|Mindfulness-based therapy: a comprehensive meta-analysis
|Khoury B, Lecomte T, Fortin G, Masse M, Therien P, Bouchard V, Chapleau MA, Paquin K, Hofmann SG
This review concluded that mindfulness-based therapy was moderately to largely effective for a variety of psychological problems, particularly for reducing anxiety, depression and stress. The authors’ conclusions may be overstated, given the poor quality and wide variation between studies.
To assess the effects of mindfulness-based therapy.
PubMed and PsycINFO were searched for relevant articles included up to May 2013, in any language. Search terms were reported.
Eligible was any study evaluating the before-and-after or controlled effects of mindfulness-based therapy for physical or medical conditions, for psychological disorders, or for a non-clinical population. Studies were excluded if they compared mediation styles or mediators, examined the non-direct effects of mindfulness, used mindfulness as a component of another treatment, or were based on meditation instruction, induction or retreats. Studies had to report sufficient information to calculate effect sizes for clinical (physical or psychological) or mindfulness outcomes.. Case studies were excluded.
In the included studies, a variety of mindfulness therapies were used, including mindfulness-based awareness processes, stress reduction, cognitive therapy, relapse prevention and yoga. Controlled studies used either a treatment or waiting list, as the control. The most common disorders were mood and cancer (some studies included caregivers), anxiety, pain, alcohol or substance abuse, fibromyalgia, overweight or obesity, social anxiety or social phobia, HIV, post-traumatic stress disorder, and headache. Many studies included more than one disorder. Outcomes were measured by a variety of instruments (details in the paper).
One reviewer selected studies for inclusion and decisions were checked by a second reviewer. Disagreements were resolved through discussion, or by contacting the original study's authors.
Assessment of study quality
Study quality was assessed, based on a modified Jadad scale. The criteria covered adherence to an established protocol, administration of measures at follow-up, use of validated mindfulness measures, and clinical and mindfulness training of therapists; the maximum score was 5. For controlled studies, the additional criteria covered: randomisation to groups, comparability of groups, and blinding; the maximum score appeared to be 11.
It seems that five reviewers assessed study quality.
Means and standard deviations were extracted, or where these were unavailable, other statistics were extracted, to calculate effect sizes (Hedges' g), 95% confidence intervals, and 95% prediction intervals, for physical, psychological and mindfulness measures. For within-group studies, where data were not available, a conservative estimate for missing or unclear data was made, according to Rosenthal's methods.
The authors did not report how many reviewers extracted the data.
Methods of synthesis
Pooled Hedges' g, with 95% confidence and prediction intervals, were calculated using a random-effects model. Statistical heterogeneity was assessed using the X² and I²; an I² of 25% was considered low, 50% was moderate, and 75% or more was high.
Meta-regression was used to assess the relationship between the pooled effect size and various moderators – treatment length, duration of home practice, therapist clinical training, therapist mindfulness training, study quality score, age of participants, and year of publication. A clinical significance analysis assessed the clinical implications of mindfulness-based therapy, for psychological outcomes, using several psychological and anxiety measures.
Publication bias was assessed using Rosenthal’s fail-safe N, and in a funnel plot.
Results of the review
There were 209 studies, with 12,145 participants; 109 were randomised controlled trials (RCTs), 26 were non-randomised controlled studies, 72 were uncontrolled before-and-after studies, and two only reported follow-up data. The controlled studies had a mean quality score of 4.84 (out of 11). The before-and-after studies had a mean quality score of 2.93 (out of 5). In studies with follow-up, it ranged from three weeks to three years. Attrition ranged from none to 81.5%.
For clinical outcomes, at the end of treatment, mindfulness-based therapy was significantly more effective than psychological education (Hedges' g 0.61, 95% CI 0.27 to 0.96; nine studies; I²=83%), supportive therapy (Hedges' g 0.37, 95% CI 0.17 to 0.57; seven studies; I²=64%), relaxation procedures (Hedges' g 0.19, 95% CI 0.03 to 0.35; eight studies; I²=59%), and imagery or suppression techniques (Hedges' g 0.26, 95% CI 0.10 to 0.53; two studies; I²=0).
There were no significant differences between mindfulness-based therapy and traditional cognitive-behavioural therapy or other behavioural therapy (nine studies), or pharmacological treatments (three studies).
Analysing before-and-after and waiting-list controlled studies separately, at the end of treatment the effect sizes for mindfulness-based therapy were larger for treatment for psychological disorders, than for physical or medical conditions.
There was a large, significant effect for anxiety in 10 before-and-after studies (Hedges' g 0.89, 95% CI 0.71 to 1.08; I²=14%), and four waiting-list controlled studies (Hedges' g 0.96, 95% CI 0.67 to 1.24). There was a moderate, significant effect for depression in five before-and-after studies (Hedges' g 0.69, 95% CI 0.52 to 0.86) and eight waiting-list controlled studies (Hedges' g 0.53, 95% CI 0.32 to 0.73). There was a moderate to high, significant effect for non-clinical populations in 18 before-and-after studies (Hedges' g 0.65, 95% CI 0.51 to 0.80) and 16 waiting-list controlled studies (Hedges' g 0.62, 95% CI 0.42 to 0.82). Statistical heterogeneity was reported to be high (details not given).
The results at follow-up were similar to those at the end of treatment (details reported).
For mindfulness outcomes, the effects of mindfulness-based therapy, at the end of treatment were lower in 23 treatment controlled trials (Hedges' g 0.42, 95% CI 0.27 to 0.57) than in 28 waiting-list controlled studies (Hedges' g 0.53, 95% CI 0.42 to 0.65) and 42 before-and-after studies (Hedges' g 0.69, 95% CI 0.59 to 0.80). Statistical heterogeneity was reported to be moderate (details not given). The effects were similar at follow-up. Other results were reported.
There was no evidence of publication bias. The clinical significance analyses corroborated the findings, showing that the severity of the initial anxiety and depression was reduced after treatment. The results of the meta-regression were reported.
Mindfulness-based therapy was moderately to largely effective for a variety of psychological problems, especially for reducing anxiety, depression and stress.
The review question and inclusion criteria were clearly defined. Only two databases were searched, but there were no language restrictions and it appears that unpublished data were included, reducing the potential for publication bias. The formal assessment of publication bias found no evidence of it. Study quality was assessed and scores were reported for each study. It appears that quality was generally low for controlled studies; it was better for before-and-after studies, but this design is open to several sources of bias. Many studies were uncontrolled, which prevents definitive conclusions.
Appropriate methods to reduce reviewer error and bias were used for selecting studies and assessing quality, but it was unclear whether they were used for data extraction. The methods of analysis seem to have been appropriate and various moderator analyses were conducted. Statistical variation was high for a number of analyses. The authors acknowledged the wide variation between the studies in their design, interventions, participants, outcomes and quality. Attrition rates were high in some studies.
The authors’ conclusions may be overstated given the poor quality and wide variation between studies.
Implications of the review for practice and research
Practice: The authors stated that mindfulness was a central component of treatment effectiveness, and that the mindfulness of participants and therapists was a strong predictor of effective therapy.
Research: The authors stated that further studies, with sound methods, were needed to establish the effectiveness of mindfulness-based therapy, compared with, or in addition to, standard treatments, particularly cognitive-behavioural therapy. Studies were needed to thoroughly examine and quantify the moderators and mediators of effective treatment.
Khoury B, Lecomte T, Fortin G, Masse M, Therien P, Bouchard V, Chapleau MA, Paquin K, Hofmann SG. Mindfulness-based therapy: a comprehensive meta-analysis. Clinical Psychology Review 2013; 33(6): 763-771
Subject indexing assigned by NLM
Anxiety Disorders /psychology /therapy; Depressive Disorder /psychology /therapy; Humans; Mindfulness; Stress, Psychological /psychology /therapy; Treatment Outcome
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.