International Journal of Medical Education.

2018;9:122-128
ISSN: 2042-6372
DOI: 10.5116/ijme.533f.0c41

Are alexithymia and empathy predicting factors

of the resilience of medical residents in France?
Audrey Morice-Ramat1, Lionel Goronflot1, Gilles Guihard2

Department of General Medicine, Faculty of Medicine, University of Nantes, France

1

Center for Research in Education of Nantes (CREN), University of Nantes, France

2

Correspondence: Gilles Guihard, Centre de Recherche en Education de Nantes (CREN), EA 2661, Université de Nantes, France
Email: [email protected]

Accepted: April 05, 2018

Abstract
Objectives: To explore resilience, resilience predicting fac- with resilience, r(135) = -.40, p<.001, and empathy,
tors and resilience distribution in French medical residents. r(135) = -.38, p<.001. Resilience was influenced by
Methods: A cross-sectional study was conducted in which alexithymia, β = -.284, p = .001, empathy, β = .255, p = .002,
gender (female < male), β = -.231, p = .002 and year of
general practice residents (n = 380) were asked to answer the
formation, β= .157, p = .036. Two clusters of residents were
Jefferson Scale of Physician Empathy, the Connor-Davidson characterized. They differed by their empathy and resilience
Resilience Scale, and the Toronto Alexithymia Scale. One profiles and by alexithymia trait.
hundred thirty-seven (137) responses were collected. The Conclusions: Alexithymia, empathy, gender and year of for-
scores of the different scales have been calculated. The score mation correspond to predicting factors of resilience. This
differences were examined using the Student’s t-test or anal- suggests that the resilience of vulnerable residents can be en-
ysis of variance. The correlations were estimated using the hanced by increasing their empathy and by reducing their
Pearson correlation coefficient. The relationships between alexithymia. Thus, teaching teams could sustain their stu-
scores were analysed by multiple linear regression. The het- dents’ well-being through educational programs aiming to
erogeneity of the sample was examined by non-hierarchical develop their understanding of their own emotions and those
cluster analysis. of their patients.
Results: Resilience and empathy were positively correlated Keywords: Burnout, coping, family medicine, medical for-
(r(135) = .36, p < .001). Alexithymia was negatively correlated mation, mental health, stress

Introduction
Professional burnout (PBO) represents an inadequate re- that ~15% to ~45 % of students are affected.16,17 Medical res-
sponse to chronic occupational stress.1 It results in psycho- idents are not spared from burnout. In particular, 25% of
logical and physiological consequences that can strongly im- general practice residents are impacted by PBO in France.18
pact individual well-being, quality of life and job The resilience represents the ability to show positive psy-
performance. Pathological signs of PBO include cardiovas- chophysiological outcomes despite experiencing aversive sit-
cular and gastrointestinal diseases.2,3 PBO is also accompa- uations or living in a stressful environment.19 Improving the
nied by psychological symptoms such as anxiety, depression, resilience of individuals has been evoked to prevent the PBO
motivational decrease, reduced interpersonal commitment occurrence in a broad working context.20 Resilience and PBO
and performance limitations.4 The PBO prevalence has been are characterized by a negative correlation in health profes-
estimated in US working adults (~28%),5,6 in the general pop- sions.21-24 Such a correlation is also reported for medical stu-
ulation of Sweden (~13%),7 and Germany (~4%).8 A growing dents.25,26 Numerous factors influence the resilience of indi-
body of evidence shows that PBO affects health care provid- viduals. These include age, gender, cultural environment,
ers,9-11 with a prevalence depending upon factors including living perspectives and personality traits.19,27 Alexithymia
the geographic location and the specialty of the health pro- corresponds to the impaired understanding of one’s own
fession.12,13 Burnout is also reported to impact health students emotions.28 Alexithymia is observed in patients suffering
during their training.14-16 The prevalence has been deter- from psycho-pathological diseases that reduce patient’s resil-
mined in different populations of medical students, showing ience.29,30 Negative correlations between alexithymia and
122
© 2018 Audrey Morice-Ramat et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License which permits unrestricted use
of work provided the original work is properly cited. http://creativecommons.org/licenses/by/3.0
resilience have been characterized in Chinese military per- Study tools
sonnels31 and in Iranian students.32 Empathy can also repre- The questionnaire included items related to socio-demo-
sent a factor influencing the resilience. Empathy represents graphic information (gender, age, year of the curriculum).
the ability to experience other’s emotions and to manifest a Items from the French versions of the Jefferson Scale of Phy-
cognitive adaptation to promote better interpersonal rela- sicians Empathy (f-JSPE; 25 items), the Connor-Davidson
tionships.33 Indirect clues suggest a cross-talk between empa- Resilience Scale (f-CDRISC; 21 items) and the 20-item To-
thy and resilience.34 However, the existence of a significant ronto Alexithymia Scale (f-TAS20) was also included in our
correlation between empathy and resilience is still dis- survey. These scales have demonstrated satisfactory reliabil-
cussed.35,36 ity for the assessment of resilience, empathy, and alexi-
The distribution of resilience among populations was thymia.43-45
previously assessed by using cluster analysis procedure. The scoring of f-JSPE items is based on a 7-point Likert
Pietrzak and Cook reported the existence of three clusters of scale with one corresponding to “full disagreement” and
individuals with distinct resilience levels among a sample of seven corresponding to “full agreement”.10 items needed re-
US veterans.37 Doron and colleagues identified five clusters verse scoring because of their formulation. Total f-JSPE score
of students differing by their coping strategies in response to varies from 20 to 140; a high score indicates high empathy.
stress.38 The work of Suriá Martínez indicated the existence F-CDRISC and f-TAS20 scales use a 5-point Likert scale for
of different resilience profiles among a sample of patients item scoring. For f-CDRISC, the item score varies from 0
with spinal cord injury.39 Taken together, these works suggest (“full disagreement”) to 4 (for “full agreement”), and the total
that the resilience is not distributed according to a single nor- score varies from 0 to 84. For f-TAS20, the item score varies
mal distribution. This is also the case for empathy and alexi- from 1 (for “full disagreement”) to 5 (for “full agreement”).
thymia distributions.40,41 Five items needed reverse scoring. Total f-TAS20 score varies
Although resilience measurement has been described in from 20 to 100. High scores for f-CDRISC and f-TAS20 indi-
medical students and in medical residents,25,26,42 little is cate high resilience and alexithymia traits.
known about the predicting factors of resilience. Further-
more, there is a lack of study concerning the characterization Data collection procedures
of resilience and resilience predicting factors in French All general practice residents were contacted by e-mail. They
health students. Therefore, the objective of the present study were invited to answer an electronic version of the question-
is to explore the resilience of medical students. To do so, a naire. The access to the questionnaire was granted after the
survey was conducted to measure resilience, to determine the validation of an electronic informed consent in which goals,
nature of predicting factors, and to characterize the distribu- means, and methods have been described. To ensure resi-
tion of resilience among general practice residents at Nantes dent’s anonymity, each resident encoded an identifier by us-
University (France). It was hypothesized that: i) the resilience ing the first two letters of his/her first name, a number corre-
is affected by predicting factors including empathy and alex- sponding to the day of birth (between 01 to 31), a number
ithymia, and ii) the resilience distribution among a student corresponding to the year of birth (between 00 and 99), a
sample is heterogeneous. number corresponding to the place of birth (between 00 and
101) and the first two letters of his/her mother’s given name.
Methods Data analysis
Data were collected at the end of the survey. They were ana-
Study design and participants lysed with SPSS 21, Sigma Plot 12 and R (3.2.5) software. F-
A cross-sectional study was conducted at Medical School of CDRISC scale has recently been validated for resilience
Nantes University (France). Ethical approval was obtained measurement in health students.43 The reliability of f-JSPE
from the Ethics committee of Nantes University. Project staff and f-TAS20 scales was re-assessed in our conditions. The
was not involved in the diploma allocation. All general prac- item-score correlation coefficient (rIS) was calculated. As
tice residents (n=380) were eligible to the study. The return Cronbach’s α coefficient is not a good estimator of internal
rate was ~40% (n=150). Thirteen incomplete answers or with consistency for multifactorial scales,46 greatest lower bound
outlying scores (Dixon’s test) were discarded. The final sam-
(GLB) and McDonald’s ω coefficient were calculated as rec-
ple (n=137, mean age =26.5, SD=1.3) corresponded to 94
ommended elsewhere.47 The scale appropriateness was de-
women (mean age =26.6, SD=1.2) and 43 men (mean age
duced from the Kaiser-Meyer-Olkin coefficient (KMO, opti-
=26.5, SD =1.6). The proportions of residents registered in
mal value above .8) and the Bartlett’s test (optimal p < .001).
years 1, 2 and 3 of the formation corresponded to ~30% (n =
The item-sampling adequacy was deduced from the anti-im-
41), ~39% (n=53) and ~31% (n=43). The gender ratio (F/M)
age correlation coefficient value calculated for each item
differed significantly between years 1 (25/16), 2 (43/10) and
(AIC, optimal value above .5).
3 (26/17) (χ2(2, N=137) = 6.293, p=.043). Becoming a general
practice resident was a deliberate choice for 123 respondents To confirm the 3-factor structure of the f-JSPE and f-
(~90%). TAS20 scales, a confirmatory factor analysis (CFA) was

Int J Med Educ. 2018;9:122-128 123

Morice-Ramat et al. Alexithymia, empathy and resilience of medical residents

performed by using the maximum likelihood method. The and discriminant analyses, the significance criterion was set
goodness of fit of a factor model was estimated according to at p < .001.
Byrne’s recommendations.48 The following indices were
calculated: i) the normed χ2 (χ2 /df, optimal value below 2.0), Results
ii) the standardized root mean square of residuals (sRMR, The normality of f-JSPE and f-TAS20 scores was confirmed
optimal value below .05), iii) the goodness of fit index (GFI) using the Shapiro-Wilk test. Therefore, a factor analysis
and the adjusted goodness of fit index (AGFI), both being using maximum likelihood method and an orthogonal
optimally higher than .90), iv) the root mean square error of rotation (Varimax) could be performed to determine the
approximation (RMSEA, optimal value below .08) and its indicators of reliability, as suggested by Costello and
relative p(close) for which a value above .05 indicates a good Osborne.52 As shown in Table 1, f-JSPE and f-TAS20 scales
fit, and v) the comparative fit index (CFI) (optimal value were characterized by acceptable to good appropriateness,
above .90). item-sampling adequacy, and item-score correlation. The
The normality of data distribution was verified by using internal consistency was considered from acceptable (for
the Shapiro-Wilk’s test (significance criterion p>.05). Mean f-JSPE) to strong (for f-TAS20) as demonstrated by GLB and
(M), standard deviation (SD), 95% confidence interval (95% ωvalues. A 3-factor structure was determined for both f-JSPE
CI) and the average score (i.e., the ratio between the total and f-TAS20 scales, as illustrated by the values of goodness-
score and the number of items of a scale) were calculated for of-fit indices (Table 1).
the different scales and different subgroups. A correlation Scores analysis
between two variables was deduced from the value of the As shown in Table 2, f-CDRISC was significantly best-scored
Pearson correlation coefficient. The differences were by male residents (low effect size). A significant difference in
estimated by Student’s t-test (for two-group comparison) resilience (low effect size) was observed between the
and by analysis of variance (ANOVA, for multiple-group residents of the different years of formation. However, our
comparison) in considering a risk α = .05. The threshold of analysis indicated that such a difference originated more
significance of a difference was set at p<.05. When a differ- likely from an interaction between gender and year of
ence was significant, the effect size was estimated by Cohen’s formation. Gender-related or curriculum-related differences
d coefficient (for two-group comparison) with the correction for f-JSPE or f-TAS20 scores were not significant.
of Rosnow and Rosenthal49 or by η2 coefficient (for multiple-
group comparison). Small, medium and large effect sizes Multiple linear regression analysis
were respectively characterized by d, η2 ≤ .2, .2 < d, η2 ≤ .5 Resilience and empathy were positively correlated, r(135) = .36,
and .5 < d, η2.50 p<.001. Negative correlations were observed between alexi-
thymia and resilience, r(135) = -.40, p<.001, and between alex-
Multiple linear regression analyses were performed to
ithymia and empathy, r(135) = -.38, p<.001. A regression model
test whether alexithymia and empathy can predict resilience.
describing the contribution of the different variables to the
Gender (male = 1, female = 2) and year of formation (year 1
resilience was elaborated. It was supported by a significant
= 1, year 2 = 2 and year 3 = 3) were also considered as poten-
regression equation, R2adjusted = .27, F(4,132) = 13.39, p< .001,
tial predicting factors of resilience. Standardized regression
Durbin-Watson coefficient = 1.625. The resilience was posi-
coefficient (β) and p values were calculated to estimate the
tively predicted by empathy, β = .255, t(132) = 3.19, p=.002, and
relationships between the different variables.
by year of formation, β = .157, t(132) = 2.12, p=.036. Mean-
The heterogeneity within a dataset can be ascertained by while, gender (female < male, β = -.231, t(132) = -3.14, p= .002)
cluster analysis (CA).51 A non-hierarchical CA (K-means) and alexithymia, β =-.284, t(132)=-3.57, p < .001, corresponded
was run by using standardized scores (z scores) for f-JSPE, f- to negative predicting factors. Alexithymia negatively influ-
CDRISC and f-TAS20 as clustering variables. The validity of enced empathy, β =-.270, t(132) = -3.21, p=.002. Neither alexi-
different models (from 2 to 4 clusters) was assessed. The sig- thymia nor empathy was affected by gender or by year of for-
nificance of between-cluster differences was calculated by mation. This model was validated by CFA as demonstrated
unpaired Student’s t-test (for a 2-cluster model) or by by the goodness-of-fit indices, χ2(5, N = 137)/df = .55, GFI =
ANOVA and post-hoc Bonferroni correction (for 3- and 4- .99, AGFI = .98, sRMR = .042, RMSEA < .001, p(close) = .846.
cluster models). The most likely cluster model contained the
highest number of clusters for which all z scores were signif- Cluster analysis
icantly different. The validity of the retained model was as- Different convergent solutions resulting from CA were
sessed by discriminant analysis (DA), in which z scores were observed. However, ANOVA shows that the differences
considered as independent variables, whereas the number of produced by 3- and 4-cluster models were not significant
clusters corresponds to the dependent variable. For cluster (data not shown). For this reason, the 2-cluster model was

124
Table 1. Determination of the psychometric properties of empathy and alexithymia scales

AGFI: Adjusted Goodness-of-Fit Index; AIC: anti-image coefficient; CFI: Confirmatory Fit Index; X2/df: normed X2; df: degree of freedom; GFI: Goodness-of-Fit Index; GLB: Greatest Lower Bound

Mean Indicators of internal

Confirmatory analysis
Scale KMO Bartlett’s test AIC range rIS (135) (SD) consistency
[95% CI] χ2/df sRMR GFI AGFI RMSEA CFI GLB ω [95% CI]
0.29 (0.15)
χ2(190, Ν = 137) = 598.8,
f-JSPE 0.79 0.608 - 0.872 1.32 0.07 0.87 0.84 0.047 0.88 0.87 0.70 [0.67-0.74]
p < 0.001 [0.22-0.36]

0.42 (0.16)
χ2(190, Ν = 137) = 830.9,
f-TAS20 0.814 0.610 - 0.911 1.34 0.07 0.87 0.83 0.05 0.92 0.91 0.84 [0.81- 0.88]
p < 0.001 [0.34-0.50]

coefficient; rIS: item-score correlation coefficient; KMO: Kaiser-Meyer-Olkin coefficient; ω: McDonald’s ω coefficient; RMSEA: Root Mean Square Error of Approximation; sRMR: standardized Root
Mean Square of Residuals; [95% CI]: 95% confidence interval.

considered as the most reliable solution. This was validated populations.56,57 There are contradictory results concerning a
by DA. The cluster effective calculated by DA were found to gender-related difference for alexithymia in the literature. In-
be identical to those determined by CA. A very strong corre- deed, previous observations highlight some gender-related
lation between the allocations determined by CA and by DA differences in German and Jordanian general popula-
was observed, r(135) =.98, p< .001. tions.56,58 However, other reports show that alexithymia is un-
The gender repartition determined for each cluster was affected by the gender of French or Japan individuals.45,59,60
similar (Table 3). By contrast, between-cluster differences Our study shows that the gender of French medical residents
calculated for resilience, empathy, and alexithymia were sig- does not influence their alexithymia trait.
nificant. The residents of cluster 1 were more resilient and The average score of empathy calculated in the present
more empathetic than those of cluster 2 (strong effect size). work is in good agreement with those calculated for medical
The residents allocated to cluster 2 exhibited a higher alexi- students61 and for French medical practitioners.54,62 The fact
thymia profile (strong effect size). The between-cluster dif- that empathy level is affected by gender or not, is currently
ference was also observed when the year of formation was debated. On the one hand, a higher empathy is reported for
considered as comparison criterion. In particular, residents females in samples of undergraduate students of Portugal
from 3rd year were more abundant in cluster 1, whereas clus- and USA.63,64 On the other hand, North American and Bra-
ter 2 was mainly constituted by 1st and 2nd-year residents. zilian male residents have higher empathy levels than their
female equivalents.65,66 The present study does not report any
Discussion gender-related empathy difference for French medical resi-
The objective of the present study consisted in a better un- dents. This suggests that medical formation at Nantes Uni-
derstanding of the characteristics of resilience of medical stu- versity minimizes or abolishes the empathy difference be-
dents. Our work was devoted to the measure of resilience, to tween male and female medical students during their early
the characterisation of some resilience predicting factors and training.
to the analysis of the resilience distribution in a sample of The resilience measured for French medical residents is
French general practice residents. Two working hypotheses higher than that determined for Chinese, Korean and US
have been tested: i) the resilience is affected by several pre- general populations.67-69 By contrast, it is close to that
dicting factors including empathy and alexithymia and ii) the measured for Australian nurses or Brazilian athletes daily liv-
distribution of resilience among the studied sample is heter- ing in stressful conditions70,71 and for Chinese and Turkish
ogeneous. earthquake survivors.72,73 This indicates that medical
formation provides a training environment susceptible to
Characterization of the resilience support the resilience of medical students.
To explore our first working hypothesis, f-JSPE and f-TAS20 The relationships between resilience, empathy, and
scales are used for empathy and alexithymia measurements. alexithymia have never been fully described, because of a lack
These scales have been described as reliable tools in different of concomitant measurements. It is shown in the present
French-speaking samples.44,45,53-55 However, a psychometric work that empathy and alexithymia correspond to significant
scale needs de novo validation when study conditions are predicting factors of resilience. Indeed, resilience is
changed.52 Our work confirms that f-JSPE and f-TAS20 negatively influenced by alexithymia and positively affected
correspond to reliable tools for empathy and alexithymia by empathy which is also negatively influenced by
measurements in French medical residents. alexithymia. A decline of medical students’ empathy occurs
The average score of alexithymia calculated for French as students progress in their training.74 Our findings suggest
medical residents is similar to those calculated for French that such a decline can be accompanied by the decrease of the
asymptomatic adults,45 and for German and Japan general students’ resilience, thus rendering the students more

Int J Med Educ. 2018;9:122-128 125

Morice-Ramat et al. Alexithymia, empathy and resilience of medical residents

Table 2. Resilience, empathy and alexithymia of general practice residents

Variable f-CDRISC f-JSPE f-TAS20

Overall (n=137) 3.17 (.73) [2.77 – 3.57] 5.61 (.43) [5.44 – 5.73] 2.38 (.54) [2.23 – 2.58]
Female (n=94) 2.79 (.70) [2.57 – 3.01] 5.63 (.41) [5.44 – 5.72] 2.38 (.54) [2.28 – 2.62]
Male (n=43) 3.34 (.63) [3.17 – 3.57] 5.57 (.47) [5.44 – 5.73] 2.37 (.56) [2.17 – 2.53]
Gender Comparison (d) t(135) = 2.09 , p = .038 (.12) t(135) = .69 , p = .408 t(135) = .01 , p = .918
Year-1 (41) 3.13 (.67) [2.92 – 3.34] 5.57 (.49) [5.46 – 5.71] 2.40 (.56) [2.23 – 2.58]
Year-2 (53) 2.79 (.69) [2.57 – 3.00] 5.57 (.47) [5.41 – 5.72] 2.41 (.54) [2.24 – 2.57]
Year-3 (43) 3.49 (.63) [3.29 – 3.69] 5.69 (.40) [5.56 – 5.82] 2.27 (.55) [2.09 – 2.44]
Between-year Comparison (η2) F(2, 134) = 6.10, p < .001 (.18) F(2, 134) = .60, p = .549 F(2, 134) = .47, p = .628
Gender x Year Comparison F(1, 131) = 3.70, p = .027 F(1,131) = 2.25, p = .109 F(1, 131) = 2.12, p = .124

Data correspond to average scores (SD) and 95% confidence interval ([95% CI]) calculated for the overall sample and different sub-samples.
d and η2: effect size of differences.

Table 3. Characteristics of the clusters resulting from cluster and discriminant analyses

Characteristics Cluster 1 Cluster 2

Analysis test, significance (d)
N 54 83

f-CDRISC 3.58 (.62) 2.81 (.64) t(135) = -6.99, p< .001 (1.22)

Cluster analysis f-JSPE 5.91 (.31) 5.42 (.38) t(135) = -8.06, p< .001 (1.44)
f-TAS20 1.88 (.35) 2.70 (.38) t(135) = 12.91, p< .001 (2.27)

N 54 83
f-CDRISC 3.58 (.62) 2.81 (.63) t(135)= -6.95, p< .001 (1.23)
Discriminant analysis
f-JSPE 5.91 (.31) 5.43 (.38) t(135)= -7.82, p< .001 (1.40)
f-TAS20 1.87 (.34) 2.70 (.38) t(135)= 13.15, p< .001 (2.33)

Gender (F - M) 36 - 18 58 - 25 χ2(1, N=137) =.04, p=.84

1 14 (25.9) 27 (32.5)

Year of formation (%) 2 16 (29.6) 37 (44.6) χ2(2, N=137) =7.21, p=.03

3 24 (44.5) 19 (22.9)

The data correspond to average scores (SD). P value indicates the significance. The effect size is given by Cohen’s coefficient (d).

vulnerable to PBO. Apart from training programs suggested suggests that the medical residency training increases the
for resilience enhancement,75,76 education programs devoted emotional understanding and the resilience of the residents.
to the reinforcement of students’ empathy could represent an Considering normative data,61,69 the residents of both clusters
efficient strategy aiming to support resilience. Also, helping exhibit non-pathological levels of empathy and resilience.
students in the understanding of their own emotions could However, a TAS20 score higher than 2.65 is reported to indi-
also figure a protection factor against PBO. cate a strong alexithymia propensity.56 Consequently, our
work suggests that the residents of the second cluster have a
Analysis of the distribution of resilience
high alexithymic profile. Therefore, this work corroborates
Cluster analysis has been used to describe heterogeneous dis- Shapiro’s comments concerning the promotion of students’
tributions of empathy,64 alexithymia57,77 and resilience37 alexithymia during preclinical and clinical medical for-
among pathological and non-pathological populations. mation.78 Predictive determinants of alexithymia during
However, these results have been obtained with independent early stages of medical training shall be investigated in future
measurements of empathy, alexithymia or resilience. In the studies.
present study, empathy, resilience, and alexithymia are con-
comitantly measured, and cluster analysis considers the three Limitations
variables simultaneously. Two clusters of residents are iden- Although significant datasets support our work, it presents
tified in very good conditions of confidence. The former is several limitations. It corresponds to a monocentric study
mainly composed of 3rd-year residents with low alexithymia (Faculty of Medicine of Nantes). The results are based on a
trait and high empathy and resilience. The latter is predomi- limited number of responses (49% of the population of resi-
nantly constituted by 1st and 2nd-year residents with a high dents). A selection bias may be present in the survey, as col-
alexithymia propensity and low empathy and resilience. This lected responses can originate from students with a positive
126
a priori for this study. Furthermore, medical residents are po- 11. de Paiva LC, Canário ACG, de Paiva China ELC, Gonçalves AK. Burnout
syndrome in health-care professionals in a university hospital. Clinics (Sao
tentially able to identify socially acceptable answers concern-
Paulo). 2017;72(5):305-9.
ing the level of empathy associated with medical practice. 12. Soler JK, Yaman H, Esteva M, Dobbs F, Asenova RS, Katic M, Ozvacic Z,
Fisher and Katz describe this social desirability-bias on self- et al. Burnout in European family doctors: the EGPRN study. Fam Pract.
report assessment.79 It may be responsible for some overesti- 2008;25(4):245-65.
13. Lee RT, Seo B, Hladkyj S, Lovell BL, Schwartzmann L. Correlates of phy-
mated responses and the present survey. The last limitation
sician burnout across regions and specialties: a meta-analysis. Hum Resour
concerns the design of the study which corresponds to a Health. 2013;11:48-63.
cross-sectional survey. This prevents the observation of time- 14. Ishak W, Nikravesh R, Lederer S, Perry R, Ogunyemi D, Bernstein C.
related changes in empathy, resilience, and alexithymia. Burnout in medical students: a systematic review. Clin Teach.
2013;10(4):242-5.
Consequently, further longitudinal investigations based on 15. Bullock G, Kraft L, Amsden K, Gore W, Prengle B, Wimsatt J, et al. The
larger samples obtained from different medical schools are prevalence and effect of burnout on graduate healthcare students. Can Med
necessary. Educ J. 2017; 8(3): e90–e108.
16. de Cavalcante Almeida G, de Souza HR, de Almeida PC, de Cavalcante
Conclusions Almeida B, Almeida GH. The prevalence of burnout syndrome in medical
students. Arch Clin Psychiatry. 2016;43(1):6-10.
Our study corresponds to the first concomitant analysis of 17. Dyrbye LN, Thomas MR, Huntington JL, Lawson KL, Novotny PJ, Sloan
resilience, empathy and alexithymia traits in a sample of JA, et al. Personal life events and medical student burnout: a multicentre
French general practice residents. It is shown that empathy, study. Acad Med. 2006;81(4):374-84.
18. Galam E, Komly V, Le Tourneur A, Jund J. Burnout among French GPs
alexithymia, year of formation and gender are predicting fac- in training: a cross-sectional study. Br J Gen Pract. 2013; 63(608): e217-24.
tors of resilience. Our work suggests that it is possible to sus- 19. Rutter M. Annual research review: resilience - clinical implications. J
tain the resilience of residents by acting on their empathy and Child Psychol Psychiatry. 2013;54(4):474-87.
20. Jackson D, Firtko A, Edenborough M. Personal resilience as a strategy for
alexithymia. Our study shows that residents can be seg-
surviving and thriving in the face of workplace adversity: a literature review.
mented into two subgroups of distinct profiles concerning J Adv Nurs. 2007;60(1):1-9.
their empathy and resilience and their alexithymia. Medical 21. Hao SW, Hong W, Xu HH, Zhou LL, Xie ZY. Relationship between resil-
training teams should consider these findings to improve ience, stress and burnout among civil servants in Beijing, China: mediating
and moderating effect analysis. Personal Individ Differ. 2015;83:65-71.
their teaching strategies.
22. Treglown L, Palaiou K, Zarola A, Furnham A. The dark side of resilience
and burnout: a moderation-mediation model. PLoS One.
Conflict of Interest 2016;11(6):e0156279.
The authors declare that they have no conflict of interest. 23. Zou G, Shen X, Tian X, Liu C, Li G, Kong L, Li P. Correlates of psycho-
logical distress, burnout, and resilience among Chinese female nurses. Ind
References Health. 2016;54(5):389-95.
24. Guo YF, Luo YH, Lam L, Cross W, Plummer V, Zhang JP. Burnout and
1. Salvagioni DAJ, Melanda FN, Mesas AE, González AD, Gabani FL, An- its association with resilience in nurses: a cross-sectional study. J Clin Nurs.
drade SM. Physical, psychological and occupational consequences of job 2017;1-9.
burnout: a systematic review of prospective studies. PLoS One. 25. Thompson G, McBride RB, Hosford CC, Halaas G. Resilience among
2017;12(10):e0185781. medical students: the role of coping style and social support. Teach Learn
2. Dimsdale JE. Psychological stress and cardiovascular disease. J Am Coll Med. 2016;28(2):174-82.
Cardiol.2008;51(13):1237-46. 26. Houpy JC, Lee WW, Woodruff JN, Pincavage AT. Medical student resili-
3. Huerta-Franco MR, Vargas-Luna M, Tienda P, Delgadillo-Holtfort I, ence and stressful clinical events during clinical training. Med Educ Online.
Balleza-Ordaz M, Flores-Hernandez C. Effects of occupational stress on the 2017;22(1):1320187.
gastrointestinal tract. World J Gastrointest Pathophysiol. 2013;4(4):108-18. 27. Daskalakis NP, Bagot RC, Parker KJ, Vinkers CH, de Kloet ER. The three-
4. Gunnar M, Quevedo K. The neurobiology of stress and development. Annu hit concept of vulnerability and resilience: toward understanding adaptation
Rev Psychol. 2007;58:145-73. to early-life adversity outcome. Psychoneuroendocrino 2013;38:1858-73.
5. Shanafelt TD, Boone S, Tan L, Dyrbye LN, Sotile W, Satele D, et al. Burnout 28. Taylor GJ. Recent developments in alexithymia theory and research. Can
and satisfaction with work-life balance among US physicians relative to the J Psychiatry. 2000;45(2):134-42.
general US population. Arch Intern Med. 2012;172(18):1377-85. 29. Bagby RM, Taylor GJ. Affect dysregulation and alexithymia. In: Taylor
6. Shanafelt TD, Hasan O, Dyrbye LN, Sinsky C, Satele D, Sloan J, et al. GJ, Bagby RM, Parker JD, editors. Disorders of affect regulation: alexithymia
Changes in burnout and satisfaction with work-life balance in physicians and in medical and psychiatric illness. Cambridge: Cambridge University Press;
the general us working population between 2011 and 2014. Mayo Clin Proc. 1997.
2015;90(12):1600-13. 30. Honkalampi K, Hintikka J, Koivumaa-Honkanen H, Antikainen R, Haa-
7. Norlund S, Reuterwall C, Höög J, Lindahl B, Janlert U, Birgander LS. Burn- tainen K, Viinamaki H. Long-term alexithymic features indicate poor recov-
out, working conditions and gender - results from the northern Sweden ery from depression and psychopathology. A six-year follow-up. Psychother
MONICA study. BMC Public Health. 2010;10:326-34. Psychosom. 2007;76(5):312–4.
8. Maske UE, Riedel-Heller SG, Seiffert I, Jacobi F, Hapke U. Prevalence and 31. Wei H-h, Zhang L-y, Zhou X-d, Wang Y-c, Ren Z-w, H,Chen C-x, et al.
comorbidity of self-reported diagnosis of burnout syndrome in the general A study of correlation between alexithymia and resilience in military person-
population. Psychiatr Prax. 2016;43(1):18-24. nel. Chin J Behav Med Brain Sci. 2011;20(12):1134-36.
9. van Mol MM, Kompanje EJ, Benoit DD, Bakker J, Nijkamp MD. The prev- 32. Fasihi R, Hassanzadeh R, Mahmoudi G. Correlation between attachment
alence of compassion fatigue and burnout among healthcare professionals in style, alexithymia and resiliency in university students. Int J Sci Basic Appl
intensive care units: a systematic review. PLoS One. 2015;10(8):e0136955. Res. 2013;2(10):898-902.
10. Kumar S. Burnout and doctors: prevalence, prevention and intervention. 33. Halpern J. What is Clinical Empathy? J Gen Intern Med. 2003;18(8):670-
Healthcare (Basel). 2016;4(3):37-45. 74.

Int J Med Educ. 2018;9:122-128 127

Morice-Ramat et al. Alexithymia, empathy and resilience of medical residents

34. Park KH, Kim DH, Kim SK, Yi YH, Jeong JH, Chae J, et al. The relation- 2014;68:71-6.
ships between empathy, stress and social support among medical students. 58. Hamaideh SH. Alexithymia among Jordanian university students: its
Int J Med Educ. 2015;6:103-8. prevalence and correlates with depression, anxiety, stress, and demographics.
35. Olson K, Kemper KJ, Mahan JD. What factors promote resilience and Perspect Psychiatr Care. 2018;54(2):274-280.
protect against burnout in first-year pediatric and medicine-pediatric resi- 59. Zhu X, Yi J, Yao S, Ryder AG, Taylor GJ, Bagby RM. Cross-cultural vali-
dents? J Evid Based Complementary Altern Med. 2015;20(3):192-8. dation of a Chinese translation of the 20-item Toronto Alexithymia Scale.
36. Mathad MD, Pradhan B, Rajesh SK. Correlates and predictors of resili- Compr Psychiatry. 2007;48(5):489-96.
ence among baccalaureate nursing students. J Clin Diagn Res. 60. Ling Y, Zeng Y, Yuan H, Zhong M. Cross-cultural validation of the 20-
2017;11(2):JC05-8. item Toronto Alexithymia Scale in Chinese adolescents. J Psychiatr Ment
37. Pietrzak RH, Cook JM. Psychological resilience in older US veterans: re- Health Nurs. 2016;23(3-4):179-87.
sults from the national health and resilience in veterans study. Depress anxi- 61. Hojat M, Gonnella JS. Eleven years of data on the Jefferson Scale of Em-
ety. 2013;30(5):432-43. pathy-Medical Student version (JSE-S): proxy norm data and tentative cut off
38. Doron J, Trouillet R, Maneveau A, Ninot G, Neveu D, Coping profiles, scores. Med Princ Pract. 2015;24(4):344-50.
perceived stress and health-related behaviors: a cluster analysis approach. 62. Lelorain S, Sultan S, Zenasni F, Catu-Pinault A, Jaury P, Boujut E, et al.
Health Promot Int. 2015; 30(1):88-100. Empathic concern and professional characteristics associated with clinical
39. Suriá Martínez R. Profiles of resilience and quality of life in people with empathy in French general practitioners. Eur J Gen Pract. 2013;19(1):23-8.
acquired disability due to traffic accidents. Gac Sanit. 2015;29(S1):55-9. 63. Hojat M, Gonnella JS, Mangione S, Nasca TJ, Veloski JJ, Erdmann JB, et
40. Chen J, Xu T, Jing J, Chan RC. Alexithymia and emotional regulation: a al. Empathy in medical students as related to academic performance, clinical
cluster analytical approach. BMC Psychiatry. 2011; 11(33):1-6. competence and gender. Med Educ. 2002;36(6):522-7.
41. Euler F, Steinlin C, Stadler C. Distinct profiles of reactive and proactive 64. Duarte Ferreira MI, Barata Raposo ML, Fonseca da Silva Farinha Rodri-
aggression in adolescents: associations with cognitive and affective empathy. guès PJ, Castelo Branco M. Measuring empathy in medical students, gender
Child Adolesc Psychiatry Ment Health. 2017;11:1-14. differences and level of medical education: an identification of a taxonomy of
42. Bacchi S, Licinio J. Resilience and psychological distress in psychology students. Inv Ed Med. 2016;5(20):253-60.
and medical students. Acad Psychiatry. 2017;41(2):185-8.
65. Moralle MR, Preston JS, Chen L, Berberian WS. An exploratory study of
43. Guihard G, Deumier L, Alliot-Licht B, Bouton-Kelly L, Michaut C,
empathy in resident physicians at an urban medical center. IJCS.
Quilliot F. Psychometric validation of the French version of the Connor-Da-
2016;9(2):526-33.
vidson Resilience Scale. Encephale. 2018;44(1):40-45.
66. Santos MA, Grosseman S, Morelli TC, Giuliano IC, Erdmann TR. Empa-
44. Zenasni F, Boujut E, Buffel du Vaure C, Catu-Pinault A, Tavani JL, Rigal
thy differences by gender and specialty preference in medical students: a
L, et al. Development of a French-language version of the Jefferson Scale of
study in Brazil. Int J Med Educ. 2016;7:149-53.
Physician Empathy and association with practice characteristics and burnout
67. Connor KM, Davidson JR. Development of a new resilience scale: the
in a sample of general practitioners. Int J Person Centered Med.
Connor-Davidson Resilience Scale (CD-RISC). Depress Anxiety.
2012;2(4):759-66.
2003;18(2):76-82.
45. Loas G, Corcos M, Stephan P, Pellet J, Bizouard P, Venisse JL, et al. Fac-
68. Jung YE, Min JA, Shin AY, Han SY, Lee KU, Kim TS, et al. The Korean
torial structure of the 20-item Toronto Alexithymia Scale. Confirmatory fac-
version of the Connor-Davidson Resilience Scale: an extended validation.
torial analyses in nonclinical and clinical samples. J Psychosom Res.
Stress Health. 2012;28(4):319-26.
2001;50(5):255-61.
69. Ni MY, Li TK, Yu NX, Pang H, Chan BH, Leung GM, et al. Normative
46. Crutzen R, Peters GY. Scale quality: alpha is an inadequate estimate and
data and psychometric properties of the Connor-Davidson Resilience Scale
factor-analytic evidence is needed first of all. Health Psychol Rev.
(CD-RISC) and the abbreviated version (CD-RISC2) among the general pop-
2017;11(3):242-7.
ulation in Hong Kong. Qual Life Res. 2016;25(1):111-6.
47. Peters G-JY. The alpha and the omega of scale reliability and validity: why
and how to abandon Cronbach’s alpha and the route towards more compre- 70. Gillespie BM, Chaboyer W, Wallis M. The influence of personal charac-
hensive assessment of scale quality. EHP. 2014;16(2):56-69. teristics on the resilience of operating room nurses: a predictor study. Int J
48. Byrne BM. In: Structural equation modeling with AMOS: basic Concepts, Nurs Stud. 2009;46(7):968-76.
applications, and programming, 2nd edition. Byrne BM, editor. New York: 71. Belem IC, Caruzzo NM, Andrade do Nascimento Junior JR, Lopes Vieira
Taylor & Francis Group, Routledge; 2009. JL Vieira LF. Impact of coping strategies on resilience of elite beach volleyball
49. Rosnow RL, Rosenthal R. Computing contrasts, effect sizes, and conter- athletes. Rev. Bras. Cineantropom. Desempenho Hum. 2014;16(4):447-55.
nulls on other people’s published data: general procedures for research con- 72. Karaırmak Ö. Establishing the psychometric qualities of the Connor-Da-
sumers. Psychol Methods. 1996;1:331-40. vidson Resilience Scale (CD-RISC) using exploratory and confirmatory fac-
50. Lakens D. Calculating and reporting effect sizes to facilitate cumulative tor analysis in a trauma survivor sample. Psychiatr Res.2010; 179:350-6.
science: a practical primer for t-tests and ANOVAs. Front Psychol. 2013; 73. Yu XN, Lau JT, Mak WW, Zhang J, Lui WW, Zhang J. Factor structure
4:863. and psychometric properties of the Connor-Davidson Resilience Scale
51. Bahr PR, Bielby R, House E. The use of cluster analysis in typological re- among Chinese adolescents. Compr Psychiatry. 2011;52(2):218-24.
search on community college students. NDIR. 2011;S1:67-81. 74. Neumann M, Edelhäuser F, Tauschel D, Fischer MR, Wirtz M, Woopen
52. Costello AB, Osborne JW. Best Practices in exploratory factor analysis: C, et al. Empathy decline and its reasons: a systematic review of studies with
four recommendations for getting the most from your analysis. PARE. medical students and residents. Acad Med. 2011;86(8):996-1009.
2005;10(7):1-9. 75. Farquhar J, Kamei R, Vidyarthi A. Strategies for enhancing medical stu-
53. Loas G, Fremaux D, Marchand MP. Factorial structure and internal con- dent resilience: student and faculty member perspectives. Int J Med Educ.
sistency of the French version of the twenty-item Toronto Alexithymia Scale 2018;9:1-6.
in a group of 183 healthy probands. Encephale. 1995;21(2):117-22. 76. Dyrbye LN, Shanafelt TD, Werner L, Sood A, Satele D, Wolanskyj AP.
54. Lamothe M, Boujut E, Zenasni F, Sultan S. To be or not to be empathic: The impact of a required longitudinal stress management and resilience
the combined role of empathic concern and perspective taking in under- training course for first-year medical students. J Gen Intern Med. 2017;
standing burnout in general practice. BMC Fam Pract. 2014;15:15. 32(12): 1309-14.
55. Loas G, Fremaux D, Otmani O, Verrier A. Prevalence of alexithymia in a 77. Porcelli P, Guidi J, Sirri L, Grandi S, Grassi L, Ottolini F, et al. Alexithymia
general population. Study in 183 "normal" subjects and in 263 students. Ann in the medically ill. Analysis of 1190 patients in gastroenterology, cardiology,
Med Psychol (Paris). 1995;153(5):355-7. oncology and dermatology. Gen Hosp Psychiatry. 2013;35(5):521-7.
56. Franz M, Popp K, Schaefer R, Sitte W, Schneider C, Hardt J, et al. Alexi- 78. Shapiro J. Does medical education promote professional alexithymia? A
thymia in the German general population. Soc Psychiatry Psychiatr Epi- call for attending to the emotions of patients and self in medical training.
demiol. 2008;43(1):54-62. Acad Med. 2011;86(3):326-32.
57. Ueno M, Maeda M, Komaki G. Different subgroups of high-scorers on 79. Fisher RJ, Katz JE. Social-desirability bias and the validity of self-reported
the TAS-20 based on the big five personality traits. Personal Individ Differ. values. Psychol Market. 2000;17(2):105-20
128