Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jul 30;41(4):e70095. doi: 10.1002/smi.70095

Workload, Work‐Life Conflict, and Stress Amongst Mental Health Professionals: The Moderating Role of Segmentation Preference

Ilona M McNeill 1,2,✉, Eloisa Cullington 2
PMCID: PMC12309875  PMID: 40737200

ABSTRACT

The nature of the mental health profession inherently puts its workers at risk of heightened psychological stress. This raises the importance of understanding why some mental health professionals show greater resilience when faced with common work stressors than others. One work stressor that has been consistently linked with stress is workload. Research has found that higher workload generally leads to greater work‐life conflict, which, in turn, leads to greater stress. The current study aimed to test this mediation model amongst mental health professionals and examined how individuals' preference for segmentation versus integration of work and private life moderates the strength of the pathways in this mediation model. Research participants, consisting of 152 Australian mental health professionals aged 18–69 years (M = 37.58, SD = 12.12), voluntarily completed a 30‐min survey capturing workload, work‐life conflict, segmentation preference, and stress. In line with expectations, higher workload predicted greater stress via higher levels of work‐life conflict. Furthermore, segmentation preference moderated the path between workload and work‐life conflict as well as the path between work‐life conflict and stress. Simple slopes analyses showed that a stronger (vs. weaker) preference for segmentation was associated with a stronger positive relationship between workload and work‐life conflict as well as a stronger positive relationship between work‐life conflict and stress. Segmentation preference may thus influence the workload—work‐life conflict—stress relationship through two distinct mechanisms. Improving our understanding of such mechanisms facilitates creation of targeted strategies to reduce work‐induced stress amongst mental health professionals.

Keywords: boundary theory, person environment fit, time pressure, work‐family border theory, work‐family conflict (WFC), workload

1. Introduction

The mental health profession, which includes a range of different occupations such as clinical psychologists, counsellors, social workers, and mental health nurses, inherently carries the risk of developing stress related illness, such as compassion fatigue and burnout, due to the emotionally taxing nature of the work (e.g., Bride 2004; Ray et al. 2013). This risk is further increased by the presence of other stress inducing job demands (Singh et al. 2020). One particular job demand that is generally amongst the most highly cited work‐based stressors is workload (e.g., Charyszyn and Tucker 2001; Paoli and Merllie 2001). Unfortunately, the recent COVID‐19 pandemic has exacerbated workload and associated psychological stress for mental health professionals, due to increased demands for mental health support (Billings et al. 2021; Byrne et al. 2021; Tandon 2020) combined with persistent understaffing (Northwood et al. 2021). Given that the experience of stress from high workload affects the health and wellbeing of mental health professionals as well as their ability to provide high‐quality satisfactory care to their clients (Austin et al. 2009; Phelps et al. 2009; White 2006; also see O'Connor et al. 2018), it is imperative to create a sophisticated understanding of the relationship between workload and stress amongst mental health professionals, including why some are more negatively affected by high workload than others.

Workload has been found to positively predict stress and related constructs such as burnout in a wide variety of professional settings beyond mental health professionals, ranging from prison wardens (Schiff and Leip 2018), and construction workers (Gómez‐Salgado et al. 2023) to veterinarians (Pohl et al. 2022) and early childhood educators (Brophy‐Herb et al. 2022). Furthermore, past research suggests that the relationship between workload on the one hand and stress and related constructs such as burnout and psychological wellbeing on the other is mediated by work‐life conflict (Bowling et al. 2015; Che et al. 2017; Geurts et al. 2003; Mansour and Tremblay 2016). Still, to date, no study has tested this particular mediation model amongst mental health professionals. The first aim of this study was therefore to examine the mediating role of work‐life conflict in the relationship between workload and stress within this professional domain.

Our second study aim was to improve our understanding of why the positive relationship between workload and stress might not always be equally strong. Understanding what situational and person‐based factors influence the strength of this relationship facilitates the development of targeted strategies and interventions aimed at reducing the impact of workload on stress. One psychological construct that has been conceptually or empirically related to all three constructs of workload, work‐life conflict, and stress, but has yet to be examined as a potential moderator of their relationships is segmentation preference. Segmentation preference is defined as the relative preference a person has for keeping the mental and physical activities associated with work separated in time or space from the mental and physical activities associated with their private life, as opposed to merging and blending these domains (Kreiner 2006; Yang et al. 2019; also see Ashforth et al. 2000). In the current paper, we build on Boundary Theory (Ashforth et al. 2000; Nippert‐Eng 1996) and the related Work‐Family Border Theory (Clark 2000) as well as on Person‐Environment Fit Theory (e.g., Kristof 1996; Kristof‐Brown et al. 2005) and the related concept of misfit (Judge 2007; Wheeler et al. 2007) and argue that segmentation preference may serve as a moderator, both of the relationship between workload and work‐life conflict, and of the relationship between work‐life conflict and stress. Specifically, we hypothesise that for people with stronger preference for segmentation, there is a stronger positive relationship between workload and work‐life conflict as well as a stronger positive relationship between work‐life conflict and stress. In the following, we first briefly review the literature around the role of work‐life conflict in the relationship between workload and stress. Next, we introduce Boundary Theory, Work‐Family Border Theory and Person‐Environment Fit Theory to show why segmentation preference is expected to moderate both paths in the mediation model from workload through work‐life conflict to stress.

1.1. Work‐Life Conflict

Research on the conflict between work and other life domains and its impact on stress and related constructs is not new (e.g., Parasuraman and Greenhaus 2002). Work‐life conflict indicates a loss of limited resources in one domain (e.g., family life, social life) due to interference from another domain (i.e., work). The detection of a lack of sufficient resources in either domain will cause distress. In addition, this lack of sufficient resources often results in people spending additional resources to try and manage the conflicting demands, thereby further draining these resources and inducing more distress (Liao et al. 2019). In some cases, coping responses may even take the form of health impairing habits to try and resolve the conflicting demands, such as reducing sleep, exercise, and relaxation (Moen et al. 2013). In line with this, meta‐analyses have consistently found a relationship between work‐life or work‐family conflict and stress (e.g., Allen et al. 2000; Amstad et al. 2011; Moran 2023), as well as between such conflict and stress‐related outcomes, such as depression, anxiety, and life satisfaction (Amstad et al. 2011).

Although the limited resources underlying work‐life conflict can take several forms (e.g., time, money, energy), work‐life conflict is often operationalised in terms of competing demands on time, which will also be our focus within the current study. When it comes to precursors to such work‐life conflict, research has found that workload positively predicts this conflict (Ilies et al. 2015; Michel et al. 2011). This relationship is unsurprising; as workload increases, it becomes increasingly likely that the time reserved for work will be insufficient to complete the work‐related tasks, resulting in these tasks presenting themselves during non‐work time, thereby taking this limited resource of time away from other life domains.

Taken together, the relationship between workload and stress appears to at least partially be explained by work‐life conflict. Indeed, researchers have found evidence for this mediation, with higher workload predicting greater conflict, and greater conflict predicting greater stress (Mansour and Tremblay 2016), emotional exhaustion (Bowling et al. 2015), burnout (Che et al. 2017), and psychological wellbeing (Geurts et al. 2003). However, the mediating role of work‐life conflict in the workload—stress relationship had not yet been tested amongst mental health professionals, so we set out to test this model amongst this specific working population with the following hypotheses:

H 1

Workload is positively related to stress experienced by mental health professionals.

H 2

The positive relationship between workload and stress experienced by mental health professionals is mediated by work‐life conflict, such that.

H 2a

Workload positively predicts work‐life conflict, and.

H 2b

Work‐life conflict positively predicts stress when controlling for workload.

1.2. Segmentation Preference

One individual difference factor that has been related to work stressors and psychological stress in previous research (Kreiner 2006) is the extent to which a person prefers to keep their work and private life segregated, which is referred to as segmentation preference. Segmentation preference is a construct that stems from two overlapping theories, namely Boundary Theory (Ashforth et al. 2000; Nippert‐Eng 1996) and Work‐Family Border Theory (Clark 2000). At the crux of these theories lies a set of propositions that explain why some people prefer to keep their work and private life separated with non‐permeable and non‐flexible boundaries, whereas others prefer greater integration with higher levels of permeability and flexibility of boundaries between domains. Permeability of boundaries between work and other life domains, here, refers to the extent to which demands from other domains can present themselves in the current domain (e.g., receiving a phone call from your boss outside of working hours). Flexibility, on the other hand, refers to the extent to which it is possible to stop attending to tasks in the current domain to attend to the demands from another domain as they arise (e.g., being able to answer the phone call and, upon hearing about a work issue that needs to be dealt with, being able to stop other activities to address the issue). Segmentation preference builds on these constructs in that people who prefer stronger segmentation between work and other life domains prefer boundaries with lower permeability and flexibility, whereas people who prefer weaker segmentation (greater integration) prefer higher permeability and flexibility (Ashforth et al. 2000; Clark 2000).

1.2.1. Moderating the Relationship Between Workload and Work‐Life Conflict

Both Boundary Theory and Work‐Family Border Theory propose that stronger boundaries between domains are preferred by those for whom boundary crossing will lead to greater conflict between domains, thereby serving as a safeguarding mechanism (Ashforth et al. 2000; Clark 2000). Given that boundary crossing from work demands into other life domains becomes more likely as workload increases, we expected that people who hold a stronger preference for segmentation would experience higher levels of work‐life conflict resulting from increased workload. By testing this proposition, our study extends prior research on the relationship between workload and work‐life conflict and, to the best of our knowledge, is the first to explore the moderating role of segmentation preference.

H 3a

The positive relationship between workload and work‐life conflict is moderated by segmentation preference, such that this relationship is stronger for those with a greater (as opposed to weaker) preference for segmenting work and private life.

1.2.2. Moderating the Relationship Between Work‐Life Conflict and Stress

According to Person‐Environment Fit Theory (e.g., Kristof 1996; Kristof‐Brown et al. 2005) and related work on misfit (e.g., Judge 2007; Wheeler et al. 2007), psychological stress levels are generally higher in situation where there is a greater lack of fit between a person's needs, values, and preferences on the one hand and an organisation's supply of conditions and resources that meet them on the other. To examine whether this applies to one's relative preference for segmentation between work and other life domains, Kreiner (2006) conducted a study that mapped out the extent to which a fit between desired and offered segmentation between work and private life influenced stress. Results showed that stress levels were indeed lower for those whose segmentation preferences were a closer fit to the degree of segmentation offered by the organisation. Seeing that work‐life conflict entails having demands from one domain pop up in and take resources away from another domain, it is a sign of higher levels of permeability and flexibility between domains. Perceptions of work‐life conflict could thus be interpreted by those with high segmentation preference as indicators that their preference for segmentation is not being met, with this perceived misfit resulting in greater levels of stress. In other words, although work‐life conflict is expected to positively predict stress for all, we expected this relationship to be stronger for those with a greater preference for segmentation.

H 3b

The positive relationship between work‐life conflict and stress, whilst controlling for workload, is moderated by segmentation preference, such that this relationship is stronger for those with a greater (as opposed to weaker) preference for segmenting work and private life.

H 3c

Finally, given that we expected segmentation preference to both strengthen the relationship between workload and work‐life conflict (H3a) and the relationship between work‐life conflict and stress (H3b), we hypothesised that the indirect effect of workload on stress via work‐life conflict would be stronger for those with a greater (as opposed to weaker) preference for segmenting work and private life (H3c).

1.3. Current Study

We set out to test our hypotheses using a cross‐sectional survey design amongst our target population of mental health professionals with data collected at a single timepoint. The full model that was examined in the current study is depicted in Figure 1.

FIGURE 1.

FIGURE 1

Open in a new tab

Theoretical model. Solid path c represents the total effect of workload on stress, thereby testing H1. Dashed path c’ represents the effect of workload on stress after controlling for the indirect effect via work‐life conflict and inclusion of segmentation preference in the model.

2. Method

2.1. Participants and Procedure

The present study formed part of a broader research project completed in partial fulfilment of the requirements for the second author's fourth‐year undergraduate course in psychology. The project was conducted in accordance with the Declaration of Helsinki as revised in 2013, with ethics approval obtained from the relevant Human Research Ethics Committee. Prior to starting data collection, G*Power 3.1 (Faul et al. 2009), was used to determine the targeted sample size. An a priori power analysis indicated that with alpha set at 0.05 (one‐sided), power of 0.80, and five predictors (X, M, Moderator and two interactions), a sample size of 149 would have enough power to detect moderator effect sizes comparable to those found by Kreiner (2006; f 2 = 0.042). This sample size also surpassed the minimum sample size required to detect a significant indirect effect with a strong effect for the a‐path, and a small to moderate effect for the b‐path through percentile bootstrapping (Fritz and Mackinnon 2007) as could be expected for these pathways based on prior findings.

The current sample consisted of 152 mental health professionals, who were practising in Australia. To minimise risk associated with participation, anyone who reported currently living with a stress related mental health condition was deemed ineligible for participation. Participants were recruited online between March of 2022 and January of 2023. The study was promoted through (1) advertisements on relevant social media pages (e.g., LinkedIn groups and Facebook groups for Australian mental health professionals), (2) contacting mental health organisations and asking them to circulate an advertisement to the study amongst their employees or on their social media pages, and (3) advertisement on a student research participation website, where students who were also mental health professionals could complete the survey in return for course credit.

To check eligibility, participants first completed a screener survey to ensure they were mental health professionals currently practising in Australia, and that they were not currently being treated for a stress‐related mental health condition. Out of 507 screener surveys that were completed, 260 were deemed ineligible for participation. Reasons for ineligibility were (in order of questions asked): not being an Australian resident (n = 28), being under 18 years of age (n = 2), not meeting the criterium of being a mental health professional (n = 132), being a mental health professional, but not currently practicing (n = 32), and currently living with a stress related mental health condition (n = 66). Of the remaining 247 entries, 70 entries did not successfully complete the consent statement at the start of the survey, a further 24 did not complete the final consent confirmation statement at the end of the survey, and one participant did complete this item, but did not consent to their data being used at the end of the survey. This resulted in 152 participants providing data for analysis, with 56 participants recruited through the research participation website and 96 participants recruited through the other means.

The 152 participants (M age = 37.58, SD = 12.12) consisted of 121 participants identifying as female (80%), 26 identifying as male (17%), 3 identifying as non‐binary or other (2%), and 2 preferring not to disclose gender (1%). The most frequently reported professions were clinical or counselling psychologist (n = 40, 26%) and mental health counsellor (n = 25, 16%), with the remainder including professions such as mental health or disability support workers (n = 14), general or provisional psychologist (n = 10), social worker (n = 10), and psychiatric registered nurse (n = 6). The most common employment type was part‐time (n = 58, 38%), followed by fulltime (n = 53, 35%), self‐employed (n = 24, 16%), and casually employed (n = 17, 11%). In terms of average actual hours spent working per week, 37 worked less than 20 h per week (24.3%), 35 worked between 20 and 30 h (23.0%), 57 worked between 31 and 40 h (37.5%), 18 worked between 41 and 50 (11.8%) and 5 worked more than 50 h per week (3.3%). Most participants were either married (n = 63, 41%) or in a domestic partnership (n = 41, 27%), with 39 participants reporting to be single (26%) and 8 being divorced (5%). Nearly half (n = 73, 48%) stated they were the main financial provider in their household. Finally, 26 participants had worked in their profession for less than a year (17%), 72 (47%) had worked in their profession between 1 and 5 years, 19 (13%) between 6 and 9 years, and, finally, 35 (23%) had worked in their profession for 10 years or more.

2.2. Measures

Workload and work‐life conflict were measure using items from the workload and work‐life conflict subscales of the Mental Health Professional Stress Scale (MHPSS, Cushway et al. 1996), which was specifically developed to capture work related stressors amongst our targeted population of mental health professionals. The original subscales consists of six items each. However, research on the MHPSS has revealed inconsistent factor structures across different samples (e.g., E. S. Lee et al. 2021; Mehrotra et al. 2000), with MHPSS factor structures presented in Lee et al. and Mehrotra et al. both deviating from the original measure presented by Cushway et al. This led us to critically evaluate the individual items used to capture workload and work‐life conflict across these past publications and select items in light of our conceptualisation of these constructs.

2.2.1. Workload

Workload was measured using five items from the workload subscale of the Mental Health Professional Stress Scale (MHPSS, Cushway et al. 1996; also see Mehrotra et al. 2000). The sixth item (i.e., ‘Not enough time for recreation’) from the workload subscale was used to capture work‐life conflict instead, as it aligned much more closely with our stated operationalisation of work‐life conflict (i.e., one domain taking up the limited resource of time from another domain) than it did with our conceptualisation of workload. For each item, respondents indicated the extent to which the listed stressor forms a source of work‐related pressure for them on a 4‐point Likert scale (1 = does not apply to me, 4 = does apply to me). Example items include ‘Too much work to do’ and ‘Not enough time to complete all tasks satisfactorily’. Final scores were averaged across items, and higher scores representing greater perceived workload.

2.2.2. Work‐Life Conflict

Although work‐life conflict can represent conflict stemming both from work taking up resources from other domains and other domains taking up resources from work, our focus for the current study was specifically on conflict stemming from work taking up the limited resource of time in other life domains. In line with this, it was measured using two items (‘Not enough time with family’, ‘Inadequate time for friendships/social relationships’) from the home‐work conflict subscale of the MHPSS (Cushway et al. 1996; also see Mehrotra et al. 2000), and one item that originally belonged in the workload subscale (‘Not enough time for recreation’). Similar to workload, for each of these three items, respondents indicated the extent to which the listed stressor forms a source of work‐related pressure for them on a 4‐point Likert scale (1 = does not apply to me, 4 = does apply to me), with final scores averaged across items, and higher scores representing greater perceived work‐life conflict.

2.2.3. Segmentation Preference

Preference for segmentation versus integration of work with other life roles was captured in terms of a preference for setting clear boundaries between work and other domains in terms of time. Although segmentation preference has previously been measured in terms of physical location‐based boundaries (e.g., Kreiner 2006), many people, including mental health professionals, are now combining office‐based work with working from home, making a focus on time‐based boundaries more appropriate. We used the 4‐item Segmentation Preference Scale (SPS; adapted from Kreiner's measure to focus on time instead of physical location by Yang et al. 2019). Items are captured on a 5‐point Likert‐type scale (1 = strongly disagree to 5 = strongly agree). An example item is ‘I prefer to complete my work only during work hours’. Final scores were conceived by averaging across items, with higher scores indicating a stronger preference for segmentation.

2.2.4. Stress

To measure participants' psychological stress, we used the well‐validated Perceived Stress Scale‐10 (PSS‐10; S. Cohen et al. 1983). This 10‐item self‐report measure identifies the extent to which individuals are finding their lives to be unpredictable, uncontrollable, and overloaded using a 5‐point Likert scale (1 = never, 5 = very often). A sample item is ‘In the last month, how often have you felt nervous and stressed?’. Final scores were conceived by averaging across items, with higher scores indicating higher levels of stress.

2.2.5. Demographic Variables

The survey captured the following demographic variables: age, gender, workplace, profession, employment type, average hours worked per week, relationship status, being the main financial provider in the household (yes/no), and years in current profession.

2.3. Data Analysis

All hypotheses were tested using the PROCESS macro (Hayes 2022) for SPSS. Specifically, to test hypotheses 1, 2a, 2b and 2b we used model 4, which allows for testing direct, indirect, and total effects simultaneously. To test hypotheses 3a, 3b, 3c we used model 58, which allows for simultaneous testing of moderation of the pathway of the predictor to the mediator and moderation of the pathway of the mediator to the dependent variable by the same moderator as well as testing the indirect effect of the predictor on the dependent variable via the mediator at different levels of the moderator. All variables were standardized prior to being entered into the analysis. We estimated 95% bias corrected confidence intervals for all standardized coefficients using 5000 iterations of bootstrapping. Hypothesised relationships were all tested at a one‐tailed alpha of 0.05, whereas other relationships were tested at a two‐tailed alpha of 0.05. In line with this, interactions were probed when their regression weight reached p < 0.05 (one‐tailed). Due to our moderator, segmentation preference, being skewed in its distribution, we followed recommendations from Hayes (2022) and probed interactions at the 16th, 50th, and 84th percentile of segmentation preference, but note that probing at standardized scores of ‐ 1, 0, and +1 yielded highly similar results.

3. Results

3.1. Preliminary Analyses

3.1.1. Measurement Model Fit

Given that the items we used to capture workload and work‐life conflict deviated from those specified to capture workload and work‐life conflict in two previous studies (Cushway et al. 1996; Mehrotra et al. 2000), and the fact that a third study found workload and work‐life conflict items to load onto a single factor (e.g., E. S. Lee et al. 2021), we conducted several confirmatory factor analyses in AMOS prior to running our main analyses. Specifically, we examined the fit of the two‐factor measurement model of workload and work‐life conflict based on our items as specified in our method section as well as the fit of the two‐factor measurement models for these constructs based on the items specified by Cushway et al. (1996) versus Mehrotra et al. (2000). Further, we compared the fit of our two‐factor model with the fit of a model where all eight items we used were set to load onto a single factor.

Our two‐factor model showed acceptable fit with the data (Comparative Fit Index = 0.97, Tucker Lewis Index = 0.94, Root Mean Square Error of Approximation = 0.08, χ2(df = 19) = 37.58, p = 0.007), whereas the model using the items as specified by Cushway et al. (1996; Comparative Fit Index = 0.89, Tucker Lewis Index = 0.83, Root Mean Square Error of Approximation = 0.10, χ2(df = 53) = 128.01, p < 0.001) and Mehrotra et al. (2000; Comparative Fit Index = 0.90, Tucker Lewis Index = 0.84, Root Mean Square Error of Approximation = 0.11, χ2(df = 34) = 93.25, p < 0.001) did not show an adequate fit. Further, comparing our two‐factor model to a one‐factor model using the chi‐square difference test showed that our 2‐factor model fit the data significantly better than the unidimensional model [Comparative Fit Index = 0.91, Tucker Lewis Index = 0.84, Root Mean Square Error of Approximation = 0.13, χ2(df = 20) = 68.38, p < 0.001, Δχ2(Δdf = 1) = 30.80, p < 0.001].

Taken together, this provides support for our selection of items to capture workload and work‐life conflict. As such, the analyses reported below all used our specified items to capture workload and work‐life conflict. Still, to examine the robustness of our findings, we checked whether the workload and work‐life conflict items as specified by Cushway et al. (1996) and Mehrotra et al. (2000) would have resulted in different findings, and have noted one small difference in the main analyses section, namely the moderation effect between work‐life conflict and segmentation preference on perceived stress not quite reaching significance when using Cushway et al. (1996)'s measure of work‐life conflict.

3.1.2. Descriptive Statistics

Table 1 presents descriptive statistics and intercorrelations between the main study variables, including the only two demographic variables that showed significant zero‐order relationships with psychological stress, namely age and number of years in profession. Upon entering these demographic variables into a linear regression analysis predicting psychological stress, only age remained a significant predictor, β = −0.209, p = 0.039 (number of years in profession: β = −0.044, p = 0.659). Hence, age was used as a covariate in the main PROCESS analyses. 1

TABLE 1.

Descriptive statistics and correlations for study variables.

Variable M SD 1 2 3 4 5 6
1. Age 37.58 12.12 —
2. Years in profession 5.40 3.78 0.56*** —
3. Workload 2.77 0.70 0.06 0.14 (0.80)
4. Work‐life conflict 2.54 0.90 −0.02 −0.05 0.70*** (0.85)
5. Segmentation preference 4.20 0.84 −0.29*** −0.06 0.04 0.13 (0.89)
6. Stress 2.79 0.72 −0.23** −0.17* 0.36*** 0.48*** 0.21* (0.91)
Open in a new tab

Note: N = 140–152. Cronbach's alphas are presented in parentheses.

*

p < 0.05.

**

p < 0.01.

***

p < 0.001.

3.1.3. Assumptions and Outliers

We checked assumptions for linear regression analyses as well as univariate and multivariate outliers. There was one case that was both a univariate (z‐score for segmentation preference = −3.80) and multivariate (Mahalanobis distance = 18.63 with df = 3, p < 0.001) outlier. We therefore ran the main PROCESS analyses both with and without this outlier included to evaluate its impact on results. Given that no impact was witnessed on the pattern of results, we only report results that included this participant below rather than reporting both sets of results. No other assumptions were violated.

3.2. Main Analyses

Table 2 presents the standardized regression coefficients resulting from the PROCESS mediation analysis that was run to test hypotheses 1, 2a, 2b and 2b. In line with Hypothesis 1, the total effect of workload on stress was significant and positive, β = 0.366, 95% CI [0.214, 0.519], p < 0.001 (one‐sided). In line with Hypothesis 2a, workload positively predicted work‐life conflict. Furthermore, in line with Hypothesis 2b, when both workload and work‐life conflict were entered as predictors of stress, work‐life conflict positively predicted stress, but workload did not. In addition, the indirect effect of workload on stress via work‐life conflict was significant, a*b = 0.254, bootstrapped 95% CI [0.124, 0.411]. Taken together this suggests that work‐life conflict fully mediates the relationship between workload and stress (H2).

TABLE 2.

Standardised regression coefficients of workload predicting stress through work‐life conflict whilst controlling for age.

Antecedent Mediator (WLC) Outcome (stress)
β LLCI ULCI sr 2 R 2 β LLCI ULCI sr 2 R 2
Age −0.066 −0.189 0.056 −0.066 0.48*** −0.232** −0.370 −0.083 −0.230 0.26***
Workload 0.692*** 0.569 0.814 0.690 0.112 −0.090 0.314 0.081
Work‐life conflict 0.368*** 0.166 0.570 0.266
Open in a new tab

Abbreviations: LLCI, lower level of 95% confidence interval; ULCI, upper level of 95% confidence interval; WLC, work‐life conflict.

*p < 0.05.

**

p < 0.01.

***

p < 0.001.

Table 3 presents the results of the moderated mediation analysis that was run to test Hypotheses 3a, 3b, 3c, including standardized regression coefficients and 95% confidence intervals. In line with Hypothesis 3a, the effect of workload on work‐life conflict was moderated by segmentation preference. Interaction prompts showed that the relationship between workload and work‐life conflict was significant at all levels of segmentation preference, but that this relationship became stronger at higher levels of segmentation preference, ranging from β = 0.578, bootstrapped 95% CI [0.418, 0.739], p < 0.001, at the 16th percentile of segmentation preference, through β = 0.692, bootstrapped 95% CI [0.570, 0.813], p < 0.001, at the 50th percentile to β = 0.805, bootstrapped 95% CI [0.637, 0.973], p < 0.001, at the 84th percentile (see Figure 2).

TABLE 3.

Standardised regression coefficients associated with full moderated mediation model predicting stress whilst controlling for age.

Antecedent Mediator (WLC) Outcome (stress)
β LLCI ULCI sr 2 R 2 (ΔR 2 ) β LLCI ULCI sr 2 R 2 (ΔR 2 )
Age −0.046 −0.173 0.081 −0.044 0.48*** −0.202** −0.354 −0.050 −0.192 0.29***
Workload 0.683*** 0.562 0.805 0.680 0.105 −0.096 0.305 0.076
SP 0.069 −0.058 0.196 0.066 0.101 −0.052 0.253 0.096
WLC 0.341*** 0.140 0.542 0.245
Workload × SP 0.127* 0.003 0.250 0.124 (0.02)*
WLC × SP 0.132* −0.013 0.277 0.131 (0.02) a
Open in a new tab

Abbreviations: LLCI, lower level of 95% confidence interval; SP, segmentation preference; ULCI, upper level of 95% confidence interval; WLC, work‐life conflict.

a

p < 0.10.

*

p < 0.05.

**

p < 0.01.

***

p < 0.001.

FIGURE 2.

FIGURE 2

Open in a new tab

Interaction effect between workload and segmentation preference on work‐life conflict.

The hypothesised moderation effect of segmentation preference on the relationship between work‐life conflict and stress (H3b) reached significance at p = 0.037 (one‐sided). 2 Probing the interaction through simple slope analyses showed that the relationship between work‐life conflict and stress did not quite reach significance at the 16th percentile of segmentation preference, β = 0.231, bootstrapped 95% CI [−0.011, 0.474], p = 0.061, but that this relationship became significant at the 50th percentile, β = 0.350, bootstrapped 95% CI [0.149, 0.550], p < 0.001, and grew more positive at the 84th percentile, β = 0.468, bootstrapped 95% CI [0.232, 0.704], p < 0.001 (see Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

Interaction effect between work‐life conflict and segmentation preference on stress.

To test the final hypothesis, the indirect effect of workload on stress via work‐life conflict was examined at different values of the moderator. This showed that the indirect effect did not reach significance at the 16th percentile of segmentation preference, a*b = 0.134, bootstrapped SE = 0.074, bootstrapped 95% CI [−0.020, 0.271], but it was significant at the 50th percentile, a*b = 0.242, bootstrapped SE = 0.074, bootstrapped 95% CI [0.100, 0.390], as well as at the 84th percentile, a*b = 0.377, bootstrapped SE = 0.107, bootstrapped 95% CI [0.188, 0.611].

4. Discussion

This study aimed to explore the effect of workload on stress, and whether this relationship is mediated by work‐life conflict in a population that is at increased risk of experiencing job related stress, namely mental health professionals. In addition, we sought to observe the moderating role of segmentation preference, both in the relationship between workload and work‐life conflict and in the relationship between work‐life conflict and stress.

First, workload positively predicted stress, and this relationship was fully mediated by work‐life conflict, conceptualised as work demands impacting time available for other life domains, such as family and leisure time. In fact, the effect size associated with the full mediation model was large, with workload and work‐life conflict accounting for more than a quarter of the variance in general stress experienced by the participants in our study. This aligns with results testing a similar model predicting stress amongst hospitality workers (Mansour and Tremblay 2016), as well as related models predicting burnout amongst nurses (Che et al. 2017) and wellbeing amongst medical residents, childcare workers, and bus drivers (Geurts et al. 2003). By showing that workload relates to general stress via increased work‐life conflict amongst mental health professionals, our findings support sentiments expressed by authors of these previous studies in stating that workload is a serious issue that may negatively impact workers outside of the workplace and thus needs to be taken into consideration to protect and promote health and wellbeing.

Second, segmentation preference, which was conceptualised as a preference for keeping clear boundaries between working hours and time spent on other life domains, moderated the relationship between workload and work‐life conflict. Specifically, the positive relationship between workload and work‐life conflict was stronger for those who had a stronger preference for segmentation of work and private life than for those who had a weaker preference for segmentation. This finding aligns with the theoretical propositions made in Boundary Theory and Work‐Family Border Theory (Ashforth et al. 2000; Clark 2000).

Finally, segmentation preference moderated the relationship between work‐life conflict and stress. Simple slopes analyses indicated that the positive relationship between work‐life conflict and stress was stronger for those who had a stronger preference for segmentation of work and private life than for those who had a weaker preference for segmentation. This aligns with our hypothesis based on Person‐Environment Fit Theory (e.g., Kristof 1996; Kristof‐Brown et al. 2005) and related work on misfit (e.g., Judge 2007; Wheeler et al. 2007), as well as with previous findings by Kreiner (2006), which showed that a greater fit between segmentation preference and work environment negatively predicted stress. Specifically, our findings align with our postulation that the experience of conflict indicates a lack of segmentation and therefore should be a greater misfit with those preferring stronger segmentation between work and other life domains, resulting in greater levels of stress associated with the same level of conflict.

4.1. Limitations and Direction for Future Research

The present findings do need to be considered in light of several limitations attached to the current study. Firstly, it needs noting that this was the first study to test the moderating effect of segmentation preference on the relationships between workload and work‐life conflict and between work‐life conflict and perceived stress. It was also the first study to use our specific set of items to capture workload and work‐life conflict, and to test a measurement model based on these items. Although our findings aligned with our hypotheses, and the measurement model showed an adequate fit with the data (whereas the measurement models based on items specified in past studies by Cushway et al. 1996; Mehrotra et al. 2000 did not), replication of results is warranted. In targeting replication of results, it would be beneficial to also consider the use of alternative measures of stress, such as measures that specifically capture affective and physiological stress responses. This would enable deeper insights into whether the relationships found in the current study equally apply to these different elements of the stress experience. Also, whilst it is worth noting that our findings generally appeared robust to changing the items used to capture workload and work‐life conflict to align with those specified in previous studies (i.e., Cushway et al. 1996; Mehrotra et al. 2000), using the items for work‐life conflict specified by Cushway et al. did result in the interaction between work‐life conflict and segmentation preference failing to reach significance, albeit with a minor shift in effect size (Δβ = 0.023). This difference could be due to the Cushway et al. items including a mix of items that capture work‐to‐life conflict (e.g. ‘Not enough time with family’), life‐to‐work conflict (i.e., ‘Relationship with spouse/partner affects work’) and items that do not appear to tap into conflict directly (i.e., ‘Work emphasises feelings of emptiness and/or isolation’). This might also explain the inadequate fit of the measurement model based on the Cushway et al. items.

Secondly, the cross‐sectional and correlational study design was chosen over a study containing multiple timepoints to place an as low as possible burden on our population of interest. Still, this design choice limits our ability to make any strong inferences around directionality of relationships, and it is possible that our found relationships operate in both directions. For one, although our hypotheses stipulate that workload influences work‐life conflict, which in turn influences stress, it is likely that there is a degree of circularity in some of these relationships, such as higher levels of stress possibly causing a greater sensitivity to work to life interference and therefore greater perceived conflict. Furthermore, according to Boundary Theory (Ashforth et al. 2000), there is a degree of circularity in the relationship between segmentation preference and experience of conflict between life domains, where anticipated conflict leads to greater preference for segmentation, and this greater preference for segmentation in turn leads to greater actual conflict experienced when boundary crossing occurs, which could in turn increase the safeguarding motive, thereby further strengthening one's segmentation preference. Although Che et al. (2017) examined the effect of workload on work‐life conflict and burnout using a longitudinal design with two waves, future studies utilising similar designs are needed to examine the direction of the effects observed in the current study by measuring all variables, including segmentation preference, across a minimum of three measurement time points and conducting cross‐lagged mediation and moderation analyses. Use of such longitudinal study designs would also allow for exploring alternative models, such as ones including the circular relationships suggested above.

Second, our segmentation data showed a negative skew with a generally strong preference for segmentation (M = 4.20 out of 5). This may be a reflection of the timing of data collection, which occurred shortly after the completion of two years of Covid‐19 related lockdowns in Australia. Specifically, the relatively strong preference for segmentation may have been a reaction to the high levels of integration between work and other life domains that were the norm in large areas of Australia in 2020 and 2021. Future studies should endeavour to collect more data amongst populations that may have a greater preference towards integration to establish whether the effects found in the current study are consistent across the full spectrum of segmentation preferences.

Generalisability of findings can be improved further in several ways. Firstly, there is no reason why the relationships tested in the current study are limited to mental health professionals, and future research should endeavour to examine the moderating role of segmentation preference in other professions. Secondly, although the distribution of stress scores amongst participants in the current study covered nearly the full range of possible scores (1.10–5 on a 5‐point Likert scale), the study excluded anyone currently living with a stress related mental health condition. Future research is therefore required to examine the extent to which the findings of the current study are generalisable to this population. Thirdly, the current study did not capture racial or ethnic background of participants, even though such demographics have been shown to impact the relationships between work‐related stressors and work‐life conflict (Luhr et al. 2022). We therefore recommend that future research in this area incorporates these variables to explore how cultural differences might moderate the observed relationships between workload, work‐life conflict, and stress.

Several additional questions need addressing through future research. For one, our hypothesis for the moderating role of segmentation preference on the relationship between workload and work‐life conflict was built on the premise that this preference is driven by a safeguarding motive, but this was not tested in the current study. Related to this, the current study focused specifically on workload as a job demand, but if segmentation preference indeed functions as a safeguarding mechanism that limits the impact of work stressors on other life domains, the relationship between other types of job stressors, such as role ambiguity and relational conflict at work, and stress may be similarly moderated by it.

Further, the focus of the current study was strictly on work‐related workload and work‐to‐life conflict. Still, similar patterns of results could be expected for workloads stemming from other life domains (e.g., carers load) and the life‐to‐work or inter‐domain conflict and stress that may result from this. More research is needed to examine whether greater demands or workloads in non‐work domains may lead to greater conflict and stress, but particularly so for individuals with a greater preference for segmentation of domains. Also, whilst the current study used a brief work‐to‐life conflict measure that focused specifically on time‐based conflict, there are other measures of work‐to‐life conflict that capture several additional dimensions, including behaviour, energy, and emotion‐based conflict (DeBaylo and Michel 2022; Greenhaus and Beutell 1985). For future studies, it would be advantageous to consider such measures or even qualitative methods to gain deeper insights into how the different dimensions of work‐to‐life conflict affect mental health professionals and interplay with segmentation preference. Additionally, there may be other individual differences that operate in a similar way to segmentation preference. One example might be Personal Need for Structure (Thompson et al. 2001), where individuals with a greater need for structure may experience greater stress as domains become more integrated.

Finally, research has shown that a person's stress mindset, referring to the extent to which stressors are interpretated as relatively adaptive (e.g., stress‐is‐enhancing) or maladaptive (e.g., that stress‐is‐debilitating), will influence their cognitive, emotional, and physiological responses to the stressor (Crum et al. 2013; Crum et al. 2017). It is currently unknown whether people with a stronger preference for segmentation may differ in their stress mindset from those with a weaker preference for segmentation, and whether this mindset may have influenced some of the relationships found in the current study. Recent research has revealed mixed results regarding the potential role of stress beliefs amongst healthcare professionals. For example, a study by Wekenborg et al. (2024) did not find any evidence for positive or negative stress beliefs moderating the relationship between personal workload and burnout symptoms amongst physicians during the COVID‐19 pandemic. Yet, a study by Laferton et al. (2024) found that negative stress beliefs did moderate the relationship between perceived increases in COVID‐19 related work stress on the one hand and increased levels of depressive, anxiety and distress symptoms amongst physicians and nurses on the other. Specifically, this study revealed a positive relationship between work stress and mental health symptoms for those with high to medium levels of negative stress beliefs, but not for those with low negative stress beliefs. Further research in this area is thus vital.

4.2. Potential Implications

Although our study did not directly examine the safeguarding motive underlying segmentation preference, our results are aligned with the idea that high workload may create more work‐life conflict and greater levels of stress for some than for others. Investigating this further is important to facilitate targeted interventions and strategies to reduce the negative impact of workload on those who are at higher risk of its negative impacts. Identifying at risk populations becomes even more important when considering that stronger boundary safeguarding and the resulting stronger relationships between workload, conflict and stress are likely connected to demographics already at risk of being disadvantaged in the workforce.

One plausible example relates to trends in gender and unpaid labour. Our sample consisted of 80% mental health professionals identifying as female, which aligns with recent data showing that in 2023, females comprised 72% of mental health nurses, 80% of psychologists, 85% of the mental health occupational therapists and 84% of accredited mental health social workers (Australian Institute of Health and Welfare 2023). Given that women still engage in substantially more unpaid labour outside of work than men (e.g., Australian Bureau of Statistics 2022; Baxter 2022; Langer et al. 2015; UN Women 2020), they may prefer greater segmentation between work and these other responsibilities in an attempt to avoid conflict between domains. 3 This means women may be at risk of experiencing both higher levels of conflict at the same level of workload, and higher levels of stress at the same level of work‐life conflict. In addition, the stronger relationship between workload, conflict and stress may, in turn, lead to the adoption of stress‐reducing, yet structural disadvantage increasing strategies, such as occupational downgrading. Indeed, women are more likely than men to take up part‐time roles or engage in roles below their skill levels, to reduce workload, thereby maintaining gender inequalities in the workforce (Hegewisch and Gornick 2011; D. Lee et al. 2020).

In sum, this study contributes to the existing evidence base that shows a need to consider workload as a life stressor that may negatively impact individuals not only within the workplace, but also outside of it. In addition, it highlights that some may be at greater risk of the negative impact of workload on conflict and stress than others, which is an important first step in the development of targeted strategies to reduce this negative impact, particularly amongst higher‐risk populations.

Ethics Statement

The study was approved by the ACAP Human Research Ethic Committee (reference number: EC01819).

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgements

The authors wish to thank the mental health professionals who participated in this study, and the mental health organisations who willingly shared an advertisement to our study with their employees. Open access publishing facilitated by Swinburne University of Technology, as part of the Wiley ‐ Swinburne University of Technology agreement via the Council of Australian University Librarians.

McNeill, Ilona M. , and Cullington Eloisa. 2025. “Workload, Work‐Life Conflict, and Stress Amongst Mental Health Professionals: The Moderating Role of Segmentation Preference.” Stress and Health: e70095. 10.1002/smi.70095.

Endnotes

1

Our age variable had a substantial percentage of its data missing (n = 12, 8%), However, given a significant Little MCAR, χ 2 (143) = 178.85, p = 0.023, we chose not to impute values for age. Instead, we reran the PROCESS analyses with years in profession as a covariate instead of age. This did not reveal any notable changes in model coefficients, leading us to only present the results using age as a covariate.

2

Running the analysis using Cushway et al. ’s (1996) items to capture work‐life conflict resulted in a non‐significant interaction effect, β = 0.109, p = 0.096 (one‐sided).

3

Indeed, females in our sample showed a slightly stronger preference for segmentation (M = 4.24, SD = 0.77) compared to males (M = 4.05, SD = 1.11), although this difference did not reach significance.

Data Availability Statement

The data that support the findings of this study are available upon reasonable request from the corresponding author. The full data are not publicly available due to containing information that could compromise the privacy of research participants.

References

  1. Allen, T. D. , Herst D. E., Bruck C. S., and Sutton M.. 2000. “Consequences Associated With Work‐To‐Family Conflict: A Review and Agenda for Future Research.” Journal of Occupational Health Psychology 5, no. 2: 278–308. 10.1037/1076-8998.5.2.278. [DOI] [PubMed] [Google Scholar]
  2. Amstad, F. T. , Meier L. L., Fasel U., Elfering A., and Semmer N. K.. 2011. “A Meta‐Analysis of Work‐Family Conflict and Various Outcomes With a Special Emphasis on Cross‐Domain Versus Matching‐Domain Relations.” Journal of Occupational Health Psychology 16, no. 2: 151–169. 10.1037/a0022170. [DOI] [PubMed] [Google Scholar]
  3. Ashforth, B. E. , Kreiner G. E., and Fugate M.. 2000. “All in a Day's Work: Boundaries and Micro Role Transitions.” Academy of Management Review 25, no. 3: 472–491. 10.2307/259305. [DOI] [Google Scholar]
  4. Austin, W. , Goble E., Leier B., and Byrne P.. 2009. “Compassion Fatigue: The Experience of Nurses.” Ethics and Social Welfare 3, no. 2: 195–214. 10.1080/17496530902951988. [DOI] [Google Scholar]
  5. Australian Bureau of Statistics . 2022. “How Australians Use Their Time, 2020‐21 Financial Year | Australian Bureau of Statistics.” (July). https://www.abs.gov.au/statistics/people/people‐and‐communities/how‐australians‐use‐their‐time/2020‐21.
  6. Australian Institute of Health and Welfare . 2023. Workforce ‐ Mental Health ‐ AIHW. Australian Institute of Health and Welfare. https://www.aihw.gov.au/mental‐health/topic‐areas/workforce. [Google Scholar]
  7. Baxter, J. 2022. Couples Sharing Childcare: The COVID‐19 Experience. [Conference presentation]. Australian Institute of Family Studies Conference. https://aifs2022.paperlessevents.com.au/share/Baxter‐97. [Google Scholar]
  8. Billings, J. , Biggs C., Ching B. C. F., et al. 2021. “Experiences of Mental Health Professionals Supporting front‐line Health and Social Care Workers During COVID‐19: Qualitative Study.” BJPsych Open 7, no. 2: e70. 10.1192/bjo.2021.29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bowling, N. A. , Alarcon G. M., Bragg C. B., and Hartman M. J.. 2015. “A meta‐analytic Examination of the Potential Correlates and Consequences of Workload.” Work & Stress 29, no. 2: 95–113. 10.1080/02678373.2015.1033037. [DOI] [Google Scholar]
  10. Bride, B. E. 2004. “The Impact of Providing Psychosocial Services to Traumatized Populations.” Stress, Trauma, and Crisis 7, no. 1: 29–46. 10.1080/15434610490281101. [DOI] [Google Scholar]
  11. Brophy‐Herb, H. E. , Brincks A., Cook J. L., et al. 2022. “Stress Intensity and Exhaustion Among Infant and Toddler Teachers: Descriptive Analysis and Associations With Sources of Stress and Coping Strategy Use.” Early Education & Development 34, no. 7: 1–20. 10.1080/10409289.2022.2151399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Byrne, A. , Barber R., and Lim C. H.. 2021. “Impact of the COVID ‐19 Pandemic – A Mental Health Service Perspective.” Progress in Neurology and Psychiatry 25, no. 2: 27. 10.1002/pnp.708. [DOI] [Google Scholar]
  13. Charyszyn, S. , and Tucker P.. 2001. “New Light on Working Hours.” Ergonomist 7, (May). [Google Scholar]
  14. Che, X. X. , Zhou Z. E., Kessler S. R., and Spector P. E.. 2017. “Stressors Beget Stressors: The Effect of Passive Leadership on Employee Health Through Workload and Work–Family Conflict.” Work & Stress 31, no. 4: 338–354. 10.1080/02678373.2017.1317881. [DOI] [Google Scholar]
  15. Clark, S. C. 2000. “Work/Family Border Theory: A New Theory of Work/Family Balance.” Human Relations 53, no. 6: 747–770. 10.1177/0018726700536001. [DOI] [Google Scholar]
  16. Cohen, S. , Kamarck T., and Mermelstein R.. 1983. “A Global Measure of Perceived Stress.” Journal of Health and Social Behavior 24, no. 4: 385–396. 10.2307/2136404. [DOI] [PubMed] [Google Scholar]
  17. Crum, A. J. , Akinola M., Martin A., and Fath S.. 2017. “The Role of Stress Mindset in Shaping Cognitive, Emotional, and Physiological Responses to Challenging and Threatening Stress.” Anxiety, Stress & Coping 30, no. 4: 379–395. 10.1080/10615806.2016.1275585. [DOI] [PubMed] [Google Scholar]
  18. Crum, A. J. , Salovey P., and Achor S.. 2013. “Rethinking Stress: The Role of Mindsets in Determining the Stress Response.” Journal of Personality and Social Psychology 104, no. 4: 716–733. 10.1037/a0031201. [DOI] [PubMed] [Google Scholar]
  19. Cushway, D. , Tyler P. A., and Nolan P.. 1996. “Development of a Stress Scale for Mental Health Professionals.” British Journal of Clinical Psychology 35, no. 2: 279–295. 10.1111/j.2044-8260.1996.tb01182.x. [DOI] [PubMed] [Google Scholar]
  20. DeBaylo, P. , and Michel J. S.. 2022. “Identifying Energy and Emotion‐Based Conflict: Development of a Refined Work‐Life Conflict Scale.” Stress and Health 38, no. 5: 961–977. 10.1002/smi.3148. [DOI] [PubMed] [Google Scholar]
  21. Faul, F. , Erdfelder E., Buchner A., and Lang A.‐G.. 2009. “Statistical Power Analyses Using G*Power 3.1: Tests for Correlation and Regression Analyses.” Behavior Research Methods 41, no. 4: 1149–1160. 10.3758/brm.41.4.1149. [DOI] [PubMed] [Google Scholar]
  22. Fritz, M. S. , and Mackinnon D. P.. 2007. “Required Sample Size to Detect the Mediated Effect.” Psychological Science 18, no. 3: 233–239. 10.1111/j.1467-9280.2007.01882.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Geurts, S. A. E. , Kompier M. A. J., Roxburgh S., and Houtman I. L. D.. 2003. “Does Work–Home Interference Mediate the Relationship Between Workload and Well‐Being?” Journal of Vocational Behavior 63, no. 3: 532–559. 10.1016/s0001-8791(02)00025-8. [DOI] [Google Scholar]
  24. Gómez‐Salgado, C. , Camacho‐Vega J. C., Gómez‐Salgado J., et al. 2023. “Stress, Fear, and Anxiety Among Construction Workers: A Systematic Review.” Frontiers in Public Health 11: 1226914. 10.3389/fpubh.2023.1226914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Greenhaus, J. H. , and Beutell N. J.. 1985. “Sources of Conflict Between Work and Family Roles.” Academy of Management Review 10, no. 1: 76–88. 10.2307/258214. [DOI] [Google Scholar]
  26. Hayes, A. F. 2022. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression‐Based Approach. 3rd ed. Guilford Press. [Google Scholar]
  27. Hegewisch, A. , and Gornick J. C.. 2011. “The Impact of Work‐Family Policies on Women’s Employment: A Review of Research From OECD Countries.” Community, Work & Family 14, no. 2: 119–138. 10.1080/13668803.2011.571395. [DOI] [Google Scholar]
  28. Ilies, R. , Huth M., Ryan A. M., and Dimotakis N.. 2015. “Explaining the Links Between Workload, Distress, and Work–Family Conflict Among School Employees: Physical, Cognitive, and Emotional Fatigue.” Journal of Educational Psychology 107, no. 4: 1136–1149. 10.1037/edu0000029. [DOI] [Google Scholar]
  29. Judge, T. A. 2007. “The Future of Person‐Organization Fit Research: Comments, Observations, and a Few Suggestions.” In Perspectives on Organizational Fit, edited by Ostroff C. and Judge T. A., 419–445. Lawrence Erlbaum Associates Publishers. [Google Scholar]
  30. Kreiner, G. E. 2006. “Consequences of Work‐Home Segmentation or Integration: A Person‐Environment Fit Perspective.” Journal of Organizational Behavior 27, no. 4: 485–507. 10.1002/job.386. [DOI] [Google Scholar]
  31. Kristof, A. L. 1996. “Person‐Organization Fit: An Integrative Review of Its Conceptualizations, Measurement, and Implications.” Personnel Psychology 49: 1–49. 10.1111/j.1744-6570.1996.tb01790.x. [DOI] [Google Scholar]
  32. Kristof‐Brown, A. L. , Zimmerman R. D., and Johnson E. C.. 2005. “Consequences of Individuals’ Fit at Work: A Meta‐Analysis of Person–Job, Person–Organization, Person–Group, and Person–Supervisor Fit.” Personnel Psychology 58, no. 2: 281–342. 10.1111/j.1744-6570.2005.00672.x. [DOI] [Google Scholar]
  33. Laferton, J. A. C. , Schiller S., Conrad D., Fischer D., and Zimmermann‐Viehoff F.. 2024. “Stress Beliefs Moderate the Impact of COVID‐19 Related Work Stress on Depressive, Anxiety and Distress Symptoms in Health Care Workers.” Stress and Health 40, no. 1: e3410. 10.1002/smi.3410. [DOI] [PubMed] [Google Scholar]
  34. Langer, A. , Meleis A., Knaul F. M., et al. 2015. “Women and Health: The Key for Sustainable Development.” Lancet 386, no. 9999: 1165–1210. 10.1016/S0140-6736(15)60497-4. [DOI] [PubMed] [Google Scholar]
  35. Lee, D. , Jalal S., Nasrullah M., Ding J., Sanelli P., and Khosa F.. 2020. “Gender Disparity in Academic Rank and Productivity Among Public Health Physician Faculty in North America.” Cureus Journal of Medical Science 12, no. 6: e8553. 10.7759/cureus.8553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lee, E. S. , Ryu V., Lee J. H., Hong H., Han H., and Park S.. 2021. “Psychometric Properties of the Korean Version of the Mental Health Professionals Stress Scale.” Frontiers in Psychiatry 12: 685423. 10.3389/fpsyt.2021.685423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Liao, E. Y. , Lau V. P., Hui R. T.‐y., and Kong K. H.. 2019. “A Resource‐Based Perspective on Work–Family Conflict: Meta‐Analytical Findings.” Career Development International 24, no. 1: 37–73. 10.1108/CDI-12-2017-0236. [DOI] [Google Scholar]
  38. Luhr, S. , Schneider D., and Harknett K.. 2022. “Parenting Without Predictability: Precarious Schedules, Parental Strain, and Work‐Life Conflict.” RSF: The Russell Sage Foundation Journal of the Social Sciences 8, no. 5: 24–44. 10.7758/RSF.2022.8.5.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mansour, S. , and Tremblay D. G.. 2016. “Workload, Generic and Work–Family Specific Social Supports and Job Stress: Mediating Role of Work–Family and Family–Work Conflict.” International Journal of Contemporary Hospitality Management 28, no. 8: 1778–1804. 10.1108/IJCHM-11-2014-0607. [DOI] [Google Scholar]
  40. Mehrotra, S. , Rao K., and Subbakrishna D. K.. 2000. “Factor Structure of the Mental Health Professionals Stress Scale (MHPSS) Among Clinical Psychologists in India.” International Journal of Social Psychiatry 46, no. 2: 142–150. 10.1177/002076400004600207. [DOI] [PubMed] [Google Scholar]
  41. Michel, J. S. , Kotrba L. M., Mitchelson J. K., Clark M. A., and Baltes B. B.. 2011. “Antecedents of Work‐Family Conflict: A Meta‐Analytic Review.” Journal of Organizational Behavior 32, no. 5: 689–725. 10.1002/job.695. [DOI] [Google Scholar]
  42. Moen, P. , Fan W., and Kelly E. L.. 2013. “Team‐Level Flexibility, Work‐Home Spillover, and Health Behavior.” Social Science & Medicine 84: 69–79. 10.1016/j.socscimed.2013.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Moran, R. C. D. 2023. “Influence of Remote Work on the Work Stress of Workers in the Context of the Covid‐19 Pandemic: A Systematic Review.” Sustainability 15, no. 16: 12489. 10.3390/su151612489. [DOI] [Google Scholar]
  44. Nippert‐Eng, C. 1996. Home and Work. University of Chicago Press. [Google Scholar]
  45. Northwood, K. , Siskind D., Suetani S., and McArdle P.. 2021. “An Assessment of Psychological Distress and Professional Burnout in Mental Health Professionals in Australia During the COVID‐19 Pandemic.” Australasian Psychiatry 29, no. 6: 628–634. 10.1177/10398562211038906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. O'Connor, K. , Muller Neff D., and Pitman S.. 2018. “Burnout in Mental Health Professionals: A Systematic Review and meta‐analysis of Prevalence and Determinants.” European Psychiatry 53: 74–99. 10.1016/j.eurpsy.2018.06.003. [DOI] [PubMed] [Google Scholar]
  47. Paoli, P. , and Merllie D.. 2001. Third European Survey on Working Conditions 2000. European Foundation for the Improvement of Living and Working Conditions. Office for Official Publications of the European Communities. [Google Scholar]
  48. Parasuraman, S. , and Greenhaus J. H.. 2002. “Toward Reducing Some Critical Gaps in Work‐Family Research.” Human Resource Management Review 12, no. 3: 299–312. 10.1016/s1053-4822(02)00062-1. [DOI] [Google Scholar]
  49. Phelps, A. , Lloyd D., Creamer M., and Forbes D.. 2009. “Caring for Carers in the Aftermath of Trauma.” Journal of Aggression, Maltreatment & Trauma 18, no. 3: 313–330. 10.1080/10926770902835899. [DOI] [Google Scholar]
  50. Pohl, R. , Botscharow J., Böckelmann I., and Thielmann B.. 2022. “Stress and Strain Among Veterinarians: A Scoping Review.” Irish Veterinary Journal 75, no. 1: 15. 10.1186/s13620-022-00220-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Ray, S. L. , Wong C., White D., and Heaslip K.. 2013. “Compassion Satisfaction, Compassion Fatigue, Work Life Conditions, and Burnout Among Frontline Mental Health Care Professionals.” Traumatology 19, no. 4: 255–267. 10.1177/1534765612471144. [DOI] [Google Scholar]
  52. Schiff, M. , and Leip L.. 2018. “The Impact of Job Expectations, Workload, and Autonomy on Work‐Related Stress Among Prison Wardens in the United States.” Criminal Justice and Behavior 46, no. 1: 136–153. 10.1177/0093854818802876. [DOI] [Google Scholar]
  53. Singh, J. , Karanika‐Murray M., Baguley T., and Hudson J.. 2020. “A Systematic Review of Job Demands and Resources Associated With Compassion Fatigue in Mental Health Professionals.” International Journal of Environmental Research and Public Health 17, no. 19: 6987. 10.3390/ijerph17196987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Tandon, R. 2020. “COVID‐19 and Mental Health: Preserving Humanity, Maintaining Sanity, and Promoting Health.” Asian Journal of Psychiatry 51: 102256. 10.1016/j.ajp.2020.102256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Thompson, M. M. , Naccarato M. E., Parker K. C. H., and Moskowitz G.. 2001. “The Personal Need for Structure (PNS) and Personal Fear of Invalidity (PFI) Scales: Historical Perspectives, Present Applications and Future Directions.” In Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition, edited by Moskowitz G., 19–39. Erlbaum. [Google Scholar]
  56. UN Women . 2020. Progress of the World’s Women 2019‐2020: Families in a Changing World. Chapter 5. Caring Families, Caring Societies. 140–173. https://progress.unwomen.org.
  57. Wekenborg, M. K. , von Dawans B., Gienger N. T., Wierick A., Weitz J., and Dobroschke J. C.. 2024. “Beliefs About Stress Moderate the Association Between COVID‐19‐related Work Demands and Burnout Symptoms in Physicians.” Stress and Health 40, no. 1: e3278. 10.1002/smi.3278. [DOI] [PubMed] [Google Scholar]
  58. Wheeler, A. R. , Coleman Gallagher V., Brouer R. L., and Sablynski C. J.. 2007. “When Person‐Organization (Mis)Fit and (Dis)Satisfaction Lead to Turnover.” Journal of Managerial Psychology 22, no. 2: 203–219. 10.1108/02683940710726447. [DOI] [Google Scholar]
  59. White, D. 2006. “The Hidden Costs of Caring.” Health Care Manager 25, no. 4: 341–347. 10.1097/00126450-200610000-00010. [DOI] [PubMed] [Google Scholar]
  60. Yang, J. , Zhang Y., Shen C., Liu S., and Zhang S.. 2019. “Work‐Family Segmentation Preferences and Work‐Family Conflict: Mediating Effect of Work‐Related ICT Use at Home and the Multilevel Moderating Effect of Group Segmentation Norms.” Frontiers in Psychology 10: 834. 10.3389/fpsyg.2019.00834. [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 upon reasonable request from the corresponding author. The full data are not publicly available due to containing information that could compromise the privacy of research participants.

Articles from Stress and Health are provided here courtesy of Wiley

RESOURCES