Horvath A.O., Del Re A.C., Flückiger C. et al.
Psychotherapy: 2011, 48(1), p. 9–16.
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 Horvath at email@example.com.
This comprehensive meta-analytic review commissioned by the American Psychological Association finds that the relationship between psychotherapist and client is one of the largest and most consistent indicators of outcomes. Authoritative practice recommendations will aid substance use counsellors and therapists.
Summary Updated in 2018. See Effectiveness Bank analysis.
[Though not specific to patients with drug and alcohol problems, studies in the analyses described included such patients, and the principles are likely to be applicable to these disorders among others, not least because substance use problems generally form part of a complex of broader psychosocial problems.]
This review is one of several in a special issue of the journal Psychotherapy devoted to evidence-based, effective therapist-client relationships. It reports on a research synthesis of the links between outcomes of individual psychotherapy and the alliance between therapist and client. 'Alliance' has been variously defined as a bond between the two which holds the client in therapy or as a collaborative working relationship, and is sometimes seen as mainly working at the unconscious level, sometimes at the conscious. In practice, in each study the alliance is defined by which of the diverse questionnaires or other methods are used to measure it. Four 'core' measures California Psychotherapy Alliance Scale, (CALPAS), Helping Alliance Questionnaires (HAq), Vanderbilt Psychotherapy Process Scale (VPPS), and Working Alliance Inventory (WAI). accounted for about two-thirds of the data. Often they deliver varying assessments so are clearly measuring somewhat different facets of the alliance, but have been found to share the central common theme of a confident, collaborative relationship.
The review incorporated meta-analyses A study which uses recognised procedures to combine quantitative results from several studies of the same or similar interventions to arrive at composite outcome scores. Usually undertaken to allow the intervention's effectiveness to be assessed with greater confidence than on the basis of the studies taken individually. synthesising results from relevant studies to provide estimates of the overall strength of the link between outcomes and alliance, and to be able to probe for influences on the strength of that link. Strength is expressed as effect sizes A standard way of expressing the magnitude of a difference (eg, between outcomes in control and intervention groups) applicable to most quantitative data. Enables different measures taken in different studies to be compared or (in meta-analyses) combined. Based on expressing the difference in the average outcomes between control and experimental groups as a proportion of how much the outcome varies across both groups. By convention, in the behavioural sciences an effect size of the type used in the featured analyses is considered a small effect at 0.10, a medium effect at 0.30, and a large effect at 0.50. using the 'r' metric, which can be squared to calculate how much of the difference in outcomes can be attributed to differences in the therapy dimension being investigated. The assumption was made that there is no single, true strength of the link between outcomes and alliance which appears to vary only because of methodological differences, but that instead the strength really might vary across the studies included in the analyses.
The analyses included studies of five or more adult participants receiving genuine (as opposed to analogue, pretend, deliberately ineffective or placebo) treatments, where the author(s) referred to one of the variables assessed for its impact on outcomes as "alliance", "therapeutic alliance", "helping alliance", or "working alliance". In contrast to previous meta-analyses, the search was extended to material in Italian, German or French, as well as English. The 201 studies in the current analysis (offering 190 separate estimates of the alliance-outcomes link) is roughly double those in the latest previous meta-analysis. It seems fair to claim that the data is a reasonable representation of alliance-outcome research to date.
Overall the strength of the link between alliance and psychotherapy outcomes equated to an effect size of 0.275, a statistically significant link representing a moderate but highly reliable relationship which accounts for about 7.5% of the variance in outcomes. In other words, the more solid the working relationship or bond between therapist and client, the better the outcomes. Though it accounts for a relatively modest proportion of variance in treatment outcome, the alliance-outcome relationship is one of the strongest and most robust predictors of treatment success research has been able to document.
However, effect size varied across the studies more than would be expected by chance. Possible reasons for this variation were explored: how the alliance was measured; from whose perspective (client, therapist, or observer); how long in to therapy; the type of outcome; the type of treatment; and whether/how the study was published. Within all the resulting 22 categories, the alliance-outcome relation remained highly statistically significant, strongly indicating that it is ubiquitous irrespective of how alliance is measured, from whose perspective, when, how the outcome is evaluated, and the type of therapy. However, within each category there remained substantial variation in the strength of the link and, with one exception, the link was not significantly stronger in one category than another, indicating that other important but unknown influences were at work. Another possible influence on the link is whether both alliance and outcome are assessed by the same person; if they are, this might inflate the link because raters unconsciously think if one is good (or bad), the other will be too. But though diminished slightly to 0.25, the link remained statistically significant even when alliance and outcome were rated by different people.
The alliance is one of the most important influences on psychotherapy outcomes.
Development and fostering of the alliance is not separate from the interventions being delivered; it is influenced by and is an essential and inseparable part of everything that happens in therapy. How the therapist does the work of treatment indirectly and inevitably forges or weakens the alliance with the client. Alliance quality reflects the degree of mutual and collaborative commitment to therapy – how well therapist and client work together.
A 'good enough' alliance early in therapy is vital for success. It retains clients and creates a 'working space' with within which to try new ways of addressing the client's concerns.
Early in therapy, adjusting methods and tasks to suit the client helps build the alliance. Melding the client's expectations and personal resources with what the therapist believes is best is an important and delicate task. A strong alliance emerges, in part, as a result of the smooth coordination of these elements.
Therapist and client do not necessarily agree on the state of the alliance, raising the possibility that therapists will make the wrong assumptions about how the client sees it. Active monitoring of the client-perceived alliance is a recommended corrective.
Alliance strength can be expected to fluctuate as therapists address difficult issues, misunderstandings, transference, and so forth. As long as they are attended to and resolved, these variations are associated with good outcomes.
Responding non-defensively to a client's negativity or hostility is critical to a good alliance. Therapists have to learn neither to internalise nor ignore these emotions.
Therapists vary in the degree to which they foster an alliance with their clients, suggesting that alliance development is a skill and/or capacity therapists can develop.
commentary This article was in a special issue of the journal Psychotherapy devoted to effective therapist-client relationships. For other Findings entries from this issue see:
Evidence-based psychotherapy relationships: Psychotherapy relationships that work II
Evidence-based psychotherapy relationships: The alliance in child and adolescent psychotherapy
Evidence-based psychotherapy relationships: Alliance in couple and family therapy
Evidence-based psychotherapy relationships: Cohesion in group therapy
Evidence-based psychotherapy relationships: Empathy
Evidence-based psychotherapy relationships: Goal consensus and collaboration
Evidence-based psychotherapy relationships: Positive regard
Evidence-based psychotherapy relationships: Congruence/genuineness
Evidence-based psychotherapy relationships: Collecting client feedback
Evidence-based psychotherapy relationships: Repairing alliance ruptures
Evidence-based psychotherapy relationships: Managing countertransference
Evidence-based psychotherapy relationships: Research conclusions and clinical practices
The special issue which contained the article featured above was the second from the task force. The first was a special issue of the Journal of Clinical Psychology. While the second aimed to identify elements of effective therapist-client relationships ('What works in general'), the first aimed to identify effective ways of adapting or tailoring psychotherapy to the individual patient ('What works in particular'). For Findings entries from this first special issue see this bulletin. Both bodies of work have also been summarised in this freely available document from the US government's registry of evidence-based mental health and substance abuse interventions.
Last revised 26 May 2011. First uploaded