Association between emotional intelligence and empathy among medical students: a single center cross-sectional study, Brazil, 2019

Introduction: Emotional intelligence (EI) and empathy are two essential skills for person-centered Medicine. Objectives: To evaluate the association between EI and empathy and to assess whether sociodemographic factors and year at the medical school influence the level of EI and empathy. Methods: Cross-sectional study carried out in medical students from a private educational institution in the city of São João del-Rei, Minas Gerais, Brazil. EI levels were assessed using the Schutte Self-report Emotional Intelligence Test and empathy levels were assessed using the Jefferson Scale of Empathy (student version). All volunteers signed the Informed Consent Form before inclusion in the study. The statistical analysis used mean values, standard deviation, frequency distribution, Student’s t test, Pearson’s correlation, and linear regression. A significance level of 0.05 was considered. Results: From August 5 to 30, 2019, 193 volunteers, corresponding to 85.8% of the total population, agreed to participate in the study. The total EI (129.8 ± 13.3) and empathy (121.2 ± 11.6) observed scores were high. EI scores were influenced only by age (padjusted = 0.018). Students attending more advanced semesters had higher total empathy scores (padjusted = 0.013). Students whose parents did not have a higher education degree also had a higher total empathy score (padjusted = 0.031). A moderate positive correlation was observed between the total EI and empathy scores (ρ=0.304, p<0.001) and between the total empathy score and the EI domain Managing Others’ Emotions (ρ=0.300, p<0.001). A weak positive correlation was also observed between the total EI score and most of the empathy domains. Conclusion: A positive correlation between emotional intelligence and empathy was observed. The age influenced EI and the year of medical school and parental schooling influenced empathy.


INTRODUCTION
In the medical profession, emotional intelligence and empathy are key concepts for organizing the interpersonal and communication skills 1,2 . Both emotional intelligence and empathy are the basis for the capacity to recognize emotions and affective messages in others, build a therapeutic alliance, effectively communicate with patients and their family members and create a good doctor-patient relationship. For this reason, these concepts have currently become so important, when compassionate, humanitarian and quality health care is sought 3 .
The idea of emotional intelligence has its roots in the concept of social intelligence, first described by Thorndike in 1920, who defined that intelligence as "the capacity to perceive internal states and one's own reasons and behaviors, as well as the others' , and to act on them appropriately based on that information" 4 . Salovey and Mayer were the first to suggest the name emotional intelligence (EI) to refer to people's ability to deal with their emotions and defined EI as "a subset of social intelligence, which comprises the capacity to monitor people's feelings and emotions, of one's own and others' , to discriminate between them and use that information to guide thoughts and actions" 4 .
In Medicine, EI has been considered very important, especially regarding the careful management of emotions, which is necessary in the practice and daily care of patients 5 . During their training, medical students witness situations of pain and suffering on a daily basis, experienced by their patients and family members, and must be able to face these situations, helping them, but at the same time understanding and differentiating the pain of the other from their own pain. Therefore, it is important that during their training they are able understand the emotions and feelings of the other, show that they understand them and how they understand them, but also that they can perceive, feel and experience these emotions and feelings without being overwhelmed by them, while always maintaining dominance and control of their own emotions 1 .
Empathy is another important skill for medical practice. According to its modern concept, empathy is the capacity to understand and share the feelings of others 6 . Hojat defined empathy in the health care context as being predominantly a cognitive emotional attribute that involves understanding the patient's pain, experiences, interests and perspectives, combined with the capacity to communicate this understanding and the ability to help 7 .
Patients see doctors who have more empathy as better professionals. A doctor can be very competent in clinical diagnosis and will still be considered ineffective if they lose the connection between patient satisfaction, adherence to medical instructions and empathy 8 . EI and empathy are related but are distinct constructs.
Among the different models of EI, the Bar-On and Goleman models consider empathy as one of its basic components, suggesting a very close association between these two constructs 1 . Salovey  Some studies have simultaneously assessed emotional intelligence and empathy, but not the association between the two constructs. Austin et al. evaluated 273 medical students in the first, second and fifth years of medical school, aiming at comparing the levels of empathy, assessing gender differences in empathy and EI and investigating whether EI and empathy were related to academic success. Higher levels of empathy were observed among women. Sparse associations were found between academic performance and EI and there was no association between academic performance and empathy 9 .
Studies that specifically assessed the association between EI and empathy are scarce. A study carried out in Japan between 2008 and 2011 evaluated 415 students starting medical school, who were asked to answer questionnaires regarding EI, empathy and personality traits. The results indicated a weak positive correlation between EI and empathy 6 .
A cross-sectional study carried out with 329 nursing students from a university in Iran showed a strong positive correlation between empathy and EI, indicating that the increase in emotional intelligence scores is directly associated with the increase in empathy scores 10 .
In recent years, it has been recognized in the medical literature that interpersonal skills and the management of emotions are very important for doctors, having a primary role in the quality of doctor-patient relationship. Considering that these are skills, they can be learned, trained and improved throughout the medical program, aiming at the formation of more humanistic doctors and being person-centered care. No study conducted in Brazil that assessed the association between emotional intelligence and empathy in medical students was identified in the literature review. Given this reality, the main objective of this study was to assess the association between emotional intelligence and empathy in this population.

METHODS
A cross-sectional study was carried out in medical students at Centro Universitário Presidente Tancredo de Almeida Neves (UNIPTAN), which is a private educational institution, located in the city of São João del-Rei, state of Minas Gerais, Brazil. UNIPTAN was founded in 2000, and the medical program was created in 2015 and follows a mixed teaching methodology. The target population consisted of all students enrolled in the second semester of 2019.
The inclusion criteria were: to be regularly enrolled in the medical program in the second semester of 2019 and to sign the Informed Consent Form (ICF). Students were excluded if they: participated in any research using one of the instruments utilized in this study in the last 2 years, did not sign the ICF, or did not complete the questionnaire. All students in the fourth and fifth semesters were excluded for having participated in a study using the questionnaire for the assessment of the EI in 2018. In the second semester of 2019, the program did not have a seventh semester, as there was no selection process corresponding to that semester.
The potential population, therefore, consisted of 225 students, 140 (62.2%) women and 85 (37.8%) men. A nonprobabilistic convenience sample was used. All students who met the inclusion criteria were invited to participate. The students were approached by the researchers at the end of a class and received detailed explanations about the study, clarifying doubts and explaining that participation would be totally voluntary. The researchers made it clear that volunteers could withdraw from participation at any time and their privacy would be fully respected and the confidentiality of their personal information would be guaranteed. Students who agreed to participate signed the ICF, received the questionnaire, filled it out and returned it to a researcher.

Data collection
Empathy was assessed using the students version of the Jefferson Scale of Empathy (JSE-s) 3,11 and EI by the Schutte Selfreport Emotional Intelligence Test (SSREIT) 12,13 .
The JSE-s is a 20-item self-completed instrument designed to measure empathy in the context of patient care and is used worldwide as a measure of empathy in medical students. Since students have contact with patients since the first year of medical school in basic health units and under the supervision of Family and Community Health Doctors, it was decided to include students in the basic cycle of medical program. The items are answered according to a seven-point Likert scale, ranging from 1 "strongly disagree" to 7 "strongly agree". The answers to questions 1, 3, 6, 7, 8, 11, 12, 14, 18 and 19 have a reverse score (strongly agree = 1, strongly disagree = 7) and their values must be inverted before the final score is calculated 11 . The values obtained from each question are added and the final score represents the empathy score. The global score ranges from 20 to 140 points. Higher scores indicate higher levels of empathy.
In addition to the total score, the questionnaire assesses three domains 3 : Perspective Taking (seven items -7 to 49 points), Compassionate Care (11 items -7 to 77 points) and the Ability to Stand in the Patient's Shoes (two items -7 to 14 points).
The SSREIT was developed and validated as an instrument to measure the level of EI based on the original instrument by Salovey and Mayer 12 . It is a 33-item, self-administered questionnaire, which uses a five-point Likert scale, in which 1 means "strongly disagree" and 5 means "strongly agree". The total score obtained is calculated by adding the reverse score on items and Utilization of Emotion (6 items -6 to 30 points) 13 .
The main outcomes were the global empathy and EI scores and the secondary outcomes were the domain scores for each of the scales. The independent variables were gender, age, parental schooling level, previous higher education degree, history of severe personal or family illness, self-reported history of depression or mental illness, program semester and intended medical specialty (clinical or surgical).

Statistical Analysis
The database was created using Microsoft ® Excel ® (Microsoft, USA) and statistical analysis was performed using IBM ® SPSS ® Statistics v. 19

Ethical Aspects
This study is in accordance with the current version of the Declaration of Helsinki, Resolution 466/2012 of CONEP and its updates and Resolution 510/2016 of CNS and was approved by the Research Ethics Committee of Universidade José do Rosário Vellano (Opinion 3,361,657).

RESULTS
From August 5 to 30, 2019, a total of 193 volunteers agreed to participate in the research. This sample corresponded to 85.8% of the total population of students ( Figure 1). It took them 15 to 20 minutes to complete the questionnaire. Table 1 shows the sociodemographic and personal characteristics of the 193 students. There was a predominance of female (126 -65.3%) and single (186 -96%) students. The second semester contributed with the highest number of students (42 -21.8%) and the eighth semester with the lowest (20 -10.4%). Most students had both parents with a high education degree (88 -45.6%), 61 (31.6%) had only one of them with a high education degree and 42 (21.8%) had neither with a high education degree. Fourteen students had a previous higher education degree (7.3%). Regarding the intended specialty, the majority indicated a clinical specialty (105 -54.4%). Forty-two students (21.8%) reported the existence of some severe illness in the family. Only seven (3.6%) reported having any serious illness and 57 (29.5%) self-reported the existence of a previous mental disorder. The mean age of the participants was 22.6 ± 4.1 years, ranging from 17 to 40 years. Most students (181 -93.8%) were aged up to 30 years old.   Table 3 shows the results of the bivariate and multivariate analyses related to the total emotional intelligence score and the sociodemographic and personal variables. Only age showed an association with EI levels. An increase in EI levels was observed with increasing age (p adjusted = 0.018).
When assessing empathy, a statistically significant difference was observed only in relation to the program semester and parental schooling level. Students from more advanced  semesters had a higher total empathy score (p adjusted = 0.013).
Volunteers whose parents did not have a higher education degree showed higher empathy levels than those who had at least one parent with a high education degree (125.0 ± 9.37 versus 120.3 ± 11.83; p adjusted = 0.031), as shown in Table 4. Pearson's correlation showed a positive association between total EI and empathy scores and their domains (Table 5).
There was also a positive and statistically significant association between all domains of EI and empathy, with the exception of the domain Use of Emotion of EI with the Compassionate Care and Ability to Stand in the Patient's Shoes empathy domains.
Although the correlations are statistically significant, only the associations between the total score and between the total empathy score and the domain Managing Others' Emotions of EI showed a moderate correlation (ρ ≥ 0.30), as shown in Figure 2.

DISCUSSION
The main objective of this study was to evaluate the association between emotional intelligence and empathy in medical students from the UNIPTAN. The influence of sociodemographic and personal factors, including the program semester, on EI and empathy was also assessed. The results showed that, in general, the EI and empathy scores were high, about 80.0% of the total possible points.  There was no statistical difference between the total empathy score between men and women. This finding contradicts the results of many international studies on empathy  Parental schooling level significantly influenced the total empathy score, with the highest scores found in students whose parents did not have a higher education degree (p adjusted = 0.038). Perhaps students born to parents who had fewer opportunities to study might feel closer to and have more intuitive identification with the patients they care for, who may be more "similar" to their own parents.
There was no association between the program semester and the levels of EI, as in the study by Coury  The main limitation of this study is its cross-sectional design, which does not show the real changes that occur with students throughout medical program. This limitation is inherent to the used design and limits the conclusions related to the program semester, as there is an implicit inference that students from the early semesters will behave like the students from more advanced semesters in the future, which is not necessarily true. Another limitation is the fact that the study was carried out in a single medical school, at a private institution, which restricts the generalization of findings to similar contexts. Moreover, this is a relatively new educational institution, whose medical program was created not long ago, with no students at the supervised internship and no students that have already graduated. It is also noteworthy that the use of self-administered questionnaires can overestimate the measures, mainly due to the effect of social desirability. As a last point, it should be considered that, although the sample represents 85.6% of the potential population, its size may not have been sufficient large to identify some associations, especially those with little effect.
Despite these limitations, no other study was identified in the national literature that assessed the association between EI and empathy in medical students or other health programs, which makes it an unprecedented research. Its findings are consistent with those of the current literature and indicate that there is a significant correlation between EI and empathy, two essential skills for good professional medical practice.

CONCLUSION
The results indicated a positive correlation between the global scores of EI and empathy and several of their domains.
Higher levels of empathy were observed in students from the more advanced semesters of the program, as well as a positive correlation between age and EI. Students with parents without a high education degree showed higher levels of empathy.