Skip to main content

Table 1 Description of studies retained by the literature review

From: Effect of burnout among physicians on observed adverse patient outcomes: a literature review

Study Country Design Sample N Response rate Measured dimensions of burnout Observed measure on quality of care Association with burnout
Fahrenkopf et al. 2008 [22] USA Observational prospective cohort study Paediatric residents 246 50% EE and DPa Missing or wrong prescription of drugs Not associated with adverse patient treatment
Zantinge et al. 2009 [23] Netherlands Observation of video-recorded consultations and questionnaire GPs 142 89% EE, DP and PAb Length of consultations, level of verbal communication, eye contact, and focus on psychosocial issues PA associated with GPs communicating less affectively, being less patient-centred and less eye contact. EE and DP not associated with adverse patient treatment
Kushnir et al. 2014 [24] Israel Cross-sectional observational study GPs/Primary care 136 99% EE, DP and PAc Number of referrals for diagnostic imaging, specialized health services or nurse sensitive treatments Associated with more referrals
Yuguero Torres et al. 2015 [25] Spain Prospective observational study (1 year) GPs 217 50% EE, DP and PAd Number of prescribed sick leaves Not associated with prescribing more or longer sick leaves
Garrouste-Orgeas et al. 2015 [26] France Prospective observational study (2-year) Doctors, nurses and care workers in intensive care unit 1988 (330 doctors) 77% EE, DP and PAe Medical errors (i.e., error of execution or error of planning). Adverse events were patient harms caused by medical interventions. Not associated with adverse patient treatment
Welp et al. 2015 [19] Switzerland Observational study Doctors and nurses in intensive care unit 1425 (243 doctors) Not specified EE, DP and PAf Length of stay in hospital and standardized mortality ratio Associated with increased standardized mortality ratio, but not length of stay
Pedersen et al. 2015 [27] Denmark Register study and questionnaire GPs 835 72% EE, DP and PAg Number of requisitions for PSA among healthy male patients Not associated with increased requisitions for PSA
Kwah et al. 2016 [28] USA Prospective observational study (1 year) First-year residents in internal medicine 54 98% (initial), 59% (cohort) EE and DPh Medication prescription errors with potential for adverse drug effects Not associated with increase in medical errors
Sun et al. 2017 [29] USA Cross-sectional observational study Primary care 102 Not specified EE, DP and PAf Antibiotic prescriptions for acute respiratory infections Not associated with increase in prescriptions
Noroxe et al. 2019 [20] Denmark Prospective observational study (6 months) GPs 781 50.2% EE, DP and PAi Conditions not requiring hospitalization in the case of appropriate intervention in primary care (ambulatory care sensitive conditions) Associated with increased frequency of hospitalizations of ambulatory care sensitive conditions
Noroxe et al. 2019 [21] Denmark Prospective observational study (6 months) GPs 409 50.2% EE, DP and PAi Change of GP (unrelated to change of address) Associated with increased likelihood of changing GPs
  1. Abbreviations: EE emotional exhaustion, DP depersonalization, PA personal accomplishment (reduced), GP general practitioner, PSA prostate-specific antigen, MBI Maslach Burnout Inventory
  2. aDefined burnout as EE > 27 and DP > 10
  3. bEE, DP and PA were dichotomized into low and high scores, and outcomes reported for each dimension
  4. cOverall burnout was measured as an average of the responses to items EE and DP (continuous variable)
  5. dScores for each category were divided into low, moderate and high
  6. eBurnout measured as
  7. (1) a combination of high EE and DP with low PA (dimensions dichotomized), and
  8. (2) defined as over a cut-off of a global MBI score of 9
  9. fEach dimension used as a continuous variable
  10. gBurnout defined as high level of emotional exhaustion > 26
  11. hBurnout defined as a high EE or high DP subscore (dichotomized variables)
  12. iBurnout measured
  13. 1) in four quartiles for each dimension
  14. 2) As a composite score by adding up points corresponding to the quartile of each subscale (reversed score for personal accomplishment);
  15. one point for scores in the first quartile, and two, three and four points for subsequent quartiles, respectively. The composite score was categorized into five groups, with increasing score indicating higher burnout