Access to this full-text is provided by Wiley.
Content available from Anatomical Sciences Education
This content is subject to copyright. Terms and conditions apply.
254
|
Anat Sci Educ. 2025;18:254–263.wileyonlinelibrary.com/journal/ase
Received: 16 June 2024
|
Revised: 11 December 2024
|
Accepted: 20 December 2024
DOI: 10.1002/ase.2551
RESEARCH REPORT
The impact of COVID- 19 pre- university education on
first- grade medical students. A performance study of students
of a Department of Histology
José Manuel García1,2 | David Sánchez- Porras1,2 | Miguel Etayo- Escanilla1,2 |
Paula Ávila- Fernández1,2 | Olimpia Ortiz- Arrabal1,2 | Miguel- Ángel Martín- Piedra1,2 |
Fernando Campos1,2 | Óscar- Darío García- García1,2 | Jesús Chato- Astrain1,2 |
Miguel Alaminos1,2
1Tissue Engineering Group, Department of
Histology, Medical School, Universit y of
Granada, Granada, Spain
2Instituto de Investigación Biosanitaria ibs.
GRANADA, Granada, Spain
Correspondence
Óscar- Darío García- García and Jesús
Chato- Astrain, Tissue Engineering Group,
Department of Histology, Medic al School,
University of Granada, Avenida Doctor
Jesus Candel Fabregas, 11, Granada
E18016, Spain.
Email: ogarcia@ugr.es and jchato@ugr.es
Funding information
Tissue Engineering Group of the
University of Granada, Grant/Award
Number: CTS- 115
Abstract
The recent coronavirus disease (COVID- 19) forced pre- university professionals to
modify the educational system. This work aimed to determine the effects of pan-
demic situation on students' access to medical studies by comparing the performance
of medical students. We evaluated the performance of students enrolled in a subject
taught in the first semester of the medical curriculum in two pre- pandemic academic
years (PRE), two post- pandemic years (POST), and an intermediate year (INT) using
the results of a final multiple- choice exam. Consistency analysis among periods was
performed using the Cronbach alpha coefficient (α), the difficulty index with random
effects correction (DI), and the point- biserial correlation index (PB). The five exams
were homogeneous and had similar α, DI, and PB difficultness. Performance signifi-
cantly decreased in POST students compared with PRE students, with a correlation
between performance and the academic years (PRE- POST). A significant decrease in
the percentage of correct answers was detected in the academic years, with POST
students showing lower results than PRE students, but not in the percentage of ques-
tions answered incorrectly. Significantly higher percentages of unanswered questions
were found among POST students. These results confirm the negative impact of the
POST pre- university educational system on the performance of students accessing
medical school and suggest that POST students could have a higher degree of uncer-
tainty. Specific education programs should be implemented during the first years of
the medical curriculum to tailor this effect and increase students' self- confidence and
knowledge, which may be associated with confidence.
KEYWORDS
COVID- 19, medical education, performance
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2025 The Author(s). Anatomical Sciences Education published by Wiley Periodicals LLC on behalf of American Association for Anatomy.
José Man uel García and Da vid Sánchez- Porras cont ributed equal ly to this study.
|
255
GARCÍA et al .
INTRODUCTION
Medical education is a complex process that requires continuous
evaluation and adaptation to new situations, which may involve
rethinking attitudes and reorienting teaching objectives, with
the ultimate goals of improving the students' knowledge and
skills.1
In this context, the recent COVID- 19 pandemic emergency
that forced education authorities and professionals to adapt the
teaching systems to the unprecedented new situation was an
enormous challenge for students worldwide.2 In medical schools,
modifications were made at different levels. At the clinical level,
the pandemic transformed the clinical rotations of residents and
medical students and the training strategies applied to surgery,3
radiology,4 and mental health,5 among other medical specialties.
In most cases, direct contact with patients was replaced by a
web- based educational system in which simulation and e- learning
strategies were implemented.6–8 In undergraduate medical edu-
cation, most medical schools opted to implement several changes
to the medical curriculum or adopt different types of student- led
educational activities.9 In general, medical teachers were forced
to modify their teaching strategies and the way they conducted
classes, adopting online distance learning and virtual teach-
ing instead of traditional in- person teaching methods.7,1 0 These
changes might have substantially altered the teaching- learning
process in medical education, as previously demonstrated.11–1 3
However, the impact of the new situation on the development of
students enrolled in pre- university studies who subsequently ac-
cessed medical studies at medical schools has not been studied
in depth; hence, the possible influence of this educational change
on the performance of these students at medical schools requires
further research.
In Spain, pre- university education consists of three distinct lev-
els: 6 years of basic studies (primary education), followed by 4 years
of secondary education (ESO) and 2 years of high school (bachiller-
ato). The last level of pre- university education is optional and focuses
on preparing students for university studies with specific training
methods for students who wish to enroll in health science education
at the university level. As for medical school students, high school
students were substantially affected by the COVID- 19 pandemic,
and numerous adaptations were required at this educational level.14
Previous reports have demonstrated that educational disruptions
affecting high school are significantly associated with students' de-
creased academic performance.15 Whether this situation affected
the performance of students enrolled in medical schools has not yet
been adequately studied.
In the present study, we evaluated the academic performance
of students enrolled in the first academic year of the medical school
in Granada, Spain, corresponding to students who received pre-
pandemic and post- pandemic high school formations, to determine
whether the pandemic might have influenced the performance of
medical students enrolled in a subject taught in the first semester of
the medical curriculum.
METHODS
Sample and data source
In this work, we evaluated the scores obtained by students enrolled
in the subject “Cell Biology and Principles of Human Genetics and
Cytogenetics” which is taught in the first semester of the first aca-
demic year (AC) of the Degree in Medicine at the Medical School of
the University of Granada, Spain. These scores correspond to each
student's performance on the final global examination taken at the
end of the AC, which corresponds to a teaching year. The results
corresponding to the last five AC were evaluated for: 2019–2020
(AC19- 20), 2020–2021 (AC20- 21), 2021–2022 (AC21- 22), 2022–
2023 (AC22- 23), and 2023–2024 (AC23- 24). The teachers and
teaching methods used during these five ACs were the same, and
the subject contents did not vary among ACs, with the only differ-
ence being that pandemic teaching was performed using online re-
sources only for AC20- 21, whereas face- to- face learning was used
for the rest of the AC.
This exam included 60 multiple- choice questions (items) re-
lated to Cell Biology and Principles of Human Inheritance taught
during the AC. For each item, the students were asked to select
the correct option from four possibilities (A, B, C, or D), with only
one bei ng co rre c t. Sel ection of the cor rec t option wa s sco red with
1 point over a total of 60 points, incorrect options were penalized
at −0.33 points, whereas unanswered questions did not affe ct the
final score. The final scores were then adapted to a scale ranging
from 0 to 10 points by dividing the final result (over 60 points)
by 6.
Consistency analysis
To determine the homogeneity of the results obtained by the stu-
dents in the different exams, we performed several consistency
analyses. First, reliability was assessed for each test by determining
the alpha coef ficient of Cronbach (α). This coefficient was calculated
as previously reported16:
𝛼
=k
k−1
(
1−
Σn
j=1𝜎2
j
𝜎2
x
)
where k is the number of
items in the test,
𝜎2
x
is the global variance of all values, and
∑n
j
=1𝜎
2
j
is
the sum of variances of the options for each question. The α coeffi-
cient can range from 0 to 1, and the highest consistency corresponds
to the highest values of α. The difficulty index with random effects
correction (DI) was then calculated for each test question using the
following formula:
DI
=A−
E
k−1
N
, where A is the number of students
who gave a correct answer to each question, E is the number of stu-
dents who gave an incorrect answer to the question, k is the number
of possible answers to each question, and N is the total number of
students in the study. Questions with DI > 0.5, are considered easy,
whereas questions with DI < 0.5, are considered difficult.16,17 Finally,
the scores assigned by the students to each question were analyzed
to determine the point- biserial correlation index (PB) with the fol-
lowing formula18:
PB
=𝜇p−𝜇q
𝜎x√
ID
1−ID , where μp is the mean score of
the students who correctly answered to each item, μq is the mean
256
|
GARCÍA et al.
score of the students who incorrectly answered to the item, σx is the
global standard deviation of all test scores, and ID is the difficulty
index of the item determined as the number of students correctly
answering the item versus the total number of students.
Performance analysis
To determine the performance of students enrolled in each AC, the
average and standard deviation values of the scores obtained in the
final examination were calculated for AC19- 20, AC20- 21, AC21- 22,
AC22- 23, and AC23- 24. Averages and standard deviations were
also calculated for the students who received pre- COVID- 19 pre-
university education (PRE), corresponding to AC19- 20 and AC20-
21, students enrolled in post- COVID- 19 pre- university education
(POST), corresponding to AC22- 23 and AC23- 24, and the group of
students who received a mixed intermediate (INT) pre- university
education system (1 year with the pre- COVID- 19 system and 1 year
with the post- COVID- 19 system), corresponding to AC21- 22.
Statistical analysis
All distributions were fir st analyzed for normality using the Shapiro–
Wilk test. The results of this test demonstrated that the criteria for
the use of parametric comparison tests were not fulfilled; therefore,
non- parametric tests were used. The Kruskal–Wallis test was used
to compare the results of several distributions at the same time (e.g.,
the results of the five ACs). For post hoc pairwise comparisons of
two specific groups (e.g., the results of AC19- 20 vs. AC20- 21), we
used the Mann–Whitney test. The correlation between two vari-
ables (for example, the AC and students' scores) was carried out
using Kendall's tau correlation test, and the time- related trend of a
variable was determined by the R2 linear trend of the variable. We
compared the scores obtained by all students, and male and female
students independently, the percentage of items answered correctly
and incorrectly, and the percentage of unanswered items. In addi-
tion, a binary logistic regression analysis was performed to predict
the association between the dichotomized type of sample (PRE or
POST) and students' final scores, percentages of correct and incor-
rect answers, and percentages of unanswered questions. Statistical
comparisons were performed using Real Statistics Resource Pack
software (Release 7.2) (Dr. Charles Zaiontz, Purdue University, West
Lafayette, IN, USA). To correct for multiple testing, a Bonferroni-
adjusted p- value of <0.001 was considered statistically significant.
Although not significant, the values between 0.05 and 0.001 were
considered marginally significant.
RESULTS
Consistency analysis of the test results
A total of 1246 students were enrolled in these five AC, with an
average of 249 ± 16 students per AC (32% men and 68% women).
Analysis of the different exams using Cronbach's alpha coeffi-
cient (Table 1) showed a high- reliability index in all cases, with results
ranging from α = 0.8833 for AC20- 21 to α = 0.9549 for AC19- 20.
When the difficulty of the items contained in each test was
analyzed using the difficulty index with random effects correction
(Table 1), we found that the percentage of items showing low diffi-
culty (DI > 0.50) was above 50% in all cases. Differences between the
different exams were not statistically significant (Fisher's exact test
p > 0.05 for all comparisons), and the correlation analysis between DI
and the AC showed that the association between both variables was
not statistically significant (Kendall p = 0.2122, r = 0.0645).
In addition, point- biserial correlation index analysis (Table 1)
revealed very high values in all cases, with more than 90% of the
items showing very good results for this parameter. As in the pre-
vious case, non- significant differences were found when PB was
compared among the different exams (Fisher's exact test p > 0.05
for all comparisons), and the correlation analysis was not statistically
significant (Kendall p = 0.5689, r = 0.0295).
Performance of the students in each academic year
Analysis of the results obtained by the students enrolled in each study
group revealed an average score value ranging from 5.17 ± 1.72 in
the academic year AC22- 23 to 6.29 ± 1.71 in AC19- 20 (Figure 1 and
Table 2), with global significant differences among the five groups com-
pared (p < 0.0001 for the Kruskal–Wallis test). The differences were
statistically significant for the pairwise comparison of AC21- 22 versus
AC22- 23 and marginally significant for the comparison between AC19-
20 versus AC20- 21, and AC22- 23 versus AC23- 24. A significant corre-
lation was found between AC and the scores obtained by the students
TAB LE 1 Analysis of the consistency of the five exams analyzed in the present study. For each academic year (AC), Cronbach's alpha
coefficient (α), the difficulty index with random effects correction (DI), and the point- biserial correlation index (PB) are shown.
AC19- 20 AC20- 21 AC21- 22 AC22- 23 AC23 - 24
Cronbach's alpha coefficient (α)0.9549 0.8833 0.9542 0.9354 0.9539
Difficulty index with random effects correction (DI) 68.33% 56.67% 63.33% 55.00% 56.67%
Point- Biserial Correlation Index (PB) 96.67% 93.33% 91.67% 95.0 0% 10 0%
Note: For the DI and PB, the percentage of items showing a value >0.5 is shown. AC19- 20: Academic year 2019–2020, AC20- 21: Academic year
2020–2021, AC21- 22: Academic year 2021–2022, AC22- 23: Academic year 2022–2023, and AC23- 24: Academic year 2023–2024.
|
257
GARCÍA et al .
(Kendall's tau p < 0.0001, r = 0.1318, and linear trend R2 = 0.6241).
When results were analyzed only for male students, results ranged
between 5.40 ± 1.73 for AC23- 24 and 6.43 ± 1.33 for AC20- 21, with
global differences among the five ACs, although only the comparison
between AC21- 22 and AC22- 23 was marginally significant for the
pairwise comparisons. For female students, the lowest results (5.07 to
±1.68) were obtained in AC22- 23, and the highest scores (6.28 ± 1.61),
in AC19- 20. Global differences among all ACs were statistically signifi-
cant, and the pairwise comparisons of AC21- 22 versus AC22- 23, AC19-
20 versus AC20- 21, and AC22- 23 versus AC23- 24 were only marginally
significant. For both genders, a significant correlation between the AC
and student performance was found (p < 0.0001 and r = 0.1763 for men
and p < 0.0001 and r = 0.1299 for women).
When PRE students were overall compared with POST students,
we found that the average scores corresponding to the PRE group
were significantly higher than those of the POST group (p < 0.0001).
Interestingly, the scores of the PRE students were similar to those of
the INT students, with non- significant differences between the two
groups (Figure 1 and Table 2). The correlation between students'
scores and the type of study previously conducted (PRE or POST) was
statistically significant (Kendall tau p < 0.0001, r = 0.1658, and linear
trend R2 = 0.7821), and the binary logistic regression analysis was sta-
tistically significant (p < 0.0001). Similar results were obtained when
the results were analyzed for a specific gender, with significant differ-
ences between PRE and POST students, but not between PRE and INT
students, and with a significant correlation between the scores and
the type of study (Kendall tau p = 0.0002 and r = 0.1527 for men, and
Kendall tau p < 0.0001 and r = 0.1671 for women).
Analysis of correct, incorrect, and unanswered items
When the percentage of items answered correctly by the students
was analyzed (Figure 2 and Table 3), we found significant overall dif-
ferences among the five ACs compared (p < 0.0001 for the Kruskal–
Wallis test), with significant differences for the pairwise comparison
of AC21- 22 versus AC22- 23, AC19- 20 versus AC20- 21, and AC22- 23
versus AC23- 24 being marginally significant. Values obtained in the
FIGURE 1 Performance of the students enrolled in each academic year (AC). Results are shown as average values for students enrolled
in each AC (panel A) and the PRE, INT, and POST groups (panel B), for the whole group of students (ALL), and for male (M) and female (F)
students only. Error bars correspond to standard deviations. PRE: Students who received pre- COVID- 19 pre- university education; INT:
Students who received a mixed intermediate pre- university education system; POST: Students enrolled in post- COVID- 19 pre- university
education. The black dotted lines show the linear trend in panels A and B. The comparison groups showing significant p values of <0.001
are shown above each specific group highlighted in green color, whereas comparison groups showing p values between 0.05 and 0.001 are
shown in yellow.
TAB LE 2 Performance of the students enrolled in each academic
year (AC) expressed as final scores.
Scores
Final score
All Males Females
A C 1 9 - 2 0 6.29 ± 1.71 6.32 ± 1.93 6.28 ± 1. 61
AC20- 21 5.81 ± 1.71 6.43 ± 1.33 5.77 ± 1.65
AC21- 22 6.02 ± 1.65 6.28 ± 1.76 5.89 ± 1.59
A C 2 2 - 2 3 5.17 ± 1.72 5.43 ± 1.79 5.07 ± 1.68
A C 2 3 - 2 4 5.53 ± 1.8 5.4 ± 1.73 5.59 ± 1.84
PRE 6.05 ± 1.73 6.38 ± 1.64 6.03 ± 1.65
INT 6.02 ± 1.65 6.28 ± 1.76 5.89 ± 1.59
POST 5.35 ± 1.77 5.42 ± 1.75 5.31 ± 1.77
AC19- 20 vs. AC20- 21 0.0020M0.9038 0.0065M
AC20- 21 vs. AC21- 22 0.3097 0.7965 0.8213
AC21- 22 vs. AC22- 23 <0.0001* 0.0023M<0.0001*
AC22- 23 vs. AC23- 24 0.0198M0.7652 0.0040M
PRE vs. POST <0.0001* <0.0001* <0.0001*
PRE vs. INT 0.6815 0.7788 0.2531
POST vs. INT <0.0001* 0.0001* 0.0010M
Note: Results are shown as average ± standard deviations and
correspond to the final score results obtained by the students included
in each study group (from 0 to 10 points). Results are shown for the
whole group of students (ALL), for male and female students, and for
students enrolled in each AC and in the PRE, INT, and POST groups.
The last rows show the p values for the statistical comparisons carried
out in the present work. PRE: Students who received pre- COVID- 19
pre- university education; INT: Students who received a mixed
intermediate pre- university education system; POST: Students enrolled
in post- COVID- 19 pre- university education. Statistically significant p
values of <0.001 are highlighted with asterisks (*), whereas p values
between 0.05 and 0.001 are labeled with “M”.
258
|
GARCÍA et al.
PRE group were significantly higher than those in the POST group but
comparable to those in the INT group. A significant correlation was
found between the AC and the percentage of correct answers (Kendall
tau p < 0.0001 and r = 0.1523), and the binary logistic regression analy-
sis was statistically significant (p < 0.0001). Similar results were found
for only female students, with significant differences for the same pair-
wise comparisons and a significant correlation with the AC (Kendall tau
p < 0.0001 and r = 0.1485). However, the analysis carried out for male
students only revealed marginally significant pairwise differences for
the comparison of AC21- 22 versus AC22- 23, and the correlation with
the AC was not statistically significant (p = 0.6743).
For the percentage of items answered incorrectly (Figure 2 and
Table 3), overall differences were found among the five ACs compared
(p < 0.0001 for the Kruskal–Wallis test), with pairwise comparisons
revealing significant differences only for AC19- 20 versus AC20- 21
and AC20- 21 versus AC21- 22. However, the differences were not
significant when comparing the PRE, POST, and INT groups. The cor-
relation of the percentage of incorrect answers with the AC was not
statistically significant (p = 0.6257), and the same trend was found in
the binary logistic regression (p = 0.1070). A similar pattern was ob-
served for the group of female students, with significant overall differ-
ences among the five ACs. Pairwise comparisons were significant for
AC19- 20 versus AC20- 21 and AC20- 21 versus AC21- 22, with AC22-
23 versus AC23- 24 being marginally significant, although the correla-
tion with the AC was not statistically significant. For men, however,
we found that overall comparisons between the five ACs were non-
significant, as were pairwise comparisons. The percentage of incorrect
answers was only marginally higher in the POST group than in the PRE
group, and the correlation between this percentage and the AC was
also marginally significant (Kendall tau p = 0.0203 and r = 0.0879).
Regarding unanswered questions (Figure 2 and Table 3), our
analysis showed significant overall differences among all the ACs
(p < 0.0001 for the Kruskal–Wallis test) and pairwise differences for
AC20- 21 versus AC21- 22 and AC21- 22 versus AC22- 23, with AC19-
20 versus AC20- 21 being marginally significant. The comparisons of
the PRE versus POST and POST versus INT groups were statistically
FIGURE 2 Percentage of questions answered correctly (panels A and B), incorrectly (panels C and D) and unanswered (panels E and F)
by each student. Results are shown as average percentages for students enrolled in each academic year (panels A, C, and E) and the PRE,
INT and POST groups (panels B, D, and F), for the whole group of students (ALL), and for male (M) and female (F) students only. Error bars
correspond to standard deviations. PRE: Students who received pre- COVID- 19 pre- university education; INT: Students who received a
mixed intermediate pre- university education system; POST: Students enrolled in post- COVID- 19 pre- university education. The black dotted
lines correspond to linear trends.
|
259
GARCÍA et al .
TAB LE 3 Performance of the students enrolled in each academic year (AC) shown as percentage of correct, incorrect and unanswered questions.
Scores
Correct answers Incorrect answers Unanswered
All Males Females All Males Females All Males Females
A C 1 9 - 2 0 68.31 ± 14. 81 69.0 6 ± 16 67.98 ± 14.31 16.09 ± 10. 29 17.65 ± 12.55 15.41 ± 9.1 15.61 ± 11.79 13.29 ± 10.65 16.6 ± 12.14
AC20- 21 65.23 ± 14.28 66.02 ± 15.12 64.8 ± 13.84 21.25 ± 11.8 21.33 ± 12.13 21.21 ± 11.66 13.16 ± 10.67 12.65 ± 10.68 13.43 ± 10.69
AC21- 22 65.95 ± 14.48 68.73 ± 15 64.47 ± 14.02 17.1 2 ± 10.02 17.69 ± 10.73 16.81 ± 9. 6 4 16.93 ± 12.29 13.58 ± 11.33 18.72 ± 12.44
A C 2 2 - 2 3 57. 8 5 ± 14.67 60.43 ± 15.03 56.84 ± 14.4 4 18 .41 ± 11.48 18.36 ± 12.23 18.43 ± 11.21 23.75 ± 12.32 21.21 ± 11.86 24.73 ± 12.39
A C 2 3 - 2 4 60.88 ± 15.6 59.81 ± 14.9 61.42 ± 15.97 16.81 ± 10.64 17. 29 ± 10.81 16.57 ± 10.58 22.31 ± 12.38 22.9 ± 12.15 22.01 ± 12.53
PRE 66.71 ± 14.61 6 7. 37 ± 15.55 66.39 ± 14.15 18.77 ± 11.38 19.7 ± 12.42 18.32 ± 10.84 14.33 ± 11.28 12.93 ± 10.64 15.01 ± 11. 53
INT 65.95 ± 14.4 8 68.73 ± 15 64.47 ± 14.02 17.1 2 ± 10.02 17.6 9 ± 10.73 16.81 ± 9. 6 4 16.93 ± 12.29 13.58 ± 11.33 18.72 ± 12.44
POST 59. 33 ± 15.19 60.1 ± 14.91 59 ± 15.33 17. 6 3 ± 11.09 17.7 9 ± 11.46 1 7. 55 ± 10.94 23.04 ± 12.36 22.11 ± 12 23.45 ± 12. 51
AC19- 20 vs. AC20- 21 0.0140M0.9651 0.0394M<0.0001* 0.2086 <0.0001* 0.0198M0.3838 0.0124M
AC20- 21 vs. AC21- 22 0.7807 0.7577 0.5638 <0.0001* 0.4388 0.0001* 0.0004* 0.3758 <0.0001*
AC21- 22 vs. AC22- 23 <0.0001* 0.0009* <0.0001* 0.1832 0.8202 0.134 0 <0.0001* 0.0001* <0.0001*
AC22- 23 vs. AC23- 24 0.0204M0.7739 0.0044M0.0689 0.7396 0.0409M0.14 02 0.5341 0.0367M
PRE vs. POST <0.0001* 0.0017M<0.0001* 0.2257 0.0011M0.6123 <0.0001* <0.0001* <0.0001*
PRE vs. INT 0.36 41 <0.0001* 0.0910 0.0792 0.0024M0.14 05 0.0068M0.0351M0.0011M
POST vs. INT <0.0001* <0.0001* 0.0006* 0.6305 0.9776 0.5253 <0.0001* <0.0001* 0.0001*
Note: Results are shown as average ± standard deviations and correspond to the percentage of items answered correctly in each study group (Correct answers), the percentage of items answered
incorrectly in each group of study (Incorrect answers), and the percentage of unanswered items in each group of study (Unanswered), from 0 to 10 points. Results are shown for the whole group of students
(ALL), for male and female students, and for students enrolled in each AC and the PRE, INT, and POST groups. The last rows show the p values for the statistical comparisons carried out in the present
work. PRE: Students who received pre- COVID- 19 pre- university education; INT: Students who received a mixed intermediate pre- university education system; POST: Students enrolled in post- COVID- 19
pre- university education. Statistically significant p values of <0.001 are highlighted with asterisks (*), whereas p values between 0.05 and 0.0 01 are labeled with “M”.
260
|
GARCÍA et al.
significant, whereas that of PRE versus INT was marginally significant.
The lowest percentages corresponded to the PRE group and the high-
est to the POST group. A significant correlation was found between
the percentage of unanswered questions and the AC (Kendall tau
p < 0.0001 and r = 0.2086), with a significant p- value in the binary lo-
gistic regression analysis (p < 0 0.0001). A similar trend was observed
in the female group, although the comparison between AC22- 23
and AC23- 24 was marginally significant. For all students, a signifi-
cant correlation with the AC was found (Kendall tau p < 0.0001 and
r = 0.1923). For male students, significant overall differences among all
the ACs were found (p < 0.0001 for the Kruskal–Wallis test), but pair-
wise comparisons were only significant for AC21- 22 versus AC22- 23,
and not for the rest of the comparisons. For women and all students,
differences were found between the PRE and POST and the POST
and INT groups.
DISCUSSION
In the present work, we conducted a performance analysis of the re-
sults obtained by a group of students of the medical school who re-
ceived pre- COVID- 19 education before enrolling in medical school,
and we compared these results with those of students who received
post- COVID- 19 high school education, revealing several crucial dif-
ferences between these groups. An important issue is whether the
type of examination and the difficulty of the exams used to evaluate
the students' performance during five ACs were similar and whether
teaching was conducted by the same teachers. To answer this ques-
tion, we used a multifactorial consistency analysis approach including
three variables (α, DI, and PB) to increment the potency of the analy-
sis.16 The results demonstrated that all the exams had high reliability
and that their difficulty was similar. In this regard, it is important to
note that the evaluation test applied to the students included in the
present study consisted of a multiple- choice question exam whose
characteristics, types of questions, and difficulty have remained un-
modified for the last five ACs. Therefore, it was expected that the α,
DI, and PB values could offer similar results for the five exams included
in the present study, and the differences among ACs should not be at-
tributed to the characteristics of the evaluation exams.
When students' performance was evaluated, we found that the
scores of students enrolled in a subject taught in the first semester of
the medical curriculum differed significantly during the last five ACs.
Specifically, the results significantly decreased for students who re-
ceived POST instruction before enrolling in medical school, and this
trend was consistent for both male and female students. These results
are in line with previous reports suggesting that POST teaching adap-
tations might have reduced medical students' knowledge and prepara-
tion for becoming future medical doctors.19–21 However, a recent study
showed that the changes implemented during the pandemic did not
affect specific skills, such as the ability of dental students to interpret
craniofacial radiologic images.22 Interestingly, a recent randomized trial
was designed to determine the outcomes of postgraduate health and
medical students subjected to online and on- site teaching.23 Although
the results are not yet available, this trial is expected to shed light on the
effects of non- presential teaching on medical students.
On the other hand, we aimed to determine specific factors that
could be associated with the different performances of PRE and POST
students. For this purpose, we analyzed the percentage of items that
the students answered correctly, incorrectly, or left unanswered and
found significant differences among the groups. Strikingly, our results
revealed a significant trend toward a lower percentage of questions
correctly answered over time, with significant differences between
PRE and POST students, and a correlation with the AC, with few dif-
ferences between genders. In contrast, the percentage of items that
were incorrectly answered remained constant and varied very little
among ACs, suggesting that PRE and POST students answered incor-
rectly in approximately the same proportion in both periods. Finally,
the analysis of unanswered questions showed a significant correlation
with the AC and a significant increase in the percentage of blank an-
swers among POST students, with few differences between men and
women. These results suggest that the performance differences be-
tween PRE and POST students could be related to the higher tendency
of PRE students to answer doubted items, whereas POST students
tended to leave more questions unanswered; however, the percentage
of failure was approximately the same across all student groups.
An important factor influencing the teaching- learning process is the
method used during learning and the teaching tools and resources im-
plemented by teachers.24 The relevance of the teaching modifications
adopted during the pandemic, including the adoption of e- learning and
digital resources in medical education,25 remains to be determined,
but these changes likely influenced the results of the present study.
Although we mainly focused on the analysis of student performance,
we cannot exclude the possibility that the differences found between
the PRE and POST groups could be related to different teaching tools
used in teaching Cell Biology and Principles of Human Inheritance
in these groups because of pandemic restrictions. However, as non-
presential teaching with classes given through the website was only
used for one of the PRE years (AC20- 21), we might hypothesize that
the effect of virtual teaching on students' performance could not be
critical. Future studies should determine the effects of teaching tools
on the capability of medical students to acquire skills and knowledge
related to this subject in the medical curriculum.
Interestingly, our study showed that students who received a
mixed model of pre- university teaching (INT group) obtained simi-
lar results to those of the PRE group, suggesting that pre- university
teaching changes affected students' performance only when the en-
tire high school period was affected (POST group). As the PRE group
received presential onsite teaching at the medical school, 1 year of
presential teaching, and 1 year of virtual teaching at the pre- university
level, it is probable that the effects of non- presential learning were not
strong enough to alter students' performance unless the entire period
of pre- university learning was affected by this type of teaching.
Multiple factors can influence the likelihood of students leav-
ing their questions unanswered on examination tests. In research
questionnaires, participants with lower previous formation and lit-
eracy failed to answer questionnaire items with a higher probability
|
261
GARCÍA et al .
than participants with higher formation.26 In medical students, the
lack of certainty is the main factor promoting avoidance of decision-
making, and students with lower levels of confidence and higher
aversion to ambiguity tend to leave more questions unanswered and
obtain lower final scores.27 This phenomenon is relevant because
confidence and confrontation with uncertainty are important re-
quirements of medical practice, where doctors must commonly face
the uncertainty of diagnosis, therapy, and patient outcomes.28 The
fact that POST students may have lower confidence in their own
knowledge and capabilities is worrisome and requires specific train-
ing programs to reinforce their own potential. It has been suggested
that this specific aspect of medical education is essential to the
medical curriculum.29 The reasons POST students may have lower
confidence and a higher degree of uncertainty remain unknown and
should be determined in future studies. However, we might hypoth-
esize that POST high school education, in which students had to
confront the uncertainty of the school system, examination meth-
ods, evolution of the disease, and many other sources of anxiety and
ambiguity, could have critically impacted the way students faced
their current duties at the medical school. This is in agreement with
several reports suggesting that post- COVID- 19, university students
have high degrees of uncertainty and that interventions targeted
at controlling intolerance to uncertainty should be implemented.30
In addition, male and female students may differ in their tendency
to answer uncertain questions, and women may have a higher ten-
dency to leave uncertain questions unanswered.31 Although the re-
sults of our study suggest that gender differences are not significant,
future studies should specifically assess this issue.
The present study has several limitations. First, this was a ret-
rospective study in which students were analyzed at different time
points (AC). Despite the same programs and teachers, it is clear that
differences may exist over different academic years. Although our
analyses revealed that the exams were similar across years, spe-
cific differences might have influenced the students' performance.
Moreover, the different teaching tools and strategies used in each
AC could also have influenced the results, especially for students
subjected to non- presential teaching for the entire academic year.
In summary, the present work demonstrates that POST students
tend to show lower performance in the exams conducted at the be-
ginning of their medical school curriculum and that this reduction
is mainly related to a lower percentage of answered questions that
could be associated with higher degrees of uncertainty. Although
multiple factors could affect the results and percentage of students'
answers, these findings shed light on the potential effects of the
pandemic on students accessing medical schools. Future formative
programs may be implemented for POST medical students to rein-
force their skills and capabilities.
AUTHOR CONTRIBUTIONS
José Manuel García: Conceptualization; investigation; supervision;
writing – original draft. David Sanchez- Porras: Conceptualization;
investigation; supervision; writing - original draft. Miguel Etayo-
Escanilla: Formal analysis. Paula Ávila- Fernández: Formal
analysis. Olimpia Ortiz- Arrabal: Formal analysis. Miguel- Ángel
Martín- Piedra: Formal analysis. Fernando Campos: Formal analy-
sis. Óscar- Darío García- García: Data curation; formal analysis; in-
vestigation; writing – review and editing. Jesús Chato- Astrain: Data
curation; formal analysis; investigation; writing – review and edit-
ing. Miguel Alaminos: Conceptualization; investigation; supervision;
writing – original draft.
FUNDING INFORMATION
This study was supported by Tissue Engineering Group of the
University of Granada (CTS- 115).
CONFLICT OF INTEREST STATEMENT
The authors declare that they have no competing interests.
DATA AVAIL ABILIT Y STAT EME NT
Quantitative data can be accessed at the open- access European
Research Data Repository Zenodo (h t t p s : / / z e n o d o . o r g / r e c o r d s /
10687862).
ETHICS APPROVAL AND CONSENT TO PARTICIPATE
This study was approved by the Ethics and Research Committee of
the University of Granada (ref. 4030/CEIH/2024).
ORCID
Miguel- Ángel Martín- Piedra https://orcid.
org/0000-0002-4639-3175
Miguel Alaminos https://orcid.org/0000-0003-4876-2672
REFERENCES
1. Guilbert JJ. Educational handbook for health personnel. Geneva,
Switzerland: World Health Organization; 1987. 360 p.
2. Guldner G, Wells J, Ayutyanont N, Iyengar R, Sprenger S, Siegel
JT, et al. COVID- 19 related disruptions to medical education and
perceived clinical capability of new resident physicians: a nation-
wide study of over 1200 first- year residents. Med Educ Online.
2023;28:2143307.
3. Gerk A, Naus A, Carroll M, Ponte L, Salgado LS, Marrazzo E, et al.
Perceived impact of coronavirus disease 2019 on surgical training:
a Brazilian survey. J Surg Res. 2024;295:619–30.
4. Alhasan AS, Alahmadi SM, Altayeb YA, Daqqaq TS. Impact of
COVID- 19 pandemic on training and well- being in radiology resi-
dency: a national survey of diagnostic radiology trainees in Saudi
Arabia. Acad Radiol. 2021;28:1002–9.
5. Alshdaifat E, Sindiani A, Khasawneh W, Abu- Azzam O, Qarqash A ,
Abushukair H, et al. The impact of COVID- 19 pandemic on training
and mental health of residents: a cross- sectional study. BMC Med
Educ. 2021;21:208.
6. Calcagnile T, Sighinolfi MC, Sarchi L, Assumma S, Filippi B,
Bonfante G, et al. COVID- 19 and slowdown of residents' activity:
feedback from a novel e- learning event and overview of the litera-
ture. Urologia. 2021;88:332–6.
7. Patel OU, Bonner A, Sisk M, Grant MB. Dynamic changes in
medical education amidst the COVID- 19 pandemic: adapting
to virtual ophthalmology residency interviews. Ann Med Surg.
2022;2012(76):103417.
8. Tao Z, Zhu E, Sun X, Sun J. Comparative effects of three differ-
ent physical education teaching modes on college students'
262
|
GARCÍA et al.
physical fitness during the COVID- 19 pandemic: a longitudinal
study. Heliyon. 2024;10:e28305.
9. Lee IR, Kim HW, Lee Y, Koyanagi A, Jacob L, An S, et al. Changes in
undergraduate medical education due to COVID- 19: a systematic
review. Eur Rev Med Pharmacol Sci. 2021;25:4426–34.
10. Ayivi- Vinz G, Tremblay M, Gadio S, Dofara SG, Daniel SJ, Talbot D,
et al. Impact of educational activity formats, online or in- person, on
the intention of medical specialists to adopt a clinical behaviour: a
comparative study. J CME. 2024;13:2363550.
11. Abati E, Nelva Stellio L, Manini A, Moroni F, Azzalini L, Vilca LM. A
cross- sectional sur vey study of the impact of COVID- 19 pandemic
on the training and quality of life of Italian medical residents in the
Lombardy region. Ann Med. 2022;54:2326–39.
12. Bergeron L, Décary S, Djade CD, Daniel SJ, Tremblay M, Rivest L- P,
et al. Factors associated with specialists' intention to adopt new be-
haviors after taking web- based continuing professional development
courses: cross- sectional study. JMIR Med Educ. 2022;8:e34299.
13. Hilliard B, Reilkoff RA, Baum KD. Rapid implementation of a medi-
cal student rotation in health systems operations and remote patient
care in response to COVID- 19. Med Educ Online. 2022;27:2067024.
14. Mody M, Hodges A , Gibson KE, Mantey J, Gontjes KJ. Does high
school health education class satisfaction influence student per-
ceptions of public health and pandemic response? Am J Infect
Control. 2024;52(6):719–25.
15. Kuhfeld M, Soland J, Lewis K. Test score patterns across three
COVID- 19- impacted school years. Educ Res. 2022;51:500–6.
16. Baladrón J, Sánchez Lasheras F, Romeo Ladrero JM, Villacampa
T, Curbelo J, Jiménez Fonseca P, et al. The MIR 2018 exam: psy-
chometric study and comparison with the previous nine years.
Medicina (Kaunas). 2019;55:751.
17. Baladrón J, Sánchez- Lasheras F, Romeo- Ladrero JM, Curbelo J,
Villacampa- Menéndez P, Jiménez- Fonseca P, et al. Evolución de los
parámetros dificultad y discriminación en el ejercicio de examen
MIR. Análisis de las convocatorias de 20 09 a 2017. FEM Rev Fund
Educ Médica. 2018;21:181–93.
18. Lord FM. Applications of item response theory to practical testing
problems. New York: Routledge; 1980 288 p.
19. Hamamoto Filho PT, Cecilio- Fernandes D, Norcia LF, Sandars J,
Anderson MB, Bicudo AM. Reduction in final year medical stu-
dents' knowledge during the COVID- 19 pandemic: insights from an
interinstitutional progress test. Front Educ. 2022;7:1033732.
20. Podubinski T, Jessup B, Kirschbaum M, Bailie J, Heaney S, Sheepway L,
et al. Perspectives of work readiness among Australian health students
trained during the COVID- 19 pandemic. BMC Med Educ. 2024;24:1053.
21. Rajesh E, Subramaniam S, Pasupathy P, Suresh T, Gopichandran V.
Lived experiences of medical students of online learning: lessons
for adopting virtual learning in medical education. BMC Med Educ.
2024;24:982.
22. Majewski J, Qarashouli D, Kazibudzka Z, Ciach K, Kurzawa P,
Orłowska K, et al. Ability assessment of the craniofacial structures
radiological anatomy on panoramic radiography among dental
student during pandemic of Covid- 19. Abilit y assessment on OPG
among dental student during pandemic of Covid- 19. Folia Med
Cracov. 2024;64:29–40.
23. Raffing R, Konge L, Tønnesen H. Learning effect of online versus
onsite education in health and medical scholarship—protocol for a
cluster randomized trial. BMC Med Educ. 2024;24:927.
24. Moreno- Marin AC, Pardo Rios M, Lopezosa- Reca E, Molina García
C, Díaz- Miguel S, Gómez- Martín B, et al. Resources for innovative
learning of anatomy and foot ossification: graphic design and vir-
tual reality. J Foot Ankle Res. 2024;17:e70008.
25. Lewis KO, Popov V, Fatima SS. From static web to metaverse: re-
inventing medical education in the post- pandemic era. Ann Med.
2024;56:2305694.
26. Firmino RT, Fraiz FC, Montes GR, Paiva SM, Granville- Garcia AF,
Ferreira FM. Impact of oral health literacy on self- reported missing
data in epidemiological research. Community Dent Oral Epidemiol.
2018;46:624–30.
27. Han PK J, Klein WMP, Arora NK. Varieties of uncertainty in health
care: a conceptual taxonomy. Med Decis Mak. 2011;31:828–38.
28. Leite- Mendes F, Delgado L, Ferreira A, Severo M. The effect of
the attitude towards risk/ambiguity on examination grades: cross-
sectional study in a Portuguese medical school. Adv Health Sci
Educ Theory Pract. 2024;29(4):1309–21. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 /
s 1 0 4 5 9 - 0 2 3 - 1 0 3 0 5 - z
29. Geller G, Faden RR, Levine DM. Tolerance for ambiguity among
medica l stude nts: im pl ication s fo r their sel ect io n, tr ai ning and pr ac-
tice. Soc Sci Med. 1990;1982(31):619–24.
30. Charbonnier E, Montalescot L, Puechlong C , Goncalves A, Le
Vigouroux S. Relationship between fear of COVID- 19, intolerance
of uncertainty, and coping strategies on university students' mental
health. Nutrients. 2023;15:4938.
31. Iriberri N, Rey- Biel P. Competitive pressure widens the gender gap
in performance: evidence from a two- stage competition in mathe-
matics. Econ J. 2019;129:1863–93.
AUTHOR BIOGRAPHIES
José Manual García, MD, PhD, is a full professor at the
Department of Histology in the Faculty of Medicine at University
of Granada. He teaches in human cytology and histology to un-
dergraduate medical students.
David Sánchez-Porras, BSc, MSc, PhD, is a postdoctoral fellow
at the Department of Histology in the Faculty of Medicine at
University of Granada. His research includes tissue engineering,
biomaterials and cell culture.
Miguel Etayo-Escanilla, BSc, MSc, is a research fellow at the
Histology Department in the Faculty of Medicine at University
of Granada currently pursuing a PhD in Biomedicine.
Paula Ávila-Fernández, BSc, MSc, is a research fellow at the
Histology Department in the Faculty of Medicine at University
of Granada currently pursuing a PhD in Biomedicine.
Olimpia Ortiz-Arrabal, BSc, MSc, PhD, is a researcher with ex-
pertise in cell culture, hydrogels and growth factors.
Miguel-Ángel Martín-Piedra, BDS, MSc, PhD, is an Associate
professor at the Department of Histology in the Faculty of
Medicine at University of Granada. He teaches human histology
to undergraduate dentistry students.
Fernando Campos, BSc, MSc, PhD, is an Associate professor at the
Department of Histology in the Faculty of Medicine at University
of Granada. He teaches human histology to undergraduate health
science students. His research involves the development of new
technologies for tissue engineering applications.
|
263
GARCÍA et al .
Óscar-Darío García-García, BSc, MSc, PhD, is an Assistant
Professor at the Department of Histology in the Faculty of
Medicine at University of Granada. He teaches human histology
and cytology to undergraduate medical, pharmacy, and physio-
therapy students.
Jesús Chato-Astrain, BSc, MSc, PhD is an Associate Professor
at the Department of Histology. He teaches human histology,
cytology and genetics to undergraduate medical, pharmacy, and
physiotherapy students. His research involves tissue engineering
techniques and methods and novel teaching strategies for health
science students.
Miguel Alaminos, MD, PhD, BSc, MSc, PhD, is a full professor
at the Department of Histology in the Faculty of Medicine at
University of Granada. He teaches human histology and genet-
ics to undergraduate medical students. His interest includes
the development of tissue engineering substitutes for clinical
applications.
How to cite this article: García JM, Sánchez- Porras D,
Etayo- Escanilla M, Ávila- Fernández P, Ortiz- Arrabal O,
Martín- Piedra M-Á, et al. The impact of COVID- 19 pre-
university education on first- grade medical students. A
performance study of students of a Department of Histology.
Anat Sci Educ. 2025;18:254–263. https://doi.org/10.1002/
ase.2551
