Burnout and health: expanding our knowledge

Burnout is often conceptualized as an affective reaction to ongoing stresses at work. Burnout's core content is the gradual depletion over time of individuals' intrinsic energetic resources; it is reflected primarily in emotional exhaustion, physical fatigue and cognitive weariness (cf. Shirom, 2003). This Special Issue focuses on employee burnout and its relationship with the different facets of physical health. The major objective of this essay is to provide readers with an overview of the specific ways in which the papers included in the Special Issue augmented our knowledge on the relationship between employee burnout and health. There are several reasons justifying the focus of this Special Issue on burnout and health. In some countries, burnout is a legitimate base for a compensation claim (Schaufeli, 2003) and represents a major reason for sickness absences and work disability (Bekker, Croon, MBI) were found to predict medically certified sickness absences for both men and women (Ahola et al., 2008) and disability pensions in two independent samples (Ahola et al., 2009; Ahola, Toppinen-Tanner, Huuhtanen, Koskinen, Peterson et al., 2008). Burnout may transfer from one employee to another, either directly or indirectly, and may thereby adversely impact the health and well-being of one's partner (Bakker, 2009). Thus, accumulated evidence supports the proposition that burnout at work can be regarded as a major public health problem and a cause for concern for health-care policy-makers. Complementing and extending a recent review of the aforementioned body of knowledge (Melamed, Shirom, Toker, Berliner, Vinokur, Pierce, DeSalvo, Bloser, Reynolds, He, therefore, one's feeling that one's physical, cognitive and emotional or interpersonal resources have been depleted is likely to impact one's SRH. The obverse may also be correct: one's perceived state of health is an important coping resource influencing one's coping effectiveness and recovery process following temporarily lost energetic resources (Hobfoll, 2002). Therefore, a change in SRH is likely to have an impact on one's level of burnout. Based on a longitudinal design and using structural equation modelling, Vinokur et al. (2009) were able to demonstrate that across time, perceived health predicted a decrease in burnout and burnout predicted a decrease in perceived health, providing support to the coexistence of both types of effects. However, they were also able to find considerable support to their expectation that the effect of perceived health on burnout is stronger than the effects of burnout on perceived health. A major conclusion that could be formulated based on Vinokur et al. (2009) is that in studies examining burnout's linkages with physical health, the reverse causation hypothesis, namely that individuals' initial health status impact their feelings of burnout, should always be investigated. In yet another study included in the Special Issue, Bakker (2009) was able to establish a bridge between two fields of research on burnout: burnout and health, and burnout's transfer or spillover among employees in the same unit or between intimate partners. Bakker's (2009) major finding that burnout may transfer from employees to their intimate partners thereby indirectly impacting their partners' SRH establishes the potential contribution of making the theoretical and empirical tie-in between the previously mentioned two subfields of burnout research. Earlier, I referred to the accumulated evidence documenting that burnout is implicated in the incidence of CVD (cf. Melamed et al., 2006). Analogously, the Job Demands–Control–Support (JDC–S) model proposed by Karasek and Theorell (1990) has been found in a meta-analytic study (Kivimaki et al., 2006) to predict CVD, and a qualitative review (Belkic, Landsbergis, Schnall, it was found to be a precursor of CVD. Additionally, the JDC–S model was found in many studies (cf. Armon, 2009) to predict burnout and insomnia. However, the relationship between burnout and the JDC–S model in the prediction of insomnia have not yet been investigated. Following the rationale mentioned earlier, Armon's paper (2009) focused on the incremental additive effects of the initial levels of burnout to the prediction of subsequent changes in the levels of insomnia over and above those of the JDC–S model. Armon's paper (2009) adds to our understanding of the impact of the work nexus on employee health in the following way. Burnout could be viewed as a proxy variable representing the accumulated effects of a variety of work-related stresses on health outcomes. These stresses include chronic stresses, hassles and critical job events. It follows that burnout is regarded as a proxy variable of these types of stresses because it reflects the extent to which one's energetic coping resources have been depleted because of one's exposure to them (cf. Melamed et al., 2006). Yet, there are several theoretical models, including the JDC–S model (Karasek Melamed, 2009) investigated the relationships between MS pain and burnout. MS pain is highly prevalent among workers in most countries. Furthermore, as documented in the two papers mentioned earlier, it represents a major cause of absenteeism, chronic disability, frequent utilization of health services and early retirement. The few past studies that found burnout and MS pain to be associated were all based on a cross-sectional design (e.g. Honkonen et al., 2006), whereas Melamed (2009) and Grossi et al. (2009) used longitudinal design studies conducted in two different countries independently. As noted by Melamed (2009), several reviews of the area have suggested that psychosocial factors are powerful predictors of MS pain; however, most past studies used a variety of psychosocial stresses in relation to MS pain, such as those included in the JDC–S model. Additionally, the findings of some past cross-sectional studies, based on the MBI (e.g. Miranda, Viikari-Juntura, Heistaro, Heliövaara, this issue is further elaborated in the epilogue to the Special Issue. In contrast, Melamed (2009) and Grossi et al. (2009) used the Shirom–Melamed Burnout Measure, an instrument whose total score is theoretically meaningful and has been empirically verified (cf. Shirom, Vinokur, & Nirel, in press). The two studies support the view that burnout is implicated in the aetiology of MS pain. Because their longitudinal design better support causal assertions, and because they controlled a variety of confounders representing alternative explanatory factors of MS pain (for example, Grossi et al., 2009, controlled for psychological distress and the components of the JDC–S model in addition to baseline levels of burnout), the two studies add an important new facet to the burnout and health literature. Grossi et al. (2009) also found that burnout exerts an effect on changes over time in MS pain over and above the effects of the JDC–S model. Therefore, their study (Grossi et al., 2009) joins hands with the Armon (2009) study reviewed earlier in providing considerable support to this hypothesis that burnout, rather than the popular JDC–S model, could be regarded as a representative of one's overall exposure to a variety of work-related stresses at work. I will use the term fatigue to refer to chronic or prolonged fatigue because these terms are used interchangeably in the relevant literature. Fatigue is a common patient complaint in a variety of medical settings, but when it appears without a known co-morbidity (such as anaemia, joint pains or any of the malignant diseases), it is very difficult to identify an organic condition explaining it. The prevalence rates of fatigue bear considerable similarity to those of burnout. There are several major similarities between the fields of study of burnout and of fatigue. Conceptually, the core content of both has been noted to be the depletion of a person's energetic resources (Shirom, 2003). Additionally, both are considered persistent and stable across time (Leone, Huibers, Knottnerus, & Kant, 2008), possibly because of the similar physiological processes underlying them (cf. Melamed et al., 2006). Most past studies on the relationships between burnout and fatigue have been based on a cross-sectional design. The major findings presented by Leone, Huibers, Knottnerus and Kant (2009) advance our understanding of the relationships between burnout and fatigue in several ways, most notably by finding that burnout and prolonged fatigue predict each other in time. Moreover, there seems to be a dose-response relationship between the two conditions. Taken together, the results by Leone et al. (2009) suggest that burnout and prolonged fatigue influence each other in the manner of a downward spiral. This implies that early intervention is important to prevent the conditions from co-occurring and the outcome from worsening. The series of studies on burnout and health in the Special Issue are characterized by higher levels of research rigour relative to past studies in this area, particularly in that most of them are based on longitudinal designs, systematically controlled for confounders and use advanced statistical techniques such as structural equation modelling to analyse their data. Additionally, some of them (van Doornen et al., 2009) directly addressed the underlying biological mechanisms linking burnout and health. Hopefully, they will pave the way to additional studies in this area because many major research questions, such as the extent to which burnout has an impact on subsequent mortality after controlling for other negative affective states (e.g. depressive symptoms), remain unanswered. Collectively, the papers included in the Special Issue offer policy-makers and practitioners in the area of occupational health evidence-based avenues of action to minimize the deleterious impact of burnout on several key aspects of employee physical health, including hospitalization for CVD and mental disorders, insomnia and MS pain. The authors acknowledge the support of the Israel Science Foundation for the project on which this study is based (Grant 962/02–1).