Send email for updates


About updates
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our analysis of a study considered particularly relevant to improving outcomes from drug or alcohol interventions in the UK. The original study was not published by Findings; click Title to order a copy. Free reprints may be available from the authors – click prepared e-mail. Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text The Summary conveys the findings and views expressed in the study. Below is a commentary from Drug and Alcohol Findings.

Copy title and link | Comment/query |

Patient reactance as a moderator of the effect of therapist structure on posttreatment alcohol use.

Karno M.P., Longabaugh R., Herbeck D.
Journal of Studies on Alcohol and Drugs: 2009, 70, p. 929–936.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Karno at karno@ucla.edu. You could also try this alternative source.

Confirmation from the US Project MATCH alcohol treatment trial that too explicitly imposing structure on therapy risks relatively poor outcomes among patients reluctant to relinquish control and who react against direction – and a further indication that this pattern is not universal, but depends on the context.

Summary In the 1990s, the US Project MATCH study tested whether different types of alcohol-dependent patients would respond better to 12-step based counselling, cognitive-behavioural therapy, or to an approach based on motivational interviewing. It was intended to be a definitive test of this kind of matching of patients to therapies. Few and minor matching effects were found, but researchers at some clinics in the trial were able to use session recordings to probe for a different kind of matching – not of therapies, but of therapeutic (or more broadly, interpersonal) styles.

MATCH therapies were either standalone treatments or 'aftercare' immediately following residential or intensive day care. At one of the clinics in the aftercare arm of the study, patients who seemed Based on ratings of their videoed behaviour in their first therapy session. at least moderately reluctant to relinquish control and who reacted against direction ('reactive' patients) did badly when therapists took the lead in structuring the sessions. Relative to other patient-therapist combinations, in the following year they drank and drank heavily more frequently. For less reactive patients, the degree to which therapists imposed structure was unrelated to subsequent drinking. These results suggested that for patients beyond a certain level of reactance, therapist-led initiation of topics, teaching, and providing information were associated with worse outcomes.

The featured report aimed to test whether these findings would be replicated at another of the MATCH aftercare sites (among 122 patients), and at three of the sites (among in total 125 patients) where MATCH therapies had been provided as standalone treatments. As in the previous study, patients were divided in to high, medium and low reactance based on ratings of their videoed behaviour in their first therapy sessions. Their therapists too were rated on how they behaved with the client across If four sessions had been attended. the first three and final sessions – specifically Another way of imposing structure – asking closed questions – was in this study found unrelated to outcomes, seemingly because it was highly correlated with asking open-ended questions. Omitting this dimension from the analysis of the earlier study did not materially affect the findings from that study. the degree to which they took the lead in structuring sessions by providing information or instruction, or introducing or changing topics.

Main findings

As in the previous study at an aftercare site, at the new aftercare site too, over the following year medium or highly reactive patients drank and drank heavily more often if their therapists had been relatively directive in imposing structure on therapy sessions. Less reactive patients actually did slightly better when therapists imposed structure. No such findings emerged when the therapies were standalone treatments; highly or less reactive patients responded equally well, regardless of whether their therapists explicitly structured sessions. These patterns were consistent across the entire year of the follow-up. Across all patients, they summed to a finding that highly structuring therapists were relatively counterproductive when the therapy was aftercare, but overall had slightly better outcomes when delivering standalone therapies. The structure-reactance interaction did not, however, affect how long patients lasted before lapsing to drinking or relapsing to heavy drinking.

The authors' conclusions

The featured study suggests that when psychosocial therapy for alcoholism follows an intensive treatment episode (that is, plays an aftercare role), the degree to which therapists structure sessions impacts differentially on patients more or less willing to relinquish control and accept direction: the more structure, the less well reactive patients do after therapy. The findings replicate those from another MATCH aftercare site, showing they transcend the particular nature of that clinic, its patients, and its therapists.

This pattern not, however, apply to standalone therapies, and in this role structure was overall beneficial, while in the aftercare role it was overall detrimental. Why this happened is unclear. Perhaps patients starting a new treatment – even those who normally react badly to direction – expect and are receptive to treatment structure. In contrast, patients emerging from intensive treatment in to an aftercare phase may expect a less structured approach; when they find the opposite, those prone to react badly to direction react as expected, and end up drinking more than when treatment meets their expectations.

For therapists in an aftercare or continuing care role, the implications are that generally they should avoid highly structuring therapy in the form of adopting a 'teaching' style, providing information, and controlling which topics are discussed, especially with reactive patients. To avoid this, they may as part of their assessments wish to ask patients to complete one of the validated questionnaires which measure reactance.

It should be remembered, however, that these suggestions emerged from observing how well patients do when therapists happen to be relatively directive or non-directive in the degree to which they structure therapy. A study which deliberately assigned patients to therapists who were directive or not would be in a stronger position to establish whether this was indeed an active ingredient affecting the success of therapy for different types of patients.


Findings logo commentary Across substance use studies, the type of patient-therapist dimensions investigated in the featured study evidence a remarkably consistent pattern. In terms of substance use outcomes, non-directive therapeutic styles work best for clients characterised by anger, defensiveness, or resistance, or who like to take control – the 'reactive' patients of the featured study. In contrast, more structured and directive approaches may profit calmer clients, those who welcome being given a lead, and those already committed to the course of action being directed. While the featured study highlighted the risks of directively imposing structure, other studies have shown that for some patients in some circumstances, being non-directive is counter-productive.

A similar pattern has been observed in psychotherapy in general in an analysis which included data from MATCH: patients who characteristically exhibit low levels of resistance or reactance respond better to directive types of treatment, while patients prone to be reactive or resistant respond best to non-directive treatments. Together with the finding that reactive patients tend to benefit least from therapy, it led the experts to recommend that highly reactive patients should be offered treatment which de-emphasises the therapist's authority and guidance, employs tasks designed to bolster patient control and self-direction, and de-emphasises the use of rigid homework assignments. In general, therapists should avoid counterproductively stimulating the patient's level of resistance.

But as the featured study found, such patterns can be context-specific – in this case, apparent when therapy was aftercare, but not when it was the primary treatment. Patients' expectations of how much structure to expect or how much was appropriate were, the researchers suspected, what made the difference in the two settings. Similarly, a Findings review has cautioned that, for example, non-directiveness from a probation officer to an offender can seem less than genuine, even to the officer, as can biting one's tongue when it would have been natural and caring to be direct about the risks a client faces. In these circumstances, even if being non-directive might generally suit the client's character, following this guideline could adversely affect one of the features of effective therapy – that the therapist comes across as 'genuine'. Outcomes might also be worse if for the sake of not being directive, therapists failed to address the emotional state of highly distressed patients.

The authors of the featured study suggested that standard psychological tests could assess how far clients resist or welcome direction and act as a guide to how tightly therapy should be structured. For their study, these assessments were made by observers on the basis of the first session video, suggesting that the therapist too could observe the patient and adjust accordingly. Feedback from early counselling sessions through recordings assessed by supervisors or peers, or through short 'de-briefing' questionnaires given to the clients, could also be used to assess when there is a mismatch between therapist and client interactional styles.

The degree to which they structure therapy, and directiveness more broadly, is just one dimension which therapists might bear in mind. For more on matching alcohol treatments to patients, see this Findings entry offering an introduction to the topic and a one-click search for relevant Findings analyses.

Last revised 19 December 2011

Comment/query
Give us your feedback on the site (one-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates


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

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

STUDY 2006 Adjust therapist directiveness to client resistance

REVIEW 2006 My way or yours?

STUDY 2005 Therapist directiveness is an important influence on outcomes

STUDY 1999 Project MATCH: unseen colossus

REVIEW 2018 Cohesion in group therapy: a meta-analysis

MATRIX CELL 2017 Alcohol Matrix cell B4: Practitioners; Psychosocial therapies

STUDY 2009 Therapist behavior as a predictor of black and white caregiver responsiveness in multisystemic therapy

REVIEW 2005 The motivational hallo

STUDY 2004 Clues on to how to match clients to therapeutic styles