Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jan 31;41(1):e70006. doi: 10.1002/smi.70006

Work‐Stressors and Depression and Anxiety—A Longitudinal Study of the Moderating Role of Self‐Compassion

E de Krijger 1,2,✉, P M ten Klooster 2, E Geuze 1,3, S M Kelders 2, E T Bohlmeijer 2
PMCID: PMC11784929  PMID: 39888660

ABSTRACT

Self‐compassion has been defined as being open to one's suffering, not avoiding or disconnecting from it, coupled with the kind and caring motivation to alleviate one's suffering. There is increasing evidence that self‐compassion might function as a buffer against the negative mental health effects of experiencing work‐stressors. However, while this moderating role of self‐compassion has been demonstrated when measures of subjective stress are used, different studies that use measures of more objective potential stressors failed to demonstrate a moderating role of self‐compassion. Furthermore, while cross‐sectional studies offer increasing support for this moderation effect of self‐compassion, few studies have examined this in longitudinal designs which may provide more robust insight into the role of self‐compassion on the relation between work stress experiences and symptoms of depression and anxiety. The aims of the current study were to examine whether self‐compassion moderates the concurrent and prospective association between a measure of potential work‐stressors and depressive and anxious symptomatology. The method consisted of a longitudinal survey study in a sample of 246 military personnel with three measurements, half a year apart from each other. Latent moderated structural equation modelling was conducted to test the potential moderating effect of self‐compassion. Self‐compassion was shown to significantly moderate the association between work‐stressors and depressive and anxiety symptomatology, both cross‐sectionally and prospectively after 6 and 12 months. Specifically, the experience of work‐stressors was positively associated with symptoms of depression and anxiety when self‐compassion was low and this association became weaker when self‐compassion was at a medium or high level. The results of the current study suggest that higher levels of work‐related stress covary more strongly with symptoms of depression and anxiety over time in personnel with lower levels of self‐compassion.

Keywords: anxiety, depression, job stress, positive psychology

1. Introduction

Since the number of workers exposed to work‐stress has increased over the past decades, work‐stress poses an increasing threat to public mental health (Melchior et al. 2007; Virtanen et al. 2007). Both stress in general, and work‐stress specifically, have been associated with the onset and exacerbation of mental illness (Esch et al. 2002; Karyotaki et al. 2020; Melchior et al. 2007, Virtanen et al. 2007). In effect, people who are exposed to high psychological job demands have been demonstrated to be twice as likely to suffer from a depression or anxiety disorder, compared to people with low job demands (Melchior et al. 2007). Based on these observations, there has been an increasing call to reduce work stress levels or help workers cope with work stress (Melchior et al. 2007), and one factor that might be of interest in this regard is self‐compassion.

Self‐compassion has been described as an adaptive way of responding to oneself in times of hardship or suffering (K. Neff 2003; Strauss et al. 2016). Neff defined it as ‘being touched by and open to one's suffering, not avoiding or disconnecting from it, generating the desire to alleviate one's suffering and to heal oneself with kindness’ (2003, 87). Research on the mental health benefits of self‐compassion has blossomed over the past decades (Lou, Wang, and Minkov 2022; MacBeth and Gumley 2012; Marsh, Chan, and MacBeth 2018; Zessin, Dickhäuser, and Garbade 2015), and self‐compassion is increasingly suggested to function as a resilience mechanism or a ‘buffer’ against the negative mental health effects of stress or other adverse events (Trompetter, de Kleine, and Bohlmeijer 2017). People that score high on self‐compassion are thought to respond to stress and stressful events in a more adaptive manner that lessens the negative mental health effects of stress. Self‐compassionate people might therefore prove to be more resilient in the face of stress, compared to less self‐compassionate people. This might be because self‐compassion ensures that the awareness of stress or stressful events is met with understanding, kindness, and care, as opposed to self‐blame or harsh self‐condemnation. Instead of causing themselves extra distress with these negative ways of relating to themselves, self‐compassionate people are therefore expected to moderate their response to stressful events with kindness and care in a way that has been described as ‘social support turned inwards’ (Bluth et al. 2016). Furthermore, self‐compassion includes turning one's attention towards the experience of stress instead of avoiding or disconnecting from it, and this has been found to allow for more adaptive emotional processing (Aldao, Nolen‐Hoeksema, and Schweizer 2010; Chawla and Ostafin 2007). In the case of intense stressful events this allows for a natural exposure process (Thompson and Waltz 2008). In general, improved awareness, exploration and understanding of the stressful experience enables more adaptive regulation strategies (Inwood and Ferrari 2018).

Self‐compassion has indeed been demonstrated to buffer the negative mental health effects of stress (Abdollahi et al. 2021; Beshai et al. 2022; Chan, Yung, and Nie 2020; Hsieh et al. 2019; Keng and Hwang 2022; Lathren, Bluth, and Park 2019; Stutts et al. 2018), and there have also been studies that demonstrate this buffering effect specifically for work‐stress or (academic) burn‐out (Cheraghian et al. 2016, Lee and Lee 2022; McCade, Frewen, and Fassnacht 2021; Wu, Cao, and Du 2023). These studies demonstrate that self‐compassion moderates the overall positive association between work‐, or other forms of stress, in the sense that this association is weaker for people that score high on self‐compassion. Unfortunately, most of these studies used cross‐sectional designs, which limits inferences about the role of self‐compassion in the prospective association between stress and mental health. Since self‐compassion and outcome measures like depression were measured at one point in time, there is a chance that low self‐compassion was merely found to be associated with higher levels of depression because people that are depressed might be less likely to rate themselves high on a measure of self‐compassion at the same timepoint. Regarding stress that is not specifically work‐related, there is some support from longitudinal studies which demonstrated that self‐compassion moderated the effects of perceived stress on mental illness prospectively after two (Keng and Hwang 2022) or 6 months (Stutts et al. 2018). Since in these studies self‐compassion was measured at a different point in time than the outcome measures, these studies offer additional support for self‐compassion as a moderator of the positive association between stress and mental illness. Overall, however, studies that used longitudinal designs remain scarce and have, to the best of our knowledge, not been used to study self‐compassion in relation to specifically work‐stress and mental illness.

Furthermore, most of the studies that offer support for self‐compassion as a moderator of the association between stress and mental illness, have operationalised stress as the subjective experience of stress (e.g., ‘I'm having feelings of being overwhelmed, overworked or overburdened’). A potential limitation of these studies is that subjective reports of stress and self‐reported mental health status contain so much conceptual overlap that the correlation between them might lack theoretical or practical significance (Lazarus et al. 1985). This conceptual overlap can be resolved by operationalising stress as more objective potential stressors (e.g., ‘I had a serious accident’) but many studies that used measures of potential stressors did not find evidence for a moderating effect of self‐compassion (Dev, Fernando, and Consedine 2020; Ford et al. 2017; Kroshus, Hawrilenko, and Browning 2022; Salinger and Whisman 2021). Salinger and colleagues suggested that potential stressors do not trigger a similar subjective experience of stress for everyone, and that self‐compassion is only deployed when somebody subjectively experiences stress or suffering (Salinger and Whisman 2021). On the other hand, one recent cross‐sectional study that focused on measures of potential work stressors did find significant moderation effects for self‐compassion (Wu, Cao, and Du 2023), but to the best of our knowledge longitudinal designs focusing on objective potential stressors have failed to demonstrate this moderating role of self‐compassion (Kroshus, Hawrilenko, and Browning 2022).

To summarise, though there is growing evidence that self‐compassion buffers the impact of work‐related stress, few studies used a longitudinal design preventing more robust insight into the role of self‐compassion on the relation between work stress experiences and symptoms of depression and anxiety. Also, while self‐compassion has been demonstrated to buffer the association between the subjective experience of stress and depressive and anxious symptomatology, results regarding objective potential stressors remain mixed and positive results have to the best of our knowledge only been demonstrated cross‐sectionally. The present study therefore employed a longitudinal design with three measurements, half a year apart from each other, and focused on a measure of potential work‐stressors. We hypothesised that self‐compassion will act as a moderator of the overall positive association between potential work‐stressors and depressive and anxious symptomatology, in the sense that this association will be weaker for participants that score high on a measure of self‐compassion compared to participants that score low on self‐compassion.

This hypothesis was tested in a population of active‐duty military personnel, because compared to civilian workers, military service is associated with both increased reports of suffering from work stress (Pflanz and Sonnek 2002) and increased rates of mental illness (Stevelink et al. 2018). The increased report of work‐stress among military personnel appears to be caused by increased common work‐stressors, such as changes in work hours, work responsibility or type of work, instead of military specific stressors such as deployment or exposure to combat (Pflanz and Sonnek 2002). Pflanz and Sonnek suggested that this might be explained by the military's emphasis on discipline, obeying orders and respecting rank hierarchy, which might lower the experience of autonomy and control, and thereby increase suffering from these common work‐stressors (2002). Given this possibly challenging work context, that might increase the risk of experiencing both increased work stress (Pflanz and Sonnek 2002) and increased mental illness (Stevelink et al. 2018), we were interested in investing in the positive mental health and resilience of military personnel. We started with exploring self‐compassion as a possible means to do so because of the amount of research offering support for a buffering effect of self‐compassion (Abdollahi et al. 2021; Beshai et al. 2022; Chan, Yung, and Nie 2020; Cheraghian et al. 2016; Hsieh et al. 2019; Keng and Hwang 2022; Lathren, Bluth, and Park 2019; Lee and Lee 2022; McCade, Frewen, and Fassnacht 2021; Stutts et al. 2018; Wu, Cao, and Du 2023). Furthermore self‐compassion was also chosen specifically for this population because research suggests that instead of self‐compassionately approaching and coping with emotional distress, military personnel tend towards concealment and suppression of distress to maintain their strong military identity (Green et al. 2010, Lorber and Garcia 2010, McAllister, Callaghan, and Fellin 2019; Neilson et al. 2020), which has been related to increased psychopathology (Neilson et al. 2020).

2. Method

The current study was not pre‐registered, but the data were specifically collected with the intent to study the moderating role of self‐compassion on the association between stress and depression and anxiety. At the same time additional data was collected for two validation studies, one of which was focused on the validation of the Sussex Oxford Compassion Scale‐Self (de Krijger et al. 2022).

2.1. Participants

Participants were recruited by sending an invitation for the survey to a random selection of 1200 Dutch active‐duty military personnel. Participants were invited to join the study with both an email and a letter. Reminders were sent after 3 and 6 weeks. Of these 1200 recipients, a total of 927 either replied that they were not willing to join or did not respond at all. Another 23 recipients were excluded because they indicated that they were currently receiving treatment for mental health issues, which was formulated as an exclusion criterion for the present study. This exclusion criterion was formulated because we aimed to study a non‐patient population because we were interested in investing in self‐compassion as a universal wellbeing intervention for military personnel (instead of investing in self‐compassion as treatment for current mental illness). Another four recipients could not be included because they failed to provide informed consent. Potential participants were asked about their current mental health status, and to provide informed consent, at the start of the survey. If they failed to provide informed consent or indicated that they were receiving treatment for mental health issues at the time of the study, the survey was closed and participants received a message to thank them for their willingness to participate but that they were excluded from participation for one of these reasons.

2.2. Measures

2.2.1. Self‐Compassion Scale‐Short Form (SCS‐SF)

Self‐compassion was measured with the Self‐Compassion Scale‐Short Form (SCS‐SF), a 12‐item measure on a 7‐point scale ranging from ‘rarely or never’ to ‘almost always’. The Dutch version of the SCS‐SF demonstrates good reliability and validity (Raes et al. 2011). For the current study separate sum scores were calculated for the six positively worded items of the SCS‐SF and the six negatively worded items, as previous research has shown that the positively formulated items (or compassionate ways of self‐responding) and negatively formulated items (or uncompassionate ways of self‐responding) constitute relatively distinct factors of self‐compassion (Brenner et al. 2017; Coroiu et al. 2018; Muris, Otgaar, and Petrocchi 2016, Muris et al. 2018; López et al. 2015; Muris and Petrocchi 2017; Babenko and Guo 2019). For the current sample good internal consistency was demonstrated for both the positively formulated items of the SCS‐SF (Cronbach's α = 0.82) and the negatively formulated items (Cronbach's α = 0.87).

2.2.2. Patient Health Questionnaire‐9 (PHQ‐9)

Depression was measured with the Patient Health Questionnaire‐9 (PHQ‐9). The PHQ‐9 is a nine‐item measure that asks respondents to indicate the extent to which they experienced DSM‐5 symptoms of depression in the past 2 weeks, on a scale ranging from 0 (not at all) to 3 (nearly every day). The total score consists of the sum of the nine items. The Dutch version of the PHQ‐9 demonstrates good reliability and validity (Zuithoff et al. 2010), and for the current sample good internal consistency was demonstrated as well (Cronbach's α for the three measurements between the range of 0.80–0.86).

2.2.3. Generalised Anxiety Disorder‐7 (GAD‐7)

Anxiety was measured with the Generalised Anxiety Disorder‐7 (GAD‐7) (Spitzer et al. 2006). The GAD‐7 is a seven‐item measure that asks respondents to indicate the extent to which they experienced DSM‐5 symptoms of generalised anxiety in the past 2 weeks, on a scale ranging from 0 (not at all) to 3 (every day). The total score consists of the sum of the seven items. The Dutch version of the GAD‐7 that was used for the present study demonstrates good reliability and validity (Donker et al. 2011), and in the current sample good internal consistency was demonstrated as well (Cronbach's α for the three measurements between the range of 0.85–0.88).

2.2.4. The Questionnaire on the Experience and Evaluation of Work (VBBA)

Objective potential work stress was measured with three subscales of the Questionnaire on the Experience and Evaluation of Work (VBBA; Van Veldhoven and Meijman 1994). The VBBA is a Dutch questionnaire that measures different domains of work experience. The three subscales that were used for the present study are: ‘work pressure’ (11 items), ‘emotional burden’ (7 items) and ‘mental burden’ (7 items). The items of these scales are specifically focused on job characteristics that can potentially trigger the subjective experience of stress, instead of focusing on subjective stress itself like is the case for many other measures of stress. Examples of items for the three subscales are:

  • –

    Work pressure: ‘do you have to work really fast?’, ‘do you have a lot of work to do?’ and ‘do you have too little work to do?’.

  • –

    Emotional burden: ‘is your job emotionally strenuous?’, ‘are you confronted with things that personally impact you during your work?’ and ‘do you feel personally attacked or threatened during your work?’.

  • –

    Mental burden: ‘does your work require a lot of focus?’, ‘do you have to keep track of many things at the same time during your work?’ and ‘does your work require great attention to detail?’.

The items are measured with a 4‐point scale ranging from ‘never’ to ‘always’. The least favourable answer (which is ‘always’ for most of the questions, but ‘never’ for a few reverse formulated questions), receives a score of 3 and the most favourable answer receives a score of 0. The VBBA demonstrates good reliability and validity (Van Veldhoven and Meijman, 1994), and in the current sample the items of the three subscales that were used, together, demonstrate good internal consistency as well (Cronbach's α = 0.87).

2.3. Statistical Analysis

The potential moderating effect of self‐compassion on the relationship between work‐stress and symptoms of depression and anxiety was tested using the latent moderated structural equations (LMS) method (Klein and Moosbrugger 2000) as implemented in Mplus 8.3 (Maslowsky, Jager, and Hemken 2015; Muthén and Muthén 2017). Latent variable interaction has several advantages over traditional regression‐based methods using observed indicators such as scale scores. Most importantly, by their ability to explicitly model measurement error in observed variables, latent interaction models maximise power to detect interaction effects over traditional regression models with observed variables (Lodder et al. 2019; Schoemann and Jorgensen 2021). Additionally, structural equation modelling as implemented in Mplus uses full information maximum likelihood (FIML) estimation, meaning all available data is used in the analysis and only people with no observations on any of the indicators in the model are deleted.

Both cross‐sectional and longitudinal latent interaction models were estimated (see Figures 1 and 2). For the cross‐sectional models, only data from the first (T0) measurement were used. In this model, the latent independent variable was work stress, as indicated by the observed sum scores of the work pressure, mental burden and emotional burden subscales of the VBBA. The latent moderator variable self‐compassion was indicated by the mean observed scores of the positive and negative items of the SCS‐SF. The latent dependent variable ‘psychopathology’ was indicated by observed sum scores of anxiety (GAD‐7) and depression (PHQ‐9). For the two longitudinal models, the independent latent psychopathology variables were calculated using GAD‐7 and PHQ‐9 at T1 and T2, respectively, while latent T0 psychopathology was added as an additional independent (control) variable.

FIGURE 1.

FIGURE 1

Open in a new tab

Baseline cross‐sectional latent moderated structural equations model. Coefficients are standardised parameter estimates. Residual variances are omitted for clarity. ***p < 0.001.

FIGURE 2.

FIGURE 2

Open in a new tab

Longitudinal latent moderated structural equations models (controlling for baseline psychopathology). Upper panel: psychopathology at T1; lower panel: psychopathology at T2. Coefficients are standardised parameter estimates. Residual variances are omitted for clarity. *p < 0.05; **p < 0.01; ***p < 0.001.

As is common in LMS, the cross‐sectional and longitudinal moderation models were estimated in a two‐step approach (Maslowsky, Jager, and Hemken 2015). First, a measurement model was established, which in LMS is the main effects model without the latent interaction term. Next, the structural model was estimated which included the latent interaction term. In the structural models, all indicators were standardised prior to analysis to obtain standardised beta regression coefficients (parameter estimates). As model fit indices are not available for LMS models, model fit was examined for the measurement models only (Maslowsky, Jager, and Hemken 2015). Since the latent interaction term does not have a mean or variance parameter and no covariance parameters with other variables, adding the latent interaction term should not negatively affect the fit of the model (Asparouhov and Muthén 2019). Goodness of fit of the measurement models was assessed by the root mean square error of approximation (RMSEA), standardised root mean squared residual (SRMR), comparative fit index (CFI), and Tucker–Lewis index (TLI). For RMSEA and SRMR values of ≤ 0.08 and ≤ 0.05, and for CFI and TLI values of ≥ 0.90 and ≥ 0.95 were, respectively, considered as reflecting acceptable and good model fit (Browne and Cudeck 1992; Hu and Bentler 1999). After ensuring the fit of the measurement models, the structural models were estimated using the XWITH command of Mplus for the two latent variables of work stress and self‐compassion. Likelihood ratio (−2 log likelihood chi‐square) tests were performed to compare the model fit of the structural models against their respective measurement models. A significant likelihood test indicates that the addition of the latent interaction term improved model fit.

A significant parameter estimate for latent ‘psychopathology’ (symptoms of depression and anxiety) on the interaction term in the structural model(s) indicates a moderating effect of latent self‐compassion. In case of a significant latent interaction term in the structural model, simple slopes were plotted to facilitate interpretation of the interaction effect.

As relationships with symptoms of depression and anxiety may be confounded by demographic or professional characteristics of the participants, additional sensitivity analyses were conducted for the structural model by controlling symptoms of psychopathology for gender (female vs. male), age (continuous), history of deployment (yes vs. no), history of mental health treatment (yes vs. no) and rank (private/corporal v. non‐commissioned or commissioned officer).

3. Results

3.1. Participants

A total of 246 participants completed the first survey, and they were sent an invitation via email for the second and third survey 6 and 12 months after completing the first survey, respectively. A total of 202 participants completed the second survey and the final survey was completed by 163 participants. See Table 1 for demographical information on the sample in each of the three waves.

TABLE 1.

Sample demographics.

Demographics T0 (n = 246) T1 (n = 202/82%) T2 (n = 163/80%)
Male 210 (85.4%) 171 (84.7%) 141 (86.5%)
Female 33 (13.4%) 28 (13.9%) 19 (11.7%)
Mean age (SD) 43.6 (10.8) 44.1 (10.4) 44.4 (10.2)
Private 9 (3.7%) 5 (2.5%) 3 (1.8%)
Corporal 13 (5.3%) 10 (5.0%) 7 (4.3%)
Non‐commissioned officer 121 (49.2%) 98 (48.5%) 79 (48.5%)
Officer 104 (42.3%) 89 (44.1%) 74 (45.4%)
History of deployment 198 (80.5%) 165 (81.7%) 133 (81.6%)
History of mental health treatment 66 (26.8%) 54 (26.7%) 44 (27.0%)
Open in a new tab

3.2. Descriptive Statistics

Table 2 reports the descriptives and zero‐order intercorrelations of the major variables. Both positive and negative self‐compassion demonstrated significant correlations with anxiety and depression at T0, T1 and T2. Work pressure at T0 was furthermore significantly associated with negative self‐compassion at T0, and with anxiety and depression at T0, T1 and T2.

TABLE 2.

Descriptive statistics and pearson correlations.

Variable M SD 1 2 3 4 5 6 7 8 9 10
1. T0 work pressure 11.68 4.71
2. T0 mental burden 13.60 4.09 0.456**
3. T0 emotional burden 5.88 3.08 0.420** 0.428**
4. T0 positive self‐compassion 29.39 7.40 −0.122 0.060 0.079
5. T0 negative self‐compassion 15.68 8.10 0.226* 0.104 0.121 −0.385**
6. T0 depression 2.13 2.68 0.202** 0.121 0.078 −0.313** 0.435**
7. T0 anxiety 1.93 2.71 0.270** 0.213** 0.112 −0.321** 0.455** 0.771**
8. T1 depression 2.62 3.45 0.237** 0.184** 0.175* −0.223** 0.426** 0.622** 0.647**
9. T1 anxiety 2.36 3.12 0.287** 0.171* 0.106 −0.278** 0.406** 0.595** 0.679** 0.863**
10. T2 depression 2.48 3.13 0.237** 0.122 0.164* −0.256** 0.369** 0.651** 0.591** 0.769** 0.695**
11. T2 anxiety 2.18 2.94 0.288** 0.198* 0.135 −0.242** 0.405** 0.593** 0.664** 0.724** 0.734** 0.803**
Open in a new tab

*p < 0.05, **p < 0.01.

3.3. Latent Moderated Structural Equations

3.3.1. Model Fit

In the first step of the analysis a measurement model was established by estimating a main effects model without the latent interaction term, for both the baseline cross‐sectional model, and for the longitudinal models with depression and anxiety measured at T1 and for T2. Measurement models showed adequate to good fit according to all fit indices for both the cross‐sectional and longitudinal measurement models (see Table 3).

TABLE 3.

Fit indices for cross‐sectional and longitudinal measurement models.

χ 2 (df) RMSEA (90% CI) SRMR CFI TLI
Cross‐sectional (T0) 22.710 (11) 0.066 (0.025–0.104) 0.047 0.975 0.953
Longitudinal (T0–T1) 42.610 (21) 0.065 (0.036–0.093) 0.040 0.976 0.959
Longitudinal (T0–T2) 47.226 (21) 0.071 (0.044–0.099) 0.042 0.966 0.941
Open in a new tab

3.3.2. Latent Interaction Results

In the second step of the analysis the structural models were estimated which included the latent interaction term. This was done for both the baseline cross‐sectional model and for the longitudinal models with depression and anxiety measured at T1 and T2. See Figures 1 and 2 for a graphical representation of the different structural equation models including the latent interaction term. Likelihood ratio tests were significant for both the cross‐sectional (−2 LL difference = 5444.216, p < 0.001) and the longitudinal structural models with depression and anxiety measured at T1 (−2 LL difference = 6301.038, p < 0.001) and T2 (−2 LL difference = 6032.638, p < 0.001) indicating that including the latent interaction term improved model fit.

The latent interaction analyses showed that there was a significant negative interaction between the experience of work stressors and self‐compassion, both cross‐sectionally (β = −0.443, SE = 0.022, p < 0.001), and longitudinally with symptoms of depression and anxiety measured at T1 (β = −0.531, SE = 0.085, p < 0.001) and at T2 (β = −0.474, SE = 0.141, p = 0.001). This indicates that self‐compassion significantly buffered the positive effect of T0 work‐stressors, on T0 symptoms of depression and anxiety but also prospectively on T1 and T2 symptoms of depression and anxiety. Sensitivity analysis controlling for gender, age, history of deployment, history of mental health treatment and rank yielded very similar and significant interaction effects for the cross‐sectional model (β = −0.555, SE = 0.037, p < 0.001) and for the longitudinal model at T1 (β = −0.605, SE = 0.127, p = 0.005). For symptoms of depression and anxiety at T2, the interaction effect was no longer significant (β = −0.018, SE = 0.147, p = 0.902).

3.3.3. Simple Slopes Analysis

To interpret the significant negative interaction between work stress and self‐compassion in the unadjusted models, simple main effects of work stress were examined at low (one standard deviation below the mean), medium (mean) and high (one standard deviation above the mean) levels of self‐compassion. When self‐compassion was low, work‐stress demonstrated a positive correlation with symptoms of psychopathology, both cross‐sectionally and prospectively after 6 and 12 months. However, the simple slopes analyses demonstrate that this correlation between work‐stress and psychopathology became significantly weaker when self‐compassion was medium or high. The simple slopes for the cross‐sectional and longitudinal models are shown in Figure 3.

FIGURE 3.

FIGURE 3

Open in a new tab

Simple slopes plots for baseline cross‐sectional model (A), T1 longitudinal model (B) and T2 longitudinal model (C).

4. Discussion

There is growing evidence that self‐compassion moderates the overall positive association between work‐stress and mental illness. However, the current lack of longitudinal studies prevents more robust insight into the role of self‐compassion on the relation between work stress experiences and symptoms of depression and anxiety. This study aimed to examine whether self‐compassion prospectively moderated the association between potential work‐stressors and depressive and anxiety symptomatology among military personnel using a longitudinal design. In accordance with our hypothesis the current results demonstrate that self‐compassion significantly moderates the association between experiencing potential work‐stressors and symptoms of depression and generalised anxiety, both cross‐sectionally and prospectively after 6 and 12 months. The experience of potential work‐stressors was demonstrated to be positively associated with symptoms of depression and anxiety when self‐compassion was low, and this association became significantly weaker when participants reported medium or high levels of self‐compassion.

The results of the current study suggest that people with higher levels of self‐compassion might have a lower risk to develop depressive and anxiety symptomatology over time in the context of work‐related stress. These results are consistent with earlier research offering support for a moderating effect of self‐compassion on the association between work‐stress and mental illness (Cheraghian et al. 2016, Lee and Lee 2022; McCade, Frewen, and Fassnacht 2021; Wu, Cao, and Du 2023) and adds to this research by demonstrating this effect prospectively as well. Furthermore, the results are in line with earlier studies showing an association between self‐compassion and adaptive coping and emotion‐regulation (Ewert, Vater, and Schröder‐Abé 2021; Finlay‐Jones 2017). Self‐compassion might buffer the negative mental health effects of work‐related stress because a high level of self‐compassion helps people to adaptively regulate and cope with this stress. Primarily it is likely that for people with adequate levels of self‐compassion, the experience of stress is met with the comforting perspective that suffering is common to all human beings. Furthermore, feelings of kindness and care have a soothing effect in the context of potentially stressful circumstances. This might help self‐compassionate people to maintain feelings of positivity and safety in the face of stress, instead of activating extra feelings of threat and negativity with self‐blame and self‐condemnation, which in turn might lead to increased anxiety and depressive symptomatology (Finlay‐Jones 2017). Secondly, self‐compassion has been suggested to be associated with more adaptive coping and regulation of stress because of the openness to suffering that it involves (Finlay Jones 2017). This open and accepting attitude, together with a kind, instead of self‐blaming, response to suffering, might help self‐compassionate people to pay attention to and work with emotional distress instead of avoiding or suppressing it. While in certain situations avoidance and suppression might be adaptive, these strategies are in general associated with increased anxiety and depressive symptomatology (Aldao, Nolen‐Hoeksema, and Schweizer 2010). Self‐compassion likely enables people to flexibly switch between either more avoidance‐based or more approach‐based strategies, and there is increasing recognition for the notion that such flexibility is needed for adaptive emotion regulation (Aldao and Nolen‐Hoeksema 2012).

Furthermore, the current study demonstrates a moderating effect of self‐compassion on the association between depressive and anxiety symptomatology and potential stressors, instead of subjective stress. This is consistent with the results from one recent cross‐sectional study that also focused on potential work stressors instead of subjective distress (Wu, Cao, and Du 2023), and adds to these results by demonstrating this effect prospectively as well. This is important because while previous studies focusing on subjective stress did demonstrate a moderating role of self‐compassion both cross‐sectionally (Cheraghian et al. 2016; Lee and Lee 2022; McCade, Frewen, and Fassnacht 2021) and longitudinally (Stutts et al. 2018), these studies are potentially limited because of the conceptual overlap between subjective stress and mental distress. Demonstrating this effect for potential stressors therefore strengthens the notion that this concerns a theoretically and practically significant association between the experience of stress and depressive and anxious symptomatology. Our results differ from the results of various previous studies that focused on potential stressors, and that did not demonstrate a moderating effect of self‐compassion (Dev, Fernando, and Consedine 2020; Ford er al. 2017; Kroshus, Hawrilenko, and Browning 2022; Salinger and Whisman 2021). The inconsistency of the results of studies focusing on potential stressors might be explained by the type of stressors that were measured in these different studies.

All the mentioned studies that focused on potential stressors included stressful work or academic events. However, while the current study and the study performed by Dev et al. (2020) and Wu et al. (2023) focused solely on work‐related stressors, the other studies mixed this with other sort of stressors, including stressors in the interpersonal domain (e.g., ‘separation from people you care about’ or ‘conflicts with friends’) (Ford et al. 2017; Kroshus, Hawrilenko, and Browning 2022; Salinger and Whisman 2021). This might be relevant because interpersonal conflict has been identified as particularly stress‐provoking, especially for young adults (Jackson and Finney 2002), the population that the studies generating non‐significant findings focused on (Ford et al. 2017; Kroshus, Hawrilenko, and Browning 2022; Salinger and Whisman 2021). Furthermore, it has been suggested that particularly stress‐provoking stressors might deplete the cognitive resources to respond with self‐compassion (Kroshus, Hawrilenko, and Browning 2022). Therefore, mixing work‐ or academic related stress with interpersonal conflict, might explain the failure to demonstrate a moderating role of self‐compassion in the studies performed by Ford et al. (2017), Kroshus, Hawrilenko, and Browning (2022) and Salinger and Whisman (2021). For the population these studies focused on, young adults, interpersonal conflict might block a self‐compassionate response because it is so stress‐provoking that the cognitive resources needed to respond with self‐compassion are depleted.

Interestingly, although the buffering effect of self‐compassion remained significant and of similar magnitude in the longitudinal analyses, the direct association between the latent variables work stress and psychopathology symptoms was clearly weaker than in the cross‐sectional baseline analysis. Although we can only speculate on the reasons behind this finding, this may well point to a truly diminishing effect of work stress on depressive and anxious symptomatology over a longer time period. It may, however, also reflect an unknown confounding third variable at baseline or artificially inflated associations as a result of common method variance (Lindell and Whitney 2001), a commonly acknowledged issue in especially cross‐sectional (survey) studies. This finding further highlights the additional value of longitudinal analyses for studying the potential buffering effect of self‐compassion the relation between stress and mental health.

In sum, the results of the current study strengthen the notion that investing in self‐compassion of workers may be an effective way to promote adaptive coping with work‐related stressors and reduce the risk of depressive and anxiety symptoms over time. Enhancing self‐compassion might be especially relevant for military personnel or emergency workers who are specifically trained to operate in emergency situations. In this type of jobs, a certain level of work‐stress is likely to remain and investing in factors that buffer the negative mental health effects of work‐stress, such as self‐compassion, might therefore be particularly relevant. To enhance self‐compassion, different self‐compassion interventions and therapies have been developed (Kirby, Tellegen, and Steindl 2017). In addition to studies demonstrating that these interventions yield significant positive effects on mental health (Ferrari et al. 2019; Egan et al. 2022), a review by Dodson and Heng also emphasises the promise that these interventions hold for organisational implementation (2022). This review for example describes studies that show that offering mindful self‐compassion (MSC) training (K. D. Neff and Germer 2013), mindfulness‐based stress reduction (MBSR; Grossman et al. 2004) or compassion cultivation training (CCT; Jazaieri et al. 2013), within a workplace context increases self‐compassion levels up to 6 months after the programme (Dodson and Heng 2022). There is also increasing support for the potential health benefit and acceptability of self‐compassion interventions for military or veteran population, especially in the case of trauma‐related psychopathology (Steen et al. 2021). The results of the current study suggest that in addition to offering self‐compassion interventions in the case of trauma‐related psychopathology, self‐compassion interventions could be beneficial to offer as a universal wellbeing intervention for military populations. Future research is needed to study how a self‐compassion intervention should ideally look like for a universal military population.

4.1. Strengths and Limitations

Strengths of the current study are its longitudinal design and the focus on potential work‐stressors, because together they enable stronger conclusions on the prospective moderating effect of self‐compassion on the association between work‐related stress and depressive and anxiety symptomatology. Furthermore, a strength is the use of latent moderated structural equation modelling.

A limitation is the fact that while the study was performed among Dutch military personnel, participants consisted mainly of non‐commissioned officers and staff officers (over 90%), and only few corporals and privates (< 10%). While a random selection of military personnel received an invitation for the study, we believe that this uneven distribution can be explained by the different work contexts of personnel in different ranks, with officers generally spending more time in the office which might be a more convenient setting for filling out questionnaires. Consequently however, these results cannot automatically be assumed to generalise to the entire military organisation.

Furthermore, the measure of stress that was used as a measure of potential stressors in the current study focuses on job characteristics that can potentially trigger the subjective experience of stress, like having to work fast or having to keep track of many things at the same time. While this differs from measures that focus on the subjective experience of stress itself, we do feel it is important to note that this measure is not as objective as a measure of exposure to stressors like divorce, financial difficulties and health problems. This is important to note because we acknowledge that the results might be different with a measure of exposure to such stressors.

Finally, while the current results offer further support for a moderating role of self‐compassion in the association between the experience of potential work‐stressors and the depressive and anxiety symptomatology, we did not study possible mechanisms of change through which self‐compassion might bring about this moderating role. We offer different suggestions as to how this might work, but this is limited by the fact that research on mechanisms of change has not been done. Research that assesses different possible mechanisms of change is needed because this will both increase our understanding of the concept of self‐compassion and its association with mental health and illness.

4.2. Conclusions and Future Research

The results of the current study highlight the value of self‐compassion regarding resilience in the face of work conditions that can be experienced as stressful. Self‐compassion was demonstrated to significantly buffer the negative mental health effects of potential work‐stressors, both cross‐sectionally and prospectively after 6 and 12 months. These findings suggest that investing in self‐compassion might be an effective way to improve resilience regarding work‐stress, especially for people with lower levels of self‐compassion.

Ethics Statement

Ethical approval was provided by University of Twente, faculty BMS (request number 191275).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

References

  1. Abdollahi, A. , Taheri A., and Allen K. A.. 2021. “Perceived Stress, Self‐Compassion and Job Burnout in Nurses: The Moderating Role of Self‐Compassion.” Journal of Research in Nursing 26, no. 3: 182–191. 10.1177/1744987120970612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aldao, A. , and Nolen‐Hoeksema S.. 2012. “When Are Adaptive Strategies Most Predictive of Psychopathology?” Journal of Abnormal Psychology 121, no. 1: 276–281. 10.1037/a0023598. [DOI] [PubMed] [Google Scholar]
  3. Aldao, A. , Nolen‐Hoeksema S., and Schweizer S.. 2010. “Emotion‐Regulation Strategies Across Psychopathology: A Meta‐Analytic Review!” Clinical Psychology Review 30, no. 2: 217–237. 10.1016/j.cpr.2009.11.004. [DOI] [PubMed] [Google Scholar]
  4. Asparouhov, T. , and Muthén B.. 2019. Latent Variable Interactions Using Maximum‐Likelihood and Bayesian Estimation for Single‐and Two‐Level Models Mplus Web Notes No. 23. http://www.statmodel.com/examples/webnotes/webnote%2023.pdf.
  5. Babenko, O. , and Guo Q.. 2019. “Measuring Self‐Compassion in Medical Students: Factorial Validation of the Self‐Compassion Scale–Short Form (SCS‐SF).” Academic Psychiatry 43, no. 6: 590–594. 10.1007/s40596-019-01095-x. [DOI] [PubMed] [Google Scholar]
  6. Beshai, S. , Salimuddin S., Refaie N., and Maierhoffer J.. 2022. “Dispositional Mindfulness and Self‐Compassion Buffer the Effects of COVID‐19 Stress on Depression and Anxiety Symptoms.” Mindfulness 13, no. 12: 3028–3042. 10.1007/s12671-022-02008-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bluth, K. , Roberson P. N. E., Gaylord S. A., et al. 2016. “Does Self‐Compassion Protect Adolescents From Stress?” Journal of Child and Family Studies 25, no. 4: 1098–1109. 10.1007/s10826-015-0307-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brenner, R. E. , Heath P. J., Vogel D. L., and Credé M.. 2017. “Two is More Valid Than One: Examining the Factor Structure of the Self‐Compassion Scale (SCS).” Journal of Counseling Psychology 64, no. 6: 696–707. 10.1037/cou0000211. [DOI] [PubMed] [Google Scholar]
  9. Browne, M. W. , and Cudeck R.. 1992. “Alternative Ways of Assessing Model Fit.” Sociological Methods & Research 21, no. 2: 230–258. 10.1177/0049124192021002005. [DOI] [Google Scholar]
  10. Chan, K. K. S. , Yung C. S. W., and Nie G. M.. 2020. “Self‐Compassion Buffers the Negative Psychological Impact of Stigma Stress on Sexual Minorities.” Mindfulness 11, no. 10: 2338–2348. 10.1007/s12671-020-01451-1. [DOI] [Google Scholar]
  11. Chawla, N. , and Ostafin B.. 2007. “Experiential Avoidance as a Functional Dimensional Approach to Psychopathology: An Empirical Review.” Journal of Clinical Psychology 63, no. 9: 871–890. 10.1002/jclp.20400. [DOI] [PubMed] [Google Scholar]
  12. Cheraghian, H. , Faskhodi B. Z., Heidari N., and Sharifi P. Y.. 2016. “Self‐Compassion as a Relationship Moderator Between Academic Burnout and Mental Health in Students.” International Journal of Academic Research in Progressive Education and Development 5, no. 2: 111–121. 10.6007/IJARPED/v5-i2/2131. [DOI] [Google Scholar]
  13. Coroiu, A. , Kwakkenbos L., Moran C., et al. 2018. “Structural Validation of the Self‐Compassion Scale With a German General Population Sample.” PLoS One 13, no. 2: e0190771. 10.1371/journal.pone.0190771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. de Krijger, E. , Willems R., Ten Klooster P., et al. 2022. “Further Validation of a Dutch Translation of the Sussex Oxford Compassion for the Self Scale in Samples of Crisis Line Volunteers, Military Personnel and Nursing Students.” Frontiers in Psychology 13: 895850. 10.3389/fpsyg.2022.895850. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dev, V. , Fernando A. T., and Consedine N. S.. 2020. “Self‐Compassion as a Stress Moderator: A Cross‐Sectional Study of 1700 Doctors, Nurses, and Medical Students.” Mindfulness 11, no. 5: 1170–1181. 10.1007/s12671-020-01325-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dodson, S. J. , and Heng Y. T.. 2022. “Self‐Compassion in Organizations: A Review and Future Research Agenda.” Journal of Organizational Behavior 43, no. 2: 168–196. 10.1002/job.2556. [DOI] [Google Scholar]
  17. Donker, T. , van Straten A., Marks I., and Cuijpers P.. 2011. “Quick and Easy Self‐Rating of Generalized Anxiety Disorder: Validity of the Dutch Web‐Based GAD‐7, GAD‐2 and GAD‐SI.” Psychiatry Research 188, no. 1: 58–64. 10.1016/j.psychres.2011.01.016. [DOI] [PubMed] [Google Scholar]
  18. Egan, S. J. , Rees C. S., Delalande J., et al. 2022. “A Review of Self‐Compassion as an Active Ingredient in the Prevention and Treatment of Anxiety and Depression in Young People.” Administration and Policy in Mental Health and Mental Health Services Research 49, no. 3: 1–19. 10.1007/s10488-021-01170-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Esch, T. , Stefano G. B., Fricchione G. L., and Benson H.. 2002. “The Role of Stress in Neurodegenerative Diseases and Mental Disorders.” Neuroendocrinology Letters 23: 199–208. [PubMed] [Google Scholar]
  20. Ewert, C. , Vater A., and Schröder‐Abé M.. 2021. “Self‐Compassion and Coping: A Meta‐Analysis.” Mindfulness 12, no. 5: 1063–1077. 10.1007/s12671-020-01563-8. [DOI] [Google Scholar]
  21. Ferrari, M. , Hunt C., Harrysunker A., Abbott M. J., Beath A. P., and Einstein D. A.. 2019. “Self‐Compassion Interventions and Psychosocial Outcomes: A Meta‐Analysis of RCTs.” Mindfulness 10, no. 8: 1455–1473. 10.1007/s12671-019-01134-6. [DOI] [Google Scholar]
  22. Finlay‐Jones, A. L. 2017. “The Relevance of Self‐Compassion as an Intervention Target in Mood and Anxiety Disorders: A Narrative Review Based on an Emotion Regulation Framework.” Clinical Psychologist 21, no. 2: 90–103. 10.1111/cp.12131. [DOI] [Google Scholar]
  23. Ford, J. , Klibert J. J., Tarantino N., and Lamis D. A.. 2017. “Savouring and Self‐Compassion as Protective Factors for Depression.” Stress and Health: Journal of the International Society for the Investigation of Stress 33, no. 2: 119–128. 10.1002/smi.2687. [DOI] [PubMed] [Google Scholar]
  24. Green, G. , Emslie C., O’Neill D., Hunt K., and Walker S.. 2010. “Exploring the Ambiguities of Masculinity in Accounts of Emotional Distress in the Military Among Young Ex‐Servicemen.” Social Science & Medicine 71, no. 8: 1480–1488. 10.1016/j.socscimed.2010.07.015. [DOI] [PubMed] [Google Scholar]
  25. Grossman, P. , Niemann L., Schmidt S., and Walach H.. 2004. “Mindfulness‐Based Stress Reduction and Health Benefits: A Meta‐Analysis.” Journal of Psychosomatic Research 57, no. 1: 35–43. 10.1016/S0022-3999(03)00573-7. [DOI] [PubMed] [Google Scholar]
  26. Hsieh, C. C. , Yu C. J., Chen H. J., Chen Y. W., Chang N. T., and Hsiao F. H.. 2019. “Dispositional Mindfulness, Self‐Compassion, and Compassion From Others as Moderators Between Stress and Depression in Caregivers of Patients With Lung Cancer.” Psycho‐Oncology 28, no. 7: 1498–1505. 10.1002/pon.5106. [DOI] [PubMed] [Google Scholar]
  27. Hu, L. T. , and Bentler P. M.. 1999. “Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives.” Structural Equation Modeling 6, no. 1: 1–55. 10.1080/10705519909540118. [DOI] [Google Scholar]
  28. Inwood, E. , and Ferrari M.. 2018. “Mechanisms of Change in the Relationship Between Self‐Compassion, Emotion Regulation, and Mental Health: A Systematic Review.” Applied Psychology: Health and Well‐Being 10, no. 2: 215–235. 10.1111/aphw.12127. [DOI] [PubMed] [Google Scholar]
  29. Jackson, P. B. , and Finney M.. 2002. “Negative Life Events and Psychological Distress Among Young Adults.” Social Psychology Quarterly 65, no. 2: 186–201. 10.2307/3090100. [DOI] [Google Scholar]
  30. Jazaieri, H. , Jinpa G. T., McGonigal K., et al. 2013. “Enhancing Compassion: A Randomized Controlled Trial of a Compassion Cultivation Training Program.” Journal of Happiness Studies 14, no. 4: 1113–1126. 10.1007/s10902-012-9373-z. [DOI] [Google Scholar]
  31. Karyotaki, E. , Cuijpers P., Albor Y., et al. 2020. “Sources of Stress and Their Associations With Mental Disorders Among College Students: Results of the World Health Organization World Mental Health Surveys International College Student Initiative.” Frontiers in Psychology 11: 1759. 10.3389/fpsyg.2020.01759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Keng, S. , and Hwang E.. 2022. “Self‐Compassion as a Moderator of the Association Between COVID‐19 Stressors and Psychological Symptoms: A Longitudinal Study.” Behaviour Change 39, no. 4: 263–274. 10.10117/bec.2022.2. [DOI] [Google Scholar]
  33. Kirby, J. N. , Tellegen C. L., and Steindl S. R.. 2017. “A Meta‐Analysis of Compassion‐Based Interventions: Current State of Knowledge and Future Directions.” Behavior Therapy 48, no. 6: 778–792. 10.1016/j.beth.2017.06.003. [DOI] [PubMed] [Google Scholar]
  34. Klein, A. , and Moosbrugger H.. 2000. “Maximum Likelihood Estimation of Latent Interaction Effects With the LMS Method.” Psychometrika 65, no. 4: 457–474. 10.1007/BF02296338. [DOI] [Google Scholar]
  35. Kroshus, E. , Hawrilenko M., and Browning A.. 2022. “Stress, Self‐Compassion, and Well‐Being During the Transition to College.” Social Science & Medicine 269: 113514. 10.1016/j.socscimed.2020.113514. [DOI] [PubMed] [Google Scholar]
  36. Lathren, C. , Bluth K., and Park J.. 2019. “Adolescent Self‐Compassion Moderates the Relationship Between Perceived Stress and Internalizing Symptoms.” Personality and Individual Differences 143: 36–41. 10.1016/j.paid.2019.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lazarus, R. S. , DeLongis A., Folkman S., and Gruen R.. 1985. “Stress and Adaptational Outcomes: The Problem of Confounded Measures.” American Psychologist 40, no. 7: 770–779. 10.1037/0003-066X.40.7.770. [DOI] [PubMed] [Google Scholar]
  38. Lee, K. J. , and Lee S. M.. 2022. “The Role of Self‐Compassion in the Academic Stress Model.” Current Psychology 41, no. 5: 3195–3204. 10.1007/s12144-020-00843-9. [DOI] [Google Scholar]
  39. Lindell, M. K. , and Whitney D. J.. 2001. “Accounting for Common Method Variance in Cross‐Sectional Research Designs.” Journal of Applied Psychology 86, no. 1: 114–121. 10.1037//0021-9010.86.1.114. [DOI] [PubMed] [Google Scholar]
  40. Lodder, P. , Denollet J., Emons W. H. M., et al. 2019. “Modeling Interactions Between Latent Variables in Research on Type D Personality: A Monte Carlo Simulation and Clinical Study of Depression and Anxiety.” Multivariate Behavioral Research 54, no. 5: 637–665. 10.1080/00273171.2018.1562863. [DOI] [PubMed] [Google Scholar]
  41. López, A. , Sanderman R., Smink A., et al. 2015. “A Reconsideration of the Self‐Compassion Scale’s Total Score: Self‐Compassion Versus Self‐Criticism.” PLoS One 10, no. 7: e0132940. 10.1371/journal.pone.0132940. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Lorber, W. , and Garcia H. A.. 2010. “Not Supposed to Feel This: Traditional Masculinity in Psychotherapy With Male Veterans Returning From Afghanistan and Iraq.” Psychotherapy: Theory, Research, Practice, Training 47, no. 3: 296–305. 10.1037/a0021161. [DOI] [PubMed] [Google Scholar]
  43. Lou, X. , Wang H., and Minkov M.. 2022. “The Correlation Between Self‐Compassion and Depression Revisited: A Three‐Level Meta‐Analysis.” Mindfulness 13, no. 9: 2128–2139. 10.1007/s12671-022-01958-9. [DOI] [Google Scholar]
  44. MacBeth, A. , and Gumley A.. 2012. “Exploring Compassion: A Meta‐Analysis of the Association Between Self‐Compassion and Psychopathology.” Clinical Psychology Review 32, no. 6: 545–552. 10.1016/j.cpr.2012.06.003. [DOI] [PubMed] [Google Scholar]
  45. Marsh, I. C. , Chan S. W. Y., and MacBeth A.. 2018. “Self‐Compassion and Psychological Distress in Adolescents‐A Meta‐Analysis.” Mindfulness 9, no. 4: 1011–1027. 10.1007/s12671-017-0850-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Maslowsky, J. , Jager J., and Hemken D.. 2015. “Estimating and Interpreting Latent Variable Interactions: A Tutorial for Applying the Latent Moderated Structural Equations Method.” International Journal of Behavioral Development 39, no. 1: 87–96. 10.1177/0165025414552301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. McAllister, L. , Callaghan J. E. M., and Fellin L. C.. 2019. “Masculinities and Emotional Expression in UK Servicemen: ‘Big Boys Don’t Cry’?” Journal of Gender Studies 28, no. 3: 257–270. 10.1080/09589236.2018.1429898. [DOI] [Google Scholar]
  48. McCade, D. , Frewen A., and Fassnacht D. B.. 2021. “Burnout and Depression in Australian Psychologists: The Moderating Role of Self‐Compassion.” Australian Psychologist 56, no. 2: 111–122. 10.1080/00050067.2021.1890979. [DOI] [Google Scholar]
  49. Melchior, M. , Caspi A., Milne B. J., Danese A., Poulton R., and Mofitt T. E.. 2007. “Work Stress Precipitates Depression and Anxiety in Young, Working Women and Men.” Psychological Medicine 37, no. 8: 1119–1129. 10.1017/S0033291707000414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Muris, P. , Otgaar H., and Petrocchi N.. 2016. “Protection as the Mirror Image of Psychopathology: Further Critical Notes on the Self‐Compassion Scale.” Mindfulness 7, no. 3: 787–790. 10.1007/s12671-016-0509-9. [DOI] [Google Scholar]
  51. Muris, P. , and Petrocchi N.. 2017. “Protection or Vulnerability? A Meta‐Analysis of the Relations Between the Positive and Negative Components of Self‐Compassion and Psychopathology.” Clinical Psychology & Psychotherapy 24, no. 2: 373–383. 10.1002/cpp.2005. [DOI] [PubMed] [Google Scholar]
  52. Muris, P. , van den Broek M., Otgaar H., Oudenhoven I., and Lennartz J.. 2018. “Good and Bad Sides of Self‐Compassion: A Face Validity Check of the Self‐Compassion Scale and an Investigation of Its Relations to Coping and Emotional Symptoms in Non‐Clinical Adolescents.” Journal of Child and Family Studies 27, no. 8: 2411–2421. 10.1007/s10826-018-1099-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Muthén, L. K. , and Muthén B. O.. 2017. Mplus User’s Guide. 8th ed. Los Angeles, CA: Muthén & Muthén. [Google Scholar]
  54. Neff, K. 2003. “Self‐Compassion: An Alternative Conceptualization of a Healthy Attitude Toward Oneself.” Self and Identity 2, no. 2: 85–101. 10.1080/15298860309032. [DOI] [Google Scholar]
  55. Neff, K. D. , and Germer C. K.. 2013. “A Pilot Study and Randomized Controlled Trial of the Mindful Self‐Compassion Program.” Journal of Clinical Psychology 69, no. 1: 28–44. 10.1002/jclp.21923. [DOI] [PubMed] [Google Scholar]
  56. Neilson, E. C. , Singh R. S., Harper K. L., and Teng E. J.. 2020. “Traditional Masculinity Ideology, Posttraumatic Stress Disorder (PTSD) Symptom Severity, and Treatment in Service Members and Veterans: A Systematic Review.” Psychology of Men & Masculinities 21, no. 4: 578–592. 10.1037/men0000257. [DOI] [Google Scholar]
  57. Pflanz, S. , and Sonnek S.. 2002. “Work Stress in the Military: Prevalence, Causes, and Relationship to Emotional Health.” Military Medicine 167, no. 11: 877–882. 10.1093/milmed/167.11.877. [DOI] [PubMed] [Google Scholar]
  58. Raes, F. , Pommier E., Neff K. D., and Van Gucht D.. 2011. “Construction and Factorial Validation of a Short Form of the Self‐Compassion Scale.” Clinical Psychology & Psychotherapy 18, no. 3: 250–255. 10.1002/cpp.702. [DOI] [PubMed] [Google Scholar]
  59. Salinger, J. M. , and Whisman M. A.. 2021. “Does Self‐Compassion Moderate the Cross‐Sectional Association Between Life Stress and Depressive Symptoms?” Mindfulness 12, no. 4: 889–898. 10.1007/s12671-020-01554-9. [DOI] [Google Scholar]
  60. Schoemann, A. M. , and Jorgensen T. D.. 2021. “Testing and Interpreting Latent Variable Interactions Using the semTools Package.” Psych 3, no. 3: 3. 10.3390/psych3030024. [DOI] [Google Scholar]
  61. Spitzer, R. L. , Kroenke K., Williams J. B., and Löwe B.. 2006. “A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD‐7.” Archives of Internal Medicine 166, no. 10: 1092–1097. 10.1001/archinte.166.10.1092. [DOI] [PubMed] [Google Scholar]
  62. Steen, M. P. , Di Lemma L., Finnegan A., Wepa D., and McGhee S.. 2021. “Self‐Compassion and Veteran’s Health: A Scoping Review.” Journal of Veterans Studies 7, no. 1: 86–130. 10.21061/jvs.v7i1.219. [DOI] [Google Scholar]
  63. Stevelink, S. A. M. , Jones M., Hull L., et al. 2018. “Mental Health Outcomes at the End of the British Involvement in the Iraq and Afghanistan Conflicts: A Cohort Study.” British Journal of Psychiatry: Journal of Mental Science 213, no. 6: 690–697. 10.1192/bjp.2018.175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Strauss, C. , Taylor B. L., Gu J., et al. 2016. “What is Compassion and How Can We Measure It? A Review of Definitions and Measures.” Clinical Psychology Review 47: 15–27. 10.1016/j.cpr.2016.05.004. [DOI] [PubMed] [Google Scholar]
  65. Stutts, L. A. , Leary M. R., Zeveney A. S., and Hufnagle A. S.. 2018. “A Longitudinal Analysis of the Relationship Between Self‐Compassion and the Psychological Effects of Perceived Stress.” Self and Identity 17, no. 6: 609–626. 10.1080/15298868.2017.1422537. [DOI] [Google Scholar]
  66. Thompson, B. L. , and Waltz J.. 2008. “Self‐Compassion and PTSD Symptom Severity.” Journal of Traumatic Stress 21, no. 6: 556–558. 10.1002/jts.20374. [DOI] [PubMed] [Google Scholar]
  67. Trompetter, H. R. , de Kleine E., and Bohlmeijer E. T.. 2017. “Why Does Positive Mental Health Buffer Against Psychopathology? An Exploratory Study on Self‐Compassion as a Resilience Mechanism and Adaptive Emotion Regulation Strategy.” Cognitive Therapy and Research 41, no. 3: 459–468. 10.1007/s10608-016-9774-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Van Veldhoven, M. , and Meijman T.. 1994. Het Meten van Psychosociale Arbeidsbelasting Met een Vragenlijst: De Vrogenliist Beleving en Beoordeling von de Arbeid (VBBA). Amsterdam, The Netherlands: NIA, Nederlands Instituur voor Arbeidsomstandigheden. [Google Scholar]
  69. Virtanen, M. , Honkonen T., Kivimäki M., et al. 2007. “Work Stress, Mental Health and Antidepressant Medication Findings From the Health 2000 Study.” Journal of Affective Disorders 98, no. 3: 189–197. 10.1016/j.jad.2006.05.034. [DOI] [PubMed] [Google Scholar]
  70. Wu, Q. , Cao H., and Du H.. 2023. “Work Stress, Work‐Related Rumination, and Depressive Symptoms in University Teachers: Buffering Effect of Self‐Compassion.” Psychology Research and Behavior Management 16: 1557–1569. 10.2147/PRBM.S403744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Zessin, U. , Dickhäuser O., and Garbade S.. 2015. “The Relationship Between Self‐Compassion and Well‐Being: A Meta‐Analysis.” Applied Psychology. Health and Well‐Being 7, no. 3: 340–364. 10.1111/aphw.12051. [DOI] [PubMed] [Google Scholar]
  72. Zuithoff, N. P. , Vergouwe Y., King M., et al. 2010. “The Patient Health Questionnaire‐9 for Detection of Major Depressive Disorder in Primary Care: Consequences of Current Thresholds in a Crosssectional Study.” BMC Family Practice 11, no. 1: 98. 10.1186/1471-2296-11-98. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Stress and Health are provided here courtesy of Wiley

RESOURCES