Empathy Gender in Dental Students in Latin America: an Exploratory and Cross-sectional Study

Background: It is well-founded that empathy is an attribute that increases the likelihood of good

The professional relationship between health practitioner and patient must be considered a two person interaction of which both have different personal interests [1], constitutes a human encounter.This relationship contains, itself, an eminent subjectivity and inter-subjectivity that goes beyond the purely clinic dimension of a treatment [2].
The empathy during health care can be understood as a cognitive and behavioral attribute, which implies the capacity to understand how the experiences and feelings of a patient can influence and can be influenced by illness and its symptoms, and the capacity to communicate that understanding to the patient [3].This constitutes one of the elements needed to develop basic communication skills for the human relationship that is performed, voluntarily in form [4].
Investigations in the professional area of health show that the empathy has been related, theoretically and empirically, to diverse attributes, such as the pro-social behavior, the ability to obtain clinical history, has increased the level of patient and doctor satisfaction, as well as good clinical results [2] [5] [6].
Various authors have proposed that women show higher levels of empathy compared to men [7]- [9] and those empathy measurements, of those investigations, have been performed with different instruments developed for the general population and the medical field.On the other hand, some authors have proposed, with theoretical basis and less empirical evidence, that empathy could be a "variable" that is subjected to the influence of several factors, in addition to the gender [10]- [12], as well as age, intent about the specialty to follow in the future, the current course the student is taking, structure and family environment, personality, empathetic experiences, socio-cultural environment, scale of ethical and moral values, among others; which could act as independent "variables" or confounding variables, and at the same time, could contribute to explain the variability observed in empathetic orientation levels found in some research [13]- [16].
There are many researches in which empirical results have been found that contradict the fact that women are more empathetic than men [17]- [19].However, more empirical evidence is required to demonstrate the real existence of this contradiction.The purpose of this investigation is to determine if indeed the levels of empathy are higher in women, in relation to the male students of 18 dental faculties of six countries in Latin America.

Materials and Methods
This is an exploratory, non-experimental, descriptive, transversal and ex post facto cause-effect investigation, bio-ethically governed by the rules of Helsinki(was approved by the Ethics Committee of Research, Development University and German Clinic with code CAS-UDD approval: 2011-64 in Santiago de Chile).The studied population was composed of students of first through fifth year from 18 dental faculties of six countries in Latin America (Dominican Republic, Costa Rica, Panama, Colombia, Argentina and Chile) (n total = 4407) (    Stratified samples by gender were obtained from this population.The collection of data was performed between July and August 2012.Since the students were able to visit different clinic areas, attend classes in different places, in addition to be absent from classes, among other circumstances, it was not possible to apply the scale to all the students.The scale was not applied to the ones who were absent, (due to the reasons before mentioned) nor on the second time in order to avoid possible skewed answers.The Jefferson Scale of Empathy (JSE) in the Spanish version for students of Medicine (version S) was used in each of the participating countries, based on criteria Alcorta-Garza et al. [10] and Rivera et al. [20], was applied to those participating students in classrooms or clinic rooms, using just one anonymous and confidential measurement by a neutral operator.Before the Jefferson Scale of Empathy (JSE) was applied, it was submitted to board (a committee composed of five relevant academic members of the Psychology and Dentistry field or related to), in order to verify the cultural and content validity [2].A pilot study was created for the purpose of checking the students comprehension of the scale culturally adapted.There was not judgment of exclusion, since the objective was to evaluate the variable of interest of as many students as possible.

Statistical Analysis
The primary and original data from the empathy levels of each university examined were submitted to the Cronbach's alpha test (reliability by internal consistency).The sum of the primary data score, obtained using the scale previously mentioned, was initially submitted to the Kolmogorov-Smirnov normality test (K-S) in both types and to the Homoscedasticity Levene test.Descriptive statistics were estimated, arithmetic mean, absolute deviation and standard deviation of these sum.Comparisons between both genders mean were performed by Student's t-test, considering the presence of equal variances.The effect size was measured by the Hedges g.Data was processed by the statistical program SPSS 20.0 TM.The significance level used was ≤0.05 in each case.

Results
The K-S test was not significant (p > 0.05), in any of the data groups, which means that the data observed were distributed normally.The Cronbach's alpha test estimated for each gender and university fluctuated between 0.768 and 0.834, which shows that the data have values of acceptable internal consistency.Table 1 shows the results of the descriptive statistics estimation of empathy levels, estimated for each gender and each of the studied universities, and Table 2 shows the results of the total mean estimation in both examined gender.This last table shows that women have superior empathy levels compared to the men, and that the student's t-test was highly significant (p < 0.0001); however, the value statistically adjusted was 0.199, which shows that the effect size is low, therefore, the magnitude of the differences between means is small.The results of estimation of equal variances between the studied gender in each university, the comparison of the means between the gender in each university and the effect size corresponding to each of the comparisons are presented in Table 3.The F-Test was not significant (p > 0.05) in all cases, with the exception of the comparison between the gender variances in the San Martin University (Barranquilla, Colombia) (p < 0.01).
The student's t-test was not significant (p > 0.05) in the following universities: Antofagasta (Chile), Andrés Bello (Viña del Mar branch, Chile), Universidad Latina and from Panama (Panama), Universidad Metropolitana and San Martin (Barranquilla, Colombia), Universidad Nacional Cayetano Heredia and Nacional Mayor de San Marcos (Lima, Peru), Universidad de Costa Rica (San José, Costa Rica) and Universidad Central del Este (Dominican Republic).
Significant differences (p < 0.05) were found in the rest of compared universities.Of all the universities where no significant differences were found, five of these absolute values of the means were higher in women; in three of them the absolute values were higher in men and in two of them, the means were practically the same (Table 1).
The effect size sample was low (lower then 0.2) in six of them and in the rest the effect was medium-sized (between 0.2 and 0.5) (Table 3).In the case of the universities where statistical differences were found, in all cases the values of the empathy levels were higher in women than in men (Table 1).However, from these statistical differences, the only one that had a high effect size was the University Católica de Córdoba (g = 0.632); meanwhile in the rest of the universities the value of the effect size sample fluctuated between 0.209 and 0.438 which is considered medium-sized (Table 3).

Discussion
The studies of gender empathy level distribution initiated with works of Block [21] and followed by Hoffman [22].The first one did not find differences between men and women; however, the second one found that women had more points than men regarding Affective Empathy (AE), but the opposite happened regarding Cognitive Empathy (CE).Later on, Eisenberg and Lennon [23] had the same results as Hoffman [22].Those differences were attributed to the gender role stereotypes.Other studies have supported those conclusions [24]- [26], but using different measurement tools.Fernández-Pinto et al. [27], suggested two possible generalizations of these results: 1) that the empathy seems to be connected to individual differences, such as personality and gender differences, and 2) the results of the research made seem insufficient to come to a conclusion about the interrelationships named, since the results come from the measurement of empathy using different tools and, therefore, underlie different concepts of conceptualizations of the empathy behind each one of the measurement types.This situation makes these results incomparable, stimulates the ambiguity and even its apparent contradiction.
Additional studies developed with students from different health and geographic areas [28]- [30], agree with the fact that women score higher empathy measures than men, with the exception of a study realized with dental students in Malaysia, in which differences between both genders were not found and men scored absolute values higher than women [31].In Latin America (LA) some studies were developed about the empathy levels in different schools of dentistry and the results around the gender differences are contradictory.Studies have been performed by Gutierrez et al. [32], Salcedo et al. (results sent to publish) in Perú; Carrasco et al. [33], Huberman et al. [34], López et al. (results not published), González et al. (results not published), and Silva et al. [35] en Chile; Silva et al. [36] in Dominican Republic; Bilbao et al. (results sent to publish), Erazo et al. [37], González et al. (data sent to publish), Pérez et al. (data sent to publish) in Colombia; Howard et al. [38], Sánchez et al. [39] in Costa Rica, Bullen et al. (results sent to publish); Gordon et al. (results not published) in Panamá, Varela et al. [40] in Argentina.All these results show that there is variability in the answer of empathy levels in both students outside and inside from all the different universities considered in this study [41].The totality of authors previously mentioned developed their research in LA with the Jefferson Scale of Empathy (JSE), adapted to the health area researched, therefore, the difference of tools seems not to be the variability source and they can be compared not only by the scale used, but by the same methodological conceptions and statistics with which these studies were developed.
Vera [41], analyzing some of these studies about empathy in dental students from LA, sets three generalizations: 1) the empathy represents a variable behavior; 2) there is a tendency to show higher levels of empathy in female and in higher courses and 3)these variables do not explain more than 20% of the variation found.The authors from this study completely agree with the generalization 1) and 3), but the data observed in this current study shows variability in the gender and, therefore, it is not possible to coincide with the generalization 2), in relation to the "gender" factor.The studies previously mentioned, show that the variability is necessary to be explained, mostly, if required to make interventions that try to increase empathy levels in students of health science, by active processes of empathy teaching-learning.
The differences found could have some sources of explanation.Mercadillo et al. [42] propose that compassion is a moral emotion that determines a help behavior and the observation of the brain activity, after the students were submitted to watch images triggered to compassion; it showed that the compassion experience triggers by the experience of physical pain or illness, and it is associated to the experience of dislike, anxiety and dominance, characteristics of negative emotions.On other hand, the brain maps of this experience [42] show that women expressed a higher and more diverse activity in the basal and limbic zones of the brain, such as parahippocampal and temporal cortex (anger and sadness) and frontal areas involved in processing learned information and carrying out (intentions and making decisions).Men showed predominant activity in the orbitofrontal cortex (learning of moral concepts and social rules), but the behavioral results of their emotional experience did not show differences, which creates contradiction between what the students said what they felt and what their brains reflected.
Two proposals exist to explain this difference.Men as well as women feel compassion in similar ways, but feelings are processed by different ways in the brain: compassion felt by women goes with an empathy that can benefit the spread, and in men the feeling of compassion and their decision to help is directed by the moral judgment of the situation: "when someone suffers one has to help".The described differences can be explained by two proposals: 1) the evolutionist in which a woman owns a more sensible empathic system that assists the nurturing system; the high levels of oxytocin in women produce reactions like inhibiting pain faced with intruders, increases the anger to protect her children and builds an emotional bond towards them [43] and 2) the cultural proposal indicates that both the family and social education teach women and men to express their emotions in different ways.Both proposals do not have to be observed mostly as contradictions.Mercadillo et al. [42] comes to the conclusion that "even though the brain and body are different between genders, those differences do not determine the way we behave, they just lead the members of a gender to respond in an easier way to specific types of situations." These situations are valued as concepts, rules and ways of expression that get into our brains and we learn during our daily lives, culturally and can create ways of response and perception throughout human development [42].However, to obtain a major approach of which is the empathic behavior between gender, longitudinal studies are required, but not only related to location, since the result of this type would explain the local situation and what can be valid for a population not necessarily is valid for others, because of the possible influxes of some factors regarding the conformation of empathy.The social stereotypes (sexual role) influence the answers, since these stereotypes assign women the tendency of caring and supporting weak people, a bigger capacity to detect feelings and nonverbal signs and a major concern about the social aspects of interaction and other´s feelings [43] [44].
On the other hand, other authors proposed that empathy has an impact in the emotional health and in the social field throughout culture [4] [43]- [46], besides it is correlated with the pro-social behavior and altruism and also inhibits the unsocial and aggressive behavior [47] [48].There are circumstances that negatively influence empathy: levels of anxiety, claustrophobia, obesity, depression and stress [47]- [49].In summary, the variability of observed empathy levels in this study, related to gender, cannot be explained.This variability constitutes empirical evidence that the empathy construct is difficult to elucidate and that some of the possible explanations that have been used so far, do not have a general character (for example, gender role).

Conclusions
It is still early to make generalizations regarding the distribution of the empathy levels in gender, because of the possible following reasons: 1) the observed variability between the genders in different universities of LA can be just a limited fact of the studied countries and, therefore, other studies are required in order to confirm that this variability is a general characteristic in LA, or is only endorsed to the six countries studied in this document, and even, it could be just a characteristic of the students of dentistry; 2) at least, in the studied region in this current document, intervention with teaching-learning processes associated with empathy should not occur without a prior study of what are the factors of influence, how they influence and by how much.

Table 1 .
Estimation results of the descriptive statistics of the levels of gender empathy in each of the universities studied.

Table 2 .
Estimation results of the total median for each gender.

Table 3 .
Results of the comparison of means of both genders in each of the universities studied and estimation of the effect size (g).