Volume 73, Issue 4 p. 1893-1915
ORIGINAL ARTICLE
Open Access

Who has the most to lose? How ICT demands undermine health-oriented leadership

Laura Klebe

Corresponding Author

Laura Klebe

Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Hamburg, Germany

Correspondence

Laura Klebe, Department of Work, Organizational and Economic Psychology, Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Holstenhofweg 85, 22043 Hamburg, Germany.

Email: [email protected]

Search for more papers by this author
Jörg Felfe

Jörg Felfe

Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Hamburg, Germany

Search for more papers by this author
Annika Krick

Annika Krick

Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Hamburg, Germany

Search for more papers by this author
Dorothee Tautz

Dorothee Tautz

Helmut Schmidt University/University of the Federal Armed Forces Hamburg, Hamburg, Germany

Search for more papers by this author
First published: 17 March 2024
Citations: 2

Funding information: The study was funded by dtec.bw – Digitalization and Technology Research Center of the Bundeswehr.

Abstract

Research shows positive effects of Health-oriented Leadership (HoL) on followers' health. However, irritation elicited by ICT hassles may reduce leaders' capacity to engage in staff care. This study examines whether ICT hassles are associated with staff care (i.e., health-promoting follower-directed leadership) via irritation and whether particularly those engaging in self-care suffer more or less from ICT demands. A moderated mediation model was tested at three measurement points (N = 582 leaders).

As expected, results show more irritation for leaders with more ICT hassles which is further associated with less staff care. Moreover, the positive relationship between ICT hassles and irritation was stronger for leaders displaying high self-care. In the same vein, the negative relationship between irritation and staff care was stronger for leaders engaging in self-care.

Findings provide the first evidence that ICT hassles are negatively related to staff care via leaders' irritation. Leaders who engage in self-care show less irritation and higher staff care but suffer more from ICT demands. To promote leaders' health and staff care in digital working contexts, organizations should provide reliable IT equipment and technical support. The study ties in with research on digital leadership and its antecedents and offers a new view on the interplay of demands and resources.

INTRODUCTION

Advancing digitalization and the possibility of working from home has not only brought new opportunities but also new risks that threaten leader and follower health (Ipsen et al., 2021; Niebuhr et al., 2022). To promote health in the workplace, research increasingly focuses on health-promoting leadership which specifically accounts for leader and follower health. The Health-oriented Leadership concept (HoL; Franke et al., 2014) builds upon the Conservation of Resources Theory (COR; Hobfoll et al., 2018) and refers to both follower-directed leadership in terms of staff care (i.e., the extent to which leaders value, are aware of and promote their followers' health at work) and self-directed leadership in terms of self-care (i.e., the extent to which leaders and followers value, are aware of and promote their own health, respectively). Empirical findings underline the positive relationships of health-oriented leadership with health, well-being, and work-related attitudes in traditional working contexts (Arnold & Rigotti, 2021; Franke et al., 2014; Kaluza et al., 2021; Santa Maria et al., 2019; Teetzen et al., 2023).

A recent study by Klebe et al. (2023) could initially replicate the positive effects of health-oriented leadership on follower health and engagement in the digital working context. However, results also revealed that frequent Information Communication Technology (ICT) hassles such as freezing displays or program breakdowns weaken the effectiveness of staff care, underlining the risks of ICT for leadership. Up to now, it is unknown whether ICT hassles only impair the effectiveness of staff care or whether they also directly impair the levels of staff care as an antecedent. Past research already showed that ICT hassles lead to an increase in irritation (i.e., subjective perceived emotional and cognitive strain; Mohr et al., 2006) and that leaders' irritation, in turn, is related to lower staff care levels (Dragano & Lunau, 2020; Klebe et al., 2022; La Torre et al., 2019; Pischel et al., 2022). Accordingly, it seems plausible that ICT hassles are negatively associated with staff care via leaders' irritation.

Moreover, it is also unclear if and how the potential health-protecting function of leaders' self-care may come into play when staff care levels are impaired by higher leader irritation (Klebe et al., 2022; Pischel et al., 2022). Self Preservation Theory (Dickerson & Kemeny, 2004) postulates that individuals primarily aim at maintaining and promoting their own status so it is conceivable that particularly self-caring leaders with a strong focus on themselves would refrain from staff care and shift their focus on their own health when they are highly irritated.

Regarding ICT hassles, it is open to whether self-care can buffer the negative relationship with irritation as suggested in the Job Demands-Resources Model (JD-R; Bakker & Demerouti, 2007). Whereas a buffering effect on negative relationships between challenge stressors and irritation was already shown for staff care (Krick et al., 2022), recent studies show that the negative relationships between hindrance stressors such as ICT hassles or inefficient technical support and irritation (i.e., stress and strain) are even stronger for individuals with more resources (e.g., Bregenzer & Jimenez, 2021; Day et al., 2012). As we will argue, self-care may barely compensate for ICT demands and self-caring leaders may experience more cognitive dissonance in situations when they would like to care for themselves, while solving technical issues is outstanding.

The aim of this study is to identify further potential risks for leadership arising from ICT usage. To this end, this study examines indirect associations between ICT hassles and staff care via leader irritation (H1), interactional effects between ICT hassles and leaders' self-care regarding their irritation (H2), and interactional effects between leaders' irritation and leaders' self-care regarding staff care (H3) from a leaders' perspective (see Figure 1) in a survey study with three measurement points.

Details are in the caption following the image
Conceptual research model of the relationships between ICT hassles, leaders' irritation, staff care, and leaders' self-care. The bold dashed lines represent moderating effects.

The contribution of this approach is three-fold: 1) From a theoretical perspective, the study uncovers situational contingencies of health-oriented leadership in the digital working context by investigating threatening effects of demands (i.e., ICT hassles) and risk factors (i.e., irritation) on resources (i.e., HoL) in the sense of COR Theory (Hobfoll et al., 2018). Advancing theoretical knowledge on leadership in the digital context helps to answer the question of how far empirical evidence from the traditional office setting can be applied to the digital working context (Bregenzer & Jimenez, 2021; Efimov et al., 2020; Klebe & Felfe, 2023; Tautz et al., 2022). 2) From a methodological perspective, the study contributes to the further validation of the HoL concept by replicating and extending previous findings in the digital context and from the perspective of leaders. The perspective of actual leaders has often been neglected in previous research on HoL but is particularly important when investigating its antecedents (Klebe et al., 2022; Pischel et al., 2022). Using three timely independent measurement points counteracts the risk of common method bias (Podsakoff et al., 2003). 3) From a practical perspective, the study identifies potential risks for leadership in the digital working context and derives important implications for leaders and organizations, as working-from-home opportunities and the use of ICTs are likely to increase even more. Advanced knowledge of leadership in the digital context may make organizations aware of more far-reaching consequences of ICT hassles and encourage counteractions.

THE HEALTH-ORIENTED LEADERSHIP CONCEPT

Positive leadership behaviors such as transformational or servant leadership represent important workplace resources for follower health and well-being (der Kinderen et al., 2020; Montano et al., 2017; Skakon et al., 2010). However, rather general positive leadership behaviors may be too vague about health-specific effects which is why health-promoting leadership concepts came to the fore in leadership research (Franke et al., 2014).

The Health-oriented Leadership concept by Franke et al. (2014) is a health-specific leadership concept that differentiates between three empirically distinct components that contribute to leader and follower health, namely staff care, leaders' self-care, and followers' self-care. Both staff care and self-care consist of three facets: 1) value (i.e., the importance that is ascribed to health), 2) awareness (i.e., the awareness for health-related warning signals), and 3) behavior (i.e., actively reducing work-related stressors and fostering resources; Franke et al., 2014). In the sense of COR Theory (Hobfoll et al., 2018), self-care represents an internal resource helping to maintain and foster one's own health (Franke et al., 2014). Self-care is positively related to general health and negatively related to exhaustion, health complaints, engagement, presenteeism, and work–family conflicts (Grimm et al., 2021; Horstmann, 2018; Kaluza et al., 2021; Pundt & Felfe, 2017). In addition, staff care represents an important external workplace resource for followers (Franke et al., 2014). Staff care is positively related to follower health and well-being, job satisfaction, affective organizational commitment, engagement, and performance (Arnold & Rigotti, 2023; Klebe, Felfe, & Klug, 2021; Krick et al., 2022; Santa Maria et al., 2019; Teetzen et al., 2023).

Previous research has shown that health-oriented leadership in terms of staff care explains additional variance in follower health and well-being above and beyond other leadership constructs such as transformational leadership (Franke et al., 2014; Kaluza et al., 2021; Pischel & Felfe, 2023; Vincent-Höper & Stein, 2019). The validation of the HoL concept showed that staff care and self-care are distinctive, but related factors that influence leader and follower health (Franke et al., 2014). As a study by Klug et al. (2019) revealed four different profiles of health-oriented leadership with different combinations of self-care (high vs. low) and staff care (high vs. low), it is important to investigate self-care and staff care as distinctive components of healthy leadership. In the following, we will discuss ICT hassles as a risk factor for both self-care and staff care in the digital working context.

ASSOCIATIONS BETWEEN ICT HASSLES AND STAFF CARE VIA LEADERS' IRRITATION

Due to the process of digitization, followers increasingly depend on ICTs. While ICTs simplify the working life in many ways (e.g., greater temporal and spatial flexibility, support in work organization), the use of ICTs may also pose a demand for their users (e.g., handling difficulties or system crashes; Hu et al., 2021). Therefore, ICTs are sometimes rather seen as a job demand than a resource (Pansini et al., 2023).

Empirical literature shows the negative effects of ICT demands on followers' well-being (Baumeister et al., 2021; Day et al., 2012; Klebe et al., 2023). For example, a study by Day et al. (2012) found that ICT demands, and particularly ICT hassles, lead to an increase in follower stress, irritation levels, and exhaustion. Especially computer malfunctions such as program breakdowns, computer crashes, and freezing displays hinder followers in task fulfillment which negatively affects their goal achievement. According to Lazar et al. (2006), followers spend high amounts of working time dealing with frustrating computer malfunctions which leads to time losses in work processes as they need to be fixed before proceeding. Accordingly, ICT hassles elicit feelings of anger and frustration (Bessière et al., 2006; Chesley, 2014; Lazar et al., 2006), which is why they represent an important predictor of irritation in digital working contexts. Particularly in the remote working context ICT hassles can pose a significant demand, as employees often have to cope with them on their own (Klebe et al., 2023).

Leaders' irritation, in turn, has been identified as an important antecedent of leadership behavior in previous research. Studies show that leader irritation leads to an increase in negative and a decrease in positive behaviors, reasoned by lacking resources and capacities combined with a decrease in self-control (Harms et al., 2017; Kaluza et al., 2019). This was also shown for staff care: Pischel et al. (2022) revealed that leaders suffering from stressors such as time pressure and high workload have more difficulties to be aware of health-related warning signals among their followers. In the same vein, other studies found that leaders' irritation reduces staff care behavior (Klebe et al., 2022; Köppe et al., 2018; Krick et al., 2022). In accordance with COR Theory (Hobfoll et al., 2018), it is argued that job demands exhaust leaders' energy and resources so that their irritation increases. As leaders are primarily concerned with protecting their own resources to maintain their health under high stress, their awareness of their followers' health-related needs decreases and they engage less in staff care.

We are only aware of one study that investigated indirect relationships between job demands and staff care via leader health. A study by Arnold and Rigotti (2023) in the school context recently revealed that job demands such as role conflicts or interruptions are related to staff care via leaders' self-care and health. We aim to replicate this finding in a digital working context and expect a trickle-down effect of ICT hassles on staff care via leaders' irritation. According to COR Theory, resources travel in caravans so that the availability of resources enables individuals to gain new resources. However, when resources are lost, this may elicit a loss spiral so that ICT hassles enhance leader irritation which in turn is associated with lower levels of staff care. We hypothesize the following:

Hypothesis 1.ICT hassles are negatively associated with staff care via leaders' irritation.

THE INTERACTION OF LEADERS' IRRITATION AND SELF-CARE ON STAFF CARE

Besides leaders' irritation, previous research also uncovered leaders' self-care as an important antecedent of staff care (Arnold & Rigotti, 2023; Grimm et al., 2021; Klug et al., 2022). However, whether self-care can also influence the negative effects of leaders' irritation on staff care is yet unknown. On the one hand, it could be possible that self-care can buffer the negative relationship between irritation and staff care. While leaders' irritation limits their capacity and resources to care for their followers' health (Pischel et al., 2022), self-care promotes leaders' own health and creates the necessary capacity to engage in staff care (Arnold & Rigotti, 2023; Grimm et al., 2021; Klebe, Klug, & Felfe, 2021). Thus, when leaders are irritated in stressful periods, being aware of and actively coping with health-related warning signals could protect and maintain leaders' necessary resources to further promote their followers' health.

On the other hand, Self Preservation Theory suggests that individuals aim to maintain and promote their own social self. Accordingly, individuals are vigilant to threats that jeopardize their status (Dickerson & Kemeny, 2004). When leaders' irritation arises, they may perceive a goal conflict between preserving their own and their followers' health which may decrease their motivation to support their followers. This reasoning would be in line with previous studies showing that leaders refrain from staff care when strain or irritation is high (Klebe et al., 2022; Pischel et al., 2022). In this case, particularly self-caring leaders may enter a defensive mode and focus on their own health instead of their followers to preserve their own status (de Dreu & van Knippenberg, 2005; Dickerson & Kemeny, 2004). While self-care represents an important internal resource for leaders that protects them from resource loss and helps them to gain more resources, leaders who take great care of their own health also have a strong focus on self-preservation (Franke et al., 2014). When emerging irritation shortens their resources and capacities, leaders need to make more efforts to maintain their high levels of self-care and lack the necessary timely and mental resources to additionally care for their followers (Klebe et al., 2022; Pischel et al., 2022). In turn, leaders who engage less in self-care may have lesser tendencies to preserve the self. Thus, they may not shift their focus in stressful times and further engage in staff care also when irritation arises.

Supporting this notion, a recent survey study by Pischel et al. (2022) found that particularly leaders with more resources (i.e., high autonomy) experienced stronger losses in staff care with increasing work stress than leaders with fewer resources (i.e., low autonomy). The authors assume that leaders' autonomy allows them to care for their own health in stressful times, but that focusing on their own health decreases the capacity to perceive warning signals and health risks of their followers at the same time. This would be also in line with Self Preservation Theory, as autonomous leaders seem to shift their focus to their own instead of their followers' health when stress occurs (Dickerson & Kemeny, 2004).

Based on the aforementioned considerations and in line with the Self Preservation Theory (Dickerson & Kemeny, 2004), we expect that leaders with a high level of self-care shift their focus and refrain from staff care when irritation arises, while leaders with lower levels of self-care do not shift their focus and maintain their staff care levels. We postulate the following:

Hypothesis 2.Self-care moderates the negative relationship between leaders' irritation and staff care. The relationship between leaders' irritation and staff care is stronger when self-care is higher.

THE INTERACTION OF ICT DEMANDS AND SELF-CARE ON LEADERS' IRRITATION

While job demands increase irritation levels (Bakker & Demerouti, 2007), self-care prevents an increase in irritation (Franke et al., 2014). However, up to now, it is unknown whether self-care may influence the relationship between job demands and irritation. Because the Job-Demands Resources model postulates that job resources can buffer the negative effects of job demands on irritation (Bakker & Demerouti, 2007), it would be plausible that self-care can mitigate the negative effects of job demands. Accordingly, a recent study by Krick et al. (2022) already shows that health-promoting employee leadership can buffer the negative effects of job demands such as workload, multitasking, and time pressure on followers' irritation.

However, the findings regarding the buffering effects of resources on the negative effects of demands are inconsistent (Häusser et al., 2010). In contrast to the findings of Krick et al. (2022), a recent study by Bregenzer and Jimenez (2021) found that the negative relationship between job demands (i.e., inefficient technical support) and irritation (i.e., stress) was even stronger with more health-promoting leadership. This indicates that the negative effects of job demands may be even stronger for individuals with more resources.

A possible explanation may be that interactions between demands and resources differ for hindrance versus challenge demands. The challenge-stressor hindrance-stressor framework postulates a two-dimensional model of stressors. While challenge stressors (e.g., workload or time pressure) have the potential to promote personal gain and growth, hindrance stressors (e.g., role conflict or hassles) have the potential to harm personal gain or growth (J. A. LePine et al., 2005). As inefficient technical support and ICT hassles discourage employees from further work and thus harm personal gain, they can arguably be classified as hindrance demands (Klebe et al., 2023).

Previous literature has already discussed the differential effects of challenge and hindrance stressors for employee outcomes in the context of the JD-R model (Crawford et al., 2010; Schneider et al., 2017; van den Broeck et al., 2010), which may also apply to interactions (Tadić et al., 2015). However, there may be also different interactions for different hindrance demands. Whereas the study of Bregenzer and Jimenez (2021) showed amplifying effects of resources (i.e., health-promoting leadership) on the negative effects of inefficient technical support as a hindrance demand, a study by Tadić et al. (2015) showed buffering effects of resources (e.g., support and feedback) on the negative effects of hindrance demands such as role conflict or ambiguity. This indicates that the fit between demands and resources is particularly crucial when it comes to hindrance demands. While it seems plausible that feedback may help employees cope with role ambiguity (Tadić et al., 2015), leaders can barely replace or compensate for an efficiently working technical support department which should help in dealing with hindrance stressors such as ICT demands (Bregenzer & Jimenez, 2021).

Supporting the assumption that ICT demands are specific hindrance demands that cannot easily be buffered, also a study by Day et al. (2012) found that the relationship between ICT hassles as a hindrance stressor and irritation (i.e., strain) was even stronger for employees with more ICT resources (i.e., providing the latest technology, software, and upgrades). Thus, even having the best technology does not buffer the negative effects of ICT hassles when individuals need to fix hassles themselves, which increases their irritation. Also, a study by Mander and Antoni (2023) supports the notion that resources need to fit demands, as the negative effects of work overload as a hindrance stressor could not be buffered but were even stronger with more autonomy. Also, the possibility of working autonomously does not prevent negative effects if the quantity of work is too high to handle.

We argue that similar to external resources, also self-care as an internal resource cannot compensate for the negative effects of ICT hassles. High-demanding working conditions such as dealing with technical malfunctions mitigate leaders' time and mental capacities, as ICT hassles need to be fixed before proceeding with work (Lazar et al., 2006). This goes along with interruptions, insecurity about when and how to proceed, and extra effort which causes irritation (Dragano & Lunau, 2020). Engaging in self-care causes additional timely effort but does not fix the acute problem of ICT hassles. When leaders want to promote their health but instead must deal with technical issues, they may experience cognitive dissonance and may appraise ICT hassles as a particularly threatening hindrance stressor (M. A. LePine, 2022). Therefore, the fit between demands and resources may not be given so that self-care may not help in coping with ICT hassles as a hindrance stressor. Instead, the discrepancy between low and high ICT hassles is expected to be stronger for leaders highly engaging in self-care, whereas leaders engaging less in self-care are expected to experience generally high levels of irritation in both conditions. In view of the overall high-stress experience, no increase or decrease can be expected from changes in single stressors.

Considering the fit between hindrance demands and resources as well as previous findings (Bregenzer & Jimenez, 2021; Day et al., 2012; Mander & Antoni, 2023), we expect that those with a high level of self-care will particularly benefit regarding their health if there are no ICT hassles but are particularly affected when hassles occur. We propose the following:

Hypothesis 3.Self-care moderates the positive relationship between ICT hassles and leaders' irritation. The relationship between ICT hassles and leaders' irritation is stronger when self-care is higher.

METHODS

Sample and procedure

The survey was conducted as an online survey including three measurement points with a time lag of two months each. Data was collected by a market research institute. All participants gave their informed consent. The first wave of data was collected in spring 2021. At t1, 1192 leaders participated, while 750 leaders participated at t2, and 582 at t3, which corresponds to a dropout rate of 48.83%. Only complete datasets were considered in the data analysis. The final sample included N = 582 leaders, who worked at least one day a week from home at the time of the survey (12.2%; 2 days a week 20.4%; 3 days a week 25.9%; 4 days a week 14.6%; 5 days a week or more 26.8%). While 37.6% of the participants were female, 62.4% were male. The mean age was M = 46.21 years (SD = 12.13). Participants worked in several sectors such as metal and electrical industry (9.3%), IT and telecommunication (9.3%), or transport and traffic (8.3%) – either in the private sector (81.1%) or in public services (18.9%). More than half of the participants indicated that they worked in companies with 100 to 500 (30.4%) or more than 500 employees (35.4%; 10.5% up to 10, 12.4% between 11 and 49, 11.3% between 50 and 99). Most of the participants (34.5%) lived in a household with two persons (single 14.8%, 3 persons 21%, 4 persons 22%, and more 7.2%). 49.8% indicated that they never need to care for children or other relatives (12% rarely, 16.3% sometimes, 14.1% often, 7.7% almost always).

Measures

ICT hassles

ICT hassles were measured at t1 using the sub-scale ‘hassles’ of the ICT demands scale by Day et al. (2012). Leaders rated their ICT demands in terms of technical working conditions while working from home with four items. An example item is “I experience glitches with software”. ICT hassles were rated on a five-point frequency scale from 1 = never to 5 = almost always. Cronbach's Alpha was α = .88.

Leaders' irritation

Leader irritation served as an indicator of leader strain. Leaders' irritation was measured at t2 with the German version of the irritation scale by Mohr et al. (2006). For reasons of economy, we have dispensed with four items so that the scale consisted of four items. Items were “I have difficulty relaxing after work”, “From time to time I feel like a bundle of nerves”, “I anger quickly”, and “I get irritated easily, although I don't want this to happen”. The scale ranged from 1 = never to 5 = almost always. Cronbach's Alpha was α = .87.

Staff care

The amount of leaders' health-promoting follower-directed leadership was measured at t3 with the sub-scale ‘Staff Care’ of the Health-oriented Leadership instrument by Pundt and Felfe (2017). Leaders were asked to rate their own staff care within the last four weeks. The scale included 18 items (four items of the sub-facet ‘awareness’ and 14 items of the sub-facet ‘behavior’), for example, “I immediately notice when something is wrong with my followers' health”. Staff care was rated on a five-point Likert scale ranging from 1 = not at all true to 5 = completely true. Cronbach's Alpha was α = .91.

Self-care

We measured leaders' health-promoting self-directed leadership by the sub-scale ‘Self Care’ of the Health-oriented Leadership instrument by Pundt and Felfe (2017). Self-care was measured at both t1 and t2. Participants were asked to rate their own self-care within the last four weeks. The scale included 14 items (three items of the sub-facet ‘awareness’ and 11 items of the sub-facet ‘behavior’), for example, “I try to reduce my demands by optimizing my personal work routine (e.g., set priorities, care for undisturbed working, daily planning)”. The scale ranged from 1 = not at all true to 5 = completely true. Cronbach's Alpha was α = .87 for both t1 and t2.

Control variables

As some populations might have more difficulties in dealing with ICT demands, we controlled for age and gender. For example, Burton-Jones and Hubons (2005) revealed that older followers report more problems working with ICTs. Moreover, Ragu-Nathan et al. (2008) found that women experience less irritation due to the use of ICTs than men. Furthermore, we controlled for working from home intensity, as it is conceivable that leaders spending more days working from home experience more ICT demands.

We conducted a CFA to test our measurement model. As individual items have low reliability, low communality, and hyperplanes that are difficult to determine, Kishton and Widaman (1994), recommend to use of item parceling in factor analyses. Accordingly, we used items parceling based on sub-facets so that we had two indicators for self-care and staff care each (sub-facets awareness and behavior, respectively). We compared the fit of our 5-factor model with competing 4-, 3-, 2-, and 1-factor models. The hypothesized 5-factor model showed a better fit than the other models (χ2 = 186[65], p < .001; CFI = .974; RMSEA = .057; Table 1). The improvement in model fit was significant (Δχ2 = 46; p < .001), supporting the differentiation between five factors.

TABLE 1. Confirmatory factor analysis.
Model χ2 df χ2/df CFI RMSEA Δχ2 Δdf
5 factors 186.369 65 2.867 .974 .057 46*** 4
4 factors 232.818 69 3.374 .965 .064 177*** 3
3 factors 410.137 72 5.696 .929 .090 741*** 2
2 factors 1152.108 74 15.569 .772 .158 915*** 1
1 factor 2067.963 75 27.573 .579 .214
  • Note: N = 582. 5 Factors: ICT hassles, self-care t1, irritation, self-care t2, staff care. 4 Factors: ICT hassles, self-care t1 + self-care t2, irritation, staff care. 3 Factors: ICT hassles, irritation, self-care t1 + self-care t2 + staff care. 2 Factors: ICT hassles + irritation, self-care t1 + self-care t2 + staff care. 1 Factor: All items are loaded on one factor. CFI: Comparative Fit Index. RMSEA: Root Mean Square Error of Approximation.
  • *** p < .001.

Analyses

To test the interrelationships between the study variables, we calculated path models using Mplus 8 (Muthén & Muthén, 2017). Since our hypotheses specify indirect and interactional relationships between ICT hassles, leaders' irritation, leaders' self-care, and staff care, we tested the mediation hypothesis (Hypothesis 1) and the moderation hypotheses (Hypotheses 2 and 3) in one path model. Given a design with three points of measurement, we used ICT hassles at t1, leaders' irritation at t2, staff care at t3, and self-care at both t1 and t2. All independent variables and moderators were standardized.

For reasons of economy, we first tested our conceptual research model without including other relevant paths of the HoL model (Franke et al., 2014). However, the model showed a rather poor fit (Chi2 = 57.321[4], p < .001; CFI = .887; RMSEA = .151; SRMR = .038). To align the final model to the original HoL model and to improve the model fit, we sequentially added additional paths as recommended by Franke et al. (2014). Therefore, we added a direct path from self-care at t1 to staff care at t3 in a second step, as both the HoL model and previous literature report direct effects of leaders' self-care on staff care (Franke et al., 2014; Grimm et al., 2021; Klug et al., 2022). This procedure improved the model fit (Chi2 = 32.723[3], p < .001; CFI = .937; RMSEA = .130; SRMR = .030). As the HoL model also postulates a direct path from leaders' self-care to leaders' irritation, we additionally added a direct path from self-care at t2 to leaders' irritation at t2 in a third step (Chi2 = 5.677[2], p = .059; CFI = .992; RMSEA = .056; SRMR = .017). As this model is in accordance with the original HoL model and showed the best fit, we included both paths in the final model and reported the results accordingly.

RESULTS

Table 2 shows means, standard deviations, and intercorrelations for all study variables. ICT hassles were positively related to leaders' irritation (r = .35, p < .001), but negatively to leaders' self-care (r = −.21, p < .001 at t1 and r = −.16, p < .001 at t2) and staff care (r = −.13, p < .01). Leaders' irritation was negatively related to leaders' self-care (r = −.38, p < .001 at t1 and r = −.41, p < .001 at t2) and staff care (r = −.32, p < .001). Moreover, leaders' self-care (t1 and t2) and staff care were positively related (r = .56, p < .001 at t1 and r = .58, p < .001 at t2). Figure 2 shows the direct and indirect effects of the complete underlying model.

TABLE 2. Means, standard deviations, and correlations for gender, age, working from home, ICT hassles, leader irritation, staff care, and self-care.
M (SD) 1 2 3 4 5 6 7 8
1 Gender
2 Age 46.21 (12.13) −.07
3 Working from home 3.15 (0.61) .04 −.07
4 ICT hassles 2.13 (0.93) .06 −.18*** .01 (.88)
5 Leader irritation 2.41 (0.90) .06 −.13** .05 .35*** (.87)
6 Staff care 3.52 (0.59) .09* .03 .06 −.13** −.32*** (.91)
7 Leader self-care t1 3.45 (0.60) −.02 .02 .09* −.21*** −.38*** .56*** (.87)
8 Leader self-care t2 3.45 (0.58) −.02 .01 .09* −.16*** −.41*** .58*** .77*** (.87)
  • Notes: N = 582. Gender coded as 1 = male, 2 = female. Alpha coefficients across the diagonal in parentheses.
  • * p < .05,
  • ** p < .01, and
  • *** p < .001.
Details are in the caption following the image
The final model of the relationships between ICT hassles, leaders' irritation, staff care, and leaders' self-care. The bold dashed lines represent moderating effects. The thin dashed lines represent additional paths derived from model comparisons.

First, it was expected that ICT hassles would be indirectly associated with staff care via leaders' irritation (H1). As can be seen in Figure 2, our results show a positive relationship between ICT hassles (t1) and leaders' irritation (t2; B = .26, SE = .04, p < .001), and a negative relationship between leaders' irritation (t2) and staff care (t3; B = −.10, SE = .04, p < .01). The direct effect of ICT hassles on staff care was not significant (B = .03, SE = .01, p = .38). Supporting Hypothesis 1, the mediation analysis revealed an indirect association between ICT hassles (t1) and staff care (t3) via leaders' irritation (t2; B = −.03, SE = .01, 95% CI [−.05, −.01], p < .05; Table 3).

TABLE 3. Specific indirect effect of ICT hassles on staff care via leader irritation.
Effect SE LLCI ULCI p

ICT hassles ➔ leader irritation ➔ staff care

−.03 .01 −.05 −.01 .01
  • Note: N = 582. LLCI and ULCI represent the lower limit and upper limit of the 95% confidence interval.

Second, we proposed that self-care would moderate the negative relationship between leaders' irritation and staff care so that the relationship would be stronger for leaders' engaging in self-care (H2). The moderation analysis revealed a significant interaction between leaders' irritation (t2) and self-care (t2) on staff care (t3; B = −.14, SE = .03, p < .001; Table 4). Results showed that the simple slopes of high self-care were significant (B = −.17, SE = .04, p < .001), but not for low self-care (B = −.03, SE = .04, p = .42). As can be seen in Figure 3, the relationship was stronger when self-care was high. Moreover, the Johnson-Neyman-Analysis indicated that the conditional effect of irritation on staff care is significant (p < .05) if self-care (t2) is outside the interval [2.31, 3.27]. Hypothesis 2 was supported.

TABLE 4. Two-way interaction of ICT hassles and self-care on leader irritation.
B SE R2 ΔR2

ICT hassles

.26*** .04 .27*** .01*

Self-care (t1)

−.09 .06
ICT hassles x self care (t1) .08* .04
  • Note: N = 582.
  • *** p < .001,
  • ** p < .01, and
  • * p < .05.
Details are in the caption following the image
Interaction between ICT hassles and leaders' self-care on leaders' irritation.

Third, we expected that self-care would also moderate the positive relationship between ICT hassles and leaders' irritation so that the relationship would be stronger for leaders engaging in self-care (H3). The moderation analysis revealed a significant interaction between ICT hassles (t1) and self-care (t1) on leaders' irritation (t2; B = .08, SE = .04, p < .05; Table 5). Results showed that the simple slopes for both low (B = .22, SE = .04, p < .001) and high self-care (B = .30, SE = .04, p < .001) were significant. As can be seen in Figure 4, the relationship was stronger when self-care was high. Moreover, the Johnson-Neyman-Analysis was conducted to determine for which values within the moderator's range of values (i.e., for which region) the conditional regression lines are significant. Johnson-Neyman-Analysis indicated that the conditional effect of ICT hassles on irritation is significant (p < .05) if self-care (t1) is higher than 2.57. Hypothesis 3 was supported.

TABLE 5. Two-way interaction of leader irritation and self-care on staff care.
B SE R2 ΔR2

Leader irritation

−.10** .04 .41*** .02**

Self-care (t2)

.33*** .05
Leader irritation x self care (t2) −.14*** .03
  • Note: N = 582.
  • *** p < .001,
  • ** p < .01, and
  • * p < .05.
Details are in the caption following the image
Interaction between leaders' irritation and leaders' self-care on staff care.

Additional analyses

To investigate whether the interaction effects differ for the sub-facets of leaders' self-care, interactions were also tested for the sub-facets awareness and behavior. Regarding the interaction between leaders' irritation and self-care, there were significant interactions for both sub-facets awareness (B = −.14, SE = .04, p < .01) and behavior (B = −.15, SE = .04, p < .001) in the expected directions. Regarding the interaction between ICT hassles and self-care, there was a significant interaction with behavior in the expected direction (B = .14, SE = .06, p < .05), but no interaction effect for awareness (B = .07, SE = .05, p = .23).

Additionally, we investigated the unique effects of ICT hassles and re-calculated our model controlling for workload. Also in this model, the expected relationship between ICT hassles and leaders' irritation remains stable. ICT hassles are related to both irritation (B = .20, SE = .04, p < .001) and workload (B = .15, SE = .04, p < .001). This underlines the unique effect of ICT hassles as a digitization-specific stressor above and beyond the general workload. Moreover, to test for causal assumptions, we re-calculated our model controlling for autoregressive effects. However, in this case, both the indirect and interactive effects disappear, indicating that leaders' irritation (Mt1 = 2.36, SDt1 = 0.89; Mt2 = 2.41, SDt2 = 0.90), self-care (Mt1 = 3.45, SDt1 = 0.60; Mt2 = 3.45, SDt2 = 0.58), and staff care (Mt1 = 3.54, SDt1 = 0.56; Mt2 = 3.53, SDt2 = 0.56; Mt3 = 3.52, SDt3 = 0.59) remain relatively stable over time.

DISCUSSION

The aim of this study was to examine the relationship between ICT hassles and staff care via leaders' irritation, as well as the influence of self-care on these relationships. In line with previous research, findings suggest that ICT hassles are related to higher leaders' irritation which is in turn related to lower levels of staff care. Moreover, both the relationships between ICT hassles and irritation, as well as between irritation and staff care were stronger for leaders displaying high self-care. In the following, we will discuss these findings in view of their implications for leadership in digital working environments.

First, the negative relationship between ICT hassles and staff care via leaders' irritation further validates and extends previous findings (Day et al., 2012; Klebe et al., 20222023; Pischel et al., 2022). In line with previous research, results show that ICT hassles are positively related to leaders' irritation (Baumeister et al., 2021; Day et al., 2012). Validating previous findings from the followers' perspective with findings from a leaders' perspective is an important extension because particularly leaders already suffer from other stressors such as time and performance pressure. An additional stressor that prevents them from accomplishing their goals and communicating with their team might be even more problematic than for followers, as leaders are responsible not only for their own work but also for the performance of the team and the organization. Therefore, it is alarming that ICT hassles are related to leaders' irritation even two months later.

Furthermore, in line with COR Theory and previous studies (Hobfoll et al., 2018; Klebe et al., 2022; Pischel et al., 2022), leaders' irritation is in turn related to lower levels of staff care. While previous studies were either conducted as experiments with participants only imagining to be a leader or measured follower perceptions, the current study replicates and complements these findings with field data from a leaders' perspective. This finding further validates the assumptions on loss spirals as postulated in COR Theory which assumes that individuals aim to protect their own resources and enter a defensive mode in case of resource loss (Hobfoll et al., 2018). It is interesting to note that ICT hassles are even more strongly related to leaders' irritation than their general workload. This underlines the notion that ICT hassles are a particularly relevant stressor in digital working environments. However, when controlling for workload, the direct relationship between leaders' irritation and staff care disappears so that leaders' irritation does not show a specific effect on staff care. While ICT hassles have a specific effect on leaders' irritation, ICT hassles may also cause additional workload. As leaders' irritation is related to workload, the specific effect of ICT hassles can be superimposed by a more general demanding effect.

Second, the findings support the assumption that the negative relationship between leaders' irritation and staff care is stronger for health-oriented leaders who engage in self-care. This is in line with Self Preservation Theory (Dickerson & Kemeny, 2004) and accounts for both self-care awareness and behavior. As leaders who are aware of their health and engage in health-promoting behavior seem to perceive a goal conflict between maintaining their own health and caring for their followers in stressful periods (Klebe et al., 2022; Pischel et al., 2022), they rather decide to preserve their own health status instead of caring for their followers (de Dreu & van Knippenberg, 2005; Dickerson & Kemeny, 2004). In turn, leaders who engage less in self-care focus on self-preservation to a far lesser extent (Arnold & Rigotti, 2023; Franke et al., 2014; Klug et al., 2022) so that they do not shift their focus and show similar amounts of staff care also when irritation arises. As they also show lower staff care levels from the outset, the discrepancy in staff care levels between low and high irritation is smaller.

Third, we found that the relationship between ICT hassles and irritation was stronger when leaders displayed high self-care. This is an important extension of the literature, as empirical findings regarding the influence of self-care on the relationship between demands and irritation were outstanding. Particularly self-care behavior seems to be crucial for the relationship between ICT hassles and irritation. A possible explanation is that self-caring leaders may experience more cognitive dissonance in situations when they would like to care for themselves (e.g., taking a break), but think that they should solve technical issues first. This result ties in with the findings of Bregenzer and Jimenez (2021), Day et al. (2012), and Mander and Antoni (2023), and supports the reasoning that resources do not necessarily buffer the negative effects of job demands. Instead, the fit between demands and resources is crucial, particularly when it comes to hindrance stressors. These stressors harm personal gain by interrupting regular work processes (J. A. LePine et al., 2005) so that rather support in eliminating the hindrance stressor is needed. It seems plausible that the same rationale may also apply to the interactions of other hindrance stressors and resources so further research on these backfiring effects is needed.

Theoretical implications

The present study highlights the notion that leaders' health and therewith also health-oriented leadership are particularly at risk in digital working contexts, as both the health-protecting effect of leaders' self-care and staff care are threatened by ICT hassles. By investigating health-oriented leadership in the digital working context, this study adds to previous studies in more traditional working contexts, extends the existing validity of the HoL concept, and contributes to the current debates on digital leadership (Bregenzer & Jimenez, 2021; Efimov et al., 2020; Klebe & Felfe, 2023; Tautz et al., 2022) and inconsistent findings regarding the interplay between demands and resources (Bakker & Demerouti, 2007; Bregenzer & Jimenez, 2021; Day et al., 2012; Krick et al., 2022; Mander & Antoni, 2023).

It is a new insight that self-care cannot buffer the negative effects of demands as postulated in the JD-R model, but that negative effects of demands are even stronger for leaders engaging in self-care. The findings show that the negative effects of demands may be particularly strong for those who make use of their resources (Bregenzer & Jimenez, 2021; Mander & Antoni, 2023; Pischel et al., 2022). This can be explained by the fit between demands and resources, which may be of particular importance when it comes to hindrance stressors. While previous research already discussed the differential effects of hindrance and challenge stressors for employee outcomes (Crawford et al., 2010; Schneider et al., 2017; van den Broeck et al., 2010), this research extends this discussion regarding the interaction between hindrance and challenge stressors with resources. As previous studies already found opposite effects for the interaction between challenge and hindrance stressors for the same resource and a similar outcome (Bregenzer & Jimenez, 2021; Krick et al., 2022), it is likely that the fit between demands and resources is particularly important when it comes to hindrance stressors such as hassles (J. A. LePine et al., 2005). Thus, our findings offer a complementary explanation for the interaction between job demands and resources, as resources do not necessarily buffer the negative effects of demands and irritation (Bregenzer & Jimenez, 2021; Day et al., 2012; Häusser et al., 2010; Mander & Antoni, 2023). The same mechanism may also apply to other resources so that further research on differential effects for challenge and hindrance demands regarding their interplay with resources is needed.

Moreover, the findings add to the existing evidence for the HoL model. Previous research investigated health-oriented leadership mostly in traditional office settings and short-term studies (e.g., Horstmann, 2018; Krick et al., 2022; Santa Maria et al., 2019). The current study, in turn, is one of the first to investigate health-oriented leadership in a completely digitalized working environment considering a digitization-specific job demand (i.e., ICT hassles) using three measurement points. Moreover, as previous studies on staff care largely refer to follower perceptions, the study extends the methodology by including the leaders' perspective as recommended in previous research (e.g., Klebe, Felfe, & Klug, 2021). This is particularly important for shedding light on the trickle-down effects of job demands on leader behavior via their health, as followers cannot truly assess the amount of job demands and irritation that leaders experience. Furthermore, research on the influence of self-care on the relationship between leaders' health and staff care was missing. This study is the first to test this interaction and thus adds an additional path to the existing HoL model (Franke et al., 2014).

By investigating trickle-down effects of job demands on staff care via leaders' irritation and by testing the potential backfiring effects of self-care as a resource on the effects of demands and irritation, the findings also add to the further validation of COR Theory and the embedding of the HoL model (Franke et al., 2014; Hobfoll et al., 2018). Revealing that those leaders with more resources experience stronger impairments under demands and irritation (Bregenzer & Jimenez, 2021; Mander & Antoni, 2023; Pischel et al., 2022), the findings support the notion that resource loss is particularly salient and severe. While previous research shows that staff care as an external resource can buffer the negative effects of job demands (Krick et al., 2022), the health-protecting effect of self-care as an internal resource is particularly at risk under high demands. Revealing that ICT demands are negatively associated with the levels of staff care via leaders' irritation, the findings moreover support the assumptions that resources travel in caravans and that individuals enter a defensive mode when resources are lost (Hobfoll et al., 2018).

Limitations and recommendations for future research

The study has some limitations that should be considered when the results are interpreted. First, the current design is limited regarding interpretations of causal relationships. Although the sequence between ICT hassles, leaders' irritation, and staff care is theoretically justified (Klebe et al., 20222023; Pischel et al., 2022), the current design cannot assess causal effects between these factors as our model does not include autoregressive effects. However, from a theoretical perspective, it does not seem plausible that ICT hassles are influenced by leaders' irritation or one's own staff care at a later point in time, or that staff care influences leaders' irritation so that we are confident that we ordered variables in the right sequence.

Second, the dropout rate in the current study may have influenced the results. While 1092 leaders took part in the study at t1, only 582 leaders took part at t3 (dropout 48.83%). Dropout rates of 30 to 70% are often reported in longitudinal studies (Gustavson et al., 2012). To avoid biased results by data imputation or missings, the final sample in the current study only included complete data sets. Nevertheless, results may also be biased with this procedure. While the levels of self-care did not differ between the t1 and the final sample, ICT hassles were a bit higher at t1 (M = 2.29, SD = 0.99) than in the final sample (M = 2.13, SD = 0.93). The t1 sample was also a bit younger and included a few more female leaders (39.8%) than the final sample (37.6%). This indicates that particularly younger leaders, female leaders, and those with high amounts of ICT hassles dropped out. However, while means can seriously be biased by dropouts, according to Gustavson et al. (2012) associations between variables are far more robust.

Third, the appropriate time interval to measure relationships between the variables and their effects is debatable. Particularly the effects of leaders' self-care are worthwhile to investigate in future studies. For example, we hypothesized that self-care at t1 would be associated with leaders' irritation at t2 two months later. In line with our assumption, correlation analyses revealed a relationship between both variables. However, when adding a direct path of self-care at t2 on irritation at t2 in the overall model, the effect from self-care at t1 disappeared which indicates that self-care is more effective for health in the short-term.

As suggested in the HoL model (Franke et al., 2014), we added a direct path from self-care at t1 to staff care at t3 about four months later. The effect of self-care t1 on staff care was nearly as high as from self-care t2 two months later, indicating that self-care has long-term effects on staff care. Therefore, the timespan in which health-oriented leadership can unfold its effects is debatable. To shed light on the specific time interval in which health-oriented leadership unfolds its effects, future studies should test causal relationships between self-care, staff care, and health with different time lags between measurements in a diary or long-term studies.

Finally, results might be affected by common method bias (Podsakoff et al., 2003), as all variables were rated by the same individuals so subjective perceptions might have influenced the results. To counteract this problem, we measured IVs separately from the DVs. Moreover, Harman's single-factor test indicates that results are not affected by common method bias, as a single factor explained only a small amount of variance (26.07%). To investigate how far demands influence leaders' well-being and their behavior toward followers, it is important to capture the leaders' perspective with regard to their own ICT hassles, their own self-care, irritation, and their staff care. Nevertheless, from a practical point of view, it would be also interesting, how followers perceive staff care when leaders are irritated, as follower perceptions are particularly relevant when it comes to their health. Future studies could therefore test this reasoning by complementing leaders' self-reports on demands and irritation with follower perceptions of leadership behavior.

Practical implications

The current findings support the notion that work design is also important in digital contexts (Parker & Grote, 2022), as not only leader health but also health-oriented leadership is particularly at risk when ICT hassles occur while working from home. Dealing with ICT hassles increases leaders' anger and frustration, which leads to an increase in their irritation (Day et al., 2012; Klebe et al., 2023). Moreover, our study reveals that the negative effects of ICT hassles on leaders' irritation cannot be buffered by the internal resource self-care, but that also the health-protecting function of self-care is negatively affected by ICT hassles as the negative effects of ICT hassles are even stronger for those engaging in self-care. Furthermore, our study reveals that not only leaders' health and the protective function of self-care are at risk due to ICT hassles, but also leaders' health-promoting behavior toward followers' is threatened due to an increase in irritation.

These findings underline the importance of properly functioning IT equipment in digital working environments. Regarding the results of the current study, there are at least three ways in which organizations will benefit from well-functioning ICTs: First, leaders' self-care will benefit from a decrease in ICT hassles. According to the current findings, self-care is better able to unfold its health-fostering effect when demands are low and when leaders can take advantage of ICTs in terms of self-care. Second, also leader health will benefit from well-functioning equipment, as low amounts of ICT hassles go along with better health and less irritation (Day et al., 2012; Klebe et al., 2023). The combination of low ICT hassles and high self-care represents the best precondition for leaders' health (i.e., low irritation), as two resources (i.e., well-functioning ICTs and self-care) amplify the effects of each other. Third, also staff care will benefit from low ICT hassles. While leaders' irritation leads to a decrease in staff care (Klebe et al., 2022; Pischel et al., 2022), and as ICT hassles are an important antecedent of irritation (Day et al., 2012; Klebe et al., 2023), properly functioning IT equipment indirectly fosters staff care.

The combination of low ICT hassles and high self-care leads to the best leader well-being (i.e., low irritation), while the combination of low irritation and high self-care leads to the highest amount of staff care. Therefore, it is the responsibility of organizations to provide leaders with properly functioning IT equipment to foster leader health, to save the health-protecting function of self-care, as well as not to risk staff care and therewith also follower health. It is especially important that leaders have access to technical support also while working from home so that irritation remains low even when hassles occur.

CONCLUSION

This study is the first to investigate antecedents of health-oriented leadership in the digital context from the leaders' perspective and thus contributes to a deeper understanding of the implications of digitization for leadership. We identified a negative relationship between ICT hassles and staff care via leaders' irritation. Moreover, results revealed that both the positive effects of ICT hassles on irritation, as well as the negative effect of irritation on staff care are particularly strong for leaders engaging in self-care, as leaders who start higher in health and staff care run the risk of falling deeper. To protect leaders' health, self-care as an internal resource for leaders as well as staff care as an external resource for followers, organizations should provide leaders with reliable IT equipment and technical support also while working from home.

CONFLICT OF INTEREST STATEMENT

The authors do not declare any conflict of interest.

ACKNOWLEDGEMENTS

Open Access funding enabled and organized by Projekt DEAL.

    ETHICS STATEMENT

    All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki.

    DATA AVAILABILITY STATEMENT

    The data that support the findings of this study are available from the corresponding author upon reasonable request.