1. Introduction
Burnout is a serious problem among healthcare professionals, such as nurses and physicians [
1,
2,
3]. Burnout is associated with a number of deleterious consequential outcomes in the healthcare provider, such as an increased number of sick leave absences, job turnover, increased risk of psychopathology and physical complications, and poorer quality of healthcare delivery [
4,
5]. Moreover, this ineffective response to stress is likely to negatively influence patient outcomes (e.g., poor treatment compliance and less satisfaction with treatment). Furthermore, these potentially adverse outcomes can contribute to exerting additional strain on the healthcare professional, thus creating a vicious cycle [
6].
The problem of burnout appears to be particularly alarming during residency, where between 36% and 76% of residents have experienced this syndrome anytime during residency [
7,
8]. Several factors have been proposed to influence burnout during residency, including gender, age, marital status, culture, workload, physical activity, and coping strategies, among others [
9,
10,
11,
12]. Of these, workload interventions have received the most attention. However, existing treatments have had a limited impact on burnout levels [
1], and public time reduction initiatives have not solved the problem [
13].
Both clinical and organizational studies have highlighted the importance of emotion regulation strategies on individual outcomes, including burnout in healthcare professionals [
14,
15,
16,
17]. However, their role in residents’ burnout, as well as their contribution above and beyond already explored variables like workload, remains unclear. The present study aims to provide evidence on the contribution of emotion regulation strategies to residents’ burnout while controlling for important covariates. Emotion regulation is a complex phenomenon influenced by an interconnected set of neuroendocrine, neurological, genetic, personality, and situational factors [
18,
19,
20,
21,
22]. In this work, we focus on two important regulatory mechanisms embedded in this complex network of factors, namely emotional suppression and cognitive reevaluation [
23].
Emotion regulation can be defined as “the process by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” [
24]. Since Dr. Hochschild’s pioneering work [
25], emotion regulation strategies have been frequently classified as surface (superficial) or deep. In surface acting, emotional expressions are adjusted to what is desired by others, but internal feelings are not necessarily modified. Therefore, a discrepancy between internally felt and required emotions usually emerges. By contrast, in deep acting, inner feelings are actively regulated and adjusted to what is displayed externally [
26]. Emotional suppression and cognitive reevaluation are the most frequent methods of surface and deep acting, respectively. Emotional suppression is characterized by inhibiting true feelings in favor of desired emotions [
27]. As opposed to this, individuals using cognitive reevaluation reappraise the situation so that this new evaluation leads to the required emotion [
14,
28]. In general, surface acting strategies, like emotional suppression, are associated with poorer outcomes, while deep acting tends to be a recommended practice [
29,
30].
An example of emotional suppression and cognitive reevaluation strategies is provided in the context of medical residents. In organizational settings, emotion regulation tends to be understood as how employees manage emotions to adapt to work demands [
31]. Residents face high responsibility demands, deal with critical life or death situations, experience patient and family demanding pressures, and often have little time to prepare an elaborated response to daily challenges. The previous, together with the fact that residents tend to be young and generally lack professional experience, frequently leads to feelings of insecurity [
32]. In this scenario, an emotional suppression strategy would be to try to hide such feelings by trying to be seen as confident by others at any cost or by using technical words in a patient’s presence, thus showing ability and knowledge [
33]. In contrast, in this same scenario, residents may decide to remind themselves that they are still trainees who are expected to lack specific knowledge and abilities and use this to challenge themselves to improve, which is an example of a cognitive reevaluation strategy.
In the current study, we explore how these two regulation strategies, namely emotional suppression and cognitive reevaluation, are related to burnout among hospital residents and how much they contribute above and beyond previously investigated risk and protective factors, like workload and demographic variables, which will be used as covariates. In line with studies suggesting the need to explore factors that facilitate or hinder burnout processes [
34], we will also examine whether the relationship between emotion regulation and burnout is contextually determined (i.e., moderated) by these covariates.
Hypothesis 1 (H1). We expect that cognitive reevaluation will be associated with lower burnout levels when compared to emotional suppression, even after adjusting for demographic and workload variables.
Hypothesis 2 (H2). We also hypothesize that emotion regulation strategies and some covariates will interact—that is, that the contribution of emotion regulation strategies on burnout will depend on the context in which they occur, in line with previous clinical research [35]. 4. Discussion
According to previous studies, physician burnout estimates in residents are generally high and up to 70% [
41,
46]. We found similar levels of burnout in our sample, indicating that many residents struggle with negative emotions during this particular training period. Overall, our results revealed that emotion regulation strategies were associated with residents’ burnout, which is in line with past research showing that emotional management is fundamental in healthcare professionals [
17,
47,
48]. Specifically, emotional suppression was associated with depersonalization, and cognitive reevaluation was related to personal accomplishment. These associations occurred in the expected direction; that is, emotional suppression was associated with more burnout, while cognitive reevaluation strategies correlated with lower levels of this syndrome. Most importantly, for the present study, the contribution of both strategies remained significant when demographic variables and work-related factors were accounted for in a multivariate regression model.
Emotional suppression is an emotion regulation strategy characterized by inhibiting true emotions in favor of feelings that are required in a given situation [
14]. However, this is psychologically costly, as the underlying appraisals of the situation are not being changed. Previous research in organizational settings has already evidenced that emotional suppression strategies are associated with increased levels of burnout [
49]. Results in the present study indicate that this might also be the case in hospital residents. Interestingly, though, our analyses suggested that this emotion regulation strategy might contribute mostly to depersonalization. The depersonalization scale measures an individual’s tendency to be cold and unfeeling towards people they care for, provide service to, or instruct at work [
41]. What our results suggest is that hiding one’s true undesired feelings (e.g., fear of failure, insecurity) as a protective strategy may increase the emotional distance between residents and patients. Because healthcare professionals’ empathy is known to predict beneficial outcomes in the patient [
50], we interpret our results as indicating that residents’ use of emotional suppression as an emotion regulation strategy may be maladaptive due to its positive association with depersonalization.
As opposed to emotional suppression, our results support the idea that cognitive reevaluation may be an adaptive emotion regulation strategy for hospital residents. Cognitive reevaluation involves reevaluating a situation to give it a new meaning, thus changing the emotional response it produces. Even though it may require a higher initial effort to find alternative interpretations, it seems to be a better strategy in the long term. Past studies exploring the relationship between cognitive reevaluation and personal outcomes were not conclusive [
49]. However, theory on emotion regulation states that cognitive reevaluation strategies are preferred over emotional suppression efforts because only the former leads to a change in experienced emotion [
14]. Our results support this hypothesis. Specifically, what our analyses indicate is that residents using cognitive reevaluation strategies tend to present increased levels of personal accomplishment. In our opinion, it is possible that the ability to reappraise a situation in order to change one’s feelings results in a more successful adaptation to the environment. In the case of hospital residents, this appears to facilitate a feeling of personal achievement at work. Note, though, that the correlational nature of our data prevents us from drawing causal conclusions from our results.
An interesting finding from the present study is that the relationship between certain predictors of burnout (i.e., emotion regulation strategies and work-related factors) and burnout might be more complex than we thought. For example, the number of hours spent with the patient might be associated with reduced depersonalization, but they appear to synergize with cognitive reevaluation strategies: residents with higher levels of cognitive reevaluation seem to benefit more from spending more hours in direct patient care. The specific number of patients attended was not assessed. Conversely, we focused on the time spent with the patients. Perhaps less time spent with patients could be associated with greater time constraints, which could explain the reduced depersonalization levels in those who dedicate more time to patients.
Our analyses also indicated that the relationship between emotion suppression strategies and burnout might be modulated by work-related factors. Specifically, while cognitive reevaluation strategies were associated with increased levels of personal accomplishment, the strength of this association was increased when residents perceived low levels of support from their supervisors. Studies frequently indicate that having a supportive supervisor is associated with better outcomes in the employee [
51,
52]. Together with existing research, our results revealed that having a supportive supervisor may also reduce the need for the use of emotion regulation strategies, namely cognitive reevaluation. The use of this emotion regulation strategy may be more important for residents’ burnout when supervisors are not supportive.
Both socio-demographic factors and work conditions have not been consistently associated with residents’ burnout [
9]. For instance, race/ethnicity, primary language, and cultural background have been found to be associated with burnout levels in residents in some studies [
53,
54], but not in others [
55,
56]. Similarly, some studies have indicated an association between burnout and some work conditions, such as working hours and number of on-call nights per week [
57], but others have not [
55,
56], and implementing a reduction in the number of work hours has not been shown to reduce burnout [
58]. Our results might suggest an explanation for these inconsistencies. Specifically, it is possible that the contribution of these variables to burnout levels is moderated by the residents’ emotion regulation style.
The present study certainly has limitations. First, the fact that the number of nurse residents was importantly lower than that of medical residents limits the generalizability of the results to a nursing population. However, burnout problems in nurses and physicians tend to be similar [
59], so it is possible that the present study findings might be useful for both professional groups. Similar to other single-center studies, one of the limitations of the present study’s findings is that the results may not be generalizable to other contexts with a different local culture or organizational settings. Our results would need to be confirmed in different contexts. Another limitation of the current investigation is that the statistical approach used (stepwise multivariate regression) is bottom-up, data-driven, and not theoretically driven. This method was selected because of the exploratory nature of the study (no clear hypothesis about the interactions) and to obtain a more parsimonious model with a more manageable number of variables. Furthermore, causal attributions cannot be drawn because of the cross-sectional nature of the study. We want to note that supervisor support, which was a relevant predictor in our models, can be confounded with other variables, such as residents’ work histories [
60]. Future research should (a) identify the distinct effect of these two variables; (b) explore the influence of psychological and situational variables not included in our work, such as personality traits, the type of assigned tasks/workplaces, and the number of patients cared for; and (c) examine the moderating effects of other demographic variables, such as age and sex/gender.