Editorial NexoScience
The shift starts at seven in the morning and the doctor walking into it has already run out of patience for it. Not tired the way a long week makes anyone tired, but hollowed out, detached from the names on the list, and quietly convinced the work no longer amounts to much. Medicine has a name for that state, burnout, and the bill for it has almost always been counted in the careers it wrecks.
The bill does not stop at the doctor. Physicians who met the criteria for burnout were about twice as likely to be involved in a patient safety incident, things like prescribing errors, missed monitoring or potentially avoidable readmissions, as physicians who were not burned out (odds ratio 2.04). They were also just over twice as likely to show lapses in professionalism, and more than twice as likely to be rated poorly by the patients they had just treated.
The size of the analysis is what makes the pattern hard to wave away. Published in The BMJ in September 2022 by a team led by researchers at the University of Manchester, it pooled 170 observational studies covering 239,246 physicians, a headcount comparable to a mid-sized city, and all of them working doctors rather than mixed samples of health staff. Earlier reviews had looked at pieces of this question, but the data were so uneven that most of them settled for narrative summaries instead of pooled numbers.
The researchers built the picture out of work that was already published. They searched Medline, PsycINFO, Embase and CINAHL from the databases’ beginnings up to May 2021, kept only observational studies of physicians, and combined the results with random effects models, an approach that assumes the true effect varies from setting to setting instead of pretending every hospital is the same. They then emailed every original author to check the extracted figures, and about half of the research teams wrote back and confirmed them.
The career side of the ledger runs steeper than the patient side. Burnout was associated with almost four times the odds of job dissatisfaction compared with satisfaction (3.79), roughly three times the odds of intending to leave the job rather than stay, and a similar jump in regretting the choice of medicine as a career in the first place. In the authors’ reading, that is the primary damage, with the quality of care sitting downstream of it.
The two faces of burnout appear to do different kinds of harm. Emotional exhaustion, the drained, depleted dimension, tracked most strongly with wanting out of the job. Depersonalisation, the cynical distancing that turns a patient into a case, tracked most strongly with worse care and unhappier patients. The researchers also note the arrow can run backwards. Being caught up in a serious safety incident is itself a known route into burnout, a loop sometimes described as secondary trauma, which means cause and effect are tangled here rather than lined up neatly.
That tangle is the study’s central limitation, and the authors say so plainly. Most of the 170 studies were cross-sectional, a snapshot of doctors at a single moment, which the authors write leaves them unable to establish causal links between burnout and either patient care or career decisions. Heterogeneity was high in nearly every comparison, safety incidents were defined broadly, more than three quarters of the safety and professionalism data came from physicians reporting on themselves, and the team flagged both publication bias and the fact that only English-language studies were included. Their own guidance is to treat the pooled effect sizes for patient safety with caution. This is an association measured across a very large group, not proof that burning out makes a given doctor harm a given patient.
Some corners of medicine carry more of the risk than others. In the exploratory analyses, the link between burnout and safety incidents was strongest among the youngest physicians and those working in emergency medicine and intensive care, and the link with low professionalism was strongest in doctors still in training and in hospital settings. Most of those signals faded once the researchers accounted for everything at once, so they read as leads rather than conclusions.
What the field still lacks is the study that follows doctors forward in time. The authors call for prospective designs that can untangle the order of events between exhaustion, cynicism and the outcomes measured here, and for access to individual participant data rather than pooled summaries. In the meantime their practical argument is that burnout has stopped being a wellbeing perk for employers to consider and has become, on this evidence, a workforce and safety problem that health systems pay for either way.
Source: Hodkinson A, Zhou A, Johnson J, Geraghty K, Riley R, Zhou A, Panagopoulou E, Chew-Graham CA, Peters D, Esmail A, Panagioti M. “Associations of physician burnout with career engagement and quality of patient care: systematic review and meta-analysis.” BMJ, 2022;378:e070442. DOI: 10.1136/bmj-2022-070442
