When COVID-19 became a pandemic and an increasingly high number of people got infected also in the Netherlands, the Dutch government announced an intelligent lockdown. As of March 15, 2020, Dutch citizens were advised to keep 1.5 m distance from each other, gatherings with more than three persons other than household members were prohibited, and employees were urged to work from home as much as possible (Rijksoverheid, 2020). Lockdown measures worldwide have inherently created a large-scale working from home experiment, which is considered by many a big success (Kelly, 2020). In light of positive noises about working from home and expectations that working from home might become the new norm in the wake or continuation of the COVID-19 pandemic (Kamouri and Lister, 2020; Lopez-Leon et al., 2020), it is imperative to examine its implications for lifestyle and employee well-being.

Dutch observational studies have shown – and newspapers reported – that the COVID-19 lockdown has a non-negligible impact on people’s lifestyle in terms of fewer social contacts, changed eating habits, decreased physical activity, decreased sleep quality, and psychological factors (iResearch, 2020). Cross-sectional studies in countries like Australia (Stanton et al., 2020), China (He et al., 2020), Chile (Reyes-Olavarría et al., 2020), Italy (Di Renzo et al., 2020), and France (Constant et al., 2020) confirm that, during the COVID-19 lockdown, people have gained weight, engaged less in physical activity, showed more sedentary behavior, consumed more alcohol, slept less well, and changed their eating habits (both positive and negative changes in healthy nutrition have been observed). A study conducted in Spain showed that changes in health-related behaviors depend on the duration of confinement due to COVID-19 and people may gradually improve their lifestyle during confinement (López-Bueno et al., 2020). We set out to contribute to this recent stream of research on changes in lifestyle since the onset of COVID-19. Our cross-sectional study focuses on working people in the Netherlands and aims to elucidate the associations between work, happiness, and lifestyle during the COVID-19 lockdown. Ultimately, our goal is to better understand the role of lifestyle in employee well-being.

Notwithstanding the obvious benefits of working from home (e.g., flexibility, no commute), it may also pose a risk for employees’ health and subjective well-being. Previous COVID-19-related research argues that working from home is related to psychosocial risks such as exhaustion and boredom (Brooks et al., 2020) as well as social isolation in the professional sphere and blurring of boundaries (Bouziri et al., 2020). Following recent calls by Cho (2020) and O’Connor et al. (2020) for more research on the impact of remote and flexible working arrangements and on changes in role boundaries in the context of COVID-19, we focus in this study on the blurring of work-life boundaries and its consequences for employee happiness. As many are forced to work from home on a permanent basis during the COVID-19 lockdown, it may become difficult to tell when the workday begins and ends. Employees may be tempted to engage in work-family multitasking more than usual as they attend to personal or family matters during the workday. In essence, the work-from-home practice in the age of COVID-19 is associated with a fungible use of time (Borpujari et al., 2020) and evidently creates considerable challenges for work-life balance (Cho, 2020). Another expected consequence of the ongoing COVID-19 outbreak is that it “may modify, for better or for worse, lifestyle behaviors” (Balanzá-Martínez et al., 2020, p. 399). In this paper, we aim to advance our understanding of the role of lifestyle in how employees respond to blurred work-life boundaries. To this end, we propose and test a conceptual model that integrates mediating and moderating perspectives on lifestyle in the process by which blurred boundaries can reduce employee happiness.

Well-being is a multifactorial construct and can be understood broadly as a person’s eudaimonic experience of fulfillment and purpose or the hedonic experience of feeling good (Linley et al., 2009; Di Fabio and Palazzeschi, 2015). Psychological well-being, which is the eudaimonic approach to well-being, is seen as well-being related to the full functioning of a person (Ryff and Keyes, 1995). The hedonic approach focuses on subjective well-being, which is defined as the balance between positive and negative mood and the evaluation of a person’s satisfaction with life (Linley et al., 2009). In colloquial terms, subjective well-being is often referred to as happiness (Diener et al., 2018; Myers and Diener, 2018). In this study, we consider happiness an important outcome in and of itself because it is a subjective state of mind reflecting overall subjective well-being (Diener, 2000) as well as a key indicator of career sustainability among employees (De Vos et al., 2020).

We draw on border theory (Clark, 2000) to argue that a particularly likely consequence of working from home as the new norm amid the COVID-19 pandemic is blurring of boundaries between work and private life (see also Schieman and Young, 2010; Bouziri et al., 2020; Cho, 2020). With greater role integration and blurring of boundaries, family-to-work and work-to-family conflict are more likely as well as interruptions and distractions while working and role transitions (Desrochers et al., 2005). When work (persistently) encroaches on and intrudes personal or family time, it might become more difficult to psychologically detach from work. Moreover, stress and negative emotions may arise from the struggles to attain or maintain work-life balance. Central to border theory is the notion that individuals are motivated to create a desired balance between work and other domains (Clark, 2000). In accordance with this notion, it has been found that the work-home interface influences how happy and satisfied individuals are with their lives (Odle-Dusseau et al., 2012; Sirgy and Lee, 2018; Tang et al., 2020). We therefore propose that blurred work-life boundaries reduce employee happiness.

To illuminate the pathways through which blurred work-life boundaries result in reduced happiness, we focus on emotional exhaustion as a mediating mechanism. Emotional exhaustion is a component of burnout and is characterized by a negative state of physical and emotional depletion due to work (Maslach and Jackson, 1981). We propose that border creep in the work-home interface is an emotionally exhausting experience that may reduce employee happiness. Indeed, those who experience difficulties with their work-life balance are more emotionally exhausted (Umene-Nakano et al., 2013). To the best of our knowledge, emotional exhaustion has not been examined as a consequence of blurred work-life boundaries per se. However, blurring of boundaries is positively associated with work-family conflict (Desrochers et al., 2005; Glavin and Schieman, 2012) and work-family conflict has been extensively linked to emotional exhaustion (for a review, see Allen et al., 2000). Moreover, emotional exhaustion has negative effects on happiness (Eckleberry-Hunt et al., 2016; Peralta and Saldanha, 2017). Thus, we put forward the following mediation hypothesis:

In this paper, lifestyle is conceptualized as comprising four behaviors: sleep, nutrition, physical activity, and relaxation (as a psychosocial behavior to cope with stress). Studies on lifestyle have extensively focused on (a combination of) sleep (e.g., Agarwal et al., 2020; Ingram et al., 2020), physical activity (Griep et al., 2015; Larsson et al., 2019), nutrition (Wu et al., 2016; Nishi et al., 2017), and psychosocial behaviors such as relaxation (e.g., Strijk et al., 2009; Van Scheppingen et al., 2015). These four factors are also included in assessment toolkits (Reis et al., 2019) and guidelines for lifestyle (Kanji et al., 2018).

A substantial body of research demonstrates the impact of lifestyle on mental health and subjective well-being (Walsh, 2011; Prendergast et al., 2016; Atzendorf et al., 2018; Lianov et al., 2019; Wang and Geng, 2019; Ingram et al., 2020). Specifically, a healthier diet (Piqueras et al., 2011; Blanchflower et al., 2013; Rooney et al., 2013; White et al., 2013), higher quantity and quality of sleep (Shin and Kim, 2018; Zhao et al., 2019) and more physical activity (Khazaee-Pool et al., 2015; Murphy et al., 2018; Zhang and Chen, 2019) have been shown to lead to higher happiness. Thus, research has established that healthy lifestyle behaviors can make people happier.

However, the role of lifestyle in the process by which blurred work-life boundaries influence happiness remains elusive. Drawing a parallel to the stressor-detachment model (Sonnentag and Fritz, 2015), lifestyle can be considered both a mediator and a moderator in this process. The mediation model proposes that blurred boundaries (perhaps via emotional exhaustion) impair healthy lifestyle behaviors, and in turn, lifestyle risk factors negatively influence happiness. It would imply that an unhealthy lifestyle is an explanation for why blurred work-life boundaries (and emotional exhaustion) reduce happiness. The moderation model proposes that healthy lifestyle behaviors buffer the (indirect) effect of blurred work-life boundaries on happiness. The relative merits of these models are not trivial; whereas a buffering effect implies that a healthy lifestyle has the potential to prevent blurred boundaries from lowering happiness, the mediation model suggests that a most likely consequence of blurred boundaries is a less healthy lifestyle. Our conceptual model in Figure 1 integrates both perspectives.

Empirical results on the potential buffering effects of lifestyle behaviors are scant and sometimes inconsistent (cf. Gerber and Pühse, 2008; DeBoer et al., 2012). However, lifestyle has been related to functional coping (Erschens et al., 2018) and recovery (Rook and Zijlstra, 2006) in the context of work-related stress. When blurred boundaries and emotional exhaustion result in a loss of resources, additional or substitute resources can be gained through a healthy lifestyle (e.g., high sleep quality; Gombert et al., 2018). That is, healthy lifestyle behaviors can help tackle feelings of frustration and fatigue related to work-life issues such that employees are less negatively affected in terms of lower happiness. Insights from the field of psychoneuroimmunology suggest that health-promoting lifestyle behaviors can help individuals increase their psychological resilience by improving their psychoneuroimmunological functions (Kim and Su, 2020). More specifically, exercise and relaxation protect against psychological distress by decreasing the release of stress-related hormones (Moraes et al., 2018), healthy nutrition decreases distress via the gut-brain axis (Madison and Kiecolt-Glaser, 2019), and adequate sleep reduces distress through changes in inflammation-related genes (Chiang et al., 2019). In sum, lifestyle behaviors can be considered adaptive coping strategies that reduce reactivity to distress and loss of resources. We therefore hypothesize that the strength of the blurred boundaries—emotional exhaustion—happiness association varies as a function of lifestyle.

Only a few studies consider lifestyle behaviors as mediators between work factors and well-being (e.g., Allen and Armstrong, 2006). Work stress and job strain have been found to be associated with fewer health-promoting lifestyle behaviors (Lindquist and Cooper, 1999; Lee et al., 2011; Griep et al., 2015; Miron and Colosi, 2018). In response to stress, employees may be motivated to use unhealthy behaviors to manage the aversive state (Ng and Jeffery, 2003). Similarly, it stands to reason that struggles in the work-life sphere and emotional exhaustion will disincline individuals to engage in health-promoting lifestyle behaviors. In fact, it has been shown that stress related to the work-life interface is associated with a variety of unhealthy lifestyle behaviors (for a review, see Hammer and Sauter, 2013). Moreover, Wu et al. (2016) found that those with a poor work-recreation balance (as part of work-life balance) are more likely to exhibit an unhealthy lifestyle, including poor nutrition, less sports and exercise activities, and poor stress management. Emotional exhaustion has also been linked to unhealthy lifestyle choices (Petrelli et al., 2018; Padilla et al., 2019). Conceptually, blurred boundaries and emotional exhaustion may deplete resources (i.e., self-regulatory resources but also time) that could facilitate a health-promoting lifestyle. Although unhealthy behaviors such as indulgent eating can bring instant gratification, as a form of maladaptive coping (Lindquist and Cooper, 1999) it may negatively influence happiness on the longer term. We therefore put forward the following hypothesis:

Before testing the hypotheses, we take a closer look at our study variables to better understand changes associated with the COVID-19 lockdown. Descriptive statistics and correlations are presented in Table 1. In our sample, the COVID-19 lockdown period was associated with higher blurring of work-life boundaries (M = 3.50), with 63.5% of respondents reporting that blurring of boundaries is (much) more than usual. We also asked how frequently participants worked from home during this period (rarely or never, sometimes, often or always) and this between-subjects factor was significantly related to changes in blurred boundaries (F = 33.7, p < 0.001), with those working from home more often experiencing increases in blurring of work-life boundaries. Since the start of the lockdown, 73.6% of respondents had worked from home often or always, compared with 11.1% before the lockdown. While little evidence was found for a negative change in overall lifestyle (M = 2.96, with a sum of 1.1 lifestyle behaviors that deteriorated), 26.4% of participants slept worse than usual (M = 2.87), 39.5% exercised less than usual (M = 2.89), and 29.0% took less time than usual to relax (M = 2.97). A positive change in healthy nutrition was observed for 23.1% of participants (M = 3.10). These percentages beg the question why the lifestyle behaviors of some have improved while they have deteriorated for others as well as what are the consequences of such changes. As can be seen from Table 1, lifestyle during the COVID-19 lockdown was associated with favorable changes in the other study variables, which preliminarily suggests that a healthy lifestyle enhances employees’ resilience to adverse circumstances.

Table 2 provides further information by comparing the key study variables across groups per gender, household composition, and other demographic characteristics. It can be seen that blurring of boundaries was negatively associated with participants’ age (p < 0.001) and age of the youngest child (p = 0.002). Moreover, an independent samples t-test revealed that those with no children at home experienced less deteriorations in blurred boundaries (t = −3.44, p = 0.001) and lifestyle behaviors (t = 2.28, p = 0.02) compared with those with one or more children at home. ANOVAs indicated that workers on a permanent contract were worse off in terms of blurred boundaries than the self-employed (p = 0.004) and higher level of education was associated with experiences of increased blurring of boundaries (F = 18.17, p < 0.001) and decreased happiness (F = 3.12, p = 0.03).

Table 3 shows that changes in the four lifestyle behaviors are interrelated (all ps < 0.001). Lifestyle behaviors during the COVID-19 lockdown are also significantly related; the positive correlations between sleep and physical activity [χ²(1) = 6.5, p = 0.01, Φ_Cramer = 0.10], sleep and nutrition [χ²(1) = 6.0, p = 0.02, Φ_Cramer = 0.09], and physical activity and nutrition [χ²(1) = 27.8, p < 0.001, Φ_Cramer = 0.20] suggest that individuals are more likely to meet the criteria for one lifestyle behavior if they meet the criteria for another.

Results from the basic mediation analysis (PROCESS model 4) showed that blurring of work-life boundaries was positively associated with emotional exhaustion (B = 0.36, p < 0.001), which in turn was negatively related to happiness (B = −0.34, p < 0.001). The negative indirect effect of blurred boundaries on happiness through emotional exhaustion was also significant (ab = −0.12, 95% CI [−0.17, −0.08]), lending support to our mediation hypothesis (H1). The direct effect of blurred boundaries on happiness was not significant (B = −0.04, p = 0.26).

The moderated mediation analysis using PROCESS model 58 revealed that lifestyle was a significant moderator of the association between emotional exhaustion and happiness (B = 0.09, p = 0.01). This interaction is visually depicted in Figure 2. For the effect of emotional exhaustion on happiness, tests of simple slopes showed that the slopes for both lower (–1 SD) and higher (+1 SD) levels of a healthy lifestyle were statistically significant. In fact, all obtained values for lifestyle fall within the region of significance of the simple slopes, which defines the specific values of the moderator at which the slope is statistically significant. Thus, a healthy lifestyle cannot entirely prevent emotional exhaustion from decreasing happiness. However, lifestyle does mitigate the detrimental effect of emotional exhaustion on happiness, which is substantiated by results regarding conditional indirect effects; that is, the indirect effect of blurred boundaries on happiness via emotional exhaustion was less negative at higher levels (+1 SD) of a healthy overall lifestyle (ab = −0.08, 95% CI [−0.13, −0.03]) than at lower levels (–1 SD) of a healthy overall lifestyle (ab = −0.15, 95% CI [−0.22, −0.08]). Results regarding the moderating role of overall lifestyle are presented in Table 4. The main predictors and control variables together explain 21% of the variance in happiness.

Figure 3 presents the results of moderated mediation analyses when zooming in on the specific lifestyle behaviors. We found that sleep was both a significant first-stage moderator for the association between blurred boundaries and emotional exhaustion (B = −0.20, p = 0.03) and a significant second-stage moderator for the association between emotional exhaustion and happiness (B = 0.18, p = 0.01). The percentile bootstrap confidence interval for the index of moderated mediation (B = 0.11) did not include zero, 95% CI [0.04, 0.18]), indicating that the negative indirect effect of blurred boundaries on happiness was significantly less strong (but still significant) for those who met the criteria for healthy sleep (ab = −0.04, 95% CI [−0.08, −0.01]) compared with those who did not meet the criteria (ab = −0.15, 95% CI [−0.21, −0.09]). For physical activity, results showed a significant interaction with emotional exhaustion in predicting happiness (B = 0.14, p = 0.03). However, the confidence interval for the index of moderated mediation (B = 0.05) included zero, 95% CI [−0.03, 0.14]), implying that the indirect effect of blurred boundaries on happiness through emotional exhaustion was not conditional on meeting the criteria for healthy physical activity. Finally, results indicated that nutrition did not moderate the association between blurred boundaries and emotional exhaustion (B = −0.01, p = 0.95), nor the association between emotional exhaustion and happiness (B = 0.08, p = 0.19). Jointly, though, the results provide evidence for Hypothesis 2 stating that healthier lifestyle behaviors buffer the detrimental indirect effect of blurred boundaries on happiness via emotional exhaustion.

We next tested our serial mediation hypothesis (H3) using PROCESS model 6. This model decomposes the total effect into indirect effects and the remaining direct effect. Three indirect effects were estimated, namely (1) blurred boundaries—emotional exhaustion—happiness, (2) blurred boundaries—lifestyle—happiness, and (3) blurred boundaries—emotional exhaustion—lifestyle—happiness. The results presented in Table 5 indicated that blurred boundaries (B = −0.11, p = 0.002) and emotional exhaustion (B = −0.24, p < 0.001) were negatively related to a healthy overall lifestyle, which is turn was positively related to happiness (B = 0.17, p < 0.001). The effect of blurred boundaries on happiness was significantly mediated in serial by emotional exhaustion and lifestyle, 95% CI [−0.03, −0.01]. The indirect effect of blurred boundaries on happiness through lifestyle but without emotional exhaustion was also significant, 95% CI [−0.04, −0.004], in addition to the mediation path from blurred boundaries through emotional exhaustion to happiness, 95% CI [−0.15, −0.07]. The remaining direct effect of blurred boundaries on happiness was not significant (B = −0.02, p = 0.52). The model explains 22% of the variance in happiness.

Regarding the specific lifestyle behaviors, we found that the overall serial mediation involving sleep quality was significant, 95% CI [−0.03, −0.01]. With emotional exhaustion in the model, blurring of work-life boundaries was not associated with sleep quality (B = −0.07, p = 0.11) and therefore the indirect effect of blurred boundaries on happiness through sleep quality (not involving emotional exhaustion) was not significant, 95% CI [−0.03, 0.004]. The mediation path from blurred boundaries through emotional exhaustion to happiness was significant, 95% CI [−0.15, −0.07], and the remaining direct effect of blurred boundaries on happiness was not significant (B = −0.04, p = 0.34). Physical activity was also a significant mediator in the total serial mediation path, 95% CI [−0.01, −0.0001]. Blurring of boundaries was not directly associated with physical activity (B = −0.04, p = 0.32) and its indirect effect on happiness via physical activity was therefore also not significant, 95% CI [−0.02, 0.003]. The mediation path from blurred boundaries through emotional exhaustion to happiness was significant, 95% CI [−0.16, −0.08], and the remaining direct effect of blurred boundaries on happiness was not significant (B = −0.04, p = 0.32). For nutrition, the results showed that both blurred boundaries (B = −0.14, p = 0.01) and emotional exhaustion (B = −0.08, p = 0.048) were negatively related to healthy eating. However, the percentile bootstrap confidence interval for the total serial mediation path (only just) included zero and was therefore not significant, 95% CI [−0.01, 0.0003]. The indirect effect of blurred boundaries on happiness through nutrition (not involving emotional exhaustion) was significant, 95% CI [−0.02, −0.0001], as well as the mediation path from blurred boundaries through emotional exhaustion to happiness, 95% CI [−0.16, −0.08]. The remaining direct effect of blurred boundaries on happiness was not significant (B = −0.03, p = 0.38). Finally, all indirect effects involving relaxation were significant. Results indicated that the effect of blurred boundaries on happiness was significantly mediated in serial by emotional exhaustion and relaxation, 95% CI [−0.02, −0.004]. The indirect effect of blurred boundaries on happiness through relaxation but without emotional exhaustion was also significant, 95% CI [−0.03, −0.002], as well as the mediation path from blurred boundaries through emotional exhaustion to happiness, 95% CI [−0.15, −0.07]. As in the other models, the remaining direct effect of blurred boundaries on happiness was not significant (B = −0.03, p = 0.44). Jointly, the results of our serial mediation analyses provide strong support for Hypothesis 3. Figure 4 offers a parsimonious presentation of the serial mediation results involving the various lifestyle behaviors.

The COVID-19 pandemic has compelled many employees worldwide to work from home. Experiences gained during lockdown are likely to have long-lasting consequences. For instance, 57% of the participants in our sample reported they would like to work from home more often in the future. Moreover, 46% of participants expected that supervisors and co-workers would be more supportive of them working from home compared with before the lockdown. Given that the blurring of work-life boundaries gets intensified for home workers, as we have found in our study and other scholars have argued as well (e.g., Bouziri et al., 2020), we set out to examine the effect of blurred boundaries on subjective well-being (i.e., happiness). In a sample of employees in the Netherlands, we observed that increases in blurred work-life boundaries predicted negative changes in happiness through enhanced emotional exhaustion.

Yet the main goal of this study was to illuminate the role of lifestyle in the process by which blurred work-life boundaries decrease happiness (see Figure 1). We found that a healthy lifestyle acted as a buffer for the detrimental indirect effect of blurred work-life boundaries on happiness. That is, those who exhibited a healthy lifestyle (characterized by multiple health behaviors) during lockdown were less negatively affected by increases in the blurring of work-life boundaries in the form of reduced happiness. Sleep in particular was a protective lifestyle factor; both the effect of blurred boundaries on emotional exhaustion and the effect of emotional exhaustion on happiness were weakened for those whose sleep had been adequate (i.e., meeting health guidelines) during the COVID-19 lockdown. Physical activity was also a protective lifestyle behavior; those who had been physically active during lockdown were less likely to suffer from lower happiness as a consequence of emotional exhaustion. However, physical activity did not buffer the indirect effect of blurred boundaries on happiness through emotional exhaustion.

We can conclude that lifestyle is an important buffer for those who experience blurred work-life boundaries and are emotionally exhausted from work. Somewhat paradoxically, however, we observed that blurred boundaries and emotional exhaustion make a healthy lifestyle more unlikely. Those who experienced heightened levels of blurring of work-life boundaries during the COVID-19 lockdown period reported a deterioration in lifestyle behaviors. Deterioration in lifestyle in turn was related to poor subjective well-being in terms of reduced happiness. This mediation effect holds for all lifestyle behaviors in our model; that is, sleep, physical activity, nutrition, and relaxation were all directly or indirectly (via emotional exhaustion) negatively affected by increases in the blurring of work-life boundaries and they all contributed to happiness. In other words, changes in lifestyle behaviors underpin the detrimental effect of blurred boundaries on happiness.

It appears that, although the need for a healthy lifestyle increases when work-life boundaries become blurred, the most likely response to this blurring seems engaging in unhealthy behaviors. Employees who experience blurring of work-life boundaries are at risk of poor sleep and neglecting relaxation, physical activity, and healthy eating. Given our results on the protective function of a healthy lifestyle, this response can be considered maladaptive and dysfunctional. For individuals who experience blurred work-life boundaries, it is particularly important to have a good night’s rest and (to a lesser degree) move and exercise regularly. When these health-promoting behaviors are not encouraged as part of the work and family climates (Buden et al., 2017), a paradox might occur in which those employees who would benefit the most from a healthy lifestyle are less able to lead a healthy life.

Following recent calls by Balanzá-Martínez et al. (2020) and Cho (2020), we have integrated in this paper a focus on blurring of work-life boundaries and lifestyle behaviors in the context of the COVID-19 pandemic. The current research has used data on the first wave of COVID-19 in the Netherlands and found evidence for a dual role of lifestyle in the proposed pathway linking blurred work-life boundaries to lower happiness. We have established that healthy lifestyle behaviors (in particular sleep and physical activity) can buffer employees against the detrimental effects of blurred work-life boundaries and emotional exhaustion on happiness. Yet we also found that greater blurring of work-life boundaries makes it more difficult for employees to sustain a healthy lifestyle (in terms of sleep, nutrition, physical activity, and relaxation). Our results regarding the influence of work-related factors on lifestyle behaviors suggest that, for many employees, it is a matter of not being able (versus willing) to lead a healthy life. The current research has important implications for both employees and employers, as we argue that lifestyle and well-being should be considered a shared responsibility between them.

The data supporting the conclusions of this study are available via https://doi.org/10.17026/dans-xzj-hzz6.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

HP and JW contributed to the conception and design of the study. HP and JW contributed to data collection. HP organized the database and performed the statistical analyses. HP and JW discussed and interpreted the results. JW carried out the literature search. HP took the lead in writing the original draft of the manuscript with input from JW. All authors contributed to the article and approved the submitted version.

JW at the time of study was employed by company SMC Rijnland Fysiotherapeuten. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.