Relational and moral competences are key features in medical practice, as they shape the way the encounter with patients unfolds and our decision making. The clinical encounter should provide a safe and warm atmosphere where patients fully disclose their symptoms and explore fears, reservations, and concerns, embarking on a reliable and effective therapeutic project. Shared decision making in medicine is always a dialogical process and tailor-made interventions must respect both patients' autonomy and beneficence in accordance to the leges artis. Despite the relevance of relational and moral competences on clinical practice these aren't routinely measured when choosing candidates for medical school and they are also dismissed in most medical education curriculums.

First, empathy is key to relational competences in medicine, allowing the patient to experience presence, warmth, and understanding.¹ Introduced in medicine in 1963 by William Osler, empathy refers to a special form of sharing patients' experience (different from contamination or neutralization of emotions) in which it is possible to "see inside" and characterize the inner life of patients. It requires a cognitively challenging² embodied recognition of another's emotional state and the identification of behavioral correlates, personality dimensions, and emotions.³ For a tentative operational definition four levels were suggested: emotive (the ability to subjectively experience and share the other's psychological state and feelings), moral (an altruistic internal force), cognitive (the intellectual ability to identify and perceive the other's perspective in an objective way), and behavioral (communicative response in order to show insight into the other's perspective).⁴ The development of empathy across medical pre-graduate curriculums was explored in order to understand possible differences across students and which factors reduced and promoted it. Students that start their studies with lower scores appear to get a greater reduction of empathy⁵ and is evidence towards empathy fading from the first year of study.^6,7 It is postulated that the rigor of the theoretical curriculum, which models the academic life, leads students to dismiss communicative and empathic skills.⁸ Other assorted explanations for the decline of empathy include 1) "de-idealization" of medicine (expectations vs reality), 2) defense mechanisms to distance themselves from illness or death⁹ and 3) the risk of litigation, economic and time constraints.¹⁰ This last factor is part of today's health systems requirements (that favor productivity over quality and fast interventions) and leads to a guarded, impersonal, and distant communication style.¹¹ Interestingly, previous studies with Portuguese samples seem to show a conservation or even increase in empathy.¹²

Temperament also has an important role on relational competences as it shapes the way individuals interact with one another. For example, irritable temperament is associated with lower levels of empathy, anger, and violence whereas depressive temperament is characterized by perseverance and reliability.¹³ Temperament has been considered since Kraepelin and the present-day definition by Akiskal as "a hereditary core of the personality that determines reactivity, mood and energy of a given individual and that it is relatively unchangeable throughout life".¹⁴ It possesses both positive and negative aspects, impacting the development of personality and quality of life, being linked to some personality traits.¹⁵

In our study, morality construct is addressed both positively as competences in moral judgements and negatively by studying callous and unemotional traits (CUT). Moral judgment as a broad category can be clarified by the analysis of paradigmatic cases. The Moral Competence Test (MCT) is a good tool to identify moral decision making. CUT are part of the affective dimension of psychopathy, including the narcissistic traits and manipulative, impulsive, and irresponsible behavior. They commonly refer to persons who lack empathy and/or guilt. More specifically, people with higher CUT "are more likely to show deficits in their processing of negative emotional stimuli, low levels of fearful inhibitions and anxiety and decreased sensitivity to punishment cues, especially when a reward-oriented response set is primed".^16,17 Morality is paramount to medicine considering the amount of ethical decisions that are involved in everyday practice.

Our study aimed at a comprehensive larger analysis of these competences in a sample of medical students including their possible inter-relations. Our first aim was to clarify our sample empathy, morality, callous traits, and temperament scores and to identify possible differences between 1st and 5th year students. Our second aim was to identify whether empathy correlated with morality (positively) and callosity (negatively). Our last aim was to identify the correlation between specific temperaments and our measured variables (BLR, ICU and MCT)

We found no statistical evidence to support any difference in empathy levels, measured by BLRI, between 1st (N=104, Mean: 41.42, SD: 22.48) and 5th-year students (N=34, Mean: 37.35, SD: 23.35), being that t₍₁₃₆₎=0.908, p=0.366. Similarly, we found no statistical evidence to support any difference between the two medical years (Mdn(1st year)= 17.00; Mdn(5th year)= 17.00) regarding Moral competencies, as measured by MCT-T (U=1059.0, p=0.969) and callous traits, as measured by ICU-T (Mdn(1st year)=16.00, Mdn(5th year)=19.00, U=2030.5, p=0.194). To note, we found no statistical evidence for differences between the two groups regarding the subscales of BLRI, MCT-T or ICU-T. Regarding Temperament (TEMPS-A), we found no statistical evidence to support any difference between the two groups, except for the anxious temperament (t₍₁₃₆₎= 2.359, p=0.020), which was higher on first-year students. See Table 2, which presents overall scores and tests between the 1st and 5th-year students.

We found a negative correlation between empathy, as measured by BLRI, and callous traits, as measured by ICU-T (r₍₁₃₆₎=-0.444, p=0.000). Noteworthy, (1) the ICU-T subscales were negatively correlated with BLRI: unemotional (r₍₁₃₆₎=-0.306, p=0.000), uncaring (r₍₁₃₆₎=-0.402, p=0.000), callousness (r₍₁₃₆₎=-0.261, p=0.002); Regarding empathy and Moral competences, we found no correlation between the two variables (r₍₉₆₎=0.029, p=0.779).

We explored the correlation between empathy and specific temperaments (anxious, irritable, hyperthymic, depressive, and cyclothymic) and the results are presented in Table 3. Among these results, we would like the reader to notice that: (1) Cyclothymic, anxious and irritable temperaments were negatively correlated with BLRI (r₍₁₃₆₎=-0.334, p=0.000, r₍₁₃₆₎=-0.281, p=0.001, r₍₁₃₆₎=-0.400, p=0.000); (2) There was no correlation between hyperthymic or depressive temperament and BLRI or between MCT-T and BLRI (see Table 3).

Relational and moral competencies (RMC) are classically considered relevant skills in medical practice. This is the first empirical investigation of RMC competencies, including callous traits, in a sample of 1st and 5th-year medical students. Our first aim was to determine Lisbon University Medical School student's empathy, moral competencies, callous traits, and temperament scores and identify potential differences in scores between the 1st and 5th-year medical years.

As for Moral competencies, our sample (N=96, Mean(MCT)= 19.03, SD: 11.56) scored more than other youth populations, as for example, a sample of 827 high school students from urban Polish schools (N=827, Mean(MCT)=14.42, SD: 9.99)²⁷ and a sample of teens with a criminal record (N=23, Mean(MCT)= 16.97, SD: 6.2).²⁸ On the other hand, medical students scored less than doctors (N=236, Mean (MCT)=22.43, SD: 14.1)²⁹, which can lead us to think that Moral Competences may evolve with age and maturity.

Regarding callous traits, medical students scored lower on ICU-T total, callousness, and uncaring (N= 138, Mean (ICU-T): 17.43, SD: 7.12; Mean(ICU-callousness): 4.93, SD: 2.68; Mean (ICU-uncaring): 5.16, SD: 3.54) than adolescents from basic and high school (N=1443, Mean (ICU-T): 24.58, SD: 2.78; Mean (ICU-callousness): 7.29, SD:4.39, Mean (ICU-uncaring): 8.14, SD: 3.98).³⁰ We speculate that adolescents who aim for a medical curriculum could be more sensitive and have better interpersonal skills, which per se would determine their interest in working closely with people. Yet previous studies of juveniles with a criminal record suggest that our sample scores are lower in respect to callousness and uncaring traits 31 (N=500, Mean (ICU-Callousness)= 13.75, SD: 5.08; Mean (ICU-uncaring): 15.41, SD: 3.05), while similar in unemotional scores (N=500, Mean (ICU-unemotional)= 7.07, SD: 2.31). In our sample, "unemotional" traits might refer to students believing that they should be guarded in their emotions (having idealized the role of "being professional" as avoiding 'personal feelings' or 'emotions') and therefore answer that taking control of one's feelings when exposed to several impacting experiences as dissecting a cadaver, drawing a patient's blood, diagnosing a terminal illness – these are 'emotional' experiences that they must manage 'unemotionally'.

With respect to temperament, the students scored higher on the depressive, hyperthymic, and anxious dimensions (N=138, Mean TEMPS-Depressive=0.41, SD:0.15, Mean TEMPS-Hyperthymic:0.49, SD:0.20, Mean TEMPS-Anxious:0.44, SD:0.21). In fact, some of the students from our sample had a family psychiatric history and/or current psychotropic treatment that could partly explain these results. When comparing our results to the ones by Jaracz, who studied 100 civil workers and 100 nurses from Poland, we found that: (1) our sample had higher scores on depressive temperament than civil workers (N=100, Mean TEMPS-Depressive=0.26 SD:0.12) and nurses (N=100, Mean TEMPS-Depressive=0.28, SD:0.11) and (2) medical students scored higher on anxious temperament than civil workers (N=100, Mean TEMPS-Anxious=0.26, SD:0.22)  and nurses (N=100, Mean TEMPS-Anxious=0.32, SD:0.22).³² Notably, when comparing to Hinic's work with 770 healthy university students (medicine, philology, economics, humanities, sports, engineering and natural sciences, musical and visual arts) we also found that our sample scored higher on depressive and anxious temperament (N=770, TEMPS-Depressive=0.15, SD:0.20,  N= 770, TEMPS-Anxious=0.36, SD:0.25).³³  If we focus on Hinic's subset of Serbian medical students, our sample still scores higher on these specific temperaments (N=100; TEMPS-Depressive=0.16, SD:0.14, Mean TEMPS-Anxious=0.40, SD:0.2).

Regarding empathy development among the Lisbon University medical students, we found no differences between 1st and 5th-year medical students. Previous studies provide assorted evidence of either an empathy decline (key factors: stress and learning overload, disease-centred approach, lack of student-patient time, abusive tutors, and students' low self-images)^34-39 or preservation and even increase.^12,40 Interestingly, our results are along with previous studies in Portuguese populations  (t₍₁₃₆₎=0.908, p=0.366), corroborating the results from the Portuguese work by Magalhães and colleagues: "nonsignificant differences on empathy scores (...) were found between the pre-clinical (...) and clinical phases".^40,41 Although we cannot exclude our result to be a false negative it could be that Portuguese medical curriculums attend and act towards caring for the factors that would determine empathy erosion.⁴²  

Our second aim was to identify whether empathy was correlated with morality (positively) and callosity (negatively) or not. Moral reasoning wasn't related to empathic skills in our sample (no significant correlation between MCT-T and BLRI scores), and, as such, the classical idea that empathy and moral competencies aren't related could be put forward. Although these results could be due to small sample size (false negative, as only 96 students completed the MCT-T questionnaire), some literature provides evidence that empathy and morality are not in fact correlated. As an example, Kiehl's study with incarcerated psychopath offenders concluded that "psychopathy as a whole did not predict the ability to understand what is morally wrong".^43,44 Furthermore, Batson and colleagues found that empathy can even lead to immoral behavior.⁴⁵ Regarding callous traits, as we expected, we found a negative correlation between them and empathy. In line with our results, previous literature defines callous traits as disregarding/uncaring about the feeling of others, concern for the consequences of his/her actions solely on his/herself rather than their effects on others, even when they result in substantial harm to others.⁴⁶

Finally, we aimed to identify the correlation between specific temperaments and empathy, as measured by BLRI. As we postulated, our work showed a negative correlation between empathy and cyclothymic, anxious, and irritable temperaments (r₍₁₃₆₎=-0.334, p=0.000, r₍₁₃₆₎=-0.281, p=0.001, r₍₁₃₆₎=-0.400, p=0.000). These temperaments, by definition, comprise various representative aspects of negative emotionality, such as irritability, anger, and frustration.^47,48 Since negative emotionality plays a crucial role on interpersonal relationships, and therefore empathy, it explains our results. On the other hand, contrary to what we were expecting, hyperthymic temperament was not correlated with empathy. According to Akiskal, hyperthymic individuals are "habitually cheerful, sociable, self-assured, boastful, improvident, uninhibited, overtalkative".⁴⁷ These characteristics seems to have a major role on positive affect ("the display of pleasure and excitement, and the tendency to seek out and approach reward-fulfilling stimulation")⁴⁷ which have been described as essential aspects for social interaction, and therefore empathy. Although it is possible that our results are false negatives, these results hint at a nonlinear relationship between empathy, temperament and positive affect. These findings support the need for a multifactorial definition of empathy, comprising not only emotional, but also cognitive and behavioral dimensions.^3,4

Our study aimed to determine empathy, moral competencies, callous traits, and temperament scores on Lisbon University Medical School students. We also had the objective of identifying potential differences in scores between the 1st and 5th-year medical years and the correlation between empathy and morality, callosity and, specific temperaments. To the best of our knowledge, this is the first work that studies empathy, callous trains, morality, and temperament as a whole in medical students. It is also the first work that analyzes the relationship between empathy and temperament.

Regarding the overall scores for the studied variables, we would like the reader to notice that medical students scored lower than the general youth population on callous traits (as measured by ICU-T) and higher on moral competencies (as measured by MCT). These results underline the fact that medical students usually have a specific profile which makes them choose a career with close contact to people. However, we found that medical students had similar scores on unemotional traits when compared to juveniles with a criminal record.  We hypothesized students idealized the role of "being professional" as avoiding 'personal feelings' or 'emotions'. In contrast to the majority of studies on empathy in medical students, which postulate a decrease of empathy throughout medical school, we found no statistical significant difference between 1st and 5th-year medical students. Our results are in line with previous Portuguese studies on the matter, suggesting that Portuguese medical school may have specific characteristics or teaching methods that preserve empathy. On the relationship between empathy and anxious, irritable and, cyclothymic temperaments, we found that they are inversely correlated. As such, the characteristics of an individual medical student may have a major role in the way he will empathize with future patients and colleagues. Finally, we found no correlation between empathy and moral competence, corroborating the findings in previous publications, where it was shown that psychopaths have the same ability to distinguish morally wrong and good.

From the results of our work, it is of utmost importance to admit students into medical school based not only on their curriculum and grades but also on individual characteristics, such as temperament and callous traits. Based on these principles, we will be able to train doctors with higher empathy levels, providing better standards of care in the future.