Emotional intelligence (EI) is described as the ability to understand and recognize emotional states and to use this understanding to manage oneself and others; it is also considered a teachable skill (1). The BarOn Emotional Quotient Inventory (EQ-i) is a validated tool to measure EI (2). EI assessment is becoming more prevalent in the healthcare field, with recent studies related to healthcare administration (3), medical school curricula (4), and nursing training (5). The EQ-i has also been used to assess educational needs within a surgical residency program (6) and to assess anesthesiology residents’ clinical performance (7).

Many residency programs select an administrative chief resident during the senior year of residency training. It is considered an honor and sometimes comes with a stipend, as the position carries extra administrative responsibilities. The selection process varies among different residency programs; most use input from the residents. It is assumed that administrative chief residents are selected in part because they are people who are organized, listen, manage stress well, are respectful to others, and have good communication skills. In other words, they are likely to possess high emotional intelligence. Our objective was to determine whether EI is higher among residents selected as administrative chief resident than among other residents in their senior year of training.

We performed an observational, cross-sectional study, using the BarOn EQ-i. We invited residents in their senior year of training, referred to as chief residents, at the University of Texas Health Science Center at Houston (UTHSCH) within the departments of anesthesiology, emergency medicine, family medicine, obstetrics and gynecology, pathology, and psychiatry to participate in the study. We included these departments since they selected an administrative chief resident and did not require the administrative chief resident to complete an additional year of training.

We sent the initial invitation by e-mail in October 2009, followed by a reminder e-mail 1 month later, in November 2009. We allowed another month to collect de-identified data. The e-mail invitation included 1) the purpose of the study; 2) investigators’ names; 3) informed consent; 4) anonymous and confidential nature of the test results; 5) assurance that no test results were to be placed in the resident’s personal file; and 6) a sample report detailing the type of information and feedback they were to receive upon completing the test. Once the resident agreed to participate and had read the informed consent, they were referred to a link with a unique identification number and password. We also required the participant to fill out a demographic-information form. The demographic form contained their age, race, gender, specialty, previous work experience before residency training, and whether or not they had been selected to be administrative chief resident. Access to the EQ-i was through Multi-Health Systems (MHS) incorporated, which provided a confidential and secure website. The incentive provided to the resident for participation was a copy of their scores, with a development report that highlighted areas of strengths and tips for improvement in areas with low grades.

We collected the assessment of the EQ-i scores from MHS, which had no knowledge of whether or not the participant was selected to be administrative chief resident or any other demographic variables. To further reduce the introduction of bias in the analytic phase, the statistician was blinded to the study groups.

We obtained continuous-variable outcome data for each participant from MHS in the following format: total EQ-i score, 5 composite, and 15 subset scores. The five composite scores included interpersonal, intrapersonal, adaptability, stress-management, and general mood. Data from each of these scales are used to create a single composite EQ-I score. All scores are normalized to a national mean of 100. Effectively-functioning individuals score between 85 and 115 (defined as 1 standard deviation [SD] above or below the mean) (2). Scores on the EQ-i range from 50 to 150. We determined the study size, assuming a two-sided α error of 0.05 and a known SD of 15 for the EQ-I, and a ratio of 3:1, enrolling 44 residents (11 administrative chief residents and 33 non administrative chief residents) would result in 80% power to detect a 15-point difference in EQ-I score. We selected a 15-point difference in total EQ-i score because this represented one SD.

We used the statistical package STATA 10.1 (College Station, TX). We compared continuous variables by use of nonparametric analysis with the Wilcoxon rank-sum test. We used the chi-square test to compare categorical variables and a linear-regression analysis to determine possible predictors of total EQ-i scores based on demographic characteristics. We obtained IRB approval for the study from UTHSCH, and we report the data using the STROBE guidelines for observational studies.

We invited a total of 66 senior-year residents, of which 46 (69.6%), agreed to participate; 13 of 15 administrative chief residents (86.6%), and 33 of the 51 non-administrative senior residents (64.7%) participated in the study. Participation, varying between specialties, was as follows: anesthesia, 19/24 (79.2%); emergency medicine, 8/9 (89%); family medicine, 4/10 (40.0%); ob/gyn, 10/11 (90.9%); pathology, 2/7 (28.6%); and psychiatry, 3/5 (60.0%). All 46 completed the online EQ-i assessment and demographic information form. We found no statistically significant differences in the baseline demographic information between administrative and non-administrative chief residents. The average Total EQ-i score for all participants was 104.9 (SD: 12.9). Emergency medicine chief residents had the highest total EQ-I scores (110.5 [5.9]), followed by pathology (108.0 [7.0]), psychiatry (105.0 [19.6]), ob/gyn (104.7 [13.3]), anesthesiology (103.5 [14.8]), and family medicine (99.0 [12.1]).

Administrative chief residents scored higher on total EQ-i and all composite scores; however, these differences did not reach statistical significance. Using the total EQ-i score as the outcome variable and gender, age, year of residency, and history of employment as independent variables, we used a multiple linear regression to study possible predictors of EQ-i scores. None of the variables had a statistically significant association.

EI has the potential to influence two important competencies in the practice of medicine: interpersonal communication and professionalism. Both of these skills are important in residency training and are recognized as a core competency by the Accreditation Council for Graduate Medical Education (ACGME). Interpersonal communication skills and professionalism are crucial in daily interactions with patients and their families, fellow residents, faculty, ancillary staff, and other members of the healthcare team. It is especially vital for chief residents, who manage groups of residents and students, and even more so for administrative chief residents, who provide leadership within an entire program.

The strengths of this study include a high participation rate, adequate power to detect a 15-point difference between EQ-i scores, and multidepartmental representation of participants.

We recognize some weaknesses of this study. We invited residents from one institution, thereby limiting the generalizability of our results. Cross-sectional studies are limited in that they only report on an association at one particular point in time. The results of the EQ-i assessment and other similar assessments of personality have been shown to vary depending on the mood of the person assessed (8). Therefore, a study of EI that is limited to a single assessment may be overly influenced by the mood of the participant undergoing assessment at that time. The BarOn EQ-i, is a self report scale of emotional intelligence. Self-report scales, unlike an ability-based test, may overestimate the participants’ true ability in regard to EI (8). Also, we were not able to control for time when the resident took the EQ-i assessment, for example, while post-call or when tired, and we cannot say how this influenced our results. Some departments had a low response rate; limiting the conclusions drawn from their scores. Other studies with larger numbers have failed to demonstrate differences in regard to EI and medical specialties or medical-specialty selection (9). The selection process of administrative chief residents varied across the six departments that participated in our study; all senior-year residents were eligible and were voted on. Some of the administrative chief residents were selected solely by fellow residents (anesthesia, obstetrics, and psychiatry), whereas others had faculty and resident input (emergency medicine, family medicine, and pathology).

On the assumption that EI is teachable and improves with age, we felt it important to exclude those departments that required their administrative chief resident to complete an additional year of training, namely, pediatrics and internal medicine. Adding those individuals, we believe, would select out individuals with higher EI scores, although we did not see an improvement in EQ-i scores with respect to age. We also did not include surgery residents, as our surgery department does not select an administrative chief resident. We did not compare EI scores and actual clinical performance. We cannot comment on whether those with higher EQ-i scores exhibit greater interpersonal skills or professionalism while working in a clinical environment. EI has been reported to increase with age, and, in other studies, women have been shown to have higher scores (10). We did not observe these associations in our study, perhaps because of small subgroup numbers. Larger studies in the future could address demographic variables that influence EI among administrative chief residents.

We are asked to teach residents interpersonal communication and professionalism skills with little more than observation to evaluate these skills. Future studies should continue to examine the relationship of EI in regard to clinical skills, patient satisfaction, and patient care, and whether medical education fosters or impedes development of these skills. Larger, multi-institutional studies would likely need to be constructed to address these questions. Emotional intelligence, as measured by the BarOn EQ-i, was not significantly higher among those residents selected to administrative chief than among their fellow chief residents.

The authors thank the Larry C. Gilstrap, M.D., Center for Perinatal Research and Women’s Health for funding the study.