Clinician Burnout and Effectiveness of Guideline-Recommended Psychotherapies

JAMA Network Open | 2024, Vol. 7. No. 4

Clinician Burnout and Effectiveness of Guideline-Recommended Psychotherapies

Nina A. Sawyer - University of Minnesota, Minneapolis

Adam Kaplan - University of Minnesota, Minneapolis

David B. Nelson - University of Minnesota, Minneapolis

Shannon Wiltsey Stirman - Stanford University

Craig S. Rosen - Stanford University 

Clinician burnout has been linked to clinician outcomes, but its connection to patient outcomes is unclear. This cohort study took place in the US Veterans Affairs Health Care System and included licensed therapists providing trauma-focused psychotherapies who completed an online survey between May 2 and October 8, 2019, along with their patients who began trauma-focused therapy in the following year. Therapists rated burnout on a validated 5-point measure from the Physician Worklife Study, with scores of 3 or higher defined as burnout. The primary outcome was clinically meaningful improvement in PTSD symptoms on the PTSD Checklist for DSM-5. Of 180 therapists, 165 completed the burnout measure and treated 1,268 patients with outcome data. About 35% of therapists endorsed burnout, and 34% of patients showed clinically meaningful improvement. Improvement occurred in 28.3% of patients seen by therapists with burnout compared with 36.8% of those seen by therapists without burnout (OR = 0.63; 95% CI, 0.48-0.85). This held after adjusting for patient dropout (aOR = 0.56; 95% CI, 0.39-0.79) and session spacing (aOR = 0.65; 95% CI, 0.49-0.86). Each additional 3 days between sessions was tied to a 20% lower chance of improvement (OR = 0.8; 95% CI, 0.7-0.92). The authors conclude that therapist burnout was associated with reduced effectiveness of trauma-focused psychotherapies, and that understanding when and how burnout affects outcomes could guide workplace interventions.

Therapist burnout, posttraumatic stress disorder, trauma-focused psychotherapy, treatment effectiveness, Veterans Health Administration

“Burnout among US health care professionals is an epidemic that preceded and has been exacerbated by the COVID-19 pandemic. The toll of burnout includes physical and mental health symptoms as well as the intention to leave the profession, potentially worsening workforce shortages. Burnout may also affect the quality of care that clinicians who remain in the workforce provide. Clinicians experiencing burnout self-report lower quality of care, poorer communication with patients, and more medical errors. However, burnout is not consistently associated with evidence of poorer quality of care or worse outcomes. Health systems need information about whether reducing burnout may be important for treatment effectiveness.” (p. 2).

“We examined treatment outcome associated with burnout among licensed mental health professionals providing trauma-focused psychotherapies (TFPs) for posttraumatic stress disorder (PTSD) within the US Veterans Health Administration (VHA).” The primary aim of the study was to “evaluate the association between burnout and TFP effectiveness”, where the researchers hypothesized “reduced odds of clinically meaningful improvement among patients of therapists experiencing burnout.” The secondary aim was to “explain how burnout may affect outcomes if [the] first hypothesis was supported. This included an examination of whether patient dropout and measures of TFP implementation quality would mediate the association between therapist burnout and patient improvement. Patient dropout before receiving an adequate dose is common for TFPs and is associated with poorer outcomes. Alternatively, burnout might be associated with reduced effectiveness if therapists experiencing burnout diverge from the treatment protocols and deliver lower quality psychotherapy.” (p. 2).

“The need for novel and improved interventions for psychological trauma is well-recognized. However, it is equally important to address modifiable factors that limit the effectiveness of existing evidence-based interventions already in practice. These findings suggest that interventions to reduce therapist burnout might also result in more patients experiencing clinically meaningful PTSD symptom relief from guideline-recommended psychotherapies. Findings further confirm the benefit of providing an adequate dose in terms of the number of and days between sessions” (p. 10)

“How can we best take action to support clinicians and reduce burnout? Low-certainty evidence indicates that interventions to increase compassion, gratitude, or mindfulness may improve resilience of health care professionals. Yet, management interventions that support workplace control and improve staffing may be preferred over and more effective than those directed toward personal well-being. This is consistent with a view that while burnout manifests in individuals, it is rooted in systems and that system-level interventions to address the causes of occupational burnout are needed.” (p. 10)

“Strengths of this study include a prospective design, a national sample of therapists and patients, and the use of a patient-reported outcome rather than clinician self-rating of the quality of care. We also examined outcomes across clinicians delivering comparably effective interventions and controlled for patient case-mix. This study also has limitations. We used a single-item burnout measure that focused on the emotional exhaustion component of burnout and assessed burnout once before the COVID-19 pandemic. We controlled whether each therapy episode was initiated prior to, in the early stages of, or during the COVID-19 pandemic. However, a stronger design would have included repeated assessments of burnout using a more comprehensive, continuous measure and evaluated how changes in burnout were associated with therapy delivery and patient outcomes. While geographic variation in burnout is intriguing, interpretation is limited by heterogeneity within census regions. Unfortunately, there were too few therapists per medical center to examine smaller geographic levels. We also note that the therapist sample was homogeneous regarding race and ethnicity. Finally, factors other than burnout may have accounted for the difference in patient outcomes” (p. 10)