniques despite the evidence that this strategy is largely fruitless. In the psychosocial treatment of addiction there is increasing support for the existence of important change processes that are common to treatments with different names and theoretical rationales.”
Now take a look back at cell A2 where Project MATCH, UKATT’s US progenitor, was the highlighted study. Put these studies side by side and you will see that after two similar trials unprecedented in their combination of size and rigour, some of the most eminent researchers into the treatment of alcohol problems from both sides of the Atlantic had come to conclusions far distant from the expectations on which their trials were founded. Like UKATT’s researchers, Project MATCH’s team came to emphasise not treatment techniques, but influences such as empathy, an effective working alliance between therapist and client, the latter’s desire to get better and inner resources to overcome alcohol dependence, a supportive social network, and the “provision of a culturally appropriate solution to a socially defined problem”.
Can therapy really make things worse? Look back at the American Psychological Association’s warnings in the “Where should I start?” section against counsellors being confrontational or negative. In case you think their concerns overstated, note that these behaviours figure among the reasons why a substantial minority of substance use clients actually get worse (review listed above) after therapy, seemingly more so (free source at time of writing) than clients receiving therapy for other conditions. Avoiding this risk (especially the risk of provoking resistance to change) has been embodied most explicitly in motivational interviewing. Among heavy drinkers, its credentials in this respect were seemingly confirmed in a seminal trial listed above.
Proofed against counterproductive reactions, appropriate for all levels of severity, generally as effective (review listed above) as other approaches but considerably briefer, motivational interviewing has been seen as a promising starting point for substance use treatment, one at least unlikely (recalling the first maxim of medicine) to do any harm.
That may be true in the absolute sense, but not in terms of (review listed above) lost opportunities to help patients who would have benefited more from another approach. Instead of the circumventions and restraint of motivational interviewing, sometimes it really is best just to tell patients what they should do or otherwise contravene motivational interviewing’s ‘rules’ rather than to inflexibly follow manuals crystallising the approach into dos and don’ts.
Probably the best established ways therapists can both get it right and get it wrong lie along the dimension of directiveness versus non-directiveness, explored in a Findings review listed above.
Sometimes it really is best to tell patients what they should do
They can get it wrong by trying to take charge with patients who react against being led – a clash motivational interviewing aims to avoid – but also by adopting motivational interviewing’s typical, ‘It’s up to you,’ stance with patients who at that time need to be taken by the hand and given a lead. The perhaps uncomfortable truth seems to be that beyond the obvious, there are no universal rules: some people in some circumstances need to be led, others to lead; some told what to do, others to feel they have come to their own decisions; some need arousing, others soothing; – and needs can change as therapy progresses.
So when with all the authority of England’s gatekeeper to the public provision of health care interventions, NICE advises (in a document listed above) that some forms of substance use therapy “should be based on a relevant evidence-based treatment manual,” remember they mean based on, not prescribed in advance no matter what the predispositions of the patient or their current circumstances, needs and preferences.
Research may have to package; therapy does not That last comment in the section immediately above brings us to an important point (free source at time of writing) about research on psychosocial therapies and its links to practice. To pin down what caused any improvements, researchers often tightly control what is being delivered by manualising interventions and training and supervising therapists to ensure they follow the manual. But (as demonstrated in our review of motivational interviewing listed above) though this might be the best way to do research, it is not necessarily the best way to do therapy, which has to sensitively adapt to where the patient is at in solidifying their commitment to tackle their substance use problems.
Similarly, researchers generally standardise interventions to a set schedule and time period in order to limit costs, equalise time spent with therapists in a comparison therapy, and to have a set end date from which the follow-up period can begin. Twelve weeks is a frequent compromise between a manageable research intervention and one long enough to have a fair chance of working. As a result, 12-week treatments collect an evidence base around them and become recommended practice in influential guidelines, such as those developed by England’s National Institute for Health and Care Excellence (1, listed above; 2). Yet there is no reason to believe that because 12 weeks is convenient for researchers, it is also the span over which patients should be treated. Some manage well with much less, others (see cell D2) benefit more from longer term care.
What we see here is the formation of a less than virtuous circle. Research takes its ideas from practice, standardises and packages that practice in order to test it, and then tests it. Via recommendations from authorities reliant mainly on research findings, practitioners may then be persuaded that the researchers’ packages – now ‘evidence-based’ – are how they too should do therapy. We can break this circle by treating research not as a blueprint for practice, but as an aid to reflecting on and developing it along with other considerations. In respect of problem drug use, this was the approach taken by authorities from the British Psychological Society when they developed their guidance on implementing the main psychosocial therapies recommended by NICE. The authors insisted that though their framework “draws heavily upon treatment manuals, it enables a more comprehensive approach to implementation than a manual alone can provide … It allows a degree of flexibility and adaptation at the level of the individual service user. Such flexibility may not be present in a particular manual, the development of which may instead be rooted in a specific service in a particular health care setting.”
Are these always the important things to do? Start this exercise by downloading the freely available guidance listed above from the American Psychological Association.
Under the heading, “Treatment factors,” on page three of the PDF file the guidance says research “suggests that a number of specific therapeutic elements are characteristic of effective treatments”. Take a critical look at these elements. They include “explicitly helping the client restructure his or her social environment in ways that support change” – or more specifically, change in the form of abstinence. Recall that restructuring the patient’s social network to be more supportive of abstinence or moderate drinking was the main focus of the novel treatment programme developed for the British UKATT trial, our “Highlighted study”. In 2006 a report for England’s substance use treatment authority had reasoned that this kind of approach would work best for dependent drinkers.
Look back at the “Highlighted study” section and remind yourself how well such an approach actually did work relative to more basic motivational treatment when subjected to a rigorous test. ‘No better,’ seemed the justifiable conclusion. Perhaps – even probably – the guidance is perfectly correct about the effects of such an approach when implemented well, but this is difficult to engineer in practice. How many clients have the resources – psychological, social and material – to replace environments, friendships, even families and intimate partners which are conducive to drinking, with those more conducive to non-drinking? Radical social restructuring can work – witness the changes wrought by therapeutic communities in the contained environments of residential rehabilitation centres – but these gains remain vulnerable on discharge back to the less readily restructured home environment.
Next element is a “focus on client motivation for change” – possibly, suggests the guidance, through ‘decisional balance’ exercises leading them to weigh up the pros and cons of changing their substance use. But if patients are already committed to change (as many will be), maybe it is not such a good idea to encourage them to rehearse the good things about their substance use? On this question, see the alcohol treatment aftercare (study 19) in a Findings review listed above. Of that and two other studies we commented, “the puzzle is not why the least committed subjects benefited from a motivational approach (this is the expected result), but why the most committed reacted badly … Possibly the explanation is what to the patient may have seemed a backward step to re-examine the pros and cons of whether they really did want to stop using drugs or commit to treatment and aftercare, when they had already decided to do so and started the process.”
A focus on boosting client motivation is above all the province of motivational interviewing, and is thought to be what it adds to non-directive counselling after the style of Carl Rogers. But when these motivation-eliciting enhancements were stripped out (study listed above; illustration left) leaving only the bedrock of non-directive, empathic listening, no significant detriment to drinking outcomes could be detected; if anything, the reverse. Moreover, neither this bedrock nor a fully-fledged motivational approach was much better than just telling patients to curb their drinking, and leaving them to do it on their own. In this study (of which more in cell B4) the patients were relatively stable, moderately dependent drinkers, but the same non-superiority of motivational interviewing has been seen (review listed above) across studies in general. If focusing on client motivation is a key to effective therapy, why across relevant studies has motivational interviewing not proved superior to other therapies?
Next the guidance’s list of “elements … characteristic of effective treatments” moves into the territory of cognitive-behavioural therapies: “helping the client to develop awareness of repetitive patterns of thinking and behavior that perpetuate substance use … accompanied by a focus on helping the client learn alternative coping skills”. If this is a key to effective therapy, why on average do cognitive-behavioural therapies do no better (review listed above) than other approaches?
Then we learn that “Effective therapies attend to the affective [emotional/mood] experiences of the client, particularly in relation to their substance use.” Yet we know that a focus on emotions can for some patients be counterproductive.
Finally, the guidance identified “strong evidence for the role of conditioning in the development and maintenance of substance use disorders,” and argued that “repeated exposure to alcohol- or drug-related situations without using” can weaken these conditioned reactions and bolster the patient’s confidence that they can handle such situations without being prompted to use substances. Yet in respect of problem drinking, what seems the latest synthesis (listed above) of relevant research found the evidence very weak though not entirely negative: “since relatively few [cue exposure therapy] studies targeting [alcohol use disorders] were available and … these were judged as providing very low quality evidence, sounder methodological trials are necessary to draw any firm conclusions.” Similarly, a review listed above of alcohol treatment research conducted for NICE, England’s gatekeeper to the public provision of healthcare interventions, found little and generally negative evidence that cue exposure bettered other therapies. When the same authority looked at the treatment of dependence on illegal drugs, it was unable to find a single cue exposure study which met its quality criteria.
These points are made not to criticise the in general very sensible suggestions made in the guidance, but to reinforce the point that generalisations are bound sometimes to be misleading in what is essentially an individualised response to an individual set of circumstances, never before encountered in precisely the same configuration. When these individual reactions are amalgamated in research studies which in turn may be amalgamated in syntheses of those studies, a small and inconsistent effect may be found which offers little helpful guidance to the practitioner. The guidance’s search for “specific therapeutic elements” may in any event have been misguided if (as discussed above) what therapies share are in fact the dominant therapeutic force.
Thanks for their comments on this entry to David A. Haaga of the American University USA and Jim Orford of the University of Birmingham in England. Commentators bear no responsibility for the text including the interpretations and any remaining errors.