Still hard to find reasons for matching patients to therapies
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our account of a study selected by Drug and Alcohol Findings as particularly relevant to improving outcomes from drug or alcohol interventions in the UK. Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text

Copy title and link | Comment/query |

Still hard to find reasons for matching patients to therapies

After finding no overall difference in effectiveness between the two therapies it tested, the UK Alcohol Treatment Trial (UKATT) has now also found no differences for different types of patients.1 The results confounded expectations that an approach based on motivational interviewing would be preferable for the least motivated or most hostile, while bolstering supportive social networks would be particularly important for patients lacking these to begin with.

FINDINGS The trial recruited 742 patients seeking treatment for alcohol problems at seven specialist treatment services in England and Wales. They were randomly allocated either to three sessions of motivational enhancement therapy or eight of social behaviour and network therapy, each spread over eight to 12 weeks. The former was a familiar elaboration of motivational interviewing, the latter a novel therapy integrating cognitive-behavioural, community reinforcement and other elements with the aim of building social networks supportive of positive change in the patient's drinking. If possible the patient's associates were directly involved in the process.

Twelve months after therapy started, 85% of surviving participants (12 had died) were re-interviewed. Across both therapies, alcohol consumption over the past three months had fallen by 45%.2 There had also been significant improvements in the severity of alcohol dependence, alcohol-related problems, and psychological health, and savings in health and social care costs.3

The featured study1 tested whether at either the three-month or the twelve-month follow-ups, certain types of patients had responded better to one therapy than the other in terms of drinking reductions (days abstinent, amount consumed when drinking), alcohol dependence, or alcohol-related problems. It was expected that the non-confrontational style of motivational interviewing would defuse the hostility of patients prone to react angrily, and help those relatively devoid of motivation find reasons to curb their drinking. The network option was expected to particularly help patients with poor family relationships or few regularly seen associates who were not also heavy drinkers. Also tested was whether a patient's mental health or severity of dependence would affect relative responses to the therapies.

Just two of these tests for 'matching' achieved the conventional level of statistical significance. Both findings were the opposite to what was expected and (along with near misses) were dismissed as chance outcomes from among the 130 tests.

IN CONTEXT UKATT derived its hypotheses partly from the US Project MATCH study, which also found its therapies roughly equivalent and few and only minor matching effects. Together these methodologically advanced studies strongly question whether it is worth trying to match alcohol patients to different outpatient psychosocial therapies. However, alternative analyses have found statistically and clinically significant matching effects from Project MATCH and might yet do so from UKATT. Some have been based not on which therapy was delivered, but on the whether the therapist's interpersonal style matched that of the patient.4 Another tailored its analysis to a model of relapse (and its opposite) as often sudden, wholesale transitions capable of being precipitated in vulnerable individuals by minor changes in circumstances or psychological state.5

Results like these mean that the possibility of matching patients to interventions cannot yet be dismissed. Studies might have produced negative results because they mistakenly assumed it was important to match to the specific therapy rather than to non-specific, cross-cutting features such as the interpersonal style of the therapist, or because their analytic model mistakenly assumed that relapse and recovery are incremental rather than often precipitous.

In another paper UKATT found just such processes at work as the patients it studied improved.6 Asked what they thought had helped, their answers commonly revealed revelatory moments which precipitated wholesale transitions in how they saw drinking and drink and in their determination to change. Others described how an understanding listener and learning new facts made a difference. The catalysts for change often preceded treatment entry, and patients saw themselves as responsible for the changes they had made using the treatments. General influences not specific to either of the treatment approaches were cited by the patients more often than those through which the approach was supposed to distinctively exert its effects, accounts which might partly explain why these effects were equivalent across the therapies. Such processes might also explain why in Project MATCH not only were the therapies equivalent, but it seemed to make little difference to drinking outcomes whether they were attended or completed.7

Patients highlighted not just the therapies tested in UKATT, but preceding, subsequent and parallel interventions, including other treatments and facilities available at the same clinics and contact with the UKATT team itself. Their accounts question the implicit assumption that all the savings in health and social care costs could be attributed to the UKATT therapies, an assumption which yielded a ratio of £5 savings for every £1 spent.3

PRACTICE IMPLICATIONS On the basis of their own work and that of Project MATCH, the UKATT researchers suggested that therapies such as those tested could be chosen on grounds other than relative effectiveness, including cost, availability of therapists, clinical judgement, and patient preference. One strategy would be to offer the cheaper and more widely available motivational interviewing first and monitor patients to see if they required further or different therapy.

For the generality of patients of the kind recruited to treatment trials, that seems an evidence-based and efficient strategy, but perhaps not one that should be universally applied. Implemented inflexibly with unsuitable patients, motivational interviewing can be counter-productive. This may have happened in Project MATCH.5 Patients who began treatment drinking heavily and lacked confidence in their ability to resist drink reacted poorly to motivational interviewing. They drank on far fewer days after cognitive-behavioural therapy. As in other studies, perhaps these patients floundered without structure, direction and concrete anti-relapse guidance.8 Sometimes patients do much better when left to go through treatment in the normal way or given simple advice, particularly those already committed to a recovery goal and strategy or who respond counter-productively to the assessment feedback often featured in motivational interviews.9 Reactions during the session itself can indicate that this is happening. Sufficiently sensitive and skilled therapists encouraged to adapt to these signals may avoid bad reactions, but in other circumstances the risk is that patients who could have done well from the start will be sent on less positive trajectory.

Thanks to Dr George Christo of the Barnet Drug & Alcohol Service and to UKATT researchers Nick Heather, Gillian Tober and Jim Orford for comments on this entry in draft. Commentators bear no responsibility for the text including the interpretations and any remaining errors.

1 FEATURED STUDY UKATT Research Team. UK Alcohol Treatment Trial: client-treatment matching effects. Addiction: 2008, 103(2), p. 228–238.

2 UKATT Research Team. Effectiveness of treatment for alcohol problems: findings of the randomised UK alcohol treatment trial (UKATT). British Medical Journal: 2005, 331.

3 UKATT Research Team. Cost effectiveness of treatment for alcohol problems: findings of the randomised UK alcohol treatment trial (UKATT). British Medical Journal: 2005, 331.

4 Karno M.P. et al. Does matching matter? Examining matches and mismatches between patient attributes and therapy techniques in alcoholism treatment. Addiction: 2007, 102(4), p. 587–596.

5 Witkiewitz K. et al. Nonnormality and divergence in posttreatment alcohol use: reexamining the Project MATCH data 'another way'. Journal of Abnormal Psychology: 2007, 116, p. 378–94. For an informal analysis of this work see: Ashton M. Catastrophe. Unpublished, 2007.

6 Orford J. et al. The clients' perspective on change during treatment for an alcohol problem: qualitative analysis of follow-up interviews in the UK Alcohol Treatment Trial. Addiction: 2005, 101, p. 60–68.

7 Cutler R.B. et al. Are alcoholism treatments effective? The Project MATCH data. BMC Public Health: 2005, 5:75. For an informal analysis of this work see: Ashton M. It makes you think ... MATCH - the untold story. Unpublished, 2007.

8 Ashton M. My way or yours? Drug and Alcohol Findings: 2005, 15, p. 22–29.

9 Ashton M. The motivational hallo. Drug and Alcohol Findings: 2005, 13, p. 23–30.

Last revised 21 February 2017. First uploaded 09 April 2008

Open Effectiveness Bank home page

Top 10 most closely related documents on this site. For more try a subject or free text search

STUDY 2010 Initial preference for drinking goal in the treatment of alcohol problems: II. Treatment outcomes

STUDY 2008 Style not content key to matching patients to therapeutic approaches

REVIEW 2011 Alcohol-use disorders: diagnosis, assessment and management of harmful drinking and alcohol dependence

STUDY 2001 Brief motivational therapy minimises health care costs except among more problematic drinkers

STUDY 2019 Efficacy and cost-effectiveness of an adjunctive personalised psychosocial intervention in treatment-resistant maintenance opioid agonist therapy: a pragmatic, open-label, randomised controlled trial

REVIEW 2011 Behavioral couples therapy for substance abusers: where do we go from here?

STUDY 2010 Naltrexone and combined behavioral intervention effects on trajectories of drinking in the COMBINE study

STUDY 2006 Adjust therapist directiveness to client resistance

STUDY 2011 Extended telephone-based continuing care for alcohol dependence: 24-month outcomes and subgroup analyses

STUDY 2006 UK trial bolsters case for well-supervised alcohol therapy