Kataoka, Hitomi U. MD; Koide, Norio MD; Ochi, Koji MD; Hojat, Mohammadreza PhD; Gonnella, Joseph S. MD
Empathy is essential in achieving optimal outcomes in patient care. Gianakos1 referred to empathy as “the ability of physicians to imagine that they are the patient who has come to them for help.” Empathy is essential to optimal patient care because it is the basis of a meaningful patient–doctor relationship. It represents the capacity of the physician to “stand in the patient’s shoes” and view the world from the patient’s perspective. Because empathy is a necessary skill for providing better patient care, it is crucial for medical school faculty to educate students on the importance of empathy as an integral part of “professionalism” in medicine.
Although the importance of empathy has been recognized and emphasized, the concept of empathy has a long history marked by ambiguity in its definition and measurement.2 Several researchers proposed treating empathy as an attitude. Larson and Yao3 proposed that physicians consider empathy as emotional labor. Hojat and colleagues2(p 80) at Jefferson Medical College proposed the following definition of empathy in the context of patient care: “Empathy is a predominantly cognitive (rather than emotional) attribute that involves an understanding (rather than feeling) of experiences, concerns and perspectives of the patient, combined with a capacity to communicate this understanding.”
The designers of the Jefferson Scale of Physician Empathy (JSPE) created the scale to measure empathy specifically in medical students and physicians in the context of patient care.4–7 Using the JSPE, various interesting findings have been reported. For example, in the analysis of empathy scores with respect to gender and specialty, women consistently scored higher than men.5,6 With control for gender, psychiatrists scored significantly higher than did physicians specializing in anesthesiology, orthopedic surgery, neurosurgery, radiology, cardiovascular surgery, obstetrics–gynecology, and general surgery.5
In examining the relationship between empathy scores and academic performance among third-year students at Jefferson Medical College, empathy scores were significantly associated with ratings of clinical competence. Students with high empathy scores received higher clinical competence ratings in six major clerkships.7 Also, students’ JSPE scores declined significantly from the end of the second year to the end of the third.8 In a cross-sectional study of American medical students, first-year medical students obtained the highest scores on the JSPE, whereas fourth-year students scored lowest.9 Similar findings were also observed in a longitudinal study of medical students using a measure of emotional empathy.10 It is also noteworthy that medical students expressing a preference for people-oriented specialties at the beginning of medical school, prior to being exposed to formal medical education, obtained a higher average on the JSPE than those expressing a preference for technology-oriented specialties.11
We designed this study to examine psychometrics of a Japanese version of the JSPE with Japanese medical students, and to test differences in empathy scores between men and women and among students in different years of medical school. Japanese medical school has different admission practices and a different curriculum compared with the United States. The entrance examination for medical school in Japan is heavily science oriented, and students enter medical school directly after graduating from high school, usually at age 18. Students complete six years of medical school before graduating. We focus on differences between the Japanese medical education model and others that may influence empathy in Japanese medical students.
Of the 600 medical students at Okayama University Medical School in Japan, 400 volunteers completed the JSPE in 2006 and 2007. Students were not compensated for their participation.
The student version (S-Version) of the JSPE used in this study includes 20 items answered on a seven-point Likert-type scale (1 = strongly disagree, 7 = strongly agree). Satisfactory evidence in support of psychometric properties of this scale has been reported.2,4–8 The JSPE has been receiving international attention by researchers and has been translated into 25 languages including Belgian, Brazilian, Chinese, Chilean, Dutch, French, German, Greek, Hebrew, Hungarian, Italian, Korean, Lithuanian, Norwegian, Persian, Peruvian, Filipono, Polish, Portuguese, Romanian, Spanish, Taiwanese, and Turkish.2
The JSPE was first translated into Japanese by one of the authors (H.U.K.). By using the back-translation procedure, the original translated version was sent to four bilingual Japanese physicians, two in the United States and two in Japan. They were asked to translate the Japanese version of the JSPE back to English. Then, two authors (H.U.K. and M.H.) reviewed the four back-translated versions to detect inconsistencies. Inconsistencies were resolved after two iterations by the bilingual translators (copies of the Japanese version can be requested either from Dr. Kataoka or Dr. Hojat). In 2006 and 2007, we distributed the final translated version of the JSPE to the first-, second-, fourth-, and fifth-year students during their regular classes. For the third- and sixth-year students, the JSPE was administered after one of their examination sessions. Students took the test individually in the classroom. We explained that the instrument was a questionnaire about empathy and that we would use the results for research purposes. The study was approved by the university’s research ethics committee.
We calculated Pearson correlation coefficients to examine item-total score correlations. Item-total score correlations were calculated based on responses to each item and total score of the JSPE minus the corresponding item. We used factor analysis (principal component factor extraction, followed by varimax rotation) to examine the underlying components of the Japanese version of the JSPE. Then we calculated the Cronbach coefficient alpha to assess the internal consistency aspect of reliability of the instrument. Further, we compared empathy scores for men and women by using a t test, and we used analysis of variance to examine score differences in different years of medical school.
Of the 400 respondents, there were 275 men (68.75%) and 103 women (25.75%) (22 did not specify their gender). There were 50, 56, 96, 65, 47, and 86 students in the first, second, third, fourth, fifth, and sixth years of medical school, respectively. The group of respondents comprises 66.7% of all students at the school. Although the response rate is relatively low, an examination of gender composition of the sample and total class confirmed that the sample represents the total population with regard to gender. Thus, we consider that their characteristics are representative of the whole group.
Summary results of factor analysis of data for the 20 items of the JSPE are reported in Table 1. As shown in the table, five factors emerged, each with an eigenvalue greater than one, accounting for a total of 53% of variance before rotation. Factor 1, which accounted for 24% of the variance, is a major component (grand factor) that can be labeled “perspective taking” based on the content of the 10 items with factor coefficients greater than .25 (Table 1). All these items also emerged as the grand factor of perspective taking in studies of American medical students2 and physicians5 and Mexican medical students.12 The perspective taking component has been described as a major dimension of empathy in patient care.2
Eight items had coefficients greater than .25 on Factor 2, a construct of “compassionate care,” that also emerged in American and Mexican samples of medical students and physicians.2,5,12 This component has been considered an important aspect of the physician–patient relationship.2 Factor 3 can be considered “empathic understanding” based on the content of five items with factor coefficients greater than .25 for this factor. This factor did not emerge among American2,5 and Mexican12 samples. It is interesting to note that each of these five items also had factor coefficients greater than .25 on the first factor, indicating that the two constructs of perspective taking and empathic understanding represent two separate dimensions of empathy in Japanese culture though they share a somewhat common substance.
Factors 4 and 5 were trivial, with factor coefficients greater than .25 on only two and three items, respectively. Factor 4 may be considered a construct involving “difficulties in taking patient’s perspective,” and Factor 5 is similar to the factor of “ability to stand in patient’s shoes” which emerged in American samples.2,5
These findings generally confirmed the three factors of “compassionate care,” “perspective taking,” and “ability to stand in patient’s shoes” that emerged in American2,4,5 and Mexican medical students.12
Descriptive statistics at item level
An examination of the item statistics showed that scores for each item ranged from a minimum of 1 to a maximum of 7, indicating that the study participants used the full range of all 7 points on the scale for each item. The mean score for the items ranged from a low of 3.3 to a high of 6.1. The standard deviations of item scores ranged from 1.1 to 1.7 (Table 1).
Item-total score correlations were all positive and statistically significant (P < .01), ranging from a low of .16 to a high of .55 (Table 1). The median of item-total score correlations was .42. The total score in calculating item-total score correlations was the sum of all items minus the particular item used in the corresponding correlation. The results of item-total score correlations generally confirmed that the direction of scoring was appropriate (indicated by positive correlations) and that all items contributed substantially to the total empathy score (indicated by statistically significant correlations).
Descriptive statistics of the scale
The mean, standard deviation, quartile points, and reliability coefficient of the scale are reported in Table 2. As shown in the table, the scores for the entire sample ranged from 56 to 134 (possible range 20–140). The mean score was 104.3 (standard deviation = 13.1), and the median was 106. The 25th and 75th percentiles were 97 and 114, respectively. The score distribution and related cumulative percentages for the entire sample are shown in Table 3.
The Cronbach coefficient alpha, an indicator of the internal consistency reliability of the measuring instrument, was .80, which is in an acceptable range for psychological measures. A similar reliability coefficient was reported for American medical students (r = .80).2 The Cronbach coefficient alpha in our study is higher than that reported for Mexican medical students (r = .74).12
Means and standard deviations for students who specified their gender (275 men, 103 women) are reported in Table 4. As it is shown in the table, women outscored men by more than three points (mean scores for women and men were 107 and 103.7, respectively). The gender difference was statistically significant (t(376) = 2.2, P = .02). The largest gender difference was observed on item 18 regarding interest in reading nonmedical literature or the arts (see Table 1) (mean for men = 5.8, mean for women = 6.2, t(376) = 2.2, P = .02).
Comparisons in different years of medical school
Mean scores and standard deviations of empathy for students in different years of medical school are reported in Table 4. As shown in the table, the mean empathy scores increased from 98.5 in the first year to 107.8 in the last year of medical school. Analysis of variance showed that the differences on mean scores in different years of medical school were statistically significant (F(5,394) = 3.6, P = .003). Duncan post hoc mean comparison test indicated that the mean scores for years 2, 3, 5, and 6 were significantly higher than for the first year of medical school.
The scale, as translated into Japanese by bilingual experts using back-translation procedures, proved to be psychometrically sound and maintained its basic underlying components. The findings of this study support the construct validity of a Japanese translation of the JSPE (S-Version) based on findings of factor analysis that replicated the three factors that emerged in American and Mexican samples. Also, the reliability of the Japanese version was as high as its original English form.
The mean score for the Japanese sample is lower than that reported by Hojat2 for American medical students (mean = 115, SD = 10, n = 685; t(1,083) = 15.8, P < .01) and by Alcorta-Garza and colleagues12 for Mexican medical students (mean = 110.4, SD = 14.1, n = 1,022; t(1,420) = 8.2, P < .01). Variation in the selection and education of medical students in different countries, the availability of appropriate role models, and expression of empathy in different cultures2 can partially explain the empathy score disparity in different cultures. The heavily science-oriented selection system and six-year curriculum (liberal arts for two years and medicine for four years) that are characteristic of Japanese medical school may help to explain this situation. Also, because students are separated from the clinical environment during their first four years, students only rarely encounter clinical role models until their final two years of training. Cultural differences, too, may contribute to the difference in scores; most Japanese patients prefer their physicians to be calm and unemotional. These factors might affect the expression of empathy among Japanese medical students.
That female Japanese medical students scored higher than their male counterparts is consistent with the findings reported for American medical students,2,4–6 American dental students,13 and Mexican medical students.12 The gender difference in empathy has been attributed to intrinsic factors (e.g., evolutionary-biological gender characteristics) as well as extrinsic factors (e.g., interpersonal style in caring, socialization, and gender role expectation).2,5
Our findings regarding enhancement of empathy during medical school in Japanese students are not in agreement with those recently reported for American medical students8–10 and American dental students.13 Several factors can explain the inconsistency.
First, it should be noted that our findings on enhancement of empathy in Japanese medical students are based on a cross-sectional design, thus the differences could be attributed to the baseline differences among students in different years of medical school. However, there were no differences in the students’ traits and abilities for those entering medical school in different years, because we have not changed the admissions policies and the entrance examination used for these six years. A more desirable approach to examine changes in empathy during medical school is to employ a longitudinal study design to follow up a cohort during their medical education.
Second, differences in admissions, curriculum, and cultural factors can partially explain the score differences in different countries. For example, the entrance examination to medical school in Japan is heavily science oriented, prompting high school students to focus almost exclusively on science subjects (biology, chemistry, mathematics, physiology) at the expense of the humanities and arts. The 18-year-old high school graduates in Japan enter medical school directly and spend six years completing undergraduate medical education. Entrance to medical school is highly competitive in Japan. Thus, high school students who are interested in going to medical school tend to concentrate heavily on studying science, subjects with little time for extracurricular activities or for development of skills related to “professionalism” before entering medical school. We consider this as one of the reasons why the baseline score on empathy is lower at the beginning of medical school in Japan compared with American medical students. The American students often come to medical school from diverse backgrounds in humanities and arts as well as science.
The Japanese medical school curriculum is different from that of the United States. For example, first- and second-year medical students at Okayama University mostly study premedical courses such as arts and science. They spend only one day a week studying basic medical science. The first two years of medical school in Japan correspond to undergraduate education in the United States. Students can choose several courses from humanistic classes such as philosophy, ethics, economy, and literature. They should choose several courses from science classes, in addition to basic subjects such as foreign languages. The increase in empathy scores as students progress through medical school could be due to the inclusion of humanities and arts early in their medical education. In addition, several special courses might contribute to the large increase in empathy scores between first- and second-year students. Our school has an “Introduction to Medicine” course, which offers various chances for students to think about the history of medicine, bioethical issues, life and death, and communication skills. Many Japanese medical schools have such courses to motivate students. Another possibility is that empathy has not been cultivated yet for first-year students with low empathy scores, as they have just been freed from the severe struggle to pass the entrance examination.
Third-year students study basic medical science, and they have many opportunities to engage in problem-based learning. In addition, Okayama University has a unique “research internship” program. For three months, students participate in research projects in various laboratories. As many as 30% of students study abroad. Fourth-year students study clinical science, and they have limited opportunities to interact with standardized patients near the end of the year. However, lectures occupy the first nine months of the fourth year. The learning style is monotonous and passive compared with other years. After passing the OSCE examinations, fifth-year students can rotate on clinical services and participate in teams responsible for patient care. During the sixth year, students can play a more active role in their rotations. Thus, during the first four years, students have rare or limited opportunities to participate in patient care and contemplate the physician–patient relationship. The medical education curriculum at Okayama University is typical of Japanese medical schools.14 The students enjoy clinical clerkships because they feel that they finally are given the opportunity to behave as physicians. This excitement may be the reason for students’ growth in empathy.
In addition, one remarkable difference between Japanese and American students is the degree of stress during medical school. In most medical schools in the United States, students are required to pass USMLE Step 1 before beginning their clinical clerkships and USMLE Step 2 during the fourth year. They know that they must perform very well to match to the postgraduate program of their choice. These factors generate challenges and stresses that often become intense. In contrast, Japanese students have one national licensure examination at the end of the sixth year. Although they also have a national achievement exam at the end of the fourth year, it is not as critical as the USMLE Step 1 for American students. Thus, Japanese medical students may be under less stress caused by relentless evaluations and examinations when compared with their American counterparts.
Cultural traits are also important in affecting empathic engagement in Japanese students. Nonverbal communication is subtle. In general, Japanese people rely less on physical gestures and facial expression to indicate being calm and in control than do their counterparts elsewhere. Also, many Japanese patients may hesitate to show their personal feelings or emotions to others, including medical staff. It is speculated that these culture-specific characteristics can result in differences in empathy scores between American and Japanese samples. However, this speculation needs empirical verification in future research.
Partly because of biotechnological developments and partly because of the changes in the health care system, it has been argued that in the contemporary system of medical education and patient care, insufficient attention is paid to human aspects of medical education and patient care. Given this universal trend, it is important and timely to study factors that contribute to improving interpersonal relationships in the context of medical education and patient care in different countries.
Therefore, a psychometrically sound measure of empathy that can be used in different cultures will help to better understand the predictors and outcomes of empathy and to empirically examine similarities and differences in factors that contribute to the enhancement of empathy in medical education and the practice of medicine in different corners of the world.
The authors appreciate the following bilingual Japanese physicians who helped in translation and back-translation procedures of the Jefferson Scale of Physician Empathy: Dr. Takami Sato, professor, Department of Medical Oncology, Thomas Jefferson University; Dr. Akira Nishisaki, Department of Anesthesiology and Clinical Care Medicine, The Children’s Hospital of Philadelphia; Dr. Takaaki Mizushima, associate professor, Department of General Medicine, Okayama University Medical School; and Dr. Yoshio Nakamura, associate professor, Department of General Medicine, Okayama University Medical School. In addition, the authors acknowledge Dr. Takaaki Mizushima and Ms. Kimi Hashimoto, who helped in data collection and data entry. They also thank Dorissa Bolinksi for her editorial assistance.
1 Gianakos D. Empathy revisited. Arch Intern Med. 1996;22:135–136.
2 Hojat M. Empathy in Patient Care: Antecedents, Development, Measurement, and Outcomes. New York, NY: Springer; 2007.
3 Larson EB, Yao X. Clinical empathy as emotional labor in the patient–physician relationship. JAMA. 2005;293:1100–1106.
4 Hojat M, Mangione S, Nasca TJ, et al. The Jefferson Scale of Physician Empathy: Development and preliminary psychometric data. Educ Psychol Meas. 2001;61:349–365.
5 Hojat M, Gonnella JS, Nasca TJ, Mangione S, Vergare M, Magee M. Physician empathy: Definition, measurement, and relationship to gender and specialty. Am J Psychiatry. 2002;159:1563–1569.
6 Hojat M, Gonnella JS, Nasca TJ, Mangione S, Veloski JJ, Magee M. The Jefferson Scale of Physician Empathy: Further psychometric data and differences by gender and specialty at item level. Acad Med. 2002;77(suppl):S58–S60.
7 Hojat M, Gonnella JS, Mangione S, et al. Empathy in medical students as related to academic performance, clinical competence and gender. Med Educ. 2002;36:522–527.
8 Hojat M, Mangione S, Nasca TJ, et al. An empirical study of decline in empathy in medical school. Med Educ. 2004;38:934–941.
9 Chen D, Lew R, Hershman W, Orlander J. A cross-sectional measurement of medical student empathy. J Gen Intern Med. 2007;22:1434–1438.
10 Newton BW, Barber L, Clardy J, Cleveland E. Is there hardening of the heart during medical school? Acad Med. 2008;83:244–249.
11 Hojat M, Zuckerman M, Magee M, et al. Empathy in medical students as related to specialty interest, personality, and perceptions of mother and father. Pers Individ Dif. 2005;9:1205–1215.
12 Alcorta-Garza A, Gonzalez-Guerrero JF, Tavitas-Herrera SE, Rodrigues-Lara FJ, Hojat M. Validity of the Jefferson Scale of Physician Empathy among Mexican medical students [in Spanish]. Salud Mental. 2005;28:57–63.
13 Sherman JJ, Cramer A. Measurement of changes in empathy during medical school. J Dent Educ. 2005;69:338–345.
14 Rao RH. Perspectives in medical education-2. A blueprint for reform of medical education in Japan. Keio J Med. 2006;55:81–95.