Burnout Syndrome and Emotional Labor in Leaders and Subordinates: A Dyad Analysis

Abstract

Burnout Syndrome is considered a chronic response to occupational stressors in the work environment. Social interactions constitute one of the stressors at work that can generate negative feelings that trigger a process of contagion of the syndrome among workers in interdependent relationships. This study aimed to analyze whether emotional labor (emotional demands, emotional dissonance) at the level of the leader and subordinate dyad contributes to the manifestation of Burnout Syndrome. The participants included 244 leader–subordinate dyads who answered a questionnaire with sociodemographic and labor data, the Spanish Burnout Inventory, a subscale of the Questionnaire on the Experience and Assessment of Work, and a subscale of the Frankfurt Emotion Work Scale. Analyses were performed using the actor–partner interdependence model (APIM) through path analysis. The results indicate that the emotional demands of the leaders and the emotional dissonance of the subordinates predict the leader’s Burnout Syndrome. The Burnout Syndrome of subordinates was predicted only by the emotional demands of subordinates. Organizational actions are necessary for the better functioning of this dyad, aiming to mitigate the negative consequences of emotional labor on workers’ mental health.

1. Introduction

Leadership is a process of social influence characterized by a type of social interaction marked by the ability of a person to guide the goals, desires, and affections of other people who are linked in some way. In a work context, the quality of the relationship between leader-follower and leader, as well as work teams and involving multi-team networks and systems, occupies a central position in the outcome of numerous organizational processes, such as the performance of the leader, the effectiveness of teams, and the group’s affective climate (Yammarino 2013). The theory of leader–member exchange (LMX) is a way of looking at leadership that focuses on how well a leader interacts with a group of followers, rather than the leader’s personal traits or the traits of their followers, which are more traditional ways of looking at leadership (Dansereau et al. 1975).

From an LMX perspective, leadership is understood as a two-way relationship between a leader and a follower (Graen and Uhl-Bien 1995). Leaders and followers engage in diverse and continuous social exchanges that are qualitatively distinct, generating positive or negative effects for the dyad itself, the work team, and also for global organizational results (Dansereau et al. 1975).

Following this same line of understanding, the theory of social interdependence highlights that the way in which relationships of (in)dependence are structured between a group of individuals causes impacts on a large proportion of the team’s results (Watson and Johnson 1972). Strong emotional ties between leaders and followers can enhance emotional contagion (emotional contamination that is not always conscious) between individuals, resulting in behavioral synchrony and emotional stability or instability within the group and between leaders and followers (Herrando and Constantinides 2021). By analogy, it can be stated that the mechanisms involved in the process of emotional contagion between leaders and followers are similar to the processes of contagion of burnout among workers (Bakker et al. 2001; Meredith et al. 2020), which are applicable to the leader and subordinate dyad. In summary, the leader’s emotional state influences the group’s affective tone, improving or worsening the affective states of the individuals involved in the interaction (Campos et al. 2021; Chi and Ho 2014).

2. Literature Review

2.1. Burnout Syndrome

Burnout Syndrome (BS) has been defined as a chronic response to occupational stress mediated by social relationships established in the work environment (Gil-Monte 2005; Guidetti et al. 2018). Considered a consequence of chronic work stress, the syndrome results from a subjective experience of a negative nature in relation to work, composed of cognitions, emotions, and attitudes (Gil-Monte 2011).

Owing to the seriousness of its consequences, in 2019, the World Health Organization (WHO) classified BS as a work-related illness to take effect in 2022, with inclusion in the International Classification of Disease-11 (ICD-11) as a phenomenon exclusive to the occupational context (Edú-Valsania et al. 2022; World Health Organization 2019) and not as an occupational disease (Aron 2023; Shoman et al. 2021). Therefore, BS was characterized as a psychosocial phenomenon resulting from the interaction between personal characteristics and the work context (Guerrero-Barona et al. 2020).

Workers affected by the syndrome have a lower quality of life, increased risk of sleep disorders, cognitive impairment, cardiovascular diseases (Mäkikangas et al. 2021), difficulty in decision-making, reduced coping capacity, anxiety, depression, dissatisfaction with work life, low self-esteem, irritability and increased consumption of alcohol and tobacco (Salvagioni et al. 2017). Organizational consequences include professional dissatisfaction, lower performance and efficiency at work, an increase in counterproductive behaviors, a reduction in the quality of services provided, an increase in interpersonal conflicts, and significant economic losses as a consequence of absenteeism (Edú-Valsania et al. 2022), long-term absences, presenteeism and increase in early disability, and greater turnover intention (Salvagioni et al. 2017).

The model of Gil-Monte (2005) consists of four dimensions: enthusiasm toward the job, psychological exhaustion, indolence, and guilt. The first dimension, inversely evaluated to characterize the syndrome, comprises the behaviors performed by the worker toward his or her goals at work, in which the worker feels enthusiastic and involved with work activities, evaluating them as rewarding and as an important instance of personal and professional satisfaction. Psychological exhaustion is defined by the occurrence of emotional and physical exhaustion caused by interpersonal relationships in the work context, where the worker relates daily with people who have or cause problems. Indolence is marked by negative attitudes in the relationship with people, where the worker relates to the people he or she needs to relate to at work in an impersonal manner. Guilt is defined as a social emotion arising from the attitudes and behaviors present in the Indolence dimension; the worker evaluates his/her behavior as if he/she were infringing some kind of ethical code or norm derived from the prescription of his/her professional role.

By looking at the different aspects of BS, the theoretical model lets us figure out how common two different profiles are: Profile 1 (Global BS) and Profile 2 (Gil-Monte 2011; Llorca-Pellicer et al. 2021). Profile 1 is characterized by a set of feelings and behaviors linked to work stress that impact worker well-being. This discomfort does not impair professional practice; however, it negatively affects the productivity and quality of their work. Profile 2 presents the same characteristics as Profile 1, plus the guilt dimension. Professionals affected by Profile 2 present serious problems in the execution of their work, including greater cognitive, emotional, and behavioral deterioration and increased health problems (Gil-Monte 2005, 2011).

The combination of high demands and low work resources offers an important explanation for BS (Bakker et al. 2023; Lesener et al. 2019). Prolonged exposure to high work demands triggers chronic exhaustion and psychological detachment, which play an important role in the development of BS (Bakker and de Vries 2021). The same work that they once found exciting makes burnt-out individuals feel exploited and exhausted (Bakker and de Vries 2021). Work demands are aspects arising from the profession and the organization that require significant physical, cognitive, and emotional effort from workers (Bakker et al. 2023).

2.2. Emotional Demands

Emotional demands are strongly present in professional service provision activities that require social interactions, and they have effects on various aspects of the worker’s mental health. Therefore, emotions are a central feature of contemporary work experiences (Torrence and Connelly 2019).

Emotional labor, with emotional demands as one of its components, has been considered a new work stressor in modern society that can lead to BS (Costakis et al. 2021; Llorca-Pellicer et al. 2021; Yang and Chen 2021). Depending on the job, the tasks may require different efforts on the part of the worker, being greater when it becomes necessary to express frequent, intense, varied emotions for prolonged periods of time. One of the negative consequences related to this type of work is BS (Grandey et al. 2015). According to a study by Bakker et al. (2023), workers reported burnout symptoms in proportion to emotional demands. Therefore, high emotional demands increase mental risks for workers by depleting their personal resources and impairing their adaptive psychic functioning (Bakker et al. 2023).

Emotional labor is a psychological process characterized by the worker’s efforts to regulate their emotions to meet the emotional demands at work in order to manifest the emotions prescribed for the work context, even if they are contrary to the emotion felt (emotional dissonance) (Clarke et al. 2024; Edú-Valsania et al. 2022; Grandey and Melloy 2017; Hochschild 2013). Although Hochschild (1979) defined it as a phenomenon with three dimensions—emotional demands, regulatory strategies adopted by the worker to deal with such demands, and emotional performance (the quality of the emotional pattern displayed)—the literature in general addresses the first two dimensions in a broader way. Emotional demands vary in terms of intensity and form of emotion displayed and, therefore, they mobilize different regulatory strategies (Blanke et al. 2022; Diefendorff and Richards 2003).

Emotional demands are identified as work characteristics that require continuous emotional effort in interpersonal relationships (de Jonge and Dormann 2003). Emotional dissonance is characterized as a discrepancy between the emotions felt and expressed by the worker (Van Dijk and Brown 2006). Although it is recognized that tolerance toward emotional dissonance may present individual differences, it has been associated with negative consequences for health (Andela et al. 2015; Clarke et al. 2024), thereby contributing to the emergence of BS (Grandey et al. 2015; Vignoli et al. 2021).

2.3. Relations between the Constructs

Key aspects of social interactions in the work environment and potential generators of emotional demands are those that occur between leaders and subordinates. Studies on the phenomenon of leadership focused on intra-group emotions, especially those arising from the leader’s emotional contagion, which contributes to the formation of the group’s affective tone (Ashkanasy and Dorris 2017). Therefore, the relationships between leaders and subordinates are important to characterize emotional labor in work contexts. Leadership can be understood as a process of significant social influence, in which the leader sensitizes the behaviors, perceptions, and feelings of their subordinates (Kafetsios et al. 2013).

BS, for example, can emerge in the group as a result of emotional contagion. A leader may begin to feel emotionally exhausted and incapable and seek to distance himself emotionally and physically from those he leads, causing the group as a whole to experience the same feelings. Emotional contagion is characterized by a non-conscious tendency to synchronize personal affective and emotional states with those of other people (Chullen 2014; Doherty et al. 1995; Hatfield et al. 1994). In this way, key social interactions, such as those that occur between leaders and subordinates, can trigger emotional contagion between individuals, resulting in behavioral synchrony (Elfenbein 2023; Herrando and Constantinides 2021).

Leaders are expected to adopt an adequate emotional standard (Ashkanasy and Humphrey 2011; Wijewardena et al. 2017), helping their subordinates to face the various emotional manifestations required for optimal professional performance (Brotheridge and Grandey 2002; Brotheridge and Lee 2003; Eberly and Fong 2013). Leaders also contribute to supporting the healthy emotional regulation of their subordinates (Wijewardena et al. 2017) and lead to better their subordinate’s health and safety-related outcomes within workplaces (Mullen et al. 2024).

Given the effects that a leader’s emotions may have on the quality of their relationship with their subordinates (Campos et al. 2021; Wang et al. 2024) and how this may increase the risk of the onset of BS (Harms et al. 2016), it is important to understand how those under their leadership interpret a leader’s emotions.

According to a study by Ten Brummelhuis et al. (2014), BS in a leader was associated with decreased leadership support behavior and increased BS in subordinates. There is evidence of an association between the BS of leaders and the somatic complaints of their subordinates (Köppe et al. 2018). Leadership can help prevent employees’ BS and improve their engagement and satisfaction, and such prevention and improvement can lead to safety outcomes such as reduced accidents, injuries, adverse events, and unsafe behaviors (Jiang et al. 2024).

Therefore, several studies focused on investigating the dyadic relationship between leaders and subordinates (Richard et al. 2022; Scott and Zweig 2021; Wang et al. 2019). The systematic review, carried out by Kim et al. (2020), identified 62 articles that used dyadic analysis between leaders and subordinates, with only one evaluating BS.

2.4. Objective and Hypothesis

Based on the above, the present study aimed to explore whether the emotional demands and emotional dissonance of leaders and subordinates affect the occurrence of BS at the level of the leader–subordinate dyad. The study’s hypotheses were (Figure 1):

H1. 

The leader’s emotional demands are positively associated with subordinates’ BS.

H2. 

The leader’s emotional dissonance is positively associated with subordinates’ BS.

H3. 

Subordinates’ emotional demands are positively associated with BS in leaders.

H4. 

Subordinates’ emotional dissonance is positively associated with BS in leaders.

H5. 

There is a dyadic interaction between the two key aspects of emotional labor (demands and dissonance) and BS in leaders and subordinates.

3. Materials and Methods

3.1. Participants

A cross-sectional study using an explanatory observational design looked at 53 leaders and 244 subordinates from nine private companies. The sample was not based on chance and was made up of dependent dyads (Kim et al. 2020) with dyad control. The sample size followed the guidelines of Hair et al. (2002), in which each variable must cover 50 participants. In the composition of the dyads for analysis purposes, a maximum of 10 subordinates were invited for each leader. Therefore, each leader could form up to 10 dyads. The total sample included 244 dyads with a relationship of at least six months, which would allow us to infer that there was some level of proximity for the emergence of the studied phenomenon.

The average weekly workload was 43.58 h (SD = 3.50; min = 30, max = 60). With regard to teams, the average number of people leading was 45 (SD = 87; min = 2, max = 438), and there were 715 workers in the organization (SD = 755; min = 3, max = 3000). In terms of the hierarchical level, most leaders (62.3%, n = 33) belonged to the technical–operational leadership level (coordinator, supervisor, team leader), 22.6% (n = 12) were intermediate (managerial levels: area, sectors, regions), and 13.2% (n = 7) belonged to the strategic leadership group (director, superintendent, president).

In the group of leaders, the average age was 34.69 years old (SD = 10.648; min = 18, max = 62), the total working time was 16.11 years (SD = 11.00; min = 1, max = 56), and the time in the current job was 5.20 years (SD = 5.80; min = 1, max = 37). As for the weekly workload, the average was 43.92 h (SD = 5.80; min = 20, max = 72).

The work teams consisted of an average of 26 people (SD = 51; min = 3, max = 400), and the average number of workers that made up the organization of participants was 736 (SD = 855; min = 3, max = 3.200). The positions held by those who were led varied markedly, from operational professionals (auxiliaries, assistants, bricklayers, mechanics, general services, operators) to tactical professionals (analysts and technicians) to strategic professionals (coordinators and managers).

Table 1. Sociodemographic characteristics of the sample.

Characteristics		Leaders		Subordinates
		n	%	n	%
Gender	Male	37	69.8	143	58.6
	Female	16	30.2	101	41.4
Marital status	With partner	44	83.0	172	70.5
	No partner	9	17.0	67	27.5
Children	Yes	41	77.4	124	50.8
	No	10	18.9	118	48.4
Educational level	Elementary School	2	3.8	50	20.5
	High school	5	9.4	67	27.5
	University education	22	41.5	80	32.8
	Postgraduate	24	45.2	43	17.6
Other work activity	Yes	6	11.3	29	11.9
	No	47	88.7	210	86.1
Company type	Public	2	3.8	3	1.2
	Private	51	96.2	236	96.7
Segment	Industry	29	54.7	73	29.9
	Services	17	32.1	125	51.2
	Business	6	11.3	35	14.5
Work relationship	Effective	49	92.5	227	93.0
	Statutory/Contestant	0	0	1	0.4
	Self Employed/Legal Entity	2	3.8	1	0.4
Pandemic work format	Virtual/Home Office	5	9.4	24	9.8
	Onsite—in the company	35	66.0	176	72.1
	Hybrid	13	24.5	41	16.8

presents other sample characteristics expressed in frequency and percentage.

It can be seen that the majority of leaders and subordinates were male, had partners and children, did not have other work activities, worked in private companies, had permanent employment contracts, and worked onsite in the company during the pandemic period. Participants differed in terms of educational level, as the majority of leaders had a postgraduate degree while subordinates had a university education.

3.2. Instruments

Data collection was conducted using three instruments:

(1)

Questionnaire on sociodemographic data (gender, age, marital status, children, and education) and work-related information (total working time, time in the current job, weekly workload, other work activities company modality, company segment, type of employment bond, hierarchical level, number of employees in the organization, work during the pandemic); for leaders (length of time as leader, number of people led) and for subordinates (duration under this leadership, number of people in the work team).

(2)

The Spanish Burnout Inventory (SBI) by Gil-Monte (2005, 2011), adapted for use in Brazil by Gil-Monte et al. (2010). The instrument consists of 20 items distributed in four subscales. For this study, global BS was used, which is the average of the 15 items from the Enthusiasm toward job (inverted) subscales, psychological exhaustion, and indolence. Enthusiasm toward job (five items, α = 0.72, in this study α = 0.84 and ω = 0.86; e.g., I feel enthusiastic about my work); Psychological exhaustion (four items, α = 0.86, in this study α = 0.84 and ω = 0.87; e.g., I feel physically tired at work); Indolence (six items, α = 0.75, in this study α = 0.67 and ω = 0.69; e.g., I am sarcastic with some people). The items were rated on a four-point scale, with (0) representing never and (4) representing every day;

(3)

Emotional labor: Assessed using two adapted subscales for the present study. The first was (3.1) Emotional Demand—Questionnaire on the Experience and Assessment of Work (QEEW) by Van Veldhoven et al. (2002), a subscale consisting of seven items (α = 0.71, in this study α = 0.79 and ω = 0.80; e.g., Does your job require a lot of emotional effort?). The items were rated on a four-point scale, with (1) representing never and (4) representing always. The second subscale was (3.2) Emotional Dissonance—Frankfurt Emotion Work Scale (FEWS) by Zapf et al. (1999), consisting of five items (α = 0.79, in this study α = 0.83 and ω = 0.84; e.g., How often do you suppress your true feelings during work (e.g., frustration)?). The items were rated on a five-point scale, with (1) representing never and (5) representing very often.

3.3. Data Collection Instruments and Procedures

The instruments were applied in person by the first author in nine private companies located in Brazil, in the state of Rio Grande do Sul. At the time of collection, the objectives of the study were explained, and an Informed Consent Form was presented for signature. Subsequently, the form was handed out for completion by the participants. The subordinates were instructed to use their direct leadership as a reference for their answers. The leader–subordinate dyads were paired, but neither party had access to the other’s responses, thus avoiding contamination of individual responses. The data collection period was from March to June 2022. The research was approved by the Research Ethics Committee of the Brazilian Ministry of Health.

3.4. Data Analysis Procedures

Data analysis was performed using the Actor–Partner Interdependence Model (APIM) (Cook and Kenny 2005), which is recommended when it is assumed that there is a direct or indirect effect between the variables of the individuals that make up the dyad. This test model has become the standard method for dyadic data analysis, in which both pair members respond to the same measures and maintain a bond of interdependence (Kenny 2018) characterized by a common goal, mutual support, and sensitivity to needs and emotions among themselves (Krasikova and LeBreton 2012).

To perform the APIM, path analysis was used with the Maximum Likelihood with Robust standard errors (MLR) estimator. This type of model is just-identified, so the degrees of freedom of the model were equal to 0. The analysis was performed in the R 4.2.1 and RStudio 2022.07.2576 environment using the lavaan package 0.6–12 (Rosseel 2012).

4. Results

Leader BS was significantly predicted by the leader’s emotional demands (B = 2.01 [95% CI 1.34–2.68], p < 0.001) and the subordinate’s emotional dissonance (B = 0.33 [95% CI 0.04–0.62], p = 0.026). The other variables were not significant in predicting leader BS: leader’s emotional dissonance (B = 0.29 [95% CI −0.15–0.74], p = 0.194) and subordinate’s emotional demands (B = −0.20 [95% CI −0.68–0.27], p = 0.395).

The only significant predictor of BS among subordinates was their emotional demand (B = 1.51 [95% CI 0.85–2.16], p < 0.001). The other variables were not significant in predicting the BS of subordinates: emotional dissonance of subordinates (B = 0.38 [95% CI −0.02–0.79], p = 0.062), emotional demand of the leader (B = 0.10 [95% CI −0.82–1.03], p = 0.828), and leader’s emotional dissonance (B = 0.10 [95% CI −0.51–0.72], p = 0.735). Ultimately, the correlation between leader BS and subordinate BS was not significant (p = 0.240). Together, the model variables were able to explain 28% of the variance in the leaders’ BS level and 26% of the subordinates’ BS level. The tested model and its results are shown in Figure 2.

5. Discussion

The present study aimed to explore whether the emotional demands and emotional dissonance of leaders and subordinates have direct effects on the emergence of BS at the level of the leader–subordinate dyad.

The findings support the formulated hypotheses (Figure 2).

With regard to H1 and H2, the leader’s emotional demands are positively associated with subordinates’ BS, and the leader’s emotional dissonance is positively associated with subordinates’ BS. This result is consistent with the literature (Ashkanasy and Dorris 2017), which reveals the impact of the leader’s emotions as a risk factor for BS (Humphrey 2023). The same occurred with Hypotheses 3 and 4, which state that subordinates’ emotional demands are positively associated with BS in leaders, and subordinates’ emotional dissonance is positively associated with BS in leaders. In this way, H5 is more closely related to the objective of the study, which states that there is a dyadic interaction between the two key aspects of emotional labor (demands and dissonance) and BS in leaders and subordinates.

The results revealed that the leaders’ BS was significantly and positively predicted by the leaders’ emotional demands and the subordinates’ emotional dissonance. Thus, the greater the emotional demands experienced by the leader and the emotional dissonance of the subordinates, the higher the levels of BS. The results support the hypothesis formulated in the study that there is a dyadic cross-level interaction between the dimensions of emotional labor and BS in leaders and subordinates, which is consistent with evidence in the literature (Llorca-Pellicer et al. 2021; Yang and Chen 2021).

The effects of emotional demands on the BS in both leaders and subordinates can be explained by the characteristics of both positions, which demand frequent interactions and are inevitably guided, among other aspects, by the emotions transmitted and the need for constant regulation. Both leaders and subordinates reciprocally build their relationships. The leader has great potential to emotionally affect the subordinates and is taken as an emotional model reference. When the leader does not properly manage his/her affective states and the emotional demands to which he/she is subject (leaders’ emotional demands), including those coming from those being led, the effects can be felt by both members of the dyad (Moin et al. 2021).

The results indicate that the emotional dissonance of the subordinates has a direct effect on the increase of BS in leaders. The assumption is that the inconsistency between the emotional expression of the subordinate and his/her internal feelings seems to be perceived by the leader, causing him/her to mobilize more of his/her personal resources to influence the goals and objectives of this subordinate. A meta-analysis study performed by Kenworthy et al. (2014) on the relationship between BS and emotional dissonance involving 16,165 employees concluded that there was a significant relationship between the two variables, indicating that showing an emotion different from that felt at work leads to emotional exhaustion, which is a dimension of BS.

It can also be assumed that showing emotions that are not felt to meet work demands (dissonance) can lead the subordinate member to use coping strategies to overcome this emotional exhaustion, generating negative reactions toward the leader and consequences for the quality of performance at work (Jeung et al. 2018). Work environments where emotional display rules are strongly monitored may result in an increase in the harmful effects of emotional labor, including BS (Jeung et al. 2018; Mesmer-Magnus et al. 2012). Although it is recognized that there are controversial positions on the benefits and harms of emotional labor (Hülsheger and Schewe 2011; Humphrey et al. 2015), professional identification can act as a protective factor by increasing the use of functional strategies by workers to deal with challenging emotional demands (Bakker and de Vries 2021).

This can be applied to every affective bond established with work. The leader is often the representative of the rules of emotional demands in the organizational context, and the quality of the leader–subordinate bond can be affected in the sense of increasing or minimizing emotional labor (Yu and Zellmer-Bruhn 2018). As leaders and subordinates interact over time, their individual behaviors become entangled, making room for new patterns of behavior to emerge as a collective phenomenon (Beenen et al. 2022).

A study by Gatti et al. (2019), also with dyads, showed that independence in the work of subordinates depends on the willingness of their leaders to grant it. The authors found a positive association between the leader’s independence and the BS of the subordinates. The only significant predictor for the BS of those observed in this study was their own emotional demands. The approach and performance goals of leaders are related to employee BS (Sijbom et al. 2019), and work demands and pressure from superiors can increase the chance of developing BS among subordinates (Laeeque et al. 2018).

Understanding the impact of BS on leaders and how it affects the quality of management is important, given its effect on the health of subordinates (Parent-Lamarche and Biron 2022). The authors found evidence that the syndrome was associated with the psychosocial safety climate.

Together, the model variables were able to explain 28% of the variance of BS in leaders and 26% in subordinates. This result indicates a high effect magnitude according to Marôco’s (2007) parameters, especially in the case of a study that used only emotional variables. This study has important implications regarding emotional labor because it involves not only leadership, as recommended in a meta-analysis by Yang and Chen (2021), but also the dyad (Kim et al. 2020), making it clear that leader–subordinate bonds generate cross-effects that need to be carefully observed when designing leader development training.

6. Strengths, Limitations, and Implications for Practice

Three strengths of the study can be highlighted. The first was conducting a study of leader–subordinate dyads, as they are key aspects of organizational behavior that have been relatively underexplored (Kim et al. 2020). The second strength was the ability to pair leader–subordinate dyads from different organizations, allowing for cross-effect analysis. This larger number of dyads enabled a more comprehensive analysis. The third strength was the high effect magnitude result obtained, suggesting that the results are likely to be applicable to the target population.

The study presents two limitations that must be considered when reading its results. The first limitation to be highlighted is the use of self-report measures, especially to measure such a complex, dynamic, and sensitive phenomenon as emotions at work, which can lead to social desirability bias. The second limitation refers to the non-random sample of participants linked mostly to private sector organizations. The third limitation is the sample size, which did not allow for cross-validation to be carried out.

It is suggested that new studies include other variables related to emotional aspects, such as emotional regulation, coping strategies, and levels of expression of emotions (superficial and deep) in order to increase the explanatory power of their possible effects on BS and the leader–subordinate dyad. Longitudinal and qualitative research on how the dyad perceives the quality of their relationship regarding the expression of their emotions proves equally important.

The study results provide a conceptual framework that confirms and highlights the importance of emotional contagion between leaders and subordinates for BS. This can provide greater support for institutional managers to develop preventive interventions.

As implications for practice, the results can support the development of human resources development policies, especially for leaders, in order to make them more capable of exerting a positive influence on their subordinates and, thus, collaborating to ensure mental health and quality of life at work.

It is evident that the importance of developing healthier strategies for emotional regulation and coping with occupational stressors, which are increasingly becoming part of everyday work and relationships between leaders and subordinates. Autonomy in the work environment can alleviate the negative results of the frequent use of emotional labor strategies, and the leader can contribute to increasing the autonomy of the subordinate at work.

Finally, psychosocial interventions in emotion regulation based on emotional intelligence for leaders and subordinates can enable them to better deal with emotional labor. Such interventions can also help leaders become more aware of the type of leadership they exercise and the effects of their emotional expressions on subordinates.