Author Reply to Letter to the Editor regarding ‘Is burnout simply a stress reaction?’

To the Editor

We welcome the opportunity to respond to several points raised by doctors Young et al. (2022). They first argue for burnout having nosological status. Current barriers include the dominant definition comprising only three symptoms (i.e. exhaustion, empathy loss and impaired work function), and with the third possibly a consequence rather than a symptom, as well as most measures not having cutoff scores to define ‘caseness’. Such limitations contribute to definitional and diagnostic imprecision and a range of consequences such as conflicting prevalence estimates – including a meta-analysis of 182 studies of physicians quantifying burnout rates varying from 0% to 80.5% (Rotenstein et al., 2018).

Having developed a broader set of symptom criteria, we recently published an article (Parker and Tavella, 2022) arguing for burnout’s formal inclusion in classificatory systems and also offered a set of criteria in accord with DSM and ICD classificatory models. Criterion A requires a work-based stressor, but we extend allowed stressors beyond formal work-based ones, recognising that ‘carer’ burnout is common. Criterion B lists and requires the presence of all five symptoms – three (noted earlier) that have dominated burnout’s definition for decades but with some redefining of those constructs – and the addition of two extra symptom domains (i.e. cognitive impairment and insularity) as generated in empirical studies (overviewed in Parker et al., 2021). Criterion C requires a level of impairment while Criterion D requires that other causal processes (e.g. differing psychological conditions, certain medical conditions, treatments, drug side-effects) are not in play. All criteria are required to be met, with the last one being highly cogent in that most measures of burnout have acceptable sensitivity (i.e. detecting true cases) but limited specificity (in that multiple other conditions such as depression or physical illness may cause false positive assignment).

Young et al. then note the diagnostic confusion between burnout and depression – a highly salient concern as researchers in the field are rather equally divided as to whether they are synonymous or not. Such a debate was evident in the fourth century when the eight deadly sins comprised tristitia (i.e. depression) and acedia (i.e. a state of non-caring, sloth and essentially burnout) before Pope Gregory joined the two in the sixth century to create the seven deadly sins. Young et al report a study that addresses the topic indirectly (by focusing on suicidal ideation and medical error). Their interpretative conclusion (that ‘Burnout may therefore be unique to depression’) is somewhat enigmatic, while they later note the reality that the ‘literature still seems unclear whether burnout could indeed be a prodromal depressive state, a subtype of depression or a discrete construct’.

We argue that depression and burnout are not synonymous and note a few arguments in support. First, a meta-analysis by Koutsimani et al. (2019) quantified only a modest correlation (of 0.52) between burnout and depression and not high enough to position the two as synonymous. Second, in developing a measure of burnout we evaluated 106 descriptor items including 37 depression items. A bifactor analysis of questionnaire items was undertaken and no depression factor emerged. In addition, we have detailed differences detailed by those who have experienced both conditions (Parker et al., 2021) and tabulated a large number of parameters for distinguishing burnout from both melancholic and non-melancholic depression (Parker and Tavella, 2021). Thus, we argue against the two states being synonymous but allow that those with burnout may develop depressive symptoms concurrently (in response to certain stressors) or as a consequence of their burnout.

We agree that both organisational and individual-focused interventions are likely to be required, reflecting the need to address the ‘stressor’ (for doctors the most commonly nominated burnout stressor is the electronic medical record) and introducing de-stressing interventions. In addition, and allowing that burnout can be viewed as a diathesis-stress condition, we argue for management addressing any personality predilection (with burnout most likely to be experienced by those who are diligent, reliable, overly conscientious, if not perfectionistic).