Wang et al.

BMC Medical Education (2019) 19:341

https://doi.org/10.1186/s12909-019-1788-3

RESEARCH ARTICLE Open Access

Empathy, burnout, life satisfaction,

correlations and associated socio-
demographic factors among Chinese
undergraduate medical students: an
exploratory cross-sectional study
Qinghua Wang1, Lie Wang2, Meng Shi1, Xuelian Li3, Rong Liu4, Jie Liu4, Min Zhu5 and Huazhang Wu5*

Abstract
Background: Medical education is widely known to be a demanding process that may cause various mental health
problems, such as burnout, which can lead to lowered levels of life satisfaction among medical students. Research
shows that empathy is negatively correlated with burnout, but there are few studies on the relationship among
empathy, burnout and life satisfaction in medical students. The objective of the present study is to explore the
correlations of empathy and burnout with life satisfaction and the associated socio-demographic factors among
Chinese undergraduate medical students.
Methods: In this cross-sectional study, 1271 undergraduate medical students (age 19.42 ± 1.34 years, 36% male)
from 1st to 4th grades completed questionnaires including the Interpersonal Reactivity Index Chinese version (IRI-
C), the Maslach Burnout Inventory Modified Chinese version (MBI-MC), the Satisfaction With Life Scale (SWLS) and
socio-demographic characteristics. Statistical analyses included Student’s t-test, one-way ANOVA, post hoc
Bonferroni tests, hierarchical linear regression analysis and general linear model-univariate full factorial model.
Results: Over four academic years, medical students’ empathy levels declined, but their burnout levels almost
plateaued and their life satisfaction levels witnessed an initial fall before a rebound. Empathy was correlated with
students’ age and grade, and burnout was associated with students’ maternal education. Significant differences in
life satisfaction were detected with regard to medical students’ age, academic year, the number of children in the
family, place of residence and parents’ educational levels.
Conclusions: Empathy explained 0.6% of the variance in life satisfaction in contrast to 13.7% of the variance
explained by burnout in life satisfaction. Although empathy did not have a main effect on life satisfaction, there
was an interaction effect of empathy and burnout on life satisfaction among students of high and low empathy
and burnout levels. Students with high levels of empathy and low levels of burnout were most satisfied with life.
Medical institutions and related authorities need to find effective measures to enhance students’ empathy levels
and reduce burnout to improve their life satisfaction.
Keywords: Empathy, Burnout, Life satisfaction, Medical students

* Correspondence: [email protected]
5
Department of Healthcare Management, School of Humanities and Social
Sciences, China Medical University, No. 77 Puhe Road, Shenyang North New
Area, Shenyang, Liaoning Province, People’s Republic of China
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Wang et al. BMC Medical Education (2019) 19:341 Page 2 of 10

Background American and Brazilian medical students in terms of men-

It is widely acknowledged that medical professionals are tal health, life quality, empathy and burnout. The results
subjected to heavy workloads, long work hours, a stress- indicated that US medical students scored significantly
ful work environment, high job pressure, and an intense higher in empathy as measured by the Empathy, Spiritual-
physician-patient relationship, which lead to various ity, and Wellness in Medicine survey (ESWIM) than their
mental health problems, such as anxiety, depression and Brazilian counterparts whereas American medical stu-
burnout. These psychological problems exert detrimen- dents were more likely to suffer from exhaustion as mea-
tal effects on the life quality and life satisfaction of sured by the Oldenburg Burnout Inventory (OLBI) [18].
health professionals, which in turn impact the quality of Studies consistently showed that burnout is signifi-
care they provide to patients [1, 2]. cantly negatively related to life satisfaction among med-
One often cited factor related to occupational burnout ical students and health professionals. Life satisfaction,
among medical professionals is compassion fatigue, as a concept concerning the cognitive aspect of subject-
which is closely associated with empathy [3, 4]. Empathy ive wellbeing, is defined as an individual’s conscious
refers to the ability to stand in others’ shoes, to under- evaluation of his or her life based on self-set standards
stand as well as to share other people’s emotions and [19]. The research conducted among 452 pharmacy stu-
feelings and the ability to communicate this understand- dents from five South Korean universities using struc-
ing [5]. Empathy is an essential element in the tural equation modelling (SEM) found that exhaustion
physician-patient relationship and affects clinical out- and cynicism, which are two facets of the Maslach Burn-
comes, patient compliance and satisfaction [6, 7]. There- out Inventory—Student Survey (MBI-SS), were both sig-
fore, great importance is attached to empathy in nificantly negatively correlated with life satisfaction [20].
undergraduate medical education, the critical phase in A three-wave seven-year prospective study among 3255
nurturing competent physicians, and the development of Finnish dentists revealed that burnout predicted depres-
empathy in medical students is considered a goal in sive symptoms and lowered life satisfaction [21]. Demer-
medical education [8]. outi and colleagues detected a significantly negative
The literature indicates that occupational burnout may relationship between exhaustion and disengagement,
have its root in the medical schools where health profes- which are two facets of the Oldenburg Burnout Inven-
sionals were trained as students. In addition to stress, anx- tory (OLBI), and life satisfaction among 109 German
iety and depression, burnout is known to be another nurses [22]. Although previous studies have confirmed
mental health problem prevalent in medical students. the associations among empathy, burnout and life satis-
Burnout is defined as a syndrome of prolonged emotional faction, to the best of our knowledge, there has been no
exhaustion, depersonalization (cynicism) and a feeling of study exploring the possible interaction effect of em-
professional inefficacy [9]; possible reasons that medical pathy and burnout on life satisfaction in medical stu-
students are prone to burnout problems are the intensive dents. Additionally, whereas some previous studies have
medical curriculum, lack of social support and inappropri- examined the associated socio-demographic factors such
ate coping styles. Studies show that burnout can occur as age, gender and grade with empathy, burnout and life
from the outset of undergraduate medical education dur- satisfaction, no research has examined the correlations
ing preclinical years [10], and the problem becomes more of socio-demographic factors such as place of residence
serious with the increase of grade levels [11]. and parents’ educational levels with empathy, burnout
The negative association of empathy with burnout and life satisfaction simultaneously in Chinese under-
among medical professionals is recorded in a large num- graduate medical students. Therefore, the aims of the
ber of studies [1–4, 12–15]. Regarding the correlation of present study were to (i) investigate the associated socio-
empathy and burnout among medical students, cross- demographic factors with empathy, burnout and life sat-
sectional multi-institutional research conducted in Brazil isfaction and (ii) determine whether there is an inter-
concluded that personal accomplishment, one facet of the action effect of empathy and burnout on life satisfaction
Maslach Burnout Inventory (MBI), held the strongest in Chinese undergraduate medical students.
negative association with personal distress, which is a sub-
scale of the Interpersonal Reactivity Index (IRI) [16]. An- Methods
other study with a sample of 265 third-year medical Study design and subjects
students found that empathy measured by the Jefferson This cross-sectional study was conducted at China Med-
Scale of Empathy (JSE) was positively correlated with per- ical University, which is a major medical university in
sonal accomplishment while negatively associated with Northeast China. A stratified random cluster sampling
depersonalization, which are two subscales of the MBI method was adopted, and questionnaires were distrib-
[17]. Lucchetti and colleagues conducted cross-cultural uted from September to November 2018 to whole clas-
empirical research comparing the differences between ses of medical students in their 1st to 4th academic year
Wang et al. BMC Medical Education (2019) 19:341 Page 3 of 10

of studies. The purpose of the survey was explained be- scale. Each item received a score ranging from 1 (strongly
forehand, and participation was voluntary. Written in- disagree) to 7 (strongly agree), and the total score was cal-
formed consent was obtained from every subject, and culated to indicate the students’ overall level of life satis-
the study was approved by the China Medical University faction. The Chinese version of the scale demonstrated
Committee on Human Experimentation. good validity and reliability among medical students in
previous studies [28, 29], and the Cronbach’s alpha coeffi-
Measurement of empathy cient for the scale in the present study was 0.882.
The two most widely used measurement scales of em-
pathy are the Jefferson Scale of Empathy (JSE) [23] and Demographic characteristics
the Interpersonal Reactivity Index (IRI) [24]. Considering Demographic information included participants’ gender,
the large number of preclinical student participants who age, grade, race, the number of children in the family,
had little or no contact with real patients in this study, place of residence and parents’ educational levels. Stu-
we used the IRI to measure empathy. With permission dents were categorized into two age groups: 17–19 and
from the author of the IRI Mark H. Davis, Zhan [25], a 20–24 years old. With regard to Chinese culture, race
scholar from Taiwan, translated the IRI scale bi- was divided into five groups: Han, Hui, Man, Wei and
directionally and tested the validity as well as reliability Other. The number of children in the family was dichot-
of the IRI Chinese version and adapted the original 28- omized into only child or not. Place of residence fell into
item scale into the 22-item IRI scale Chinese version three categories: cities, towns, and villages. Parents’ edu-
(IRI-C). The IRI-C included four subscales: Perspective cation was divided into three levels: primary school and
Taking (PT: 5 items), Personal Distress (PD: 5 items), Fan- below, secondary school, college and above.
tasy Scale (FS: 6 items) and Empathic Concern (EC: 6
items). Each item was scored on a 5-point Likert scale Statistical analysis
from 0 (not true of me at all) to 4 (very true of me). Rong Student’s t-test and one-way ANOVA were used to
and colleagues [26] tested the IRI-C in Chinese college examine the differences in empathy, burnout and life
students and demonstrated the sound psychometric prop- satisfaction within the categorical demographic variables.
erties of the scale. The Cronbach’s alpha coefficients for Post hoc Bonferroni tests were performed to examine if
the four subscales in the present study were PT: 0.767, the difference was significant between every two groups
PD: 0.806, FS: 0.665, EC: 0.615 and the Cronbach’s alpha concerning continuous variables with regard to multi-
coefficient for the total scale was 0.760. categorical demographic variables. Hierarchical linear re-
gression analysis was utilized to explore the association
Measurement of burnout of empathy and burnout with life satisfaction. General
The Maslach Burnout Inventory Modified Chinese ver- linear model-univariate full factorial model was used to
sion (MBI-MC) was adapted by Lian [27], which was the calculate main effects, the interaction effect and simple
most widely used burnout inventory among college stu- effects of two independent variables of empathy and
dents in China. Based on the MBI, the MBI-MC reduced burnout on the dependent variable of life satisfaction.
the 22 items in the original scale to a 20-item scale that All the statistical tests were performed using SPSS 22.0
contained three subscales: Low Mood (LM: 8 items), Im- with the significance level set at p < 0.05 (two-tailed).
proper Behaviour (IB: 6 items) and Low Achievement
(LA: 6 items). Example questions in the MBI-MC are “I Results
feel exhausted after studying for a whole day” and “I can Demographic characteristics and differences in empathy,
calmly handle my emotional problems”. Each item was burnout and life satisfaction
scored on a 5-point Likert scale from 1 (not true of me The differences in the continuous variables of empathy,
at all) to 5 (very true of me), and after negatively worded burnout and life satisfaction within the socio-
questions were reverse scored, the sum score was calcu- demographic categorical variables of the subjects are pre-
lated to measure medical students’ overall level of burn- sented in Table 1. Of the 1563 invited medical students,
out. The Cronbach’s alpha coefficients for the three 1271 agreed to participate, signed the written informed
subscales in the present study were LM: 0.828, IB: 0.702, consent forms and completed the questionnaires, with a
LA: 0.664 and the Cronbach’s alpha coefficient for the response rate of 81.3%. The average age of the participants
total scale was 0.872. was 19.42 ± 1.34 years old, and the number of males was
458 (36%), while the number of females was 813 (64%). As
Measurement of life satisfaction can be seen from Table 1, the empathy level of students
Medical students’ levels of life satisfaction were mea- aged 17–19 was significantly higher than that of students
sured by the Satisfaction With Life Scale (SWLS) [19], aged 20–24 (52.78 vs 51.10, p = 0.005), and students of
which consisted of 5 items scored on a 7-point Likert various grades saw a significant difference in terms of their
Wang et al. BMC Medical Education (2019) 19:341 Page 4 of 10

Table 1 Demographic characteristics and differences in empathy, burnout and life satisfaction (N = 1271)
Variables N (%) IRI-C (M ± SD) p MBI-MC (M ± SD) p SWLS (M ± SD) p
Gender 0.116 0.091 0.052
Male 458 (36.0) 51.44 ± 10.02 53.30 ± 12.18 22.14 ± 6.78
Female 813 (64.0) 52.40 ± 10.70 52.13 ± 11.63 22.88 ± 6.44
Age 0.005 0.697 0.018
17–19 725 (57.0) 52.78 ± 10.60 52.66 ± 11.55 22.99 ± 6.27
20–24 546 (43.0) 51.10 ± 10.23 52.40 ± 12.21 22.11 ± 6.93
Academic year 0.001 0.751 0.028
First year 415 (32.7) 53.66 ± 10.12 52.33 ± 10.92 23.33 ± 6.20
Second year 341 (26.8) 51.73 ± 10.91 53.15 ± 12.98 22.26 ± 6.43
Third year 264 (20.8) 51.83 ± 10.41 52.30 ± 11.25 21.91 ± 6.78
Fourth year 251 (19.7) 50.10 ± 10.15 52.35 ± 12.31 22.67 ± 7.05
Race 0.417 0.979 0.044
Han 1030(81.0) 52.00 ± 10.62 52.56 ± 12.05 22.50 ± 6.64
Hui 18 (1.4) 48.06 ± 8.71 51.22 ± 9.16 24.00 ± 5.58
Man 108 (8.5) 52.17 ± 9.66 52.28 ± 11.53 24.13 ± 6.29
Wei 29 (2.3) 52.69 ± 7.76 52.52 ± 8.44 23.45 ± 5.58
Other 86 (6.8) 53.26 ± 10.69 53.06 ± 11.30 21.50 ± 6.41
The only child 0.078 0.302 < 0.001
Yes 745 (58.6) 51.62 ± 10.40 52.26 ± 11.95 23.28 ± 6.56
No 526 (41.4) 52.67 ± 10.54 52.96 ± 11.67 21.67 ± 6.48
Place of residence 0.214 0.209 < 0.001
Cities 609 (47.9) 52.08 ± 10.58 51.98 ± 12.45 23.53 ± 6.49
Towns 313 (24.6) 51.28 ± 10.53 52.77 ± 11.35 22.50 ± 6.77
Villages 349 (27.5) 52.71 ± 10.20 53.35 ± 11.13 21.12 ± 6.27
Paternal education 0.789 0.334 < 0.001
Primary school and below 133 (10.4) 52.60 ± 10.65 53.95 ± 11.53 20.48 ± 6.65
Secondary school 719 (56.6) 51.93 ± 10.58 52.48 ± 11.69 22.47 ± 6.36
College and above 419 (33.0) 52.10 ± 10.24 52.22 ± 12.18 23.54 ± 6.75
Maternal education 0.099 0.005 < 0.001
Primary school and below 187 (14.7) 53.57 ± 10.40 54.77 ± 10.83 20.46 ± 6.32
Secondary school 706 (55.6) 51.75 ± 10.55 52.61 ± 11.71 22.55 ± 6.39
College and above 378 (29.7) 51.87 ± 10.30 51.32 ± 12.40 23.80 ± 6.77
Total sample 1271 (100.0) 52.06 ± 10.47 52.55 ± 11.84 22.62 ± 6.57
Significance level: p < 0.05 (two-tailed); M ± SD: mean ± standard deviation; IRI-C Interpersonal Reactivity Index Chinese version, MBI-MC Maslach Burnout Inventory
Modified Chinese version, SWLS Satisfaction With Life Scale

IRI-C scores (p < 0.001). Burnout levels were associated Significant differences in empathy, burnout and life
with medical students’ maternal education (p = 0.005). Re- satisfaction within socio-demographic multicategorical
garding life satisfaction, students aged 17–19 were more variables
satisfied with life than those aged 20–24 (22.99 vs 22.11, The significant differences in empathy, burnout and life
p = 0.018), and students who were the only child in the satisfaction with regard to socio-demographic multicate-
family saw higher levels of life satisfaction (23.28 vs 21.67, gorical variables were examined by ANOVA post hoc
p < 0.001). Significant differences were also observed in life Bonferroni tests. Before the tests, the trial on equality of
satisfaction with regard to medical students’ academic year variance was performed to guarantee that equal variance
(p = 0.028), race (p = 0.044), place of residence (p < 0.001), was achieved, and the results of Bonferroni tests show
paternal education (p < 0.001) and maternal education that the empathy level of 1st_year medical students was
(p < 0.001). significantly higher than that of 4th_year students (53.66
Wang et al. BMC Medical Education (2019) 19:341 Page 5 of 10

vs 50.10, p < 0.001). For burnout levels, students whose satisfaction. As revealed in the figure, there was an over-
mothers held college and above qualifications were less all downward trend in empathy, with a significant differ-
likely to suffer from burnout problems than students ence detected between the 1st and 4th grades (53.66 vs
whose mothers’ educational level was primary school 50.10, p < 0.001). As for burnout, there was almost a lev-
and below (51.32 vs 54.77, p = 0.003). In terms of life sat- eling out since no significant differences were observed
isfaction, 1st_year medical students enjoyed life more between students of various academic years. Life satis-
than their 3rd_year counterparts (23.33 vs 21.91, p = faction saw another pattern, with the mean scores show-
0.035). Though one-way ANOVA shows that students of ing an initial fall to the lowest in 3rd_year students
different ethnic groups saw different levels of life satis- before rising in 4th_year students to a level comparable
faction (p = 0.044), no significant differences were de- to that of 1st_year students. Overall, 1st_year medical stu-
tected by post hoc Bonferroni tests. Regarding place of dents enjoyed life most and their mean score in life sat-
residence, students from cities and towns saw higher life isfaction was significantly higher than the score of
satisfaction levels than those from villages (Mcities = 3rd_year students (23.33 vs 21.91, p < 0.05).
23.53 vs Mvillages = 21.12, p < 0.001; Mtowns = 22.50 vs
Mvillages = 21.12, p = 0.020). Both maternal and paternal Results of hierarchical linear regression analysis
educational levels were correlated with medical students’ Hierarchical linear regression analysis was adopted to
life satisfaction, with students whose parents held college explore the correlation of empathy and burnout with life
and above qualifications enjoying life most (Mpaternal edu- satisfaction and the results were presented in Table 2.
cation (college and above) = 23.54 vs Mpaternal education (secondary As can be seen from the table, demographic factors of
school) = 22.47, p = 0.023; Mpaternal education (college and age and gender explained 0.9% of the variance in life sat-
above) = 23.54 vs Mpaternal education (primary school and below) = isfaction. After empathy was added in step 2, the F value
20.48, p < 0.001; Mmaternal education (college and above) = 23.80 for the model was still significant (F = 6.287, p < 0.001)
vs Mmaternal education (secondary school) = 22.55, p = 0.008; with empathy significantly positively predicting life satis-
Mmaternal education (college and above) = 23.80 vs Mmaternal edu- faction (β = 0.075, p < 0.01), which accounted for another
cation (primary school and below) = 20.46, p < 0.001). Students 0.6% of the variance in life satisfaction. The association
whose parents held secondary school qualifications between burnout and life satisfaction was significantly
enjoyed higher life satisfaction levels than those whose negative (β = − 0.371, p < 0.001), explaining additional
parents’ educational levels were primary school and 13.7% of the variance in life satisfaction. It is noticeable
below (Mpaternal education (secondary school) = 22.47 vs Mpater- that in all three models age was significantly negatively
nal education (primary school and below) = 20.48, p = 0.004; Mma- associated with life satisfaction (β = − 0.078, p < 0.01 in
ternal education (secondary school) = 22.55 vs Mmaternal education step 1, β = − 0.071, p < 0.05 in step 2 and β = − 0.074,
(primary school and below) = 20.46, p < 0.001). p < 0.01 in step 3, respectively).

The trend in mean scores of empathy, burnout and life Main effects and the interaction effect of empathy and
satisfaction of undergraduate medical students over four burnout on life satisfaction
academic years General Linear Model-Univariate Full Factorial Model
Figure 1 shows the trend of the 1st to 4th_year medical was adopted to examine the main effects and to explore
students’ mean scores in empathy, burnout and life whether there was an interaction effect of empathy and

Fig. 1 The trend in mean scores of empathy, burnout and life satisfaction of undergraduate medical students over a span of four academic years
(presented as mean ± standard deviation). Total sample: N = 1271; 1: first year students (n = 415); 2: second year students (n = 341); 3: third year
students (n = 264); 4: fourth year students (n = 251). *p < 0.05; ***p < 0.001
Wang et al. BMC Medical Education (2019) 19:341 Page 6 of 10

Table 2 Hierarchical linear regression analysis with life model explained 21.3% of the variance in life satisfaction
satisfaction as the dependent variable (adjusted R2 = 0.213).
Variables Step 1 (β) Step 2 (β) Step 3 (β)
Step 1 Simple effects of two levels of empathy and burnout on
Age −0.078 **
−0.071 *
−0.074 ** life satisfaction
Post hoc Bonferroni tests were performed to explore the
Gender 0.053 0.050 0.033
simple effects of two levels of empathy and burnout on
Step 2
life satisfaction. The results of pairwise comparisons (for
Empathy 0.075** 0.053* details, please refer to the Additional file 1) show that at
Step 3 the low burnout level, the mean difference in life satis-
Burnout −0.371*** faction between students of low empathy and high em-
F 5.833 **
6.287 ***
56.592*** pathy was significant (24.49 vs 26.82, F(1,351) = 5.970,
p = 0.015, ηp2 = 0.017), while at the high burnout level,
R2 0.009 0.015 0.152
the mean difference in life satisfaction between students
△R2 0.009 0.006 0.137
of low empathy and high empathy was nonsignificant
Note. N = 1271. *p < 0.05 (two-tailed); **p < 0.01 (two-tailed);
***
p < 0.001 (two-tailed)
(20.15 vs 18.63, F(1,351) = 2.569, p = 0.110, ηp2 = 0.007).
When the level of empathy was fixed at low, the mean
difference in life satisfaction between students of low
burnout on life satisfaction. We adopted two factors burnout and high burnout was significant (24.49 vs
with two levels of factorial design in our study, and since 20.15, F(1,351) = 20.705, p < 0.001, ηp2 = 0.056). It is no-
there was no standard for cut-off points, we used the ticeable that when the level of empathy was fixed at
lower and upper quartiles to dichotomize the levels of high, there was a statistically significant difference with a
empathy and burnout into low (< 25%) and high (> 75%). larger effect size in the mean value of life satisfaction be-
Before the test, homogeneity of variance was checked by tween low burnout and high burnout students (26.82 vs
one-way ANOVA, and the result shows that Levene Stat- 18.63, F(1,351) = 75.008, p < 0.001, ηp2 = 0.176). Figure 2
istic = 0.959, p = 0.412, so homogeneity of variances was shows the simple effects of two levels of empathy and
achieved. Table 3 demonstrates the results of between- burnout on life satisfaction by post hoc Bonferroni tests.
subjects effects tests, and as revealed in the table, burn-
out had a main effect on medical students’ life satisfac- Discussion
tion (F(1,351) = 87.008, p < 0.001, ηp2 = 0.199). Post hoc This is the first study examining the associated socio-
Bonferroni tests of the main effect revealed that life sat- demographic factors with empathy, burnout and life sat-
isfaction levels of students with low burnout were sig- isfaction and the possible interaction effect of empathy
nificantly higher than those with high burnout levels and burnout on life satisfaction in Chinese undergradu-
(Mlife satisfaction (low burnout) = 25.66 vs Mlife satisfaction (high ate medical students. As for the correlation of empathy
burnout) = 19.39, p < 0.001) (for details, please refer to the with socio-demographic characteristics, it is found that
Additional file 1). Whereas the main effect of empathy younger and lower-grade students were more empathetic
on life satisfaction did not reach significance level (F(1, than their older higher-grade counterparts. This finding
351) = 0.362, p = 0.548, ηp2 = 0.001), there was an inter- is consistent with the conclusion in a study by Li and
action effect of empathy and burnout on life satisfaction colleagues [30], but in contradiction with the result of
(F(1,351) = 8.195, p = 0.004, ηp2 = 0.023). The whole another study by Wen [31]. It is worth mentioning that

Table 3 Main effects and the interaction effect of empathy and burnout on life satisfaction
Source SS df MS F p ηp2
Corrected Model 3893.306a 3 1297.769 32.906 < 0.001 0.220
Intercept 177,449.242 1 177,449.242 4499.334 < 0.001 0.928
Empathy 14.268 1 14.268 0.362 0.548 0.001
Burnout 3431.491 1 3431.491 87.008 < 0.001 0.199
Empathy * Burnout 323.191 1 323.191 8.195 0.004 0.023
Error 13,843.088 351 39.439
Total 199,599.000 355
Corrected Total 17,736.394 354
a
R Squared = .220 (Adjusted R Squared = .213); Dependent variable: life satisfaction; SS type III sum of squares, df degree of freedom, MS mean square, ηp2 partial
eta squared. Effect size was measured by partial eta squared: small = 0.01 to 0.06, medium = 0.06 to 0.138, large> 0.138
Wang et al. BMC Medical Education (2019) 19:341 Page 7 of 10

Fig. 2 Simple effects of two levels of empathy and burnout on life satisfaction

although the sample in these two studies was Chinese adopted by medical students whose mothers were of
medical students, the scale used for measuring empathy higher educational levels when they encountered difficul-
was the student version of the Jefferson Scale of Phys- ties and felt exhausted. This problem-based coping strat-
ician Empathy (JSPE-S). Research results concerning em- egy, along with adjusted behaviours may help students
pathy in medical students from other cultures were also recover from academic and emotional fatigue; thus, they
inconsistent. A study with a sample of 320 medical stu- experienced less burnout. The finding that no significant
dents across the 1st to 6th academic years in Brazil de- difference was detected in the burnout levels of students
tected no significant differences in empathy considering in different academic years was in line with the research
students’ grade [32], whereas research in Iran indicated conducted in 22 medical schools in Brazil [16], which con-
that the overall level of empathy among medical stu- cluded that burnout problems existed in all stages of med-
dents in the preclinical period was higher than that in ical education.
the clinical period [33]. However, another study from Regarding life satisfaction, students of the 17–19 age
India showed that medical students’ empathy scores had group enjoyed life more than those in the 20–24 age
a tendency to decline initially and then rebound over group. This finding is inconsistent with the conclusion
time [34]. The contradiction in these findings may be in Shi [29], which did not show significant differences in
due to the differences in measurement tools, cultures the life satisfaction levels of Chinese medical students in
and the environment in which the empirical research different age groups. It is noteworthy that students in
was carried out. Another point worth mentioning is that the 17–19 age group also had higher levels of empathy
no significant difference in the mean empathy score was in the present study, and since empathy was found to be
found between males and females in the present study, negatively correlated with stress and burnout but posi-
which was in line with the conclusion of the study by Li tively associated with social support [16, 38], it is reason-
[30], although a number of other studies revealed that able that students with higher levels of empathy also
female medical students were more empathetic than enjoyed higher levels of life satisfaction. In the current
males [31–34]. study, 1st_year medical students had significantly higher
A new finding in the present study was that medical stu- life satisfaction scores than 3rd_year students but not 4th
dents’ burnout levels were related to maternal but not pa- -year students, which can be explained from the per-
ternal educational levels, and students whose mothers spective that the 3rd academic year in China is the
held college and above qualifications were less likely to period in which undergraduate medical students transi-
suffer from burnout problems than students whose tioned from preclinical to clinical studies, and this crit-
mothers’ educational level was primary school and below. ical transitional period may bring maladjustment and
In China, as in most other countries around the world, mental health problems such as stress, anxiety and de-
mothers still play a major role in children’s growth and pression that lead to decreased life satisfaction. In the
education and spend relatively more time with children in 4th academic year, medical students gradually adapt to
comparison to fathers. Research shows that maternal edu- the curriculum and learning environment, such as in-
cation is inextricably correlated with children’s language ternships in university-affiliated hospitals, so the levels
skills, academic performance, intellectual functioning and of life satisfaction rebound. As expected, students who
maladaptive behaviours [35–37]. Therefore, we assume were the only child in the family, who were from cities
that more adaptive coping strategies were likely to be and towns and whose parents were of higher educational
Wang et al. BMC Medical Education (2019) 19:341 Page 8 of 10

levels enjoyed higher levels of life satisfaction since they daily schedule and one’s self-efficacy were also found to be
were more likely to have better living conditions and re- related to medical students’ burnout [10]. In addition to
ceive better education. It is worth mentioning that the reforming the medical curriculum and scoring system, im-
difference in life satisfaction was marginally significant proving the learning environment, promoting social sup-
with regard to gender in this study, with females more port and functional coping strategies, nurturing students’
satisfied with life than their male counterparts (22.88 vs resilience and empathy were all found to be effective in al-
22.14, p = 0.052). This finding can be explained by the leviating medical students’ stress and combating burnout
one-child policy in effect in China for several decades; problems.
under this policy, boys were more valued and parental Our study shows that although empathy did not have
expectations for them were higher, and compared with a main effect on life satisfaction, it interacted with burn-
females, males shouldered more pressure and responsi- out and exerted an interaction effect on the levels of
bilities, as they were considered pillars of the family, so medical students’ life satisfaction. Medical students with
they were more likely to suffer from mental health prob- high levels of empathy and low levels of burnout enjoyed
lems such as depression, which was demonstrated in life most, and this finding sheds light on identifying feas-
some studies of Chinese medical students [39, 40], and ible interventions to enhance students’ empathy and re-
this may lead to lower life satisfaction in males. duce burnout problems to improve their life satisfaction.
Hierarchical linear regression analysis shows that It is noteworthy that the effect size was larger for the
burnout explained a large proportion of the variance mean difference in life satisfaction in high empathy
(13.7%) in life satisfaction, whereas empathy only groups with low and high burnout levels than in low
accounted for 0.6% of the variance in life satisfaction. empathy groups with low and high burnout levels (ηp2 =
Age was found to significantly negatively predict life sat- 0.176 vs ηp2 = 0.056). This finding indicates that students
isfaction and this result was in line with the finding in with high empathy were probably more affected by
the present study that students of the 17–19 age group burnout with regard to life satisfaction than students
were more satisfied with life than those in the 20–24 age with low empathy, which lends further support to the
group. With the increase of age and grade, medical stu- idea that enhancing students’ empathy and reducing
dents need to deal with more academic work, especially burnout need to be simultaneously addressed to meet
when they started their clinical phase study. The mal- the target of achieving high life satisfaction levels. Unlike
adjustment brought by the transitional period from pre- among health professionals, empathy may not exert such
clinical to clinical learning environment and the heavy a significant impact on burnout among undergraduate
workload may result in decreased life satisfaction levels. medical students, especially among preclinical students
The results of ANOVA show that burnout had a main who have little or no contact with real patients. How-
effect on life satisfaction, and to effectively address burn- ever, cultivating students’ empathy is critical and needs
out problems, we need to find stressors that lead to to be emphasized from the outset of medical education
burnout in medical students. First, the intensive curricu- as this has a long-term effect. At present, some com-
lum and the rigid scoring system in medical education monly known measures to enhance empathy include
were found to be a source of stress. Slavin and col- early exposure to clinical training, inter-professional
leagues [41] instituted curricular changes including a education (IPE), a curriculum emphasizing communica-
pass/fail grading system and reduced contact hours for tion and interpersonal skills, a self-reflection approach,
preclinical students and found that there was a signifi- creating a culture of humanism and teachers as role
cant decline in medical students’ stress, anxiety and de- models in medical education. Since studies show that
pression levels. Second, learning environments such as the empathy levels of medical students may erode as
clinical rotation factors for clinical students and personal they step into higher grades, there is an urgent need to
characteristics and life events were significant contribu- experiment with different methods to enhance empathy
tors to burnout. A study by Dyrbye [42] detected the as- and implement effective methods in medical education.
sociations of call frequency and clinical workload in Several limitations of the present study should be ac-
rotation with burnout among medical students in their knowledged. First, with the cross-sectional design, con-
clinical practice environment, and the study also con- clusions about causality cannot be drawn, and
cluded that personal life events, gender and learning en- longitudinal studies should be carried out in the future.
vironment were all factors associated with burnout Second, as self-administered questionnaires were distrib-
among preclinical students. Third, lack of social support uted to collect data, response bias and social desirability
and low empathy levels as well as dysfunctional coping bias cannot be avoided. Third, the present study was
strategies were detected in some studies as risk factors conducted in one medical university in China, so gener-
for burnout [43, 44]. It is worth mentioning that, in the alizations of the conclusions should be made with cau-
research by Mazurkiewicz and colleagues, control over tion and future multi-institutional research in different
Wang et al. BMC Medical Education (2019) 19:341 Page 9 of 10

cultures is suggested. Fourth, the present study analysed Funding

the effect of empathy on burnout and life satisfaction of None declared.

medical students, but in addition to empathy, other vari-

Availability of data and materials
ables such as self-efficacy and self-esteem also influenced The datasets used and/or analyzed in the present study are available from
burnout and life satisfaction levels in medical students, the corresponding author on reasonable request.
which should be examined in future studies.
Ethics approval and consent to participate
All subjects provided written informed consents and the study procedures
Conclusions were approved by the Committee on Human Experimentation of China
This study explored the relationship among empathy, Medical University.
burnout and life satisfaction and the associated socio-
demographic factors in Chinese undergraduate medical Consent for publication
students. The results show that the levels of empathy were Not applicable.
related to students’ age and grade, and levels of burnout
Competing interests
were associated with students’ maternal education. As for The authors declare that they have no competing interests.
life satisfaction, significant differences were detected with
regard to medical students’ age, academic year, the num- Author details
1
English Department, School of Fundamental Sciences, China Medical
ber of children in the family, place of residence and par- University, No. 77 Puhe Road, Shenyang North New Area, Shenyang,
ents’ educational levels. Over the four academic years, Liaoning Province, People’s Republic of China. 2Department of Social
medical students’ empathy levels saw a decline, but their Medicine, School of Public Health, China Medical University, No. 77 Puhe
Road, Shenyang North New Area, Shenyang, Liaoning Province, People’s
burnout levels almost plateaued, and their life satisfaction Republic of China. 3Department of Epidemiology, School of Public Health,
levels witnessed an initial fall before a rebound. There was China Medical University, No. 77 Puhe Road, Shenyang North New Area,
an interaction effect of empathy and burnout on life satis- Shenyang, Liaoning Province, People’s Republic of China. 4Department of
Statistics, School of Public Health, China Medical University, No. 77 Puhe
faction, and students with high empathy and low burnout Road, Shenyang North New Area, Shenyang, Liaoning Province, People’s
were most satisfied with life. Medical institutions and re- Republic of China. 5Department of Healthcare Management, School of
lated authorities need to find effective measures to en- Humanities and Social Sciences, China Medical University, No. 77 Puhe Road,
Shenyang North New Area, Shenyang, Liaoning Province, People’s Republic
hance students’ empathy levels and reduce burnout of China.
problems to improve their life satisfaction.
Received: 17 June 2019 Accepted: 3 September 2019

Supplementary information
Supplementary information accompanies this paper at https://doi.org/10.
1186/s12909-019-1788-3. References
1. Yuguero O, Melnick ER, Marsal JR, Esquerda M, Soler-Gonzalez J. Cross-
sectional study of the association between healthcare professionals’
Additional file 1: Data on main effects and simple effects of empathy empathy and burnout and the number of annual primary care visits per
and burnout on life satisfaction by ANOVA post hoc Boferroni tests. patient under their care in Spain. BMJ Open. 2018;8:e020949.
(DOCX 58 kb) 2. Yuguero O, Marsal JR, Buti M, Esquerda M, Soler-González J. Descriptive
study of association between quality of care and empathy and burnout in
primary care. BMC Med Ethics. 2017;18:54–61.
Abbreviations
3. Duarte J, Pinto-Gouveia J. Empathy and feelings of guilt experienced by
EC: Empathic Concern; ESWIM: Empathy, Spirituality, and Wellness in
nurses: a cross-sectional study of their role in burnout and compassion
Medicine survey; FS: Fantasy Scale; IB: Improper Behaviour; IPE: Inter-
fatigue symptoms. Appl Nurs Res. 2017;35:42–7.
professional education; IRI: Interpersonal Reactivity Index; IRI-C: Interpersonal
4. Silver J, Caleshu C, Casson-Parkin S, Ormond K. Mindfulness among genetic
Reactivity Index Chinese version; JSE: Jefferson Scale of Empathy; JSPE-
counselors is associated with increased empathy and work engagement
S: student version of the Jefferson Scale of Physician Empathy; LA: Low
and decreased burnout and compassion fatigue. J Genet Couns. 2018;27:
Achievement; LM: Low Mood; MBI: Maslach Burnout Inventory; MBI-
1175–86.
MC: Maslach Burnout Inventory Modified Chinese version; MBI-SS: Maslach
5. Hojat M. Empathy in patient care: antecedents, development, measurement,
Burnout Inventory-Student Survey; OLBI: Oldenburg Burnout Inventory;
and outcomes. New York: Springer; 2007.
PD: Personal Distress; PT: Perspective Taking; SEM: structural equation
6. Kim SS, Kaplowitz S, Johnston MV. The effects of physician empathy on
modelling; SWLS: Satisfaction With Life Scale
patient satisfaction and compliance. Eval Health Prof. 2004;27(3):237–51.
7. Hojat M, Spandofer J, Louis DZ, Gonnella JS. Empathic and sympathetic
Acknowledgements orientations toward patient care: conceptualization, measurement, and
The authors would like to thank the faculty who organized the distribution psychometrics. Acad Med. 2011;86(8):989–95.
and collection of the questionnaires in their classrooms as well as the 8. Shapiro J. Walking a mile in their patients’ shoes: empathy and othering in
students for their participation. medical students’ education. Philos Ethics Humanit Med. 2008;3:10–20.
9. Maslach C, Jackson SE, Leiter MP. The Maslach burnout inventory. 3rd ed.
Authors’ contributions Palo Alto, CA: Consulting Psychologists Press; 1996.
QHW was in charge of the study design, questionnaire survey, drafting and 10. Mazurkiewicz R, Korenstein D, Fallar R, Ripp J. The prevalence and
revising the manuscript. LW and MS contributed to the study design and correlations of medical student burnout in the pre-clinical years: a cross-
provided some valuable advice. XLL, RL, JL and MZ helped to collect and sectional study. Psychol Health Med. 2012;17(2):188–95.
analyze the data. HZW made substantial intellectual contributions to the 11. Liu HC, Yansane AI, Zhang YR, Fu HJ, Hong NR, Kalenderian E. Burnout and
conception of the study, analyses of the data and draft of the manuscript. study engagement among medical students at Sun Yat-sen University,
All authors read and approved the final version of the manuscript. China: A cross-sectional study. Medicine. 2018;97(15):e0326.
Wang et al. BMC Medical Education (2019) 19:341 Page 10 of 10

12. PENŠEK L, SELIČ P. Empathy and burnout in Slovenian family medicine children living in low socioeconomic areas in Khartoum. Sudan Ahfad J.
doctors: the first presentation of Jefferson scale of empathy results. Zdr 1993;10(1):44–55.
Varst. 2018;57(3):155–65. 37. Carvalho C, Cruz O. Disciplinary behavior of mothers of preschool children:
13. Lee PT, Loh J, Sng G, Tung J, Yeo KK. Empathy and burnout: a study on effects of maternal efficacy beliefs, children’s gender and age, and mothers’
residents from a Singapore institution. Singap Med J. 2018;59(1):50–4. education. Estudos de Psicologia (Campinas). 2018;35(4):433–43.
14. Wilkinsona H, Whittington R, Perry L, Eames C. Examining the relationship 38. Park KH, Kim DH, Kim SK, Yi YH, Jeong JH, Chae J, et al. The relationships
between burnout and empathy in healthcare professionals: a systematic between empathy, stress and social support among medical students. Int J
review. Burn Res. 2017;6:18–29. Med Educ. 2015;6:103–8.
15. Yuguero O, Marsal JR, Esquerda M, Vivanco L, Soler-Gonzalez J. Association 39. Shi M, Liu L, Wang ZY, Wang L. Prevalence of depressive symptoms and its
between low empathy and high burnout among primary care physicians correlations with positive psychological variables among Chinese medical
and nurses in Lleida. Spain Eur J Gen Pract. 2017;23(1):4–10. students: an exploratory cross-sectional study. BMC Psychiatry. 2016;16:3–10.
16. Paro HBMS, Silveira PSP, Perotta B, Gannam S, Enns SC, Giaxa RRB. Empathy 40. Shi M, Liu L, Yang YL, Wang L. The mediating role of self-esteem in the
among medical students: is there a relation with quality of life and relationship between big five personality traits and depressive symptoms
burnout? PLoS One. 2014;9(4):e94133. among Chinese undergraduate medical students. Personal Individ Differ.
17. Hojat M, Vergare M, Isenberg G, Cohen M, Spandorfer J. Underlying 2015;83:55–9.
construct of empathy, optimism, and burnout in medical students. Int J 41. Slavin SJ, Schindler DL, Chibnall JT. Medical student mental health 3.0:
Med Educ. 2015;6:12–6. improving student wellness through curricular changes. Acad Med. 2014;
18. Lucchetti G, Damiano RF, DiLalla LF, Lucchetti ALG, Moutinho ILD, Ezequiel 89(4):573–7.
ODS, et al. Cross-cultural differences in mental health, quality of life, 42. Dyrbye LN, Thomas MR, Harper W, Massie FS Jr, Power DV, Eacker A, et al.
empathy, and burnout between US and Brazilian medical students. Acad The learning environment and medical student burnout: a multicentre
Psychiatry. 2018;42:62–7. study. Med Educ. 2009;43:274–82.
19. Diener E, Emmons RA, Larsen RJ, Griffin S. The satisfaction with life scale. J 43. Erschens R, Lodaa T, Herrmann-Wernera A, Keifenheima KE, Stuber F,
Pers Assess. 1985;49:71–5. Nikendei C, et al. Behaviour-based functional and dysfunctional strategies of
20. Cho E, Jeon S. The role of empathy and psychological need satisfaction in medical students to cope with burnout. Med Educ Online. 2018;23:1535738.
pharmacy students’ burnout and well-being. BMC Med Educ. 2019;19:43–54. 44. Galam E, Soupault CV, Bunge L, du Vaure CB, Boujut E, Jaury P. ‘Intern life’: a
21. Hakanen JJ, Schaufeli WB. Do burnout and work engagement predict longitudinal study of burnout, empathy, and coping strategies used by
depressive symptoms and life satisfaction? A three-wave seven-year French GPs in training. BJGP Open. 2017. https://doi.org/10.3399/
prospective study. J Affect Disord. 2012;141:415–24. bjgpopen17X100773.
22. Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. A model of burnout
and life satisfaction amongst nurses. J Adv Nurs. 2000;32(2):454–64. Publisher’s Note
23. Hojat M, Gonnella JS, Nasca TJ, Mangione S, Vergare M, Magee M. Physician Springer Nature remains neutral with regard to jurisdictional claims in
empathy: definition, components, measurement, and relationship to gender published maps and institutional affiliations.
and specialty. Am J Psychiatry. 2002;159:1563–9.
24. Davis MH. Measuring individual differences in empathy: evidence for
multidimensional approach. J Pers Soc Psychol. 1983;44:113–26.
25. Zhang FF, Dong Y, Wang K, Zhan ZY, Xie LF. Reliability and validity of the
Chinese version of the interpersonal reactivity index-C. Chinese J Clin
Psychol. 2010;18(2):155–7 (in Chinese).
26. Rong X, Sun BH, Huang XZ, Cai MY, Li WJ. Reliabilities and validities of
Chinese version of interpersonal reactivity index. Chinese J Clin Psychol.
2010;18(2):158–60 (in Chinese).
27. Lian R, Yang LX, Wu LH. Relationship between professional commitment
and learning burnout of undergraduates and scales developing. Acta
Psychol Sin. 2005;37(5):632–6 (in Chinese).
28. Shi M, Liu L, Sun X, Wang L. Associations between symptoms of attention-
deficit/ hyperactivity disorder and life satisfaction in medical students: the
mediating effect of resilience. BMC Med Educ. 2018;18:164–70.
29. Shi M, Wang XX, Bian YG, Wang L. The mediating role of resilience in the
relationship between stress and life satisfaction among Chinese medical
students: a cross-sectional study. BMC Med Educ. 2015;15:16–22.
30. Li DJ, Xu HM, Kang MY, Ma SL. Empathy in Chinese eight-year medical
program students: differences by school year, educational stage, and future
career preference. BMC Med Educ. 2018;18:241–9.
31. Wen DL, Ma XD, Li HH, Liu ZF, Xian BS, Liu Y. Empathy in Chinese medical
students: psychometric characteristics and differences by gender and year
of medical education. BMC Med Educ. 2013;13:130–5.
32. Santos MA, Grosseman S, Morelli TC, Giuliano ICB, Erdmann TR. Empathy
differences by gender and specialty preference in medical students: a study
in Brazil. Int J Med Educ. 2016;7:149–53.
33. Rezayat AA, Shahini NS, Asl HT, Jarahi L, Behdani F, Shojaei SRH, et al.
Empathy score among medical students in Mashhad, Iran: study of the
Jefferson scale of physician empathy. Electron Physician. 2018;10(7):7101–6.
34. Chatterjee A, Ravikumar R, Singh S, Chauhan PS, Goel M. Clinical empathy in
medical students in India measured using the Jefferson scale of empathy–
student version. J Educ Eval Health Prof. 2017;14:33–8.
35. Conant LL, Liebenthal E, Desai A, Binder JR. The relationship between
maternal education and the neural substrates of phoneme perception in
children: interactions between socioeconomic status and proficiency level.
Brain Lang. 2017;171:14–22.
36. Washi S, Cowan D, Terry RD. The impact of mother's education on
indicators of school performance of first through third grade primary school