Flowchart of the inclusion of studies in the review.
Meta-analysis of individual study and pooled effects. Each line represents 1 study in the meta-analysis, plotted according to the standardized mean difference (SMD; roughly the difference between the mean score of participants in the intervention group and the mean score of participants in the control group). The squares show the SMD for each study, and the diamond represents the pooled SMD. Weights are from random-effects model.
Subgroup analysis of individual study and pooled effects of physician-directed and organization-directed interventions on burnout scores. Each line represents 1 study in the meta-analysis, plotted according to the standardized mean difference (SMD). The squares show the SMD for each study, and the diamond represents the pooled SMD. Weights are from random-effects model.
Funnel plot with pseudo 95% confidence intervals. The outer lines indicate the triangular region within which 95% of studies are expected to lie in the absence of both biases and heterogeneity. The funnel plot shows no substantial asymmetry (Egger regression intercept −0.28, SE = 0.16, P = .11).56
eTable 1. Completed PRISMA checklist
eMethods 1. Systematic review protocol
eMethods 2. Search strategy Medline
eMethods 3. List of excluded studies
eTable 2. Burnout measure and main findings of included studies
eFigure 1. Study ratings in the critical appraisals tool
eFigure 2. Forest plot of the effects of interventions on burnout scores in physicians with low and extensive working experience
eFigure 3. Forest plot of the effects of interventions on burnout scores in primary care and secondary care physicians
eFigure 4. Forest plot of the effects of interventions on burnout scores across studies with low risk of bias ratings
eFigure 5. Forest plot of the effects of interventions on depersonalization domain of Maslach Burnout Inventory
eFigure 6. Forest plot of the effects of interventions on personal accomplishment domain of Maslach Burnout Inventory
eTable 3. Results of subgroup analyses on depersonalization and personal accomplishment domains of MBI
Customize your JAMA Network experience by selecting one or more topics from the list below.
Panagioti M, Panagopoulou E, Bower P, et al. Controlled Interventions to Reduce Burnout in PhysiciansA Systematic Review and Meta-analysis . JAMA Intern Med. 2017;177(2):195–205. doi:10.1001/jamainternmed.2016.7674
Copyright 2016 American Medical Association. All Rights Reserved.
Are interventions for reducing burnout in physicians effective?
This meta-analysis of 20 controlled interventions on 1550 physicians found that existing interventions were associated with small and significant reductions in burnout. The strongest evidence for effectiveness was found for organization-directed interventions, but these interventions were rare.
More effective models of interventions are needed to mitigate risk for burnout in physicians. Such models could be organization-directed approaches that promote healthy individual-organization relationships.
Burnout is prevalent in physicians and can have a negative influence on performance, career continuation, and patient care. Existing evidence does not allow clear recommendations for the management of burnout in physicians.
To evaluate the effectiveness of interventions to reduce burnout in physicians and whether different types of interventions (physician-directed or organization-directed interventions), physician characteristics (length of experience), and health care setting characteristics (primary or secondary care) were associated with improved effects.
MEDLINE, Embase, PsycINFO, CINAHL, and Cochrane Register of Controlled Trials were searched from inception to May 31, 2016. The reference lists of eligible studies and other relevant systematic reviews were hand searched.
Randomized clinical trials and controlled before-after studies of interventions targeting burnout in physicians.
Data Extraction and Synthesis
Two independent reviewers extracted data and assessed the risk of bias. The main meta-analysis was followed by a number of prespecified subgroup and sensitivity analyses. All analyses were performed using random-effects models and heterogeneity was quantified.
Main Outcomes and Measures
The core outcome was burnout scores focused on emotional exhaustion, reported as standardized mean differences and their 95% confidence intervals.
Twenty independent comparisons from 19 studies were included in the meta-analysis (n = 1550 physicians; mean [SD] age, 40.3 [9.5] years; 49% male). Interventions were associated with small significant reductions in burnout (standardized mean difference [SMD] = −0.29; 95% CI, −0.42 to −0.16; equal to a drop of 3 points on the emotional exhaustion domain of the Maslach Burnout Inventory above change in the controls). Subgroup analyses suggested significantly improved effects for organization-directed interventions (SMD = −0.45; 95% CI, −0.62 to −0.28) compared with physician-directed interventions (SMD = −0.18; 95% CI, −0.32 to −0.03). Interventions delivered in experienced physicians and in primary care were associated with higher effects compared with interventions delivered in inexperienced physicians and in secondary care, but these differences were not significant. The results were not influenced by the risk of bias ratings.
Conclusions and Relevance
Evidence from this meta-analysis suggests that recent intervention programs for burnout in physicians were associated with small benefits that may be boosted by adoption of organization-directed approaches. This finding provides support for the view that burnout is a problem of the whole health care organization, rather than individuals.
Burnout is a syndrome consisting of emotional exhaustion, depersonalization, and a diminished sense of personal accomplishment, which is primarily driven by workplace stressors.1(pp191-218)2 Burnout is a major concern for physicians. Nearly half of practicing physicians in the United States experience burnout at some point in their career.3 Although there are substantial differences by specialty, physicians at the front line of care report the highest rates of burnout.4
Burnout has serious negative consequences for physicians, the health care system, and for patient outcomes. Burnout in physicians has been linked with lower work satisfaction, disrupted personal relationships, substance misuse, depression, and suicide.5,6 Within health care organizations, burnout is related to reduced productivity, high job turnover, and early retirement.7-9 Importantly, burnout can result in an increase in medical errors, reduced quality of patient care, and lower patient satisfaction.10-15 It is not surprising, therefore, that wellness of physicians is increasingly proposed as a quality indicator in health care delivery.16
Leading drivers of burnout include excessive workload, imbalance between job demands and skills, a lack of job control, and prolonged work stress.17 Recently, there has been a shift from viewing burnout as an individual problem to a problem of the health care organization as a whole, rooted in issues related to working environment and organizational culture.18 It has been suggested that reducing risk of burnout in physicians requires change in organizations, as well as support for individual physicians.19
Interventions for burnout can be classified into 2 main categories, physician-directed interventions targeting individuals and organization-directed interventions targeting the working environment.20,21Quiz Ref ID Physician-directed interventions typically involve mindfulness techniques or cognitive behavioral techniques to enhance job competence and improve communication skills and personal coping strategies. Organization-directed interventions can involve simple changes in schedule and reductions in the intensity of workload or more ambitious changes to the operation of practices and whole health care organizations. These usually involve improved teamwork, changes in work evaluation, supervision to reduce job demand and enhance job control, and increasing the level of participation in decision making.
We conducted a systematic review and meta-analysis of studies that evaluated interventions to reduce burnout in physicians. We decided to focus on burnout scores as the main outcome of this review because burnout is the best-recognized serious negative consequence of work stress in physicians18,22 and the most commonly reported, and consistently measured, outcome of work stress interventions.20,21,23 Moreover, by focusing on burnout, we established a level of homogeneity in terms of outcomes that allowed us to test our aims meta-analytically.
Our first objective was to assess the effectiveness of interventions in reducing burnout. Second, we examined what types of interventions are the most effective (organization directed, physician directed). Third, we examined whether there are any differences in the effect of interventions in different health care settings (primary care, secondary or intensive care) and in physicians with different levels of working experience. Our rationale was that physicians working in different organizational settings or physicians with different levels of experience might have diverse needs and might respond differently to burnout interventions.
The reporting of the review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (eTable 1 in the Supplement).24 The protocol is included in eMethods 1 in the Supplement.
The study population comprised physicians of any specialty in the primary, secondary, or intensive care setting including residents and fellows. Studies based on a mix of physicians and other health care professionals were included in the review if the physicians made up at least 70% of the sample.
Eligible interventions were any intervention designed to relieve stress and/or improve performance of physicians and reported burnout outcomes including physician-directed interventions and organization-directed interventions. Physician-directed interventions focused on individuals (eg, cognitive behavioral therapies, mindfulness-based stress reduction techniques, educational programs for improving communication skills) whereas organization-directed interventions introduced changes in the resources, the working environment, and/or work tasks to decrease stress (eg, changes in the intensity and/or schedule of the workload or deeper improvements in the operation of health care organizations and teamwork).
Eligible comparisons included any type of control (eg, waiting list or no intervention). Outcome was burnout measured using validated tools such as the Maslach Burnout Inventory (MBI)1 or other validated measures of burnout. Eligible study designs were quantitative intervention designs described in the Cochrane handbook including randomized clinical trials, nonrandomized trials, controlled before-after studies, and interrupted time series. Context was any health care setting including primary care and secondary care.
Interventional studies not reporting data on burnout outcomes but providing data on general stress, well-being, or job satisfaction were excluded, as was gray literature.
Five electronic bibliographic databases were searched from inception until May 31, 2016: MEDLINE, Embase, CINAHL, Cochrane Register of Controlled Trials, and PsycINFO. The search strategy included combinations of 3 key blocks of terms (burnout; physicians; interventions) using medical subject headings (MESH terms) and text words (eMethods 2 in the Supplement). Searches were supplemented by hand searches of the reference lists of eligible studies and systematic reviews.
The results of the searches were exported in Endnote and duplicates were removed. Study selection was completed in 2 stages. First, the titles and abstracts of the studies were screened and subsequently the full texts of relevant studies were accessed and further screened against the eligibility criteria. The title and abstract screening was undertaken by M. P., whereas 2 independent reviewers were involved in full-text screening. Interrater reliability was high (κ = 0.96). Disagreements were resolved through discussions.
An Excel data extraction form was developed and initially piloted in 5 randomly selected studies. Quantitative data for meta-analysis were extracted on a separate extraction sheet. Authors were contacted when data were missing or incomplete. The following descriptive information was extracted from the studies:
Study: research design, method of recruitment, and content of control
Participants: sample size, age, sex, setting and/or specialty, years of work experience
Intervention: content, delivery format, intensity, follow-up time points
Outcomes: scores in burnout including emotional exhaustion, depersonalization, and professional accomplishment.
The critical appraisal of the studies was performed using the Effective Practice and Organisation of Care (EPOC) risk of bias tool.25 It was chosen because it is appropriate for use across all types of intervention designs described in the Cochrane handbook. The EPOC tool contains 9 standardized criteria scored on a 3-point scale, corresponding to low, unclear, and high risk.
Standardized mean differences (SMDs) and associated confidence intervals for the burnout outcomes of all the studies were calculated in Comprehensive Meta-Analysis.26 The pooled SMDs and the forest plots were computed using the metaan command in Stata 14.27 The main meta-analysis evaluated the effectiveness of the interventions in reducing burnout. Quiz Ref IDThe MBI measure for burnout provides ratings in 3 domains (emotional exhaustion, depersonalization, and personal accomplishment). It is not recommended that they be combined.1 In line with previous meta-analyses, we used only the emotional exhaustion domain of MBI in the analyses.23 Emotional exhaustion is considered the most central aspect of burnout (some studies only use this domain), and other unidimensional measures of burnout focus on emotional exhaustion.23,28 To ease the interpretation of the results we “back-transformed” the pooled SMD to a mean difference for the emotional exhaustion subscale, under certain assumptions. When data were available for more than 1 follow-up assessment point, the short-term assessment points were inserted in the main analysis. Three prespecified subgroup analyses29 were carried out:
Type of interventions—we tested the effectiveness of physician-directed and organization-directed interventions.
Working experience of physicians—we examined the differential treatment effects across studies that recruited physicians with extensive working experience (mean of ≥5 years) and studies that recruited physicians with low experience (mean of <5 years). All studies classified into the low-experience category explicitly reported in the Methods that they recruited junior physicians.
Health care setting—we tested the effects of interventions separately in physicians based in primary care and in secondary care.
Two sensitivity analyses were performed. We examined the effects of interventions on the other 2 domains of MBI (depersonalization and personal accomplishment). We also examined whether effects were robust when only studies with low risk of bias scores were retained in the analyses.
Heterogeneity was assessed using the I2 statistic. Conventionally, I2 values of 25%, 50%, and 75% indicate low, moderate, and high heterogeneity.30 All analyses were conducted using a random-effects model, even if I2 was low. Random-effects models are more conservative and have better properties in the presence of any heterogeneity.31,32 The Cohen Q test of between-group variance was used to test whether the effectiveness of burnout interventions is significantly different across subgroups. Cluster randomized clinical trials were identified and the precision of analyses adjusted using a sample size/variation inflation method, assuming an intraclass correlation of 0.02. Provided that we identified 10 or more studies,33 we aimed to use funnel plots and the Egger test to assess small-sample bias (an indicator of possible publication bias).34 Funnel plots were constructed using the metafunnel command,35 and the Egger test was computed using the metabias command.36
As shown in Figure 1, the search strategy yielded 2322 articles. Following the removal of duplicates, 1723 articles were retained for title and abstract screening. Of these, 75 were relevant for full-text screening and 19 studies were included in the review.37-55 One study included a lower percentage of physicians (67%), but we retained it in the analyses to maximize the evidence base.39
The Table presents the characteristics of the 19 studies (including 20 independent comparisons on 1550 physicians; mean [SD] age, 40.3 [9.5] years). Eight studies were conducted in the United States (42%), 4 in Europe, 3 in Australia, 2 in Canada, 1 in Argentina, and 1 in Israel. An equal proportion of men and women were recruited in the majority of studies.
Seven studies recruited physicians working in primary care (mostly labeled “general practitioners”), 10 studies recruited physicians in secondary care (eg, physicians in intensive care units, oncologists, and surgeons), and 2 studies recruited a mixed sample of physicians through their registration in national medical associations. Across all interventions, the main eligibility criteria were being a physician (working in a specific setting in most cases) and willingness to take part in the study. None of the studies specifically targeted physicians with certain severity levels of burnout. The majority of studies (n = 12 [67%]) were based on experienced physicians (mean working experience of ≥5 years) whereas 7 studies were based on recently qualified physicians (mean working experience of <5 years). With the exception of 1 study,37 all used the MBI to assess the severity of burnout (eTable 2 in the Supplement).
Interventions varied considerably in their characteristics including content, duration/intensity, and length of postintervention assessment points (see Table). The majority (n = 12 [60%]) were physician-directed interventions that comprised mindfulness-based stress reduction techniques, educational interventions targeting physicians’ self-confidence and communication skills, exercise, or a combination of these features.
Within the category of organization-directed interventions, 5 studies evaluated simple workload interventions that focused on rescheduling hourly shifts and reducing workload. Only 3 studies tested more extensive organization-directed interventions incorporating discussion meetings to enhance teamwork and leadership, structural changes, and elements of physician interventions such as communication skills training and mindfulness.
The duration of the interventions ranged from 2 weeks to 9 months. Follow-up assessment points ranged from 1 day to 18 months after the intervention. All interventions were delivered in face-to-face format.
The results of the risk of bias assessment are presented in eFigure 1 in the Supplement. Eighteen comparisons were randomized clinical trials (95%) whereas 2 were controlled before-and-after studies. Fifteen comparisons (75%) fulfilled 6 of the 9 risk of bias criteria (a higher score indicates lower vulnerability to bias). Three comparisons fulfilled 8 or 9 criteria (17%) while 5 fulfilled 4 or fewer criteria (25%); most moderately accounted for the risk of bias criteria.
Interventions were associated with small, significant reductions in burnout (SMD = −0.29; 95% CI, −0.42 to −0.16; I2 = 30%; 95% CI, 0 to 60%) (Figure 2). The back-transformed emotional exhaustion score for the intervention group was 15.1 (95% CI, 13.9 to 16.5), compared with a control group score of 17.9 and assuming a standard deviation of 8.97 for the effect.
Physician-directed interventions were associated with small significant reductions in burnout (SMD = −0.18; 95% CI, −0.32 to −0.03; I2 = 11%; 95% CI, 0 to 49%; back-transformed emotional exhaustion score = 16.2; 95% CI, 14.7 to 17.3 compared with a control group score of 17.9) whereas organization-directed interventions were associated with medium significant reductions in burnout (SMD = −0.45; 95% CI, −0.62 to −0.28; I2 = 8%; 95% CI, 0 to 60%; back-transformed emotional exhaustion score = 13.9; 95% CI, 12.4 to 14.7 compared with a control group score of 17.9) (Figure 3). The effects of organization-directed interventions were significantly larger than the effects of physician-directed interventions (Cohen Q = 4.15, P = .04).
The pooled effect of interventions on burnout scores was medium and significant across studies mainly based on experienced physicians (SMD = −0.37; 95% CI, −0.58 to −0.16; I2 = 42%; 95% CI, 0 to 70%; back-transformed emotional exhaustion score = 14.6; 95% CI, 12.7 to 16.5 compared with a control group score of 17.9) and small and significant across studies on physicians with limited experience (SMD = −0.27; 95% CI, −0.40 to −0.14; I2 = 0%; 95% CI, 0 to 75%; back-transformed emotional exhaustion score = 15.5; 95% CI, 13.8 to 16.9 compared with a control group score of 17.9) (eFigure 2 in the Supplement). This group difference was nonsignificant (Q = 0.92, P = .34).
Interventions in primary care were associated with small to medium reductions in burnout (SMD = −0.39; 95% CI, −0.59 to −0.19; I2 = 4%; 95% CI, 0 to 69%; back-transformed emotional exhaustion score = 14.4; 95% CI, 12.6 to 16.2 compared with a control group score of 17.9). Interventions in secondary care were associated with small significant reductions in burnout (SMD = −0.24; 95% CI, −0.41 to −0.07; I2 = 41%; 95% CI, 0 to 65%; back-transformed emotional exhaustion score = 15.7; 95% CI, 13.9 to 17.4 compared with a control group score of 17.9) (eFigure 3 in the Supplement). This difference was nonsignificant (Q = 0.51, P = .48).
The treatment effect derived by studies at lower risk of bias (ie, scoring low on 6 of the 9 risk of bias criteria) was similar to the overall effects of the main analysis (SMD = −0.32; 95% CI, −0.49 to −0.14; I2 = 42%; 95% CI, 0 to 70%) (eFigure 4 in the Supplement).
Interventions were associated with very small significant reductions in depersonalization (SMD = −0.21; 95% CI, −0.35 to −0.06; I2 = 33%; 95% CI, 0 to 68%) (eFigure 5 in the Supplement) and small improvements in personal accomplishment (SMD = 0.30; 95% CI, 0.15 to 0.45; I2 = 0; 95% CI, 0 to 58%) (eFigure 6 in the Supplement). The subgroup analyses in these 2 domains showed similar results but were based on a smaller number of studies (eTable 3 in the Supplement).
We found no evidence of funnel plot asymmetry, which might indicate publication bias for the main, or subgroup analyses (Egger test P = .11 for main analysis) (Figure 4).
Quiz Ref IDThis meta-analysis showed that interventions for physicians were associated with small significant reductions in burnout. Organization-directed interventions were associated with higher treatment effects compared with physician-directed interventions.Quiz Ref ID Interventions targeting experienced physicians and delivered in primary care showed evidence of greater effectiveness compared with interventions targeting less experienced physicians and delivered in secondary care, but these group differences were nonsignificant.
This is a comprehensive meta-analysis of controlled interventions aimed at reducing physician burnout. Quiz Ref IDThe 2 greatest threats to the validity of meta-analysis are heterogeneity and publication bias. However, the biggest strength of this work is the large number of identified and meta-analyzed controlled comparisons (20, when approximately 11.5% of all meta-analyses include ≥10 studies), which allows us to reliably estimate and model heterogeneity levels.57 In addition, the size of the meta-analysis allowed us to assess publication bias with adequate power.33 Although publication bias tests are rarely conclusive, we did not observe any bias indications in the plot or test.
The included studies differed significantly in terms of content of interventions, study design and/or quality, and length of follow-up that limit the extent to which broad conclusions can be drawn about the overall effectiveness of physician interventions. However, estimates of heterogeneity in the pooled analyses were low to moderate by conventional thresholds and random-effects models were applied in all analyses.58 Heterogeneity was further addressed by conducting prespecified subgroup analyses (within the limits of power).59 While this is a useful approach for producing guidance to design and deliver the most effective interventions, subgroup analyses should be interpreted cautiously because other, uncontrolled differences between studies might account for the results.60,61
Three existing systematic reviews have examined the effectiveness of work stress interventions in health care professionals, with only 1 of these specifically focused on physicians.21,62,63 Our findings regarding the overall effectiveness of burnout interventions and the increased effectiveness of organizational interventions are in agreement with the most recent meta-analysis on physician burnout.63 In comparison, we narrowed our attention to controlled interventions and we undertook additional evidence-based prespecified subgroup analyses to examine whether the characteristics of interventions, physicians, and health care settings influenced the overall effect of burnout interventions. This decision was based on the recognition that controlled interventions offer the best opportunity to reach rigorous conclusions about the effectiveness of the tested interventions and that intervention studies on physician burnout are highly heterogeneous. This approach enabled us to draw informative conclusions regarding the effectiveness of burnout interventions among physicians that take into account the influence of the distinct features of interventions, physicians, and health care settings.
Even though many studies have examined risk factors for burnout in physicians, relatively few intervention programs have been developed and evaluated. Our main finding is that the treatment effects were significant but small, equal to a 3-point reduction in the emotional exhaustion domain of the MBI. At present, the low quality of the research evidence does not allow firm practical recommendations, but we offer some insights for research and clinical directions.
Organization-directed interventions were more likely to lead to reductions in burnout, but there were large variations in terms of actual approaches, intervention ingredients, and intensity. Those that combined several elements such as structural changes, fostering communication between members of the health care team, and cultivating a sense of teamwork and job control tended to be the most effective in reducing burnout.45 However, such intense organization-directed interventions were rare and were not evaluated widely. The majority of organization-directed interventions that we included in the analyses introduced simple reductions in the workload or schedule changes. Concerns about implementation and delivery costs of organization-directed interventions, especially if they involve complex and major health care system changes, might explain their scarcity.20,64 A recent example promoting healthy individual-organization relationships is the Listen-Act-Develop model implemented in Mayo Clinic.65 Large-scale cluster-randomized trials of such programs at the institutional or even at the national level that emphasize organizational culture by creating a safe space for staff to acknowledge and decrease stress are possibly an optimal framework for mitigating burnout.
Physician-directed interventions led to very small significant reductions in burnout. We found no evidence that the content (eg, mindfulness, communicational, educational components) or intensity of these interventions might increase the derived benefits based on our critical review. This finding, in combination with the larger effects of organization-directed interventions, supports the argument that burnout is rooted in the organizational coherence of the health care system.19,66 If burnout is a problem of whole health care systems, it is less likely to be effectively minimized by solely intervening at the individual level. It requires an organization-embedded approach.19 Moreover, physicians expected to deal with burnout individually and remotely from their practicing organization might view physician-directed interventions as a personal responsibility (or blame themselves for being less “resilient”) rather than as a shared resource to create a flourishing health care environment.65,67 There is some evidence that elements of the physician-directed interventions (eg, mindfulness) are effective when supported by organizational approaches.23,55 However, other unexamined factors at the process of the intervention delivery or at the participant level might account for the observed differences in the effectiveness of organization-directed and physician-directed interventions. Research programs to understand the best context for the delivery, evaluation, and implementation of burnout interventions are required.68-70
Physicians based in different health care settings or at different stages of their career might face unique challenges and have different needs. We found smaller benefits for recently qualified and secondary care physicians. The evidence indicates that young physicians are at higher risk for burnout compared with experienced physicians,4 so future research should focus on prevention among less experienced physicians. Interventions focused on enhancing teamwork, mentoring, and leadership skills might be particularly suitable for young physicians and for physicians dealing with intense work and patients with complex care needs.71-73
This meta-analysis found that physicians could gain important benefits from interventions to reduce burnout, especially from organization-directed interventions. However, this evidence is derived from interventions developed and evaluated in diverse groups of physicians and health care settings. Burnout is associated with serious risks to both physicians and patients; thus, it is imperative that physicians have access to evidence-based interventions that reduce the risk for burnout.
Accepted for Publication: September 12, 2016.
Corresponding Author: Maria Panagioti, PhD, NIHR School for Primary Care Research, Manchester Academic Health Science Centre, Oxford Rd, Williamson Bldg, Manchester M13 9PL, United Kingdom (email@example.com).
Published Online: December 5, 2016. doi:10.1001/jamainternmed.2016.7674
Author Contributions: Dr Panagioti had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Panagioti, Chew-Graham, van Marwijk, Esmail.
Acquisition, analysis, or interpretation of data: Panagopoulou, Bower, Lewith, Kontopantelis, Dawson, van Marwijk, Geraghty.
Drafting of the manuscript: Panagioti, Chew-Graham, Dawson, van Marwijk.
Critical revision of the manuscript for important intellectual content: Panagioti, Panagopoulou, Bower, Lewith, Kontopantelis,
Dawson, van Marwijk, Geraghty, Esmail.
Statistical analysis: Panagioti, Bower, Kontopantelis, Dawson.
Obtained funding: Panagioti, Chew-Graham, Esmail.
Administrative, technical, or material support: Panagioti, Dawson, Geraghty.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was funded by the UK National Institute of Health Research (NIHR) School for Primary Care Research (Study No. R119013). The Medical Research Council Health eResearch Centre grant MR/K006665/1 supported the time and facilities of Dr Kontopantelis.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, the NIHR, or the Department of Health.
Create a personal account or sign in to: