[Teaching empathy in medical education].

Abstract

To determine whether there is any difference in empathic tendencies and attitudes towards communication in medical practice among the first-year students of Adnan Menderes University Medical School, and to determine the effect of education and the relation between these two features.
The Empathic Tendency Scale and Attitudes Towards Communication in Medical Practice Questionnaire were used to collect data from 36 (55.3 % of the total) students (65.7 % female; mean age: 18.3).
Students are clustered into two groups for both scales using K-means analysis, which was used to see whether the students were homogeneous regarding empathy and communication attitudes at the beginning of the teaching period. According to the Empathic Tendency Scale results 25% of students have a high empathic tendency (HET) and 75% have a low empathic tendency (LET) and the results of the Attitudes Towards Communication in Medical Practice Questionnaire indicated that 56% have positive and 44% have negative attitudes towards communication. However, the difference between the pre-and post-teaching results of either group seems to show a tendency towards change (p< 0.10). The change has a negative and a positive tendency in the HET and LET groups, respectively.
Our results seem to verify the predicted differences among the students in their empathic tendencies and attitudes towards communication in medical practice. It seems that empathic tendency is more likely to be changed by an appropriate teaching technique when change within the groups is concerned.

Full text

Page 1

1
SUMMARY: Teaching Empathy in Medical Education
Objectives: To determine whether there is any difference in
empathic tendencies and attitudes towards communication
in medical practice among the first-year students of Adnan
Menderes University Medical School, and to determine
the effect of education and the relation between these two
features.
Method: The Empathic Tendency Scale and Attitudes
Towards Communication in Medical Practice Questionnaire
were used to collect data from 36 (55.3 % of the total)
students (65.7 % female; mean age: 18.3).
Results: Students are clustered into two groups for both
scales using K-means analysis, which was used to see
whether the students were homogeneous regarding empathy
and communication attitudes at the beginning of the teaching
period. According to the Empathic Tendency Scale results
25% of students have a high empathic tendency (HET) and
75% have a low empathic tendency (LET) and the results of
the Attitudes Towards Communication in Medical Practice
Questionnaire indicated that 56% have positive and 44%
have negative attitudes towards communication. However,
the difference between the pre-and post-teaching results of
either group seems to show a tendency towards change (p<
0.10). The change has a negative and a positive tendency in
the HET and LET groups, respectively.
Conclusion: Our results seem to verify the predicted
differences among the students in their empathic tendencies
and attitudes towards communication in medical practice. It
seems that empathic tendency is more likely to be changed
by an appropriate teaching technique when change within
the groups is concerned.
Key Words: Empathy, education, medicine, undergraduate,
communication
Türk Psikiyatri Dergisi 2005; 16(2)
Turkish Journal of Psychiatry
Teaching Empathy in Medical Education
Çiğdem DEREBOY, Hacer HARLAK, Serdar GÜREL, Ayfer GEMALMAZ, Mehmet ESKİN
INTRODUCTION
Physicians are increasingly focusing on scien-
ce, technology and medical economics and various
authors accuse medical trainers of raising students
who do not understand the circumstances of and
show adequate tenderheartedness to patients (Starr
1982, Ludmerer 1999). Wilkes and colleagues
emphasized what the trainers should do in order to
make medical students consider there is a human
being behind the curtain, at the end of the endos-
cope or in the scanner who thinks, feels and scares
(2002).
Hojat and colleagues reported that the quality of
interpersonal communication between the physici-
an and the patient was very important in adequacy
of clinical consequences (2002). They thought that
this quality depended largely on the skills of the
physician, so this characteristic should be taken
into consideration in determining clinical suffici-
ency. There are some other data consistent with
this point of view (Hojat et al. 1986). If the patient
thinks that he is been understood during the inter-
view, then he will be relaxed and express himself
easily. Consequently, diagnosis will be exact and
participation of the patient to therapeutic process
will be provided. One of the first examples was in
Maastricht University from Netherlands and they
performed necessary program changes by 1980s.
A study group which was formed by behavioral
scientists, family practitioners and specialists from
other clinical sciences trained the students about
communication skills for four years and reported
their results (Van Dalen et al. 1999, 2001). They
evaluated which factor most influenced commu-
nication skills and found that the content of the
program was most prominent (1999). Nowadays,

Page 2

2
training for communication skills takes place in
academic programs of many medical schools in
USA or other countries (Makoul 2003). In our co-
untry, training programs for communication skills
are present in medical schools of Marmara, Dokuz
Eylül, Pamukkale and Adnan Menderes Univer-
sities where training model is active, integrated
and mixed (Sayek and Kılıç 2000, Dereboy et al.
2001). Research on communication skills educati-
on revealed that content, timing, training and eva-
luation methods varied between different medical
schools (Harrison and Glasgow 1996, Hargie et al.
1998, Aspergen 1999, Sarıkaya et al. 2004).
In “Physicians of Tomorrow 2003” report of Bri-
tish Medical Council, training for communication
skills was considered as an important part of medi-
cal education programs and 12 learning objectives
were determined. Four of them were related with
positive attitudes towards learning communication
skills. Moreover, development of empathy in order
to provide satisfactory patient-physician relations-
hip was one of the proposed learning objectives
of the Association of American Medical Colleges
(AAMC Medical Sschool Objective Project, Re-
port I 1998). Empathy was defined as “conside-
ring oneself as the opposite person”. A more clear
definition was “an intellectual activity which aims
to better understand sensations and point of view
of other people” (Egan 1998). But to understand
perceptions and sensations of other people intel-
lectually was not adequate; to express empathy to
other people was complementary (Dökmen 1999).
As other individual characteristics, empathy varies
between people. Some people may have a lesser
degree of empathy as a consequence of develop-
mental, experimental, social, educational and ot-
her internal or external factors (Hojat et al. 2002)
In Adnan Menderes University, training for
communication skills is administered between
grades 1 to 5. A research project which aimed to
evaluate attitudes of medical students towards le-
arning and using communication skills in medical
practice as well as the development of listening and
empathy was carried on. In this study we aimed to
evaluate in-group differences of attitudes towards
learning communication skills and empathic ten-
dencies, whether these characteristics changed af-
ter training program or not and how they interacted
with each other among first year students of Adnan
Menderes University Medical School.
METHOD
Participants
First grade students of Adnan Menderes Uni-
versity Medical School in academic period of
2003-2004 were included in the study: the num-
bers of participants were 62 in pre- (females 49.2
%, males 50.2 %, mean age 18.4, SD=0.8) and
36 in post-education periods (females 65.7 %,
males 34.3 %, mean age 18.34, interval=17-20,
SD=0.68). Accessibility ratios according to grade
presences were 95.3 % in first and 55.3 % in last
tests.
Materials
Empathic Tendency Scale (ETS): This is a
self-report scale consisting 20 items. Each item is
scored between 1-5 according to answers as “ab-
solutely contrary”, “mostly contrary”, “hesitant”,
“mostly convenient” and “absolutely convenient”.
Some items are reverse coded in calculation of to-
tal scores. Possible lowest score is 20 and highest
score is 100 points (Dökmen 1988).
Instrument for Attitudes Towards Communica-
tion in Medicine (IATCM): This scale was develo-
ped by the authors of this study in order to evalu-
ate attitudes towards communication. It is a Likert
type self-report scale consisting 8 items. Each item
is scored between 1-5 according to answers as “I
am absolutely against”, “I do not think so”, “I do
not have any idea”, “I think so” and “I have just
the same idea”. Some items are reverse coded in
calculation of total scores. Possible lowest score is
8 and highest score is 40 points. When relatively
low number of items is taken into consideration,
internal consistency of the scale (ά=0.77) is adequ-
ate (Nunnaly and Bernstein 1994). Item-total cor-
relations vary between 0.25 and 0.65 (Table 1).
Procedure
First, the quality and necessity of the study was
described to students who were present in the class
and later ETS and IATCM scales were administe-
red. Five months later at the last lesson, same ins-
truments were administered to students who were
present in the class. Training of communication
skills included totally 30 hours; 8 hours in the first
three committees and 6 hours in the fourth com-
mittee. These lessons were arranged as interactive
presentations, exercises, role plays and dual practi-
ces. At last three lessons, following a presentation
about empathy, the students performed an exerci-

Page 3

3
se of developing an empathic reaction. Later, they
watched a video show which lasted for about 5
minutes and reflecting conflicting and empathy-
free relations between three students of whom one
was from medical school. Then the students for-
med groups consisting three of them each and as-
ked to prepare conversation texts which aimed to
reveal empathic reactions. During this preparation
procedure, information about empathy provided
by authors of this article and an education guide
about communication skills were used. Students
prepared their conversation texts with the help
of these empathy guides, then presented them to
educational staff and other students, and received
feedback.
Statistical Analysis
Data were analyzed with SPSS 10.0 software.
While internal consistency calculations and pre-
test correlations were performed with 62 partici-
pants, K-means group analysis, comparison of pre-
and post-tests, post-test correlations and analysis
of variances between groups were performed with
36 participants. The significance of the difference
between mean pre- and post-test ETS and IATCM
scores was evaluated with t test. Comparisons bet-
ween groups were evaluated with non-parametric
Mann-Whitney U test because of the decrease in
the number of participants.
FINDINGS
Differences and intersections between gro-
ups before training
In order to determine whether the students for-
med a homogenous group according to commu-
nication skills and their attitudes K-mean group
analyses were performed. Both ETS and IATCM
scores were included as independent variables in-
group analyses and the group was divided into two
separate subgroups. Total numbers and percen-
tages of students in both groups are presented in
Table 2. The differences between ETS scores high
and low groups (z=-3.9, p<0.000) and mean sco-
res of IATCM points positive and negative groups
(z=-2.46, p<0.000) were statistically significant in
post-training measurements.
When intersection between empathic tendency
low and high groups and attitude towards commu-
nication in medicine positive and negative groups
was considered, it was found that of 16 students
whose attitude was negative 14 students (87.5%)
were in low and two students (12.2%) were in high
empathy group. Of 20 students whose attitude was
positive 13 students (65%) were in low and 7 stu-
dents (35%) were in high empathy group (Figure
1).
Comparisons of pre- and post-test findings
As shown in Table 2, differences between pre-
and post-test means of student groups after ETS
and IATCM administration were not statistically
TABLE 1. Instrument for Attitudes Towards Communication in Medicine
There are some expressions about patient-physician relationship below. Following reading
each expression, check suitable choice on the scale.
1. I am absolutely against
2. I do not think so
3. I do not have any idea
4. I think so
5. I have just the same idea
1 I believe that to be in a good communication with the patient is the keystone of success in
medicine.
1 2 3 4 5
2 I think communication skills of a physician are important. 1 2 3 4 5
3Communication style of a physician does not have any influence on the treatment, appropriate
diagnosis, but medication choices are more important.
1 2 3 4 5
4Patient’s trust on physician and thoughts about to be understood are very important in treatment. 1 2 3 4 5
5Patient-physician communication is not important in determining true complaint of the patient. 1 2 3 4 5
6Better understanding psychosocial condition of the patient does not have any positive influence on
treatment.
1 2 3 4 5
7 I do not think there is any need for communication skills education in medical faculties. 1 2 3 4 5
8I think communication is not a learnable skill. 1 2 3 4 5

Page 4

4
significant. Nonetheless, the difference between
pre- and post-test means of empathy low and high
groups showed a tendency towards change despi-
te this difference was not statistically significant
(p<0.10). This change was in negative direction in
empathy high group and positive direction in em-
pathy low group.
ETS and IATCM scores did not differ accor-
ding to gender, age and socio-economic status.
The groups were similar according to these vari-
ables as well.
Correlations between scales
The correlation between pre-test ETS and
IATCM scores was calculated to be r=0.34. This
correlation coefficient was significant at 0.05 le-
vel. The correlation between post-test scores was
calculated as r=0.32. Significance value for this
calculation was just above 0.05 level (p<0.057).
DISCUSSION
In this study, findings from group analyses re-
vealed that students of Adnan Menderes Univer-
sity Medical School did not form a homogenous
group according to their attitudes towards commu-
nication skills and empathic tendencies at the be-
ginning of medical education. One of the primary
objectives of education in first grade was to lessen
presumptive differences between students in atti-
tudes and interpersonal communication skills and
try equalizing the benefit which would be recei-
ved from following training periods. Results of the
study supported the presumption about there wo-
uld be individual differences among students for
attitudes and interpersonal communication skills.
Both groups according to attitudes towards com-
munication skills were nearly equal. At the begin-
ning of the education, 20 students (56%) reported
positive attitudes towards communication and 16
students (44%) reported negative attitudes. Ro-
senthal and Ogden reported that 89.3% of first
grade students who agreed participation reported
the importance of communication skills training
(1998). But, this higher-level of agreement indica-
tes only responses to a single statement “commu-
nication skills should be emphasized more”. This
ratio might be expected to be lower in scales like
our instruments that included more complex and
detailed expressions. Moreover, our study group
did not only include students who gave consent to
participate, but 95% of students who were present
in the classroom were included. Relatively higher
level of negative attitude might be a consequen-
ce of this inclusion pattern. According to empat-
hic tendencies, ¾ of students formed a group and
¼ of them formed another. At the beginning, 27
students (75%) had low and 9 students (25%) had
high empathy scores.
In studies evaluating the place of communica-
tion skills in medical training, empathic tendency
and attitudes towards communication were taken
separately and the possible relation between them
was not questioned. In our study, we investigated
the relation between empathy tendency and attitu-
des towards communication and found that 65%
of students with positive attitude were in low em-
pathy group and this value increased to 87.5% in
negative attitude group. On the other hand, 35%
of students with positive attitude were in high
TABLE 2. Percentages, mean points and pre- and post-test comparisons of the groups
% (n) Pre-test
mean
Post-test
mean
t/zs.d.p
Empathy Tendency
Participants
High empathy group
Low empathy group
25 (9)
75 (27)
67.8 (ss=9.5)
81,1 (ss=1.8)
63,4 (ss=6.4)
69.6 (ss=8.8)
79,2 (ss=3.6)
66,4 (ss=7.7)
0.68
-1.85
-1.68
35 0.18
0.06
0.09
Attitude
Participants
Positive attitude group
Negative attitude group
56 (20)
44 (16)
35.6 (ss=4.0)
37,3 (ss=4.6)
33,6 (ss=1.6)
35.1 (ss=4.7)
36,8 (ss=3.5)
32,8 (ss=5.2)
-1.36
-0.67
-0.70
350.50
0.51
0.48

Page 5

5
empathy group, whereas this value decreased to
12.2% in negative attitude group. These ratios re-
vealed that low empathy level was more and high
empathy level was less in negative attitude group
versus students with positive attitude.
The measurements demonstrated significant
differences before and after training between both
negative-positive attitude and high-low empathy
groups. This result was interpreted as the inadequ-
acy of education in the first year of medical edu-
cation in reaching the objective of equalizing the
benefit which would be received from following
training periods. When in-group changes before
and after training was taken into consideration,
none of the mean scores differed significantly
from one another, but it might be said that empathy
level was more sensitive to change. These chan-
ges were towards decrease in high empathy group
and increase in low empathy group. As low em-
pathy group formed the biggest portion of students
(75%), increasing effectiveness of education with
some reviews might be considered. Moreover,
this tendency may reach statistically significance
level in a larger sample size. Empathy levels af-
ter education increased in some recent studies and
decreased in others (Hojat et al. 2002, 2004). In a
study which evaluated whether empathy levels of
first grade students increased or not, it was found
that empathy levels changed in many of them after
education, but no increase in expressing empathy
was present in 30% of them (Winefeld and Chur-
Hansen 2000).
A shift towards negative attitude towards com-
munication may be expected in medical students as
they are not used to deal with social sciences. But
our results did not show such a negative change in
attitudes. Rees and Sheard reported lower levels of
positive attitude in first grade students from Not-
tingham and Leicester Universities after education
for communication skills (2003). Some variables
like training techniques and methods which tend to
change from experiential to didactic style, time pe-
riod spared and teacher characteristics result with
different levels of changes in information, skills
and attitudes. Our findings demonstrated that at-
titude changes towards communication were not
adequate, but greater number of students who had
lower empathy levels was more prone to change
after education. It was reported that to change and
measure attitudes was more difficult then to deve-
lop and measure skills (Turgut 1997). Performance
of empathy education by means of role-play met-
hod might provide additional benefit in our study.
Our empathy education and measurement methods
might be thought to be taken into consideration by
trainers and trainees in psychiatry and other rela-
ted psychological fields.
There were some reports about gender diffe-
rences in negative attitudes against communicati-
on (Batenburg and Smal 1997, Rees and Sheard
2002). Batenburg and Smal reported lesser levels
of negative attitude in females both before and af-
ter education (1997). Rees and Sheard reported si-
milar results in their study with first grade students
like our group (2002). In another study with third
grade students higher empathy scores were found
to be related to female gender (Hojat et al. 2002).
But in our study, there was not any difference bet-
ween genders or age groups according to ETS and
IATCM scores. Relatively small number of our
participants might obscure possible differences
and lead to type II error.
Studies like this one include methodological li-
mitations. Relatively low level accessibility to all
students at the last test form major limitations of
our study. Majority of participants in second test
were females by chance (65.7%). When observati-
on of Hojat and colleagues (2004) as empathy po-
ints of third grade students decreased after educati-
on is taken into consideration, it may be concluded
that both the same group during education period
and all students from different grades at the same
time should be evaluated. Moreover, comparison
of students from different college programs with
or without communication skills training may pro-
vide additional benefit. Differences between pre-
and post-test designs may be a consequence of
causes other than communication skills training.
In order to prevent this problem, controlled trial
FİGURE 1. Empathy tendencies of students according to attitudes
towards communication (n=36).
Positive attitude Negative attitude
Low empathy
High empathy

Page 6

6
without selection should be performed.
ETS is a valid and reliable instrument, but it
should be remembered that it was not developed to
measure educational gain. Moreover, the validity
study of IATCM is lacking despite its well known
reliability in terms of internal consistency.
This is the first study reporting data about the
efficiency of education for communication skills
in students of medical schools in Turkey. Results
give the impression that empathic tendency may
be increased with appropriate educational met-
hods. It may be concluded that dependence of
educational programs and methods to experiential
learning principles will lead to adequate clinical
results, more healthy people and better satisfied
physicians.
REFERENCES
AAMC. (1998) Learning Objectives for Medical Student
Education: Guidelines for Medical Schools, http://www.aamc.org/
meded/msop/start.htm.
Aspergen K. (1999) BEME Guide No. 2: Teaching and learning
communication skills in medicine, Medical Teacher, 21(6): 563-570.
Batenburg V and Smal JA (1997) Does a communication skills
course influence medical students’ attitudes? Medical Teacher, 19:
263-69.
Dereboy İF, Gürel M, Erpek S, Savk Ö. (2001) Tıp Eğitiminde
Tam Entegrasyona Doğru: Menderes Deneyimi. Toplum ve
Hekim,16:194-204.
Dökmen Ü. (1999) İletişim Çatışmaları ve Empati, Sistem
Yayıncılık, İstanbul.
Dökmen, Ü. (1988) Empatinin Bir Modele Dayandırılarak
Ölçülmesi ve Psikodrama ile Geliştirilmesi. A.Ü. Eğitim Bilimleri
Fakültesi Dergisi, 6 (21):155-190.
Egan G. (1998) The Skilled Helper; A Problem-Management
Approach to Helping, 6.baskı, Brooks/Cole Publishing Company.
General Medical Council (1993) Tomorrow’s Doctors:
Recommendations on undergraduate medical education.
Hargie O, Dickson D, Boohan M et al. (1998) A survey of
communication skills training in UK schools of medicine: present
practices and prospective proposals. Med Educ, 32(1):25-34.
Harrison A, Glasgow N (1996) Communication skills training
early in the medical curriculum: UAE experience. Medical Teacher;
18(1): 35-44.
Hojat M, Gonnella JS, Mangione S et al. (2002) Empathy
in medical students as related to academic performance, clinical
competence and gender. Med Educ, 36 (6): 522-7.
Hojat M, Veloski JJ, Borenstein BD (1986) Components of
clinical competence of physicians: an empirical approach. Educational
Psychol Measurement, 46: 761-9.
Hojat M, Mangione S, Nasca TJ ve ark. (2004) An emprical study
of decline in empathy in medical school, Med Educ, 38: 934-41.
Ludmerer KM (1999) Time to Heal. New York Oxford University
Press.
Makoul G (2003) Communication skills education in medical
school and beyond. JAMA 289(1): 293.
Nunnally JC, Bernstein IP (1994) Psychometric Theory, 3. baskı.
New York: Mc Graw Hill.
Rees C and Sheard C (2002) The relationship between medical
students’ attitudes towards communication skills learning and their
demographic and education related characteristics Med Educ, 36:
1017-27
Rees C ve Sheard C (2003) Evaluating first year medical
students’ attitudes to learning communication skills before and after a
communication skills course Medical Teacher, 25: 302-307.
Rosenthal J, Ogden J (1998) Changes in medical education: The
beliefs of medical students. Med Educ, 32: 127-32.
Sarıkaya Ö, Uzuner A, Gülpınar MA et al. (2004) İletişim
becerileri eğitimi: İçerik ve değerlendirme. Tıp Eğitimi Dünyası, 14:
28-37.
Sayek İ, Kılıç B (2000) MÖTER: Mezuniyet Öncesi Tıp Eğitimi
Raporu, Türk Tabipleri Birliği.
Starr P (1982) The Social Transformation of American Medicine,
New York: Basic Books.
General Medical Council (2003) Tomorrow’s Doctors:
Recommendations on undergraduate medical Education, General
Medical Council, London.
Turgut MF (1997) Eğitimde Ölçme ve Değerlendirme Metodları,
10.baskı, Ankara: Gül yay.
Van Dalen J, Bartholomeus P, Kerkhofs E, et al. (2001) Teaching
and Assessing communication skills in Maastricht: The first twenty
years. Medical Teacher, 23(3): 245-251.
Van Dalen J, VanHout JCHM, Scherpbier AJJA ve ark. (1999)
Factors influencing the effectiveness of communication skills
training: Programme contents outweigh teacher’s skills. Medical
Teacher, 21:308-11.
Wilkes M, Milgrom E, Hoffman JR (2002) Toward more
empathic students. Medical Education, 36: 528-33.
Winefeld Chur-Hansen (2000) Evaluating the outcome of
communication skill teaching for entry-level medical students: does
knowledge of empathy increase? Med Educ, 34(2):90-4.