The moderator effect of sex on attitude toward communication, emotional intelligence, and empathy in the nursing field

ABSTRACT Objectives: to analyze differences in the variables for the object of this study (attitude toward communication, emotional intelligence, and empathy) according to sex; verify correlations among variables between men and women and analyze regression models according to sex. Method: the ATC was used to measure attitudes toward communication; the Jefferson Scale of Empathy was used to measure empathy; and the Trait Meta Mood Scale 24 was used to measure emotional intelligence. The sample was composed of 450 nurses working in 7 hospitals located in Valencia, Spain. The t-test for independent samples was used to verify whether there were statistically significant differences, together with a prior application of the Levene test to assess the equality of variances. The correlations were analyzed using Person’s coefficient. Finally, the Beta coefficients of variables predicting ATC’s dimensions were verified using hierarchical multiple linear regression according to sex. Results: There are statistically significant differences based on sex for the variables, correlations and power of prediction. Conclusions: This study presents evidence on how the levels of variables (attitudes toward communication, EI, and empathy) vary among nurses according to sex, as well as the relationships established among such variables.


Interpersonal communication between nurses
and patients is directly linked to care practices, as the basis of nursing care (1) and a central attribute of nursing care models. From this perspective, the communication skills of nurses are established as the central axis of quality of care (2) and, therefore, of patient satisfaction (3) . In this sense, the literature suggests the  (4) . Dispositional factors, or intrinsic variables, depend on the professional's characteristics: age, sex, background, years of work, attitude, empathy, and emotional intelligence (EI) (5) , to mention a few.
Communication is a dynamic, multidimensional and complex process (6) , thus, measuring it is difficult and requires multiple perspectives to be considered. Therefore, if communication with patients is considered a human behavior, its measurement should take into account attitudes. One's attitude toward communication is probably one of the main determinants of nurse communication, since there is strong correlation between attitude and behavior (7) . Additionally, communication can be influenced by a nurse's EI and empathy. EI enables nurses to properly regulate their emotions and those of others and is a very important requirement in key abilities, such as communication and empathy. Empathy is defined as one's ability to read the emotions of others, putting oneself in another's situation, understand one else's thoughts and feelings (8) . Therefore, nurses with empathic abilities can understand a patient and establish a supporting relationship (9) . Considering the relationships existing among the variables under study, there are other sociodemographic variables, such as sex, that can affect relationships among them. Therefore, we need to describe the role the sex of professionals plays in these relationships. Women generally pay greater attention to their emotions (10) and, for this reason, are more receptive to emotional support measures, while men are usually less attentive to their emotions (11)(12) .
This ability of women to pay greater attention to their emotions is an important resource for nurses, because it enables them to acquire a greater awareness of their own feelings and those of others, which is related to EI, and a greater understanding of a patient's situation, which is related to empathy, so that the support provided by women is more significant (13) . These aspects are greatly important for nurses, because, even though men are increasingly choosing nursing as a profession, it remains a highly feminized profession (14) . Despite the importance of these aspects, there is a lack of studies assessing the moderating effect of sex on the relationships among such variables. In general, studies suggest that nurses lacking EI and empathy do not have the ability to communicate effectively, with patients nor the staff, leading to an unfavorable work climate and an increase in care delivery errors (15) . Patients are able to achieve a maximum level of wellbeing only when care is performed with effective communication skills (16) . Finally, EI is related to communication skills and a high EI is the proof that empathy and social skills are present (17) .

Method
This study's population was composed of 450 nurses providing direct care to patients in 7 public hospitals in Valencia, Spain. Inclusion criteria were nurses actively working in the selected hospitals who had previously consented to participate in the study.

Data collection procedure
After obtaining authorization from head nurses, the participants signed free and informed consent forms. The nurses filled out the questionnaires (taking approximately 35 minutes) and deposited them in the boxes located in the different services. After 2 weeks, reminders were sent by email and after 3-4 weeks, the questionnaires were collected. A total of 1,124 questionnaires were distributed, 460 of which returned, and ten discarded, as less than 60% of the instrument

Data collection instruments
A self-administered instrument was used to collect data. The participants were supposed to respond to 3 instruments along with a form addressing sociodemographic data. The five-item Likert scale ranged from 1 to 5 (1=totally disagree and 5=totally agree). The instruments were the following: version was adapted by Fernández-Berrocal (19) and is intended to assess EI. It also presents appropriate psychometric properties: Emotional attention, Cronbach's alpha=0.80; Emotional clarity, Cronbach's alpha=0.87 and Emotional Repair: Cronbach's alpha=0.85.

Data analysis
The differences in the variables that are the object of this study were first analyzed according to sex, then the correlations among the variables were calculated for men and women and, afterwards, regression models were also analyzed according to sex. To confirm whether there were statistically significant differences, a t-test for independent samples was used together with a prior application of the Levene test to prove equality of variances. The correlations were analyzed using Person's coefficient. Finally, the Beta coefficients of the predictor variables of the ATC's dimensions were determined according to sex using hierarchical multiple linear regression.

Ethical aspects
The study was approved by the Institutional Review Board at the University of Valencia (H1432032268924) and at the selected hospitals. All the participants consented after having received clarification regarding the study's objectives and procedures and were ensured of the confidentiality of information provided.

Participants' sociodemographic characteristics
The participants' ages ranged from 22 to 64 years old, with an average age of 44. 13

Correlations among the ATC, JSE and TMMS24 according to sex
There is a statistically significant correlation among  Table 2).

Discussion
This study's results reveal statistically significant those reported by other studies, in which scores are slightly higher among women than among men (13,(20)(21) .
No significant differences were found between men and women in regard to the 3 dimensions of the TMMS24.
Future studies should deepen investigation of these data as the literature (19) confirms differences in terms of EI between men and women; i.e., women focus better and pay greater attention to their feelings. One potential explanation for a lack of difference between women and men in this specific case is that there are more women than men in the sample, a situation that is quite common in the context of nursing (14,22) . In the group of women, a statistically significant correlation was found among most dimensions of the scales under study, except between ATC's dimensions and TMMS24's emotional attention.
In the group of men, all the variables were significantly patients. Once again, the overall differences between men and women in regard to emotions (23) , in addition to differences assigned to the extrinsic characteristics of socialization and intrinsic characteristics of learned gender roles, may explain these results (24) .
One of the main limitations of this study is the sample. Non-probabilistic sampling procedures generally do not represent the population of nurses, and, even though a greater proportion of women was found, the results cannot be generalized. Another limitation is the use of self-reported instruments, a tool commonly used in investigations but that may introduce bias known as social desirability bias (25) .
Therefore, other types of questionnaires and/or external objectives measures are recommended to detect differences between sexes. All these limitations will be taken into account in future investigations.

Conclusion
This study presents evidence on how the levels of variables (attitudes toward communication, EI, and empathy) vary among nurses according to sex, as well as the relationships established among such variables.
These findings enable the planning and assessment of nursing training programs to improve the levels of these variables among nurses. The assessment of these variables is essential given their repercussions for the quality of nursing care, thus for patient satisfaction.