Hostname: page-component-78c5997874-94fs2 Total loading time: 0 Render date: 2024-11-14T01:27:36.723Z Has data issue: false hasContentIssue false

An Underlying Common Factor, Influenced by Genetics and Unique Environment, Explains the Covariation Between Major Depressive Disorder, Generalized Anxiety Disorder, and Burnout: A Swedish Twin Study

Published online by Cambridge University Press:  13 September 2016

Lisa Mather*
Affiliation:
Division of Insurance Medicine, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
Victoria Blom
Affiliation:
Division of Insurance Medicine, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden The Swedish School of Sport and Health Sciences, Stockholm, Sweden Department of Psychology, Stockholm University, Stockholm, Sweden
Gunnar Bergström
Affiliation:
Division of Intervention and Implementation Research, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden Centre for Occupational and Environmental Medicine, Stockholm County Council, Stockholm, Sweden
Pia Svedberg
Affiliation:
Division of Insurance Medicine, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
*
address for correspondence: Lisa Mather, Division of Insurance Medicine, Department of Clinical Neuroscience, Karolinska Institutet, Berzelius väg 3, SE-171 77 Stockholm, Sweden. E-mail: [email protected]

Abstract

Depression and anxiety are highly comorbid due to shared genetic risk factors, but less is known about whether burnout shares these risk factors. We aimed to examine whether the covariation between major depressive disorder (MDD), generalized anxiety disorder (GAD), and burnout is explained by common genetic and/or environmental factors. This cross-sectional study included 25,378 Swedish twins responding to a survey in 2005–2006. Structural equation models were used to analyze whether the trait variances and covariances were due to additive genetics, non-additive genetics, shared environment, and unique environment. Univariate analyses tested sex limitation models and multivariate analysis tested Cholesky, independent pathway, and common pathway models. The phenotypic correlations were 0.71 (0.69–0.74) between MDD and GAD, 0.58 (0.56–0.60) between MDD and burnout, and 0.53 (0.50–0.56) between GAD and burnout. Heritabilities were 45% for MDD, 49% for GAD, and 38% for burnout; no statistically significant sex differences were found. A common pathway model was chosen as the final model. The common factor was influenced by genetics (58%) and unique environment (42%), and explained 77% of the variation in MDD, 69% in GAD, and 44% in burnout. GAD and burnout had additive genetic factors unique to the phenotypes (11% each), while MDD did not. Unique environment explained 23% of the variability in MDD, 20% in GAD, and 45% in burnout. In conclusion, the covariation was explained by an underlying common factor, largely influenced by genetics. Burnout was to a large degree influenced by unique environmental factors not shared with MDD and GAD.

Type
Articles
Copyright
Copyright © The Author(s) 2016 

Mental disorders such as depression and anxiety are one of the main reasons for the increase of years lived with disability globally (Global Burden of Disease Study 2013 Collaborators, 2015). Burnout has been found to be closely related to depression (Bianchi et al., Reference Bianchi, Schonfeld and Laurent2015a) and has also been found to be associated with anxiety (Ding et al., Reference Ding, Qu, Yu and Wang2014; Toker et al., Reference Toker, Shirom, Shapira, Berliner and Melamed2005). Burnout has been defined as ‘A state of physical, emotional, and mental exhaustion caused by long-term involvement in situations that are emotionally demanding’ (Pines & Aronson, Reference Pines and Aronson1988, p. 9). There is an ongoing discussion about whether burnout is a unique condition or a form of depression (Bianchi et al., Reference Bianchi, Schonfeld and Laurent2015b). Several twin studies have found that depression and anxiety are highly comorbid due to the fact that they share genetic risk factors (Middeldorp, Cath et al., Reference Middeldorp, Cath, Van Dyck and Boomsma2005). However, less is known about the genetic risk factors for burnout and whether they are shared with depression and anxiety.

Burnout has mainly been described as a work-related phenomenon with three dimensions: exhaustion, cynicism, and a sense of ineffectiveness (Maslach et al., Reference Maslach, Schaufeli and Leiter2001). However, measurement instruments have been developed that measure burnout both among persons who are working and among persons without paid work. The Pines Burnout Measure takes a wider perspective on burnout and can be used also in non-working populations (Pines et al., Reference Pines, Aronson and Kafry1981). The Pines Burnout Measure correlates mainly with the exhaustion dimension of the Maslash Burnout Inventory (Shirom & Ezrachi, Reference Shirom and Ezrachi2003).

Major depressive disorder (MDD) and generalized anxiety disorder (GAD) are heritable, and meta-analyses of twin studies have estimated the heritabilities to be 37% and 32%, respectively (Hettema et al., Reference Hettema, Neale and Kendler2001; Sullivan et al., Reference Sullivan, Neale and Kendler2000). Previous twin studies have found a complete overlap between the genetic risk factors for depression and anxiety in women and a very large overlap in men (Kendler et al., Reference Kendler, Neale, Kessler, Heath and Eaves1992; Kendler et al., Reference Kendler, Gardner, Gatz and Pedersen2007). Candidate genes have also been found for both depression and anxiety, and the 5-HTTLPR short variant, involved in the serotonergic system was found to be involved in both depression and anxiety disorders (Gatt et al., Reference Gatt, Burton, Williams and Schofield2015). The heritability of burnout is less studied. The few studies presented so far have shown inconsistent results, and the heritability was estimated to be 13–37% (Blom et al., Reference Blom, Bergstrom, Hallsten, Bodin and Svedberg2012; Middeldorp et al., Reference Middeldorp, Cath and Boomsma2006). The variation in results may be due to the fact that there exists no generally accepted definition of burnout and hence different measurements have been used. In a previous study, we identified that burnout predicted sick leave due to mental disorders, such as depression and anxiety, because of shared genetic risk factors (Mather, Bergström et al., Reference Mather, Bergström, Blom and Svedberg2014). Another study found that the correlation between burnout and anxious depression was partly explained by genetic factors in common to both (Middeldorp et al., Reference Middeldorp, Cath and Boomsma2006). Moreover, both personal history, as well as a family history of depression have been found to predict emotional exhaustion, the key feature in burnout (Nyklicek & Pop, Reference Nyklicek and Pop2005), a finding that supports a shared genetic vulnerability.

The aim of this study was to examine to what degree the covariation between MDD, GAD, and burnout is explained by common genetic and environmental factors, in women and men, using a biometric twin design in a large sample of Swedish twins.

Materials and Methods

Sample

This study has a cross-sectional design and used data from the Swedish Twin Registry (STR). The STR is a population-based registry that contains all twins born in Sweden (Magnusson et al., Reference Magnusson, Almqvist, Rahman, Ganna, Viktorin, Walum and Lichtenstein2013). Twins that responded to the Study of Twin Adults—Genes and Environment (STAGE), performed by the STR in 2005–2006, were included. STAGE was a large web-based questionnaire sent to all twins in the STR born between 1959 and 1985 (N = 42,582), with a response rate of 59.6%. All respondents were included and hence the sample contained 25,378 twins. In the sample, there were 8,646 complete twin pairs, where 2,151 were monozygotic (MZ) females, 1,402 MZ males, 1,510 dizygotic (DZ) females, 1,000 DZ males, and 2,583 DZ opposite sex twin pairs. The sample also contained 8,086 single twins that were included in the analyses. More information about STAGE is available elsewhere (Furberg et al., Reference Furberg, Lichtenstein, Pedersen, Thornton, Bulik, Lerman and Sullivan2008; Lichtenstein et al., Reference Lichtenstein, Sullivan, Cnattingius, Gatz, Johansson, Carlstrom and Pedersen2006). The mean age was 33.6 (SD 7.7) years and the sample contained 55.6% women (Table 1). The prevalence of MDD, GAD, and burnout did not vary much with age; hence, age was not included in the analyses.

TABLE 1 Frequencies (%) of Major Depressive Disorder, Generalized Anxiety Disorder, Burnout, and Zygosity Among 25,378 Swedish Twins, Stratified on Sex

Measures

Lifetime prevalence of depression was measured with 38 questions based on the Structured Clinical Interview for DSM-IV Disorders (SCID; First et al., Reference First, Spitzer, Gibbon, Williams and Benjamin1996). SCID is based on criteria for MDD in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 2000). Criteria A, C, and E had to be fulfilled in order for the participant to be classified as having had MDD. At least five of the following symptoms had to have been present during the same 2-week period; at least one of the symptoms had to be (1) depressed mood or (2) loss of interest or pleasure (criteria A).

  1. 1. depressed mood most of the day, nearly every day;

  2. 2. markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day;

  3. 3. significant weight loss when not dieting or weight gain, or decrease or increase in appetite nearly every day;

  4. 4. insomnia or hypersomnia nearly every day;

  5. 5. psychomotor agitation or retardation nearly every day;

  6. 6. fatigue or loss of energy nearly every day;

  7. 7. feelings of worthlessness or excessive or inappropriate guilt nearly every day;

  8. 8. diminished ability to think or concentrate, or indecisiveness, nearly every day;

  9. 9. recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or specific plan for committing suicide.

The symptoms had to cause clinically significant distress or impairment in social, occupational, or other important areas of functioning (criteria C) and not be better accounted for by bereavement (criteria E; American Psychiatric Association, 2000).

Lifetime prevalence of anxiety was measured with 23 questions based on SCID (First et al., Reference First, Spitzer, Gibbon, Williams and Benjamin1996). Criteria A and C had to be present in order for the participant to be classified as having had GAD. Excessive anxiety and worry, occurring more days than not for at least 6 months, about a number of events or activities had to be reported (criteria A) and the anxiety and worry had to be associated with three or more of the following symptoms (criteria C):

  1. 1. restlessness or feeling keyed up or on edge;

  2. 2. being easily fatigued;

  3. 3. difficulty concentrating or mind going blank;

  4. 4. irritability;

  5. 5. muscle tension;

  6. 6. sleep disturbance (American Psychiatric Association, 2000).

Burnout was measured with the short form of the Pines Burnout Measure; it correlates strongly (0.90) with the full 21-item Pines Burnout Measure (Hallsten et al., Reference Hallsten, Josephson and Torgén2005). The scale includes the questions: ‘How often during the last 12 months have you felt low?’, ‘How often during the last 12 months have you felt emotionally exhausted?’ and ‘How often during the last 12 months have you felt run down?’ The answers were given on a 7-point Likert scale ranging from 1 = never to 7 = all the time. Cronbach's alpha was 0.89. The mean burnout score was calculated and a categorical variable was created: no burnout (1–2.99), risk of burnout (3–3.99) and burnout (4–7). Previous studies have used four categories (Takai et al., Reference Takai, Takahashi, Iwamitsu, Ando, Okazaki, Nakajima and Miyaoka2009; Takai et al., Reference Takai, Takahashi, Iwamitsu, Oishi and Miyaoka2011); however, as no concordant male DZ twin pairs were present in the highest burnout group, the two highest categories were collapsed.

Zygosity was assessed in STAGE using a set of questions assessing twin pair similarity; this method has been compared with genetic testing in two sub-samples of the twin registry, and proved correct in 98–99% of the pairs (Lichtenstein et al., Reference Lichtenstein, de Faire, Floderus, Svartengren, Svedberg and Pedersen2002). Sex was entered into the analysis as a dichotomous variable.

Analysis

The biometric model is a type of structural equation model that uses the variance/covariance structure to investigate the genetic and environmental underpinnings of a phenotype, that is, additive genetics (A), non-additive genetics (D), shared environment (C), and unique environment (E) (Purcell, Reference Purcell, Plomin, DeFries, Knopik and Neiderhiser2013). Including opposite-sex twins allowed us to test for qualitative sex differences, that is, if the same genes are underpinning the phenotype in women and men, by testing if the genetic correlation can be set to 0.5 in opposite sex twins. As our sample included only twins reared together, C and D could not be tested simultaneously (Rijsdijk & Sham, Reference Rijsdijk and Sham2002). In order to find the best-fitting and most parsimonious model, nested sub-models were tested against the full models or more complex sub-models using likelihood ratio test (Purcell, Reference Purcell, Plomin, DeFries, Knopik and Neiderhiser2013). As MDD and GAD were binary variables, liability threshold models were used that assumes there is an underlying normally distributed liability to the phenotypes. Answering ‘don't know/don't want to answer’ was treated as missing values. Polychoric and tetrachoric phenotypic, intrapair, and cross-twin cross-trait correlations were calculated, as they give a first impression of genetic variance and covariance structures in SAS. Analyses was performed using OpenMx software (Boker et al., Reference Boker, Neale, Maes, Wilde, Spiegel, Brick and Fox2011), run within the R environment (R Development Core Team, 2010).

Univariate Analyses

To test the assumption that thresholds do not vary between MZ and DZ twins and between twin 1 and 2 in a pair (randomly assigned), saturated models were specified and compared with a nested model, forcing the thresholds to be equal. As the difference between MZ and DZ correlations were somewhat different for men and women, we tested both ACE and ADE models for all phenotypes (Table 2). Models were specified, allowing for both qualitative and quantitative sex differences using five zygosity groups: MZ women, MZ men, DZ women, DZ men, and opposite-sex DZ pairs. As these models are not nested, model selection was based on Bayesian Information Criterion (BIC) values (Markon & Krueger, Reference Markon and Krueger2004; Raftery, Reference Raftery1995). Sub-models were based on the best-fitting model, either ACE or ADE. Sex differences were tested in the full models; first models were tested that only allow for quantitative sex differences, that is, restricting the genetic correlation between the opposite-sex DZ twin pairs to be 0.5. Subsequently, models forcing the path estimates to be equal for women and men were utilized. AE models were then built, where the D or C parameter was set to zero. E models were then created, where the A parameter was restricted to be zero as well.

TABLE 2 Polychoric (Burnout) and Tetrachoric (Major Depressive Disorder and Generalized Anxiety Disorder) Within Pair and Cross-Twin, Cross-Trait Correlations With 95% Confidence Intervals Among 8,646 Complete Twin Pairs

MZ = monozygotic, DZ = dizygotic.

Multivariate Analysis

First, saturated models were used and compared with a nested model to test equal thresholds as in the univariate analysis. In order to investigate the relationship between MDD, GAD, and burnout, we tested three different multivariate models: the Cholesky decomposition, the common factor independent pathway model, and the common factor common pathway model, with one latent factor (Purcell, Reference Purcell, Plomin, DeFries, Knopik and Neiderhiser2013). In the Cholesky model, three of each of the factors (A, C/D, E) were included. In an independent pathway model, there is one shared A,C/D, and E factor with a path to each phenotype, as well as one separate A,C/D, and E factor per phenotype. As an independent pathway and Cholesky decomposition has the same number of estimated parameters when three variables are used, the best-fitting model was chosen based on BIC value. A common pathway model was then created and compared against the independent pathway model. In a common pathway model, factors load onto a latent common factor that in turn has a path to each phenotype; that is, there is an unmeasured common factor that explains the covariation of the measured phenotypes. It also contains a factor with an independent path to each phenotype. An AE common pathway model was then compared to the full common pathway model. Further, a sub-model removing the phenotype specific to a path to MDD was compared against the AE model. Finally, an E common pathway model was tested against the AE model without a phenotype specific to a path to MDD.

The study was approved by the regional ethics committee board in Stockholm, Sweden (Dnr: 2009/2053-31/5. Date: 11/02/2010).

Results

Univariate Analyses

The models restricting the thresholds to be equal between MZ and DZ twins and twin 1 and 2 in a pair did not fit significantly worse compared to the saturated models for any of the phenotypes (MDD: p = .39, GAD: p = .30, burnout: p = .26). We found no statistically significant sex differences; for all phenotypes, removing both qualitative and quantitative sex differences did not significantly worsen the fit of the models (Table 3). Further, removing the D or C parameters did not result in significantly different fit statistics. However, when removing the A parameter, the models fit significantly worse; hence, AE models with no sex limitations were selected as the final models for all variables (Table 3). All three phenotypes were found to have moderate proportions of the variation explained by additive genetic factors (Table 4).

TABLE 3 Model Fit Statistics of the Univariate Models for Burnout, Major Depressive Disorder and Generalized Anxiety Disorder and for the Multivariate Models

Best-fitting model highlighted in bold. df = degrees of freedom, -2LL = -2 Log-likelihood, AIC = Akaike's information criterion, BIC = Bayesian information criterion. Phenotypic variation was decomposed into additive (A) and non-additive (D) genetic variation, and shared (C) and unique (E) environmental variation. GSL = general sex limitation (allows both qualitative and quantitative sex differences). CSL = common sex limitation (allows only quantitative sex differences). NSL = no sex limitation (allows no sex differences). IP = independent pathway. CP = common pathway.

TABLE 4 Proportions of Additive Genetic (a2) and Unique Environmental (e2) Effects From the Best-Fitting Univariate Models With 95% Confidence Intervals

Multivariate Analysis

The phenotypic correlations were 0.71 (0.69–0.74) between MDD and GAD, 0.58 (0.56–0.60) between MDD and burnout, and 0.53 (0.50–0.56) between GAD and burnout, while cross-twin, cross-trait correlations were similar between all three phenotypes (Table 2). As there were no statistically significant sex differences, two zygosity groups, MZ and DZ (including opposite-sex DZ twins), were used in the multivariate analysis (Table 3). ADE models were used, as the cross-twin, cross-trait correlations for MZ twins were more than double that of DZ twins (Table 2). The model restricting the thresholds to be equal between MZ and DZ twins and twin 1 and 2 in a pair did not fit significantly worse compared to the saturated model (p = .90). The common pathway model was a better fit to data than the independent pathway model and Cholesky decomposition based on BIC and likelihood ratio test (Table 3). Removing the D factor and the phenotypic specific a path to MDD (which was estimated to zero) did not significantly worsen the fit, hence, this was chosen as the final model (Table 3). The path estimates with 95% confidence intervals for the model can be found in Figure 1. The covariance components (heritability estimates) can be obtained by squaring the path estimates. Results indicate that there is a latent common factor that is mostly influenced by additive genetic effects (58%), but also by unique environment (42%), that explains the covariation between MDD, GAD, and burnout. This latent common factor explained 77% of the variation in MDD, 69% in GAD, and 44% in burnout. Both burnout and GAD were also found to have phenotype-specific additive genetic effects explaining 11% of the variance each, while MDD did not have phenotype-specific additive effects. The proportion of the variation that was explained by phenotype-specific unique environmental factors was 45% for burnout, 23% for MDD, and 20% for GAD.

FIGURE 1. Path estimates for the best-fitting model with 95% confidence intervals.

Discussion

In this cross-sectional twin study, we found that the associations among MDD, GAD, and burnout were consistent with the existence of a single latent common factor influenced mostly by genetics (58%), but also unique environment (42%). For MDD and GAD, the majority of the variation was explained by this common factor (77% and 69%, respectively); while for burnout the proportion was lower (44%). All genetic risk factors for MDD went through the common factor, while GAD and burnout each had 11% of the variation explained by additive genetic factors unique to each phenotype. The largest proportion of the variation in burnout (45%) was explained by unique environmental factors not shared with MDD and GAD, while for GAD and MDD, environmental factors unique to each phenotype explained 23% and 20%, respectively.

The fact that a common factor model gave the best fit to the data is in line with a previous twin study, examining the covariation between insomnia, fatigue, and depression, which also found that a common factor, to a large degree influenced by genetic factors, explained the covariation (Hur et al., Reference Hur, Burri and Spector2012). Both neuroticism and negative affectivity have been found to be markers for general vulnerability to internalizing disorders (Ormela et al., Reference Ormela, Jeronimusa, Kotovc, Riesea, Bosa, Hankind and Oldehinkela2013; Paulus et al., Reference Paulus, Talkovsky, Heggeness and Norton2015) and a possible explanation is that the latent factor in the present study could represent such an underlying temperament. Neuroticism has been found to correlate genetically with internalizing disorders (Mikolajewski et al., Reference Mikolajewski, Allan, Hart, Lonigan and Taylor2013) and share approximately half of its genetic risk factors with depression (Kendler et al., Reference Kendler, Gatz, Gardner and Pedersen2006a). However, Kendler et al. (Reference Kendler, Gardner, Gatz and Pedersen2007) found that only 25% of the genetic correlation between depression and anxiety were explained by neuroticism, so the latent common factor found in the current study is not likely purely reflecting neuroticism. Many researchers have moved towards looking at underlying temperaments for mental disorders such as depression and anxiety (Brown, Reference Brown2007). Transdiagnostic interventions for common mental disorders have also been developed and shown symptom improvement (Barlow, Reference Barlow2004; Ejeby et al., Reference Ejeby, Savitskij, Ost, Ekbom, Brandt, Ramnero and Backlund2014). Moreover, the molecular genetic research has begun studying pleiotropy across traditional diagnostic boundaries (Cross-Disorder Group of the Psychiatric Genomics Consortium, 2013).

Whether burnout and depression are separate entities, or whether burnout is a form of depression, is currently under debate (Bianchi et al., Reference Bianchi, Schonfeld and Laurent2015b). Our results show that even though burnout shares the majority of its genetic risk with depression, most of the environmental variance and a small amount of genetic variance were unique to burnout. This was despite the fact the Pines Burnout Measure was used, a measure that has been found to be more closely related to depression than the most commonly used measure of burnout, the Maslash Burnout Inventory (Enzmann et al., Reference Enzmann, Schaufeli, Janssen and Rozeman1998; Shirom & Ezrachi, Reference Shirom and Ezrachi2003). In a previous study, we also found that both the phenotypic and genetic correlation between burnout, measured with the Pines Burnout Measure, and sick leave due to stress-related mental disorders (0.56) was lower than between burnout and sick leave due to other mental disorders (0.68; Mather, Bergström et al., Reference Mather, Bergström, Blom and Svedberg2014). The environmental variation unique to burnout could, for example, represent work-related factors such as psychosocial work environment and work-home conflicts, or stressful life events outside work, that have all been shown to be associated with burnout, independent of genetic and shared environmental factors (Blom et al., Reference Blom, Bodin, Bergstrom, Hallsten and Svedberg2013; Blom et al., Reference Blom, Sverke, Bodin, Bergstrom, Lindfors and Svedberg2014; Mather, Blom et al., Reference Mather, Blom and Svedberg2014).

We found high heritability estimates of depression and anxiety compared with previous findings that also found sex differences, while we did not (Hettema et al., Reference Hettema, Neale and Kendler2001; Kendler et al., Reference Kendler, Gatz, Gardner and Pedersen2006b; Kendler et al., Reference Kendler, Gardner, Gatz and Pedersen2007; Middeldorp et al., Reference Middeldorp, Cath and Boomsma2006; Sullivan et al., Reference Sullivan, Neale and Kendler2000). Shared environment has been found to have no impact on depression and anxiety (Hettema et al., Reference Hettema, Neale and Kendler2001; Sullivan et al., Reference Sullivan, Neale and Kendler2000), while the findings have varied regarding effects of shared environment on burnout (Blom et al., Reference Blom, Bergstrom, Hallsten, Bodin and Svedberg2012; Middeldorp, Stubble et al., Reference Middeldorp, Stubble, Cath and Boomsma2005; Middeldorp et al., Reference Middeldorp, Cath and Boomsma2006).

Strengths of the current study include a large sample of twins from the STR, since the register is population based; generalizability of the findings is high for similar age groups. However, since the STAGE questionnaire was so extensive, there are many internal missing values. Due to the amount of missing values and the low prevalence of GAD, this may be why we were unable to find statistically significant sex differences, even though the within-pair correlations indicated there may be differences in heritability in women and men. The fact that the Pines Burnout Measure was used allowed inclusion of all participants. Previous studies have found that burnout is high in groups such as students, athletes, and family caregivers (Dyrbye et al., Reference Dyrbye, West, Satele, Boone, Tan, Sloan and Shanafelt2014; Gustafsson et al., Reference Gustafsson, Kentta, Hassmen and Lundqvist2007; Lindstrom et al., Reference Lindstrom, Aman and Norberg2010; Takai et al., Reference Takai, Takahashi, Iwamitsu, Ando, Okazaki, Nakajima and Miyaoka2009) and not only in those employed. Moreover, the ‘healthy worker effect’ often present in burnout studies has been reduced (Schaufeli et al., Reference Schaufeli, Bakker, Hoogduin, Schaap and Klader2001). Weaknesses include the somewhat low response rate and that burnout was assessed over the last year, while lifetime prevalence of MDD and GAD were measured. Moreover, the measures of MDD and GAD were based on diagnostic criteria, while the Pines Burnout Measure is not a clinical instrument used to assess a diagnosis, but rather mainly a measurement of emotional exhaustion (Shirom & Ezrachi, Reference Shirom and Ezrachi2003). The DSM criteria are meant to be assessed by clinical interview and not as a questionnaire; this may also have affected the sensitivity and specificity of these self-reported measures. There have been varying results when comparing web-based questionnaires with interviews for MDD and GAD (Carlbring et al., Reference Carlbring, Forslin, Ljungstrand, Willebrand, Strandlund, Ekselius and Andersson2002; Farvolden et al., Reference Farvolden, McBride, Bagby and Ravitz2003; Nguyen et al., Reference Nguyen, Klein, Meyer, Austin and Abbott2015). However, self-reported scales measuring depression have been found to capture the genetic variance well when compared with structured clinical interviews, which is considered the gold standard (Foley et al., Reference Foley, Neale and Kendler2001; Gjerde et al., Reference Gjerde, Røysamb, Czajkowski, Reichborn-Kjennerud, Ørstavik, Kendler and Tambs2011).

In summary, we found high correlations between MDD, GAD, and burnout (ranging from 0.53 to 0.71), which were best explained by a model containing an underlying common factor influenced by genetics (58%) and unique environment (42%). All genetic risk factors for MDD were mediated through this factor, while GAD and burnout also had unique genetic risk factors. Burnout was influenced by unique environmental factors to a larger degree than MDD and GAD.

Acknowledgments

This study was supported by grants from the Swedish Research Council for Health, Working Life and Welfare (2009-0548), Karolinska Institutet Doctoral student funding (KID), and the Swedish Society of Medicine, and Magnus Bergvall Foundation. The Swedish Twin Registry is supported by Sweden's Department of Higher Education, AstraZeneca, and the Swedish Research Council. The Study of Twin Adults: Genes and Environment (STAGE) was supported by the National Institute of Health, USA, (Grant numbers DK 066134 and CA 085739).

Conflict of Interest

None.

References

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.Google Scholar
Barlow, D. H. (2004). Psychological treatments. American Psychologist, 59, 869878.CrossRefGoogle ScholarPubMed
Bianchi, R., Schonfeld, I. S., & Laurent, E. (2015a). Burnout-depression overlap: A review. Clinical Psychology Review, 36, 2841.CrossRefGoogle ScholarPubMed
Bianchi, R., Schonfeld, I. S., & Laurent, E. (2015b). Is it time to consider the ‘Burnout Syndrome’ A distinct illness? Frontiers in Public Health, 3, 158.CrossRefGoogle ScholarPubMed
Blom, V., Bergstrom, G., Hallsten, L., Bodin, L., & Svedberg, P. (2012). Genetic susceptibility to burnout in a Swedish twin cohort. European Journal of Epidemiology, 27, 225231.CrossRefGoogle Scholar
Blom, V., Bodin, L., Bergstrom, G., Hallsten, L., & Svedberg, P. (2013). the importance of genetic and shared environmental factors for the associations between job demands, control, support and burnout. PLoS One, 8, e75387.CrossRefGoogle ScholarPubMed
Blom, V., Sverke, M., Bodin, L., Bergstrom, G., Lindfors, P., & Svedberg, P. (2014). Work-home interference and burnout: A study based on Swedish twins. Journal of Occupational and Environmental Medicine, 56, 361366.CrossRefGoogle ScholarPubMed
Boker, S., Neale, M., Maes, H., Wilde, M., Spiegel, M., Brick, T., . . . Fox, J. (2011). OpenMx: An open source extended structural equation modeling framework. Psychometrika, 76, 306317.CrossRefGoogle ScholarPubMed
Brown, T. A. (2007). Temporal course and structural relationships among dimensions of temperament and DSM-IV anxiety and mood disorder constructs. Journal of Abnormal Psychology, 116, 313328.CrossRefGoogle ScholarPubMed
Carlbring, P., Forslin, P., Ljungstrand, P., Willebrand, M., Strandlund, C., Ekselius, L., & Andersson, G. (2002). Is the internet-administered CIDI-SF equivalent to a clinician-administered SCID Interview. Cognitive Behaviour Therapy, 31, 183189.CrossRefGoogle Scholar
Cross-Disorder Group of the Psychiatric Genomics Consortium. (2013). Identification of risk loci with shared effects on five major psychiatric disorders: A genome-wide analysis. Lancet, 381, 13711379.CrossRefGoogle Scholar
Ding, Y. W., Qu, J. W., Yu, X. S., & Wang, S. (2014). The mediating effects of burnout on the relationship between anxiety symptoms and occupational stress among community healthcare workers in China: A cross-sectional study. PLoS One, 9, e107130.CrossRefGoogle Scholar
Dyrbye, L. N., West, C. P., Satele, D., Boone, S., Tan, L., Sloan, J., & Shanafelt, T. D. (2014). Burnout Among U.S. medical students, residents, and early career physicians relative to the general U.S. population. Academic Medicine, 89, 443451.CrossRefGoogle Scholar
Ejeby, K., Savitskij, R., Ost, L. G., Ekbom, A., Brandt, L., Ramnero, J., . . . Backlund, L. G. (2014). Randomized controlled trial of transdiagnostic group treatments for primary care patients with common mental disorders. Family Practice, 31, 273280.CrossRefGoogle ScholarPubMed
Enzmann, D., Schaufeli, W. B., Janssen, P., & Rozeman, A. (1998). Dimensionality and validity of the Burnout Measure. Journal of Occupational and Organizational Psychology, 71, 331351.CrossRefGoogle Scholar
Farvolden, P., McBride, C., Bagby, R. M., & Ravitz, P. (2003). A web-based screening instrument for depression and anxiety disorders in primary care. Journal of Medical Internet Research, 5, e23.CrossRefGoogle ScholarPubMed
First, M. B., Spitzer, R. L., Gibbon, M., Williams, J. B. W., & Benjamin, L. (1996). Structured clinical interview for DSM-IV-patients edition (with psychotic screen, version 2.0). New York: Biometrics Research Department, New York State Psychiatric Institute.Google Scholar
Foley, D. L., Neale, M. C., & Kendler, K. S. (2001). Genetic and environmental risk factors for depression assessed by subject-rated Symptom Check List versus Structured Clinical Interview. Psychological Medicine, 31, 14131423.CrossRefGoogle ScholarPubMed
Furberg, H., Lichtenstein, P., Pedersen, N., Thornton, L., Bulik, C., Lerman, C., & Sullivan, P. (2008). The STAGE cohort: A prospective study of tobacco use among Swedish twins. Nicotine & Tobacco Research, 10, 17271735.CrossRefGoogle ScholarPubMed
Gatt, J. M., Burton, K. L., Williams, L. M., & Schofield, P. R. (2015). Specific and common genes implicated across major mental disorders: A review of meta-analysis studies. Journal of Psychiatric Research, 60, 113.CrossRefGoogle ScholarPubMed
Gjerde, L. C., Røysamb, E., Czajkowski, N., Reichborn-Kjennerud, T., Ørstavik, R. E., Kendler, K. S., & Tambs, K. (2011). Strong genetic correlation between interview-assessed internalizing disorders and a brief self-report symptom scale. Twin Research and Human Genetics, 14, 6472.CrossRefGoogle Scholar
Global Burden of Disease Study 2013 Collaborators. (2015). Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet, 386, 743800.CrossRefGoogle Scholar
Gustafsson, H., Kentta, G., Hassmen, P., & Lundqvist, C. (2007). Prevalence of burnout in competitive adolescent athletes. Sport Psychologist, 21, 2137.CrossRefGoogle Scholar
Hallsten, L., Josephson, M., & Torgén, M. (2005). Performance-based self-esteem - A driving force in burnout processes and its assessment. Stockholm: National Institute for Working Life.Google Scholar
Hettema, J., Neale, M., & Kendler, K. (2001). A review and meta-analysis of the genetic epidemiology of anxiety disorders. American Journal of Psychiatry, 158, 15681578.CrossRefGoogle ScholarPubMed
Hur, Y. M., Burri, A., & Spector, T. D. (2012). The genetic and environmental structure of the covariation among the symptoms of insomnia, fatigue, and depression in adult females. Twin Research and Human Genetics, 15, 720726.CrossRefGoogle ScholarPubMed
Kendler, K. S., Gardner, C. O., Gatz, M., & Pedersen, N. L. (2007). The sources of co-morbidity between major depression and generalized anxiety disorder in a Swedish national twin sample. Psychological Medicine, 37, 453462.CrossRefGoogle Scholar
Kendler, K. S., Gatz, M., Gardner, C. O., & Pedersen, N. L. (2006a). Personality and major depression: A Swedish longitudinal, population-based twin study. Archives of General Psychiatry, 63, 11131120.CrossRefGoogle ScholarPubMed
Kendler, K. S., Gatz, M., Gardner, C. O., & Pedersen, N. L. (2006b). A Swedish national twin study of lifetime major depression. Am J Psychiatry, 163, 109114.CrossRefGoogle ScholarPubMed
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (1992). Major depression and generalized anxiety disorder. Same genes, (partly) different environments? Archives of General Psychiatry, 49, 716722.CrossRefGoogle ScholarPubMed
Lichtenstein, P., de Faire, U., Floderus, B., Svartengren, M., Svedberg, P., & Pedersen, N. L. (2002). The Swedish Twin Registry: A unique resource for clinical, epidemiological and genetic studies. Journal of Internal Medicine, 252, 184205.CrossRefGoogle ScholarPubMed
Lichtenstein, P., Sullivan, P. F., Cnattingius, S., Gatz, M., Johansson, S., Carlstrom, E., . . . Pedersen, N. L. (2006). The Swedish Twin Registry in the third millennium: An update. Twin Research and Human Genetics, 9, 875882.CrossRefGoogle ScholarPubMed
Lindstrom, C., Aman, J., & Norberg, A. L. (2010). Increased prevalence of burnout symptoms in parents of chronically ill children. Acta paediatrica, 99, 427432.CrossRefGoogle ScholarPubMed
Magnusson, P. K., Almqvist, C., Rahman, I., Ganna, A., Viktorin, A., Walum, H., . . . Lichtenstein, P. (2013). The Swedish Twin Registry: establishment of a biobank and other recent developments. Twin Research and Human Genetics, 16, 317329.CrossRefGoogle ScholarPubMed
Markon, K. E., & Krueger, R. F. (2004). An empirical comparison of information-theoretic selection criteria for multivariate behavior genetic models. Behavior Genetics, 34, 593610.CrossRefGoogle ScholarPubMed
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of Psychology, 52, 397422.CrossRefGoogle ScholarPubMed
Mather, L., Bergström, G., Blom, V., & Svedberg, P. (2014). The covariation between burnout and sick leave due to mental disorders is explained by a shared genetic liability: A prospective Swedish twin study with a five-year follow-up. Twin Research and Human Genetics, 17, 535544.CrossRefGoogle ScholarPubMed
Mather, L., Blom, V., & Svedberg, P. (2014). Stressful and traumatic life events are associated with burnout-a cross-sectional twin study. International Journal of Behavioral Medicine, 21, 899907.CrossRefGoogle ScholarPubMed
Middeldorp, C. M., Cath, D. C., & Boomsma, D. I. (2006). A twin-family study of the association between employment, burnout and anxious depression. Journal of Affective Disorders, 90, 163169.CrossRefGoogle ScholarPubMed
Middeldorp, C. M., Cath, D. C., Van Dyck, R., & Boomsma, D. I. (2005). The co-morbidity of anxiety and depression in the perspective of genetic epidemiology. A review of twin and family studies. Psychological Medicine, 35, 611624.CrossRefGoogle ScholarPubMed
Middeldorp, C. M., Stubble, J. H., Cath, D. C., & Boomsma, D. I. (2005). Familial clustering in burnout: A twin-family study. Psychological Medicine, 35, 113120.CrossRefGoogle ScholarPubMed
Mikolajewski, A. J., Allan, N. P., Hart, S. A., Lonigan, C. J., & Taylor, J. (2013). Negative affect shares genetic and environmental influences with symptoms of childhood internalizing and externalizing disorders. Journal of Abnormal Child Psychology, 41, 411423.CrossRefGoogle ScholarPubMed
Nguyen, D. P., Klein, B., Meyer, D., Austin, D. W., & Abbott, J. A. (2015). The diagnostic validity and reliability of an internet-based clinical assessment program for mental disorders. Journal of Medical Internet Research, 17, e218.CrossRefGoogle ScholarPubMed
Nyklicek, I., & Pop, V. J. (2005). Past and familial depression predict current symptoms of professional burnout. Journal of Affective Disorders, 88 (1), 6368.CrossRefGoogle ScholarPubMed
Ormela, J., Jeronimusa, B. F., Kotovc, R., Riesea, H., Bosa, E. H., Hankind, B., . . . Oldehinkela, A. J. (2013). Neuroticism and common mental disorders: Meaning and utility of a complex relationship. Clinical Psychology Review, 33, 686697.CrossRefGoogle Scholar
Paulus, D. J., Talkovsky, A. M., Heggeness, L. F., & Norton, P. J. (2015). Beyond negative affectivity: A hierarchical model of global and transdiagnostic vulnerabilities for emotional disorders. Cognitive Behaviour Therapy, 44, 389405.CrossRefGoogle ScholarPubMed
Pines, A. M., & Aronson, E. (1988). Career burnout: Causes and cures. New York: Free Press.Google Scholar
Pines, A. M., Aronson, E., & Kafry, D. (1981). Burnout: from tedium to personal growth. New York: The Free Press.Google Scholar
Purcell, S. (2013). Statistical methods in behavioral genetics. In Plomin, R., DeFries, J. C., Knopik, V. S. & Neiderhiser, J. M. (Eds.), Behavioral genetics (6th ed., pp. 357411). New York: Worth Publishers.Google Scholar
R Development Core Team. (2010). R foundation for statistical computing. Vienna, Austria: Author.Google Scholar
Raftery, A. E. (1995). Bayesian model selection in social research. Sociological Methodology, 25, 111163.CrossRefGoogle Scholar
Rijsdijk, F., & Sham, P. (2002). Analytic approaches to twin data using structural equation models. Briefings in Bioinformatics, 3, 119133.CrossRefGoogle ScholarPubMed
Schaufeli, W. B., Bakker, A. B., Hoogduin, K., Schaap, C., & Klader, A. (2001). On the clinical vailidty of the Maslach burnout inventory and the burnout measure. Psychology and Health, 16, 565582.CrossRefGoogle Scholar
Shirom, A., & Ezrachi, Y. (2003). On the discriminant validity of burnout, depression and anxiety: A re-examination of the Burnout Measure. Anxiety, Stress and Coping, 16, 8397.CrossRefGoogle Scholar
Sullivan, P. F., Neale, M. C., & Kendler, K. S. (2000). Genetic epidemiology of major depression: Review and meta-analysis. American Journal of Psychiatry, 157, 15521562.CrossRefGoogle ScholarPubMed
Takai, M., Takahashi, M., Iwamitsu, Y., Ando, N., Okazaki, S., Nakajima, K., . . . Miyaoka, H. (2009). The experience of burnout among home caregivers of patients with dementia: Relations to depression and quality of life. Archives of Gerontology and Geriatrics, 49, e1–e5.CrossRefGoogle ScholarPubMed
Takai, M., Takahashi, M., Iwamitsu, Y., Oishi, S., & Miyaoka, H. (2011). Subjective experiences of family caregivers of patients with dementia as predictive factors of quality of life. Psychogeriatrics 11, 98104.CrossRefGoogle ScholarPubMed
Toker, S., Shirom, A., Shapira, I., Berliner, S., & Melamed, S. (2005). The association between burnout, depression, anxiety, and inflammation biomarkers: C-reactive protein and fibrinogen in men and women. Journal of Occupational Health Psychology, 10, 344362.CrossRefGoogle ScholarPubMed
Figure 0

TABLE 1 Frequencies (%) of Major Depressive Disorder, Generalized Anxiety Disorder, Burnout, and Zygosity Among 25,378 Swedish Twins, Stratified on Sex

Figure 1

TABLE 2 Polychoric (Burnout) and Tetrachoric (Major Depressive Disorder and Generalized Anxiety Disorder) Within Pair and Cross-Twin, Cross-Trait Correlations With 95% Confidence Intervals Among 8,646 Complete Twin Pairs

Figure 2

TABLE 3 Model Fit Statistics of the Univariate Models for Burnout, Major Depressive Disorder and Generalized Anxiety Disorder and for the Multivariate Models

Figure 3

TABLE 4 Proportions of Additive Genetic (a2) and Unique Environmental (e2) Effects From the Best-Fitting Univariate Models With 95% Confidence Intervals

Figure 4

FIGURE 1. Path estimates for the best-fitting model with 95% confidence intervals.