Is public healthcare healthy? The role of emotional exhaustion

ACCEPTED VERSION OF THE PAPER Is public healthcare healthy? The role of emotional exhaustion López-Cabarcos, Mª Ángeles, López-Carballeira, Analía, Ferro-Soto, Carlos Cite: López-Cabarcos, M.Á., López-Carballeira, A. and Ferro-Soto, C. (2021), "Is public healthcare healthy? The role of emotional exhaustion", Baltic Journal of Management, Vol. 16 No. 3, pp. 386-406. https://doi.org/10.1108/BJM-04-2020-0136 Baltic Journal of Management Is public healthcare healthy? The role of emotional exhaustion Journal: Baltic Journal of Management Manuscript ID BJM-04-2020-0136.R3 Manuscript Type: Original Article Keywords: Human resource management, Job demands, Job resources, Public sector, Wellbeing, Stress http://mc.manuscriptcentral.com/bjom Baltic Journal of Management Baltic Journal of Management 1 Is public healthcare healthy? The role of emotional exhaustion Abstract Purpose: This study analyzes the mediating effect of emotional exhaustion between certain job demands (workload, role conflict, and influence at work) and employees’ work attitudes (affective commitment and turnover intention) in public healthcare. Furthermore, it analyzes the moderating effect of possibilities for development and the degree of freedom at work between the above-mentioned job demands and emotional exhaustion. Methodology: A total of 512 healthcare professionals participated in the study. The data were analyzed using structural equation modeling and a hierarchical multiple regression analysis. Findings: The results show that emotional exhaustion fully mediates the relationship between job demands (workload and role conflict) and work attitudes (affective commitment and turnover intention). Moreover, the possibilities for development and degree of freedom at work moderate the relationship between role conflict and emotional exhaustion. Practical implications: Strategies should be designed to prevent employees from becoming emotionally exhausted and lead them to feel more motivated, which results in a more effective public healthcare service. Originality: This study stresses the importance of analyzing the role of emotional exhaustion in the public healthcare context. It demonstrates the mediating role of this variable between several antecedents and consequences, and it analyzes whether other relevant variables can moderate the negative effects of emotional exhaustion. Keywords: Emotional exhaustion; Job demands; Work attitudes; Public sector; Healthcare Page 1 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 2 Article classification: Research paper Introduction Healthcare professionals are susceptible to feelings associated with emotional exhaustion, which refers to the loss of emotional and physical resources caused by interpersonal demands (Maslach and Leiter, 2016). Intense interpersonal contact with service users and their families and the need to provide a quality health service without faults or errors may trigger a loss of physical and/or emotional energy (Maslach and Leiter, 2016). The huge pressure resulting from a constantly insufficient patient/professional ratio may explain the development of emotional exhaustion in the Spanish public healthcare system. The doctor/patient ratio in Europe is 3.5 and nurse/patient ratio is 8.2 per 1,000 inhabitants, while in Spain the figures are 3.8 and 5.2, respectively (Ministry of Health, 2019). Prior research confirms that job demands entailing pressures such as role conflict (Garrosa et al., 2010) or excessive workload (Gillet et al., 2020) may be relevant in the development of emotional exhaustion among healthcare professionals. Influence at work may also be a relevant job demand, but the relationship is not so clear. In parallel, the conservation of resources theory (Hobfoll and Freedy, 1993) suggests that emotionally exhausted employees do not have sufficient energy to do their work correctly, which obliges them to formulate strategies to avoid losing more resources. Accordingly, high job demands may lead to emotional exhaustion and cause negative effects, such as less organizational commitment or a greater desire to leave the organization (Van Bogaert et al., 2014; Wright and Hobfoll, 2004). Both the vocational and public service nature of healthcare professions reveal why affective commitment and turnover intention are relevant outcomes. The mediating relationships between these variables have not been studied scarcely in human resources literature. Therefore, the first aim of this study is to analyze whether the relationships between job demands and employees’ work attitudes can be Page 2 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 3 fully explained by the effect of emotional exhaustion, which in turn may have severely negative consequences for employees, the organization, and society in general. The job demands-resources theory also suggests that job resources may moderate the relationship between job demands and emotional exhaustion (Bakker and Demerouti, 2007). In the context of public healthcare, examples include possibilities for development and the degree of freedom at work. Therefore, the second aim of this study is to analyze whether the possibilities for development as a possible motivating strategy and/or the capacity to influence a high working pace can help prevent or mitigate the effects of emotional exhaustion. The scarce previous research on the moderating role of possibilities for development has shown mixed results (Bakker et al., 2004; Hu et al., 2011) and does not confirm the moderating role of degree of freedom at work (Day et al., 2009; Peters et al., 2016). Using a study sample comprising Spanish public healthcare professionals and applying structural equation modeling and hierarchical multiple regression analysis, this research aims to provide important contributions for research and practice in the healthcare context. That is, it expands knowledge about employees’ well-being. To date, the scarce previous studies analyzing the mediating role of emotional exhaustion between job demands and employees’ work attitudes report mixed results (Alola et al., 2020; Hang-Yue et al., 2005; Hur et al., 2015; Kemp et al., 2013). Specifically, emotional exhaustion may lead employees to establish lower emotional bonds and a greater desire to leave the organization, which can negatively influence the service quality, level of responsibility assumed, prospects for promotion, variety of tasks undertaken, or choice of more appropriate working methods. All of this can lead to errors with serious consequences for employees, the organization, and society. Consequently, the moderating role of possibilities for development or influence at work that prevent or mitigate the negative effects of employees’ emotional exhaustion is extremely important. In this sense, the paper concludes that human resource managers should urgently design new strategies to Commented [RK1]: The sentence is not clear Commented [LCA2R2]: According to Editor’s suggestion, this sentence has been rewritten to make it more readable for the readers. Commented [RK3R2]: The second sentence still needs editing. Commented [LCA4R2]: Following the Editor’s recommendation, the second sentence has been reviewed and rewritten to make it more readable for the readers. Page 3 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 4 prevent employees from becoming emotionally exhausted to ensure that they are more motivated, ultimately resulting in a more effective public healthcare service. Literature review Emotional exhaustion is the dimension that best characterizes burnout, understood as a psychological syndrome that emerges as a prolonged response to chronic interpersonal stressors on the job. Emotional exhaustion is associated with feelings related to weariness, loss of energy, exhaustion, and fatigue caused by interpersonal demands (Maslach and Leiter, 2016). Shirom (1989) suggests that depersonalization and low personal accomplishment are incidental or unnecessary because they do not include the interpersonal aspects of burnout. In this sense, several authors suggest that emotional exhaustion displays a stronger relationship with certain outcome variables, such as organizational commitment (Wright and Hobfoll, 2004) or turnover intention (Van Bogaert et al., 2014), than other components of burnout. Because of its public nature, the healthcare profession can be considered complex, leading professionals to suffer emotional exhaustion. Indeed, healthcare professionals must develop a patient-centered care approach because each patient shows a combination of special characteristics such as illness, pain, uncertainty, or fear that make him or her a unique challenge (Keeling et al., 2018). Healthcare professionals have to offer service quality without faults or errors as much as possible, often with limited resources (Thanacoody et al., 2014), knowing that patients expect higher service quality (Paillé, 2013). Additionally, the intense, direct contact between professionals and patients (Schaufeli, 2007) and the coexistence of high rates of mortality (Rizo-Baeza et al., 2018) may cause professionals to develop negative feelings toward their work (Maslach and Leiter, 2016). Moreover, the insufficient number of professionals to satisfy patients’ demands (Thanacoody et al., 2014) drives healthcare professionals to endure long working hours, spend limited time with each patient, and suffer Page 4 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 5 sleep disorders due to shift work, bear enormous professional responsibility, or experience feelings of intense failure (Tziner et al., 2015). Consequently, the inherent features of healthcare activities may lead employees to emotional exhaustion. In this regard, the Spanish public healthcare system is based on the principle of universal coverage that guarantees access to a wide range of services (Esteban‐Sepúlveda et al., 2019). However, it has had to face significant challenges, such as the aging population, growing levels of disability and chronic diseases, and introduction of new medical technological developments, which increase healthcare costs (Black and Gruen, 2005). Consequently, the Spanish government has had to invest substantially in healthcare activities, improve internal operations, and increase operational control (Nelson and Sen, 2014). In this context, Spanish healthcare professionals have worked subject to overload, decreased time for patient care, increased working hours, temporary employment, intense pressure, and all of this within organizational structures that are resistant to change (Garcia‐Garcia et al., 2011; Ruiz-Fernandez et al., 2020); these factors generate a loss of energy and emotional exhaustion. The job demands-resources theory suggests that high job demands may spark a psychological process that leads to emotional exhaustion with serious consequences for the individual and the organization (Bakker and Demerouti, 2007). Employees subjected to excessive job demands make a greater effort to meet their goals at work and do not have sufficient energy for their tasks, being unable to adapt to the working environment, leading them to emotional exhaustion (Bakker and Demerouti, 2007). Prior research indicates that job demands, such as role conflict or workload, may be strong predictors of the development of emotional exhaustion in the public healthcare sector (Schaufeli, 2007). Workload, stemming from a high volume of work combined with limited time to complete it (Kahn et al., 1964), may lead to high demands on employees, which can result in fast-paced Page 5 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 6 work or the accumulation of tasks (Kozak et al., 2013). Chronic work overload makes it difficult for individuals to rest or recover and causes a loss of energy (Maslach and Leiter, 2016). Prior research indicates that workload may be a strong predictor of emotional exhaustion in healthcare (Garrosa et al., 2010; Gillet et al., 2020). On the other hand, role conflict arises from incompatibility between the subject’s expectations and job requirements (Kahn et al., 1964). A lack of suitable resources, inconsistent policies, or incongruent expectations may place employees under pressure, which generates role conflict (Kemp et al., 2013). This constant draining of energy may increase emotional exhaustion (Bakker and Demerouti, 2007). Han et al. (2015) confirmed a positive relationship between role conflict and emotional exhaustion in the healthcare sector. Another relevant job demand in the healthcare context is influence at work, which refers to the degree of control over job activities (Fernet et al., 2004). It is related to the decision-making process in working conditions regarding order, methodology, or tasks (Moncada et al., 2005). In general, centralized, hierarchical structures focus decision-making on just a few members of the organization, resulting in limited decisional autonomy and possible feelings of emotional exhaustion (Knudsen et al., 2006). Conversely, decentralized and participatory structures allow for greater decisional autonomy, which can help solve specific problems and avoid emotional exhaustion (Bondarouk et al., 2016; Gil-Monte and Peiró, 1997). Influence at work can be considered either a job demand or job resource. As a resource, it refers to freedom in decisionmaking to obtain other resources needed to complete the tasks correctly (He et al., 2017; Kowalski et al., 2010); as a job demand, decisional decentralization may increase the number of everyday decisions, culminating in emotional exhaustion (Knudsen et al., 2009). Accordingly, in public healthcare, professionals have limited decision-making autonomy due to highly standardized procedures, rules, and policies. However, they should have a high degree of autonomy and freedom of action to ensure immediate, accurate, and fault-free responses Page 6 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 7 when caring for patients. Therefore, a lack of influence at work may lead to feelings of emotional exhaustion, although excessive decision-making autonomy may cause a loss or lack of control, which may also lead to emotional exhaustion. Previous research offers mixed results regarding the relationship between influence at work and emotional exhaustion. While some authors conclude that there is a significant and negative relationship between these two variables (He et al., 2017), others find a significant and positive relationship (Knudsen et al., 2006). Others affirm that there is no relationship between them (Ito and Brotheridge, 2003). Mediating role of emotional exhaustion The job demands-resources theory (Bakker and Demerouti, 2007) indicates that high job demands can spark a stressful process that may, via emotional exhaustion, lead to negative outcomes. The conservation of resources theory suggests that emotional exhaustion implies a lack of energy to cope appropriately with work tasks, making it necessary to avoid even greater losses of resources (Hobfoll and Freedy, 1993). Emotionally exhausted employees have fewer opportunities to establish positive experiences at work, so they might suffer negative effects such as a reduction in their organizational commitment (Chen and Yu, 2014) or an increase in turnover intention (Van Bogaert et al., 2014). The vocational nature of the healthcare profession (Maslach et al., 2009) may be expressed through the affective commitment of employees, emerging as a relevant outcome of emotional exhaustion. Affective commitment is the emotional bond that expresses employees’ identification with and participation in the organization (Allen and Meyer, 1990) as well as a positive emotional attachment and relative engagement with each other (Alonderienė and Šimkevičiūtė, 2018). Accordingly, affectively committed employees accept and internalize the organization’s values and goals, and their willingness to make the necessary effort to achieve them (Mowday et al., 1979). Conversely, emotionally exhausted employees withdraw Page 7 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 14 exhaustion (β=.045, n.s.) is far from significant; thus, the mediating effects of these variables were not tested. The results indicated that workload and role conflict accounted for 32.5% of emotional exhaustion, and emotional exhaustion accounted for 14.3% of turnover intention and 5.2% of affective commitment. Mediation Table II shows the results of the mediation effect of emotional exhaustion on the four possible mediation relationships. The results showed that emotional exhaustion fully mediated the relationship between workload and affective commitment (Table IIa). Two additional models, partial mediation and direct effect, were tested to confirm this result (Baron and Kenny, 1986). The chi-square of Model 3 (full mediation) is lower than the chi-square of Model 2 (direct effect) and significantly different (Δχ2=161.235, Δdf=1); it is higher than the chi-square of Model 1 (partial mediation) but is not significantly different (Δχ2=.022, Δgl=1, n.s.). The Sobel test (1982) also supported the mediating effect of emotional exhaustion (Z=-3.826, p<.00013004). To test the mediation effect with greater accuracy, two additional analyses were performed. First, the Goodman test (1960), which unless it is undefined, produces slightly more accurate confidence intervals than the Sobel test (Z=-3.840, p<.00012298). Second, the bootstrap percentile confidence interval method, which is useful for establishing confidence intervals for estimated parameters that have no known distributions or have violated the normality assumption, was performed. Table III shows the results of the BC bootstrap method for direct and indirect effects. According to the results, H2 was supported. Likewise, emotional exhaustion was found to fully mediate the relationship between workload and turnover intention (Table IIb). The chi-square of Model 3 (full meditation) is lower than the chi-square of Model 2 (direct effect) and significantly different (Δχ2=177.555, Δdf=1); it is higher than the chi-square of Model 1 (partial mediation) but not significantly different Page 14 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 15 (Δχ2=.958, Δgl=1, n.s.). The Sobel (Z=5.120, p<.0000003) and Goodman (Z=5.136, p<.00000028) tests also supported this mediating effect. Table III lists the results of the BC bootstrap method. According to the results, H5 was supported. The results also showed that emotional exhaustion fully mediated the relationship between role conflict and affective commitment (Table IIc). The chi-square of Model 3 (full mediation) is lower than that of Model 2 (direct effect) and significantly different (Δχ2=80.649, Δdf=1); it is higher than the chi-square of Model 1 (partial mediation) but not significantly different (Δχ2=2.93, Δgl=1, n.s.). The Sobel (Z=-3.672, p<.00024018) and Goodman (Z=-3.693, p<.00022122) tests also supported this mediating effect. Table III lists the results of the BC bootstrap method. Based on these results, H3 was supported. Emotional exhaustion was also shown to fully mediate the relationship between role conflict and turnover intention (Table IId). The chi-square of Model 3 (full mediation) is lower than that of Model 2 (direct effect) and significantly different (Δχ2=86.063, Δdf=1); it is higher than the chi-square of Model 1 (partial mediation) but not significantly different (Δχ2=3.83, Δgl=1, n.s.). The Sobel (Z=4.716, p<.0000024) and Goodman (Z=4.738, p<.00000215) tests also supported this mediating effect. Table III lists the results of the BC bootstrap method. According to the results, H6 was supported. [Insert Table II about here] [Insert Table III about here] Moderation Hierarchical multiple regression (Cohen and Cohen, 1983) was used to examine the moderating effects. In particular, the following have been analyzed: the moderating effect of workload and possibilities for development on employees’ emotional exhaustion, moderating effect of Page 15 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 16 workload and degree of freedom at work on employees’ emotional exhaustion, moderating effect of role conflict and possibilities for development on employees’ emotional exhaustion, and moderation effect of role conflict and degree of freedom at work on employees’ emotional exhaustion. The reliability of the moderator variables was estimated, yielding Cronbach’s alpha values exceeding .8 in all cases (Bagozzi and Yi, 1988). After centering the predictor/moderator variables (Dawson, 2014), six different regression analyses were performed. Table IV shows that the interaction of possibilities for development and degree of freedom at work with influence at work and workload did not significantly affect employees’ emotional exhaustion. Therefore, H7, H8, H10, and H11 were not supported. However, the interaction of possibilities for development and degree of freedom at work with role conflict had a significant impact on employees’ emotional exhaustion. Therefore, H9 and H12 were supported. Figure 2 illustrates the mediation and moderation effects. [Insert Figure 2 about here] [Insert Table IV about here] Therefore, the relationship between role conflict and emotional exhaustion differs depending on the level of possibilities for development and degree of freedom at work, but it is unclear exactly how it differs. The interaction coefficients are positive, so the positive relationship between role conflict and emotional exhaustion increases as possibilities for development or degree of freedom at work increases. However, it is not easy to gauge the size and precise nature of these effects by merely examining the coefficients. Therefore, the effects were plotted for visual interpretation (Dawson, 2014). The variables, possibilities for development, and degree of freedom at work were classified into three levels, namely low, moderate, and high, to predict the relationship between role conflict and emotional exhaustion at each level. Three regression groups were identified. Figure 3 indicates that the relationship between role conflict and Commented [LCA7]: This sentence has been revised and rewritten to make it more readable for the readers. Commented [RK8R8]: Still incorrect Commented [LCA9R8]: Following the Editor’s suggestion, the sentence has been reconsidered and rewriting to improve readability for the readers. Commented [RK10]: What reliability, how tested and what ? Be more specific - Commented [LCA11R11]: We have tried to provide more information about reliability. Page 16 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 17 emotional exhaustion is always positive, but the relationship is more positive for employees with lower possibilities for development than for those with higher levels [R2 Lineal PD high=.043 (correlation value=.2073); R2 Lineal PD moderate=.094 (correlation value=.3065); R2 Lineal PD low=.199 (correlation value=.4460)]. Figure 3 also indicates that the relationship between role conflict and emotional exhaustion is always positive, but it is more positive for employees with a moderate/lower degree of freedom at work than for those with higher levels [R2 Lineal DF high=.031 (correlation value=.1760); R2 Lineal degree of DF moderate=.151 (correlation value=.3885); R2 Lineal DF low=.146 (correlation value=.3820)]. [Insert Figure 3 here] Discussion This study analyzed the role of emotional exhaustion among public healthcare professionals. Workload and role conflict are positively related to emotional exhaustion. As indicated in prior research, the large number of patients and responsibility to provide high-quality service without faults or errors and/or scarce resources may cause healthcare professionals to develop emotional exhaustion (Kowalski et al., 2010). Similarly, inconsistent expectations and policies, procedures, or systems that are poorly defined, communicated, understood, or implemented may cause healthcare employees to develop emotional exhaustion (Kozak et al., 2013). However, prior research has shown mixed results regarding the relationship between influence at work and emotional exhaustion. This may be because, in certain job environments, decisionmaking freedom may be understood as a source of motivation; while in others, it may be understood as an excessive responsibility with negative consequences for employees. In line with previous research, emotionally exhausted healthcare professionals devote less energy to establishing emotional bonds with the organization because they are less emotionally attached to it (Thanacoody et al., 2014). At the same time, they withdraw psychologically, Page 17 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 18 which may cause them to desire to leave the organization, even if they do not actually do so (Van Bogaert et al., 2014) because of the job security of public healthcare professionals. However, the fact that they desire to leave may seriously interfere with their work and affect performance and service quality. The results also show that emotional exhaustion fully mediates the relationship between job demands (workload and role conflict) and employee work attitudes (affective commitment and turnover intention). According to the job demands-resources theory, workload and/or high levels of role conflict cause emotional exhaustion among healthcare employees, who then reduce their affective bonds with the organization and consider leaving. This is especially serious in a vocational profession; therefore, if their inspiration/vocation is diluted, employees may end up regarding their work as merely routine or mechanical and lose one of the fundamental values for providing a quality service. Total mediation means that the characteristics and nature of work in public health lead employees to achieve lower levels of affective commitment and higher levels of turnover intention if they feel emotionally exhausted. Considering the characteristics and nature of public healthcare, the generation of job resources through learning/career development or managing work pace may mitigate the effect of job demands on emotional exhaustion. This research concludes that the degree of freedom at work and possibilities for development do not moderate the relationship between influence at work and emotional exhaustion. Influence at work can be considered either a job demand or job resource. The dual conception of this variable, combined with emotional exhaustion, leads to mixed results. Along the same lines, the degree of freedom at work and possibilities for development do not moderate the relationship between workload and emotional exhaustion. That is, workload may be so great in the public healthcare sector that investing in job resources, allowing people to manage their own work pace or promoting exciting career development programs is unlikely to mitigate the effects of exhausting working hours and high levels of Page 18 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 19 responsibility that lead people to feel emotionally exhausted. However, both possibilities for development and degree of freedom at work moderate the always-positive relationship between role conflict and emotional exhaustion. Thus, the influence of role conflict on emotional exhaustion can be controlled/mitigated by suitable career development programs or the ability to control work hours, with greater moderating effects of learning and career development through professional training. Managerial Implications The results provide relevant information for healthcare employees, human resource managers, and public institutions. An analysis of the factors that trigger emotional exhaustion and its consequences in healthcare induces organizations to implement complete programs to mitigate, eliminate, and prevent the causes of emotional exhaustion. This improves the health and quality of life of healthcare professionals, and consequently, the service quality. Therefore, human resource managers and public decision-makers should formulate strategies that will increase the number of professionals working in healthcare and/or adapt workloads to working hours. The inclusion of innovative technological resources may help relieve their huge workload. Furthermore, there is a need to design training and career development programs that allow healthcare professionals to develop the skills and competencies they need to work more efficiently and under better conditions. Clear, transparent policies and operating procedures are needed to solve problems faster or prevent them from arising by improving the time available to provide better quality service due to the formalized and bureaucratized nature of the public healthcare context. In this sense, providing more flexibility, reducing bureaucratic constraints, providing professionals with greater discretion to perform their jobs effectively, or promoting cooperative or extra-role behaviors among employees can enhance organizational innovativeness and adaptability, thus Page 19 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 20 reducing emotional exhaustion. In addition, in public healthcare, exciting, motivating career planning should be based more on years of service than results or achievements. In this sense, for professions that depend on updated scientific knowledge, a career path should allow the possibility of combining healthcare work with research, with each professional deciding on the percentage of time to dedicate to each one. At the same time, a flow of information using clear, transparent, and convenient means of communication can establish a climate of trust, improving the quality of healthcare services. Moreover, fair criteria for performance assessment must be adopted and communicated openly, thus enhancing the level of performance. These strategies should prevent healthcare professionals from becoming emotionally exhausted and leave them feeling more motivated and happier, thus resulting in a more effective public healthcare service. Therefore, such strategies can provide healthcare professionals with the resources needed to face the loss of energy caused by excessive demands in public health services today. Conclusions and future research The results of this study emphasize the importance of analyzing the role played by emotional exhaustion in public healthcare. Preventing emotional exhaustion among healthcare employees should be a priority for public health administrators and human resources managers who are responsible for designing preventive strategies involving the adaptation to certain job characteristics. In this sense, specific action programs are urgently needed to improve the work environment and, consequently, employees’ well-being and service quality. From the corporate social responsibility perspective, employees’ well-being within organizations largely depends on the appropriate management of job resources and demands, but these resources and demands may change over time. The goal of achieving healthy organizations in which quality and employees’ well-being are the main priorities should not change over time. Page 20 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 21 As with any empirical research, this study had some limitations. Based on the use of data collected from a single-sitting, self-reporting measure, its cross-sectional survey design could have led to common method variance bias. Several procedures have been conducted to overcome this problem. After analyzing the similarities and differences regarding the research topic between the most important European countries, only one country was considered in the study. Future studies should consider other countries to highlight their idiosyncrasies and generalize the results obtained in this research. Future research should also consider other job demands (e.g., leadership style), job resources (e.g., social support), and organizational variables (e.g., organizational justice) to understand the role of emotional exhaustion and its effects on other outcomes better (e.g., job performance). References Allen, N.J. and Meyer, J.P. (1990), “The measurement and antecedents of affective, continuance and normative commitment to the organization”, Journal of Occupational Psychology, Vol. 63, No. 1, pp. 1–18. doi:10.1111/j.2044-8325.1990.tb00506.x. Alola, U.V., Avcı, T. and Öztüren, A. (2020), “The nexus of workplace incivility and emotional exhaustion in hotel industry”, Journal of Public Affairs, p. e2236. doi:10.1002/pa.2236. Alonderienė, R. and Šimkevičiūtė, I. (2018), “Linking protean and boundaryless career with organizational commitment: The case of young adults in finance sector”, Baltic Journal of Management, Vol. 13, No. 4, pp. 471–487. doi:10.1108/BJM-06-2017-0179. Bagozzi, R.P. and Yi, Y. (1988), “On the evaluation of structural equation models”, Journal of the Academy of Marketing Science, Vol. 16, No. 1, pp. 74–94. doi:10.1007/BF02723327. Page 21 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 22 Bakker, A.B. and Demerouti, E. (2007), “The job demands-resources model: State of the art”, Journal of Managerial Psychology, Vol. 22, No. 3, pp. 309–328. doi:10.1108/02683940710733115. Bakker, A.B., Demerouti, E. and Verbeke, W. (2004), “Using the job demands-resources model to predict burnout and performance”, Human Resource Management, Vol. 43, No. 1, pp. 83–104. doi:10.1002/hrm.20004. Baron, R.M. and Kenny, D.A. (1986), “The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51, No. 6, pp. 1173–1182. doi:10.1037//00223514.51.6.1173. Black, N. and Gruen, R. (2005), Understanding Health Services, McGraw-Hill Education, UK. Bondarouk, T., Bos-Nehles, A. and Hesselink, X. (2016), “Understanding the congruence of HRM frames in a healthcare organization”, Baltic Journal of Management, Vol. 11, No. 1, pp. 2–20. doi:10.1108/BJM-02-2015-0035. Brauner, C., Wöhrmann, A.M., Frank, K. and Michel, A. (2019), “Health and work-life balance across types of work schedules: A latent class analysis”, Applied Ergonomics, Vol. 81, Vol. 81, No. 1029, 102906. doi:10.1016/j.apergo.2019.102906. Chen, C.F. and Yu, T. (2014), “Effects of positive vs negative forces on the burnoutcommitment-turnover relationship”, Journal of Service Management, Vol. 25, No. 3, pp. 388–410. doi:10.1108/JOSM-01-2012-0003. Page 22 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 23 Cohen, J. and Cohen, P. (1983), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, Lawrence Erlbaum Associates, Hillsdale, NJ. Dawson, J.F. (2014), “Moderation in management research: What, why, when, and how”, Journal of Business and Psychology, Vol. 29, No. 1, pp. 1–19. doi:10.1007/s10869-0139308-7. Day, A.L., Sibley, A., Scott, N., Tallon, J.M. and Ackroyd-Stolarz, S. (2009), “Workplace risks and stressors as predictors of burnout: The moderating impact of job control and team efficacy”, Canadian Journal of Administrative Sciences / Revue Canadienne des Sciences de l’Administration, Vol. 26, No. 1, pp. 7–22. De Simone, S., Planta, A. and Cicotto, G. (2018), “The role of job satisfaction, work engagement, self-efficacy and agentic capacities on nurses’ turnover intention and patient satisfaction”, Applied Nursing Research: ANR, Vol. 39, pp. 130–140. doi:10.1016/j.apnr.2017.11.004. Esteban‐Sepúlveda, S., Moreno‐Casbas, M.T., Fuentelsaz‐Gallego, C. and Ruzafa‐Martinez, M. (2019), “The nurse work environment in Spanish nurses following an economic recession: From 2009 to 2014”, Journal of Nursing Management, Vol. 27, No. 6, pp. 1294– 1303. doi:10.1111/jonm.12811. Fernet, C., Guay, F. and Senécal, C. (2004), “Adjusting to job demands: The role of work selfdetermination and job control in predicting burnout”, Journal of Vocational Behavior, Vol. 65, No. 1, pp. 39–56. doi:10.1016/S0001-8791(03)00098-8. Page 23 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 30 Ruiz-Fernández, M.D., Ortega-Galán, Á.M., Fernández-Sola, C., Hernández-Padilla, J.M., Granero-Molina, J. and Ramos-Pichardo, J.D. (2020), “Occupational factors associated with health-related quality of life in nursing professionals: A multi-centre study”, International Journal of Environmental Research and Public Health, Vol. 17, No. 3, pp. 982. doi:10.3390/ijerph17030982 Schaufeli, W.B. (2007), “Burnout in health care”, in Carayon, P. (Ed.), Handbook of Human Factors and Ergonomics in Health Care and Patient Safety, Lawrence Erlbaum, Mahwah, NJ, pp. 217–232. Shirom, A. (1989), “Burnout in work organizations”, in International Review of Industrial and Organizational Psychology Cooper, C.L. & Robertson, I.T. (Eds.), John Wiley & Sons, Oxford, England, pp. 25–48. Sobel, M.E. (1982), “Asymptotic confidence intervals for indirect effects in structural equation models”, Sociological Methodology, Vol. 13, No. 1982, pp. 290–312. Takase, M. (2010), “A concept analysis of turnover intention: Implications for nursing management”, Collegian, Vol. 17, No. 1, pp. 3–12. doi:10.1016/j.colegn.2009.05.001. Thanacoody, P.R., Newman, A. and Fuchs, S. (2014), “Affective commitment and turnover intentions among healthcare professionals: The role of emotional exhaustion and disengagement”, International Journal of Human Resource Management, Vol. 25, No. 13, pp. 1841–1857. doi:10.1080/09585192.2013.860389. Tziner, A., Rabenu, E., Radomski, R. and Belkin, A. (2015), “Work stress and turnover intentions among hospital physicians: The mediating role of burnout and work Page 30 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 31 satisfaction”, Revista de psicología del trabajo y de las organizaciones, Vol. 31, No. 3, pp. 207–213. doi:10.1016/j.rpto.2015.05.001. Van Bogaert, P., Adriaenssens, J., Dilles, T., Martens, D., Van Rompaey, B. and Timmermans, O. (2014), “Impact of role-, joband organizational characteristics on nursing unit managers’ work related stress and well-being”, Journal of Advanced Nursing, Vol. 70, No. 11, pp. 2622–2633. doi:10.1111/jan.12449. Van der Heijden, B., Brown Mahoney, C. and Xu, Y. (2019), “Impact of job demands and resources on nurses’ burnout and occupational turnover intention towards an agemoderated mediation model for the nursing profession”, International Journal of Environmental Research and Public Health, Vol. 16, No. 11. doi:10.3390/ijerph16112011. Van der Voet, J., Kuipers, B.S. and Groeneveld, S. (2016), “Implementing Change in Public Organizations: The relationship between leadership and affective commitment to change in a public sector context”, Public Management Review, Vol. 18, No. 6, pp. 842–865. doi:10.1080/14719037.2015.1045020. Wright, T.A. and Hobfoll, S.E. (2004), “Commitment, psychological well-being and job performance: An examination of conservation of resources (COR) theory and job burnout”, Journal of Business and Management, Vol. 9, No. 4, pp. 389–406. Xanthopoulou, D., Bakker, A.B., Dollard, M.F., Demerouti, E., Schaufeli, W.B., Taris, T.W. and Schreurs, P.J.G. (2007), “When do job demands particularly predict burnout?: The moderating role of job resources”, Journal of Managerial Psychology, Vol. 22, No. 8, pp. 766–786. doi:10.1108/02683940710837714. Page 31 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 1 Table I. Means, standard deviations, correlations and estimated reliabilities M SD IW WL RC EE AC TI PD DF IW 2.808 .89141 .762 WL 2.962 .67454 -.279** .699 RC 3.013 .71559 -.210** .327** .727 EE 3.790 1.11371 -.172** .528** .315** .820 AC 3.499 .74568 .127** -.110* -.120** -.205** .826 TI 1.650 .88851 -.024 .124** .160** .267** -.266** .827 PD 3.774 .71368 .342** -.074 -.030 -.192** .208** -.169** .721 DF 2.835 .8079 .493** -.266** -.148** -.216** .122** -.054 .194** .676 *p<.05 **p<.01 Note: N=512. IW=Influence at work; WL=Workload; RC=Role conflict; EE=Emotional exhaustion; AC=Affective commitment; TI=Turnover intention; PD=Possibilities for development; DF=Degree of freedom at work. Cronbach’s  on the diagonal. Page 32 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 2 Table II. Fit results and path coefficients for structural equation models Table IIa. Mediating role of EE between WL→AC χ2(df) GFI RMSEA AGFI NFI TLI CFI χ2/df Model 1 234.412(96) .946 .053 .924 .923 .940 .952 2.442 Model 2 395.647(98) .918 .077 .887 .869 .875 .898 4.037 Model 3 234.434(97) .946 .053 .925 .923 .942 .953 2.417 Standardized coefficients (t-values) Model 1 Model 2 Model 3 WL→AC -.014(-.018)** -.082(.009)** WL→EE .578(11.12)*** .578(11.12)*** EE→AC -.208(-3.01)** -.216(-4.08)*** Table IIb. Mediating role of EE between WL→TI χ2(df) GFI RMSEA AGFI NFI TLI CFI χ2/df Model 1 160.016(58) .954 .059 .928 .932 .940 .955 2.759 Model 2 337.571(60) .918 .077 .887 .869 .875 .898 4.037 Model 3 160.974(59) .954 .058 .929 .932 .941 .956 2.728 Standardized coefficients (t-values) Model 1 Model 2 Model 3 WL→TI -.061(-.082)** .114(.021)* WL→EE .582(11.41)*** .58(11.37)*** EE→TI .321(4.59)*** -.281(5.73)*** Table IIc. Mediating role of EE between RC→AC χ2(df) GFI RMSEA AGFI NFI TLI CFI χ2/df Model 1 158.730(83) .961 .042 .943 .944 .965 .972 1.912 Model 2 239.379(85) .943 .060 .919 .916 .930 .944 2.816 Model 3 161.656(84) .960 .043 .943 .943 .965 .972 1.924 Standardized coefficients (t-values) Model 1 Model 2 Model 3 RC→AC -.113(-1.69)** -.181(-3.02)** RC→EE .43(8.43)*** .432(8.47)*** EE→AC -.163(-2.72)** -.216(-4.08)*** Table IId. Mediating role of EE between RC→TI χ2(df) GFI RMSEA AGFI NFI TLI CFI χ2/df Model 1 93.261(48) .971 .043 .952 .958 .971 .979 1.942 Model 2 183.154(50) .947 .072 .917 .917 .917 .937 3.663 Model 3 97.091(49) .969 .044 .951 .956 .970 .977 1.981 Standardized coefficients (t-values) Model 1 Model 2 Model 3 RC→TI .122(1.87) .218(3.69)*** RC→EE .432(8.47)*** .435(8.53)*** EE→TI .226(4.19)*** .283(5.66)*** Note: Model l-Partial mediation; Model 2-Direct effect; Model 3-Full mediation. *p<.05, **p<.01, ***p<.001. Page 33 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 3 Table III. Direct and indirect effects of the mediating relationships Direct effects Effect SE p LLCI ULCI RC EE → .429 .052 .001 .320 .531 EE AC → -.223 .052 .001 -.325 -.121 RC EE → .432 .052 .001 .322 .532 EE TI → .292 .046 .001 .198 .378 WL EE → .588 .048 .001 .489 .678 EE AC → -.223 .052 .001 -.327 -.122 WL EE → .589 .048 .001 .491 .679 WL IT → .290 .046 .001 .198 .374 Indirect effects Effect BootSE p BootLLCI BootULCI RC EE AC → → -.096 .026 .001 -.152 -.048 RC EE IT → → .126 .027 .001 .075 .185 WL EE AC → → -.131 .034 .001 -.201 -.070 WL EE IA → → .171 .030 .001 .112 .231 Page 34 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 4 Table IV. Moderation effects Moderation effects of PD on the relationship between IW and EE Model 1 Model 2 Model 3 IW(β) -.157*** -.098* -.098* PD(β) -.172*** -.174*** IW(β) X PD(β) .006(.895) Adjusted R square .025 .051 .051 F 12.908*** 13.665*** 9.098*** df (residual) 510 509 508 Moderation effects of PD on the relationship between WL and EE Model 1 Model 2 Model 3 WL(β) .529*** .518*** .518*** PD(β) -.151*** -.151*** WL(β) X PD(β) .003(.945) Adjusted R square .28 .302 .302 F 197.925*** 110.174*** 73.307*** df (residual) 510 509 508 Moderation effects of PD on the relationship between RC and EE Model 1 Model 2 Model 3 RC(β) .319*** .313*** .319*** PD(β) -.179*** -.174*** RC(β) X PD(β) -.103(.013) Adjusted R square .102 .134 .144 F 57.626*** 39.262*** 28.529*** df(residual) 510 509 508 Moderation effects of DF on the relationship between IW and EE Model 1 Model 2 Model 3 IW(β) .157*** - .078 - .078 DF(β) -.160* -.161* IW(β) X DF(β) .006(.896) Adjusted R square .025 .044 .044 F 12.908*** 11.722*** 7.806*** df (residual) 510 509 508 Moderation effects of DF on the relationship between WL and EE Model 1 Model 2 Model 3 WL(β) .529*** .504*** .505*** DF(β) -.092* -.093* WL(β) X DF(β) .015(.688) Adjusted R square .280 .287 .288 F 197.925*** 102.690*** 68.401*** df(residual) 510 509 508 Page 35 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management 5 Moderation effects of DF on the relationship between RC and EE Model 1 Model 2 Model 3 RC(β) .319*** -.292*** .304*** DF(β) -.183*** -.179*** RC(β) X DF(β) -.095(.023) Adjusted R square .102 .134 .143 F 57.626*** 39.479*** 28.276*** df (residual) 510 509 508 *p<.05, **p<.01, ***p<.001. Page 36 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management Figure 1. Proposed model Page 37 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management Figure 2. Structural Model Page 38 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Baltic Journal of Management Figure 3. Regression effects between role conflict and emotional exhaustion depending on different levels of possibilities for development and degree of freedom at work RC EE EE RC Page 39 of 41 http://mc.manuscriptcentral.com/bjom Baltic Journal of Management 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60