In the teaching profession, the growing demands from policymakers and expectations from society, as well as the increasing number of tasks and bureaucratic work for which teachers lack time and resources, may lead to high external work pressure and chronic work overload, both during and after teachers’ workday (Apple, 1989; Kokkinos, 2007; Ballet and Kelchtermans, 2009; Isabel et al., 2021). When job demands exceed job resources, and certain individual predispositions are present, an obsessive need to work may occur, namely workaholism (Schaufeli, 2016; Tabak et al., 2021). Workaholism is characterized as a form of addiction that comprises the dimensions of affect (enjoying work but feeling guilty and anxious when not working), cognition (obsessing about work), and behavior (working excessive hours, even at home; Ng et al., 2007). The term workaholism was coined by Oates (1971), who described it as “the compulsion or the uncontrollable need to work incessantly” (p. 11), referring to the behavioral and cognitive components of alcoholism that may result in various stress-related health outcomes (Preckel et al., 2005; Clark et al., 2016).

Personality traits can influence how individuals perceive their work and how invested they become (Kokkinos, 2007; Schaufeli, 2016; Perera et al., 2018; Tabak et al., 2021). A growing body of research suggests that workaholism is positively related to neuroticism, as neuroticism may increase the risk of developing an obsessive need or dependence on work (Clark et al., 2016; Schaufeli, 2016). The Five-Factor Model (Costa and McCrae, 1992) describes neuroticism as being vulnerable to psychological stress, prone to unrealistic beliefs, and unable to control urges and successfully cope with stress. Furthermore, teachers with high neuroticism scores have been shown to be at higher risk for burnout based on situational anxiety related to challenging student behavior and classroom management, self-doubt, self-criticism, and work-related rumination (McIlveen and Perera, 2016).

Like neuroticism, workaholism was found to be positively related to burnout (Ng et al., 2007; Andreassen, 2014; Clark et al., 2016). A study by Nie and Sun (2016) found significant correlations between workaholism, job burnout, and depression in Chinese university teachers emphasizing how personal life and health can be affected when work becomes the priority in life. In addition, previous studies show an association between overcommitment to work (Preckel et al., 2005) and excessive workload (Schnorpfeil et al., 2002) with vital exhaustion (VE). The concept of VE has received increasing research attention for its adverse effects on health (Kop et al., 1996; Prescott et al., 2003; Balog and Konkolÿ, 2019). Characteristics of VE include excessive fatigue, lack of mental and physical energy, feelings of demoralization, and irritability (Appels and Mulder, 1989). Though the underlying causes of VE are still not fully known, it is thought that VE emerges when personal resources required to adapt to stress are depleted and can thus be considered a consequence of perceived chronic stress and burnout (Bellingrath et al., 2008; Schoch et al., 2018). Furthermore, VE has been shown to be a significant and independent risk factor for cardiovascular disease in men (Appels and Mulder, 1989). Kudielka et al. (2008) found a positive relation between VE and circulating fibrinogen levels in teachers. Increased circulating fibrinogen levels have been linked to higher cardiovascular mortality and morbidity. These findings suggest that dysregulated hemostasis may be a biological pathway through which chronic stress may negatively impact cardiovascular health in teachers (Kudielka et al., 2008).

Vital exhaustion can be assessed subjectively with the Maastricht Vital Exhaustion Questionnaire (MQ; Appels and Mulder, 1989). However, researchers have been urged to aim for multimethod approaches to assess chronic stress, to account for the distinction between self-reported and physiological chronic stress (Cohen, 1997). Self-reports may be influenced by various individual factors such as mood and personality traits like neuroticism (Christensen et al., 2019). In addition, some items used in neuroticism and subjective stress-related measures may be similar, resulting in common method variance and reduced discriminant validity. Physiological measures of stress over time, on the other hand, are generally less influenced by an individual’s immediate mood or personality and are not affected by recalling biases and social desirability (Gow et al., 2010; Stirratt et al., 2015; Christensen et al., 2019; Prado-Gascó et al., 2019). Hair cortisol concentration (HCC) analysis has been a valuable tool for chronic physiological stress measurement and has been increasingly used in the last decade (Russell et al., 2012; Stalder and Kirschbaum, 2012). Cortisol is the most common glucocorticoid (GC) in humans and an essential element of the hypothalamus-pituitary–adrenal (HPA) axis stress response. By driving the release of GCs, the HPA axis is a central player in maintaining homeostasis and alerting the organism to environmental changes (Herman and Cullinan, 1997). HCC provides a unique opportunity to quantify long-term systematic cortisol exposure retrospectively over several months (Gow et al., 2010; Stalder and Kirschbaum, 2012; Ullmann et al., 2016). Circadian rhythms and acute changes regulated by the HPA axis are thought to affect HCC less than salivary cortisol (Stalder and Kirschbaum, 2012), and several studies showed significantly higher HCC in chronically stressed individuals (Dettenborn et al., 2010; Manenschijn et al., 2011; Stalder et al., 2017; Brianda et al., 2020).

To the best of our knowledge, no study has investigated the relationship between workaholism and neuroticism with HCC in teachers. However, some studies have linked HCC to work stress, with inconsistent results (Steinisch et al., 2014; Gidlow et al., 2016; Janssens et al., 2017; Herr et al., 2018). Regarding neuroticism and cortisol, previous studies have found an association between neuroticism and higher salivary basal cortisol or area under the curve on one or more days (e.g., Nater et al., 2010; Garcia-Banda et al., 2011; Laceulle et al., 2015). In contrast to salivary cortisol, research on the relationship between neuroticism and HCC is still scarce, but with promising results: Rietschel et al. (2016) reported a modest positive association between neuroticism and long-term HCC over the past three months. A study by Engert et al. (2021) showed neuroticism predicted HCC in 1 cm hair segments. Similarly, Erickson et al. (2021) found that neuroticism predicted higher HCC over 2 months, even after controlling for other personality traits. These results support using HCC to investigate associations between chronic physiological stress indicators and traits like neuroticism.

The aim of this longitudinal study is twofold. First, we intend to examine the relationship between workaholism and teachers’ chronic stress indicators, that is, self-reported vital exhaustion (VE) and hair cortisol concentrations (HCC; measured over three-cm hair segments) at two time points 1 year apart. Second, we aim to test the moderating role of neuroticism on the relationship between workaholism and VE and workaholism and HCC. Teachers have been found to have high levels of occupational stress and are at increased risk for burnout syndrome (Gluschkoff et al., 2016; Richards et al., 2018; García-Carmona et al., 2019). Examining long-term relationships between individual teacher characteristics and chronic stress-related indicators may provide a better understanding of contributing factors to high stress levels and psychosomatic complaints among teachers.

To tackle the study’s main objectives, we tested the following four hypotheses: (H1) Workaholism and neuroticism are associated with VE at t1 and t2 1 year later. Studies have found associations between workaholism and adverse subjective health outcomes, especially burnout (Ng et al., 2007; Andreassen, 2014; Clark et al., 2016). As VE may be considered a consequence of chronic stress and burnout (Bellingrath et al., 2008; Schoch et al., 2018), it seems reasonable to expect a positive association between workaholism and VE at t1 and t2. Neuroticism is considered a risk factor for burnout (Schaufeli and Enzmann, 2020). Thus, we expect to see an association between neuroticism and VE at t1 and, since it is a relatively stable personality trait (Borkenau and Ostendorf, 1993), likewise with VE at t2 1 year later. (H2) Neuroticism moderates the relationship between workaholism and VE at t1 and t2. This hypothesis is based on the assumption that a workaholic teacher with high neuroticism may be more vulnerable to psychological stress (Costa and McCrae, 1992). Such a teacher may also experience more self-doubt, anxiety, and pressure to perform than a workaholic teacher with low neuroticism (Clark et al., 2016; Aziz et al., 2018) and may consequently be at higher risk of experiencing VE. (H3) Workaholism and neuroticism are positively associated with HCC at t1 and t2. Neuroticism and workaholism appear to be risk factors for chronic stress (Clark et al., 2016; Erickson et al., 2021), which in turn has been associated with higher HCC (Russell et al., 2012; Stalder and Kirschbaum, 2012). (H4) Neuroticism moderates the relationship between workaholism and HCC at t1 and t2. Based on the literature above, we assume that workaholic teachers with high levels of neuroticism may have higher HCC than those with low neuroticism.

The present study aimed to investigate the interaction between the two potential risk factors of workaholism and neuroticism as longitudinal predictors of vital exhaustion (VE) and hair cortisol concentration (HCC). Additionally, we explored the moderating role of neuroticism on the relationship between workaholism, VE, and HCC.

The results partly confirmed our first hypothesis (H1), which stated that workaholism and neuroticism were positively associated with VE at t1. Our findings are consistent with previous research suggesting that workaholism and neuroticism may be risk factors for adverse subjective health outcomes (Preckel et al., 2005; Ng et al., 2007; Andreassen, 2014; Clark et al., 2016). The association between workaholism and VE supports the assumption that the exhaustive and compulsive work-related behaviors of workaholism come at the expense of teachers’ health and well-being (Preckel et al., 2005; Nie and Sun, 2016). However, contrary to our expectation, workaholism and neuroticism were both unrelated to VE one year later at t2 when controlling for t1. VE showed a very high longitudinal stability of r = 0.79 over the year between the two measurements at t1 and t2. Thus, not much variance remained that other variables could explain. Direct analysis at t2 without controlling for t1 showed similar effects as at t1. Furthermore, our results showed sex differences in VE at t2. One explanation could be that women generally report higher VE than men, which may be due to psychological and physiological sex differences in the stress response (Frestad and Prescott, 2017). In addition, studies suggest that women had a more significant decrease in mental health during the COVID-19 lockdown than men (Adams-Prassl et al., 2020), which may also have influenced the sex differences in our sample, as the follow-up data collection 1 year later occurred during the pandemic.

The second hypothesis stated that neuroticism moderates the relationship between workaholism and VE at t1 and t2, which has been confirmed. One possible explanation for this finding may be that teachers with high neuroticism scores are more likely to push themselves to or beyond the limits of their work performance, which eventually results in higher VE. Several previous studies support this assumption that neuroticism positively relates to workaholism as it may create an obsessive need or dependence on work, which, in turn, can result in adverse health outcomes (Lahey, 2009; Clark et al., 2016; McIlveen and Perera, 2016; Schaufeli, 2016, Gillet et al., 2018). In addition, neuroticism was found to be associated with low self-esteem (Amirazodi and Amirazodi, 2011), and some researchers suggest that one driving force behind workaholic employees may be the desire for self-worth validation (Aziz et al., 2018). Workaholics who experience low self-esteem, high levels of guilt, anxiety, and excessively high expectations of themselves are also likely to experience high levels of work stress (Clark et al., 2016; Aziz et al., 2018). High work stress combined with neuroticism may cause teachers to devote even more effort and time to work, perpetuating the vicious cycle.

Regarding hair cortisol concentrations (HCC), no association was found with workaholism or neuroticism, neither at t1 nor t2 (H3). One possible explanation for the associations of workaholism and neuroticism with vital exhaustion but not with HCC could be a simple negativity effect, that is, individuals high in neuroticism do not necessarily feel more exhausted or experience more workaholism, but they may report their feelings as stronger. Previous work suggests that neuroticism is associated with increased reporting of somatic symptoms but not with increased experience of these symptoms (McCrae and Costa, 1986; Larsen, 1992). In addition, the relationship between HCC and subjective work stress has yielded contradictory results in the existing literature (Qi et al., 2014, 2015; Steinisch et al., 2014; Gidlow et al., 2016; Janssens et al., 2017; Herr et al., 2018). Possible reasons may be the heterogeneity of the study populations, different hair sample lengths, and different stress questionnaires for different periods (Staufenbiel et al., 2013; Gidlow et al., 2016). Finally, van der Meij et al. (2018) compared HCC in a high and a normal workload sample and found an association with HCC only in the high workload sample. These findings suggest that a certain stress threshold must be reached to cause changes in HCC. The teachers in our sample are in the middle range between mild and moderate VE levels. Therefore, another reasonable explanation for our null findings may be that our teacher sample did not reach the threshold of chronic stress exposure that was high enough to associate with HCC.

The present study has several limitations. First, due to the COVID-19 pandemic, there were changes in the everyday lives of our participants, such as lockdowns and adaptation to digital teaching, which may have affected the stress-related measurements at one-year follow-up. The first assessments were conducted before the pandemic. Second, the assessment of neuroticism can be criticized since we did not use the full 12-item scale of the NEO-FFI (Borkenau and Ostendorf, 1993) which may not capture neuroticism in its broader concept. Third, workaholism and neuroticism were measured only once, which does not allow an analysis of the constructs’ bidirectional associations. Fourth, studies that have examined the overlap between vital exhaustion and depression have yielded inconsistent results regarding the extent to which they reflect the same underlying mechanisms (van Diest and Appels, 1991; Kopp et al., 1998; Kudielka et al., 2004; Vroege et al., 2012). Though our sample consists of generally healthy teachers, we did not control for depression, which may be considered a limitation. Finally, the sample of this study is relatively small, limiting the generalizability of the results. The findings should be replicated in a large cohort study to determine the effect of workaholism and neuroticism on adverse health outcomes in the general population. This would also help determine under what circumstances a relationship between workaholism, personality traits, and hair cortisol concentrations does or does not exist. That said, this study also includes some considerable strengths. This study is (to the best of the authors’ knowledge) the first to examine workaholism and neuroticism in relation to long-term hair cortisol concentrations in teachers. The longitudinal design further underscores the value of the study’s results.

If replicated in future research, our results have practical implications. In particular, the finding that workaholism was positively associated with vital exhaustion in teachers with high levels of neuroticism may highlight the importance of raising teachers’ awareness of potential risk factors contributing to vital exhaustion. In regards to neuroticism, a meta-analysis (Jorm, 1989) has shown that rational-emotive and similar therapies can successfully reduce neuroticism levels despite the dispositional nature of neuroticism. Moreover, providing in-service training courses for teachers may help familiarize them with the concepts of work behaviors, vital exhaustion, and the role of their personality in these processes. Although such training alone is unlikely to be sufficient to ultimately reduce the burden on teachers, it may help develop an understanding of potential adverse individual and organizational consequences of maladaptive or dysfunctional behaviors and coping styles. Identifying coping styles and work behaviors that match their personality style may help teachers to achieve and obtain a more sustainable work-life balance.

Taken together, workaholism and neuroticism appear to be associated with vital exhaustion in teachers. The association was manifested in increased self-reported vital exhaustion but not in increased HCC. Our findings emphasize the importance of considering individual characteristics when investigating vital exhaustion in teachers. Furthermore, this study adds to the literature on the divergence between self-reports and HCC. Further research is necessary to investigate the applicability of HCC as a biomarker of chronic stress in the context of work.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of the Canton Bern (no. 2019–00787) Internal Review Board (IRB) of the University of Bern. The participants provided their written informed consent to participate in this study.

AW: conceptualization, methodology, formal analysis, and writing—review and editing. SS: conceptualization, investigation, and writing—original draft. GJ: formal analysis. MgH and WT: writing—review and editing. RLM: conceptualization, methodology, and writing—review and editing. All authors contributed to the article and approved the submitted version.

This work was supported by the Swiss National Science Foundation SNF, under Grant number 100019_185484, and the University of Teacher Education Bern, Grant number 16 w 0008 02.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.