The purpose of the present study was 1) to identify factors that may influence academic stress in medical students and 2) to investigate the causal relationships among these variables with path analysis.
One hundred sixty medical students participated in the present study. Psychological parameters were assessed with the Medical Stress Scale, Minnesota Multiphasic Personality Inventory, Hamilton Depression Scale, Beck Depression Inventory, and Academic Motivation Scale. Linear regression and path analysis were used to examine the relationships among variables.
Significant correlations were noted between several factors and Medical Stress scores. Specifically, Hamilton Depression Scale scores (β=0.26, p=0.03) and amotivation (β=0.20, p=0.01) and extrinsically identified regulation (β=0.27, p<0.01) response categories on the Academic Motivation Scale had independent and significant influences on Medical Stress Scale scores. A path analysis model indicated that stress, motivation, and academic performance formed a triangular feedback loop. Moreover, depression was associated with both stress and motivation, and personality was associated with motivation.
The triangular feedback-loop structure in the present study indicated that actions that promote motivation benefit from interventions against stress and depression. Moreover, stress management increases motivation in students. Therefore, strategies designed to reduce academic pressures in medical students should consider these factors. Additional studies should focus on the relationship between motivation and depression.
Medical education is associated with high stress.
Various factors contribute to academic stress in medical students. Previous studies have suggested that stress-related factors may be interlinked, forming an intricate psychological structure that may precipitate and perpetuate academic stress. For instance, academic stress is associated with academic performance,
Previous studies have focused on academic motivation much less than on other factors. One of the most widely acknowledged theories on this topic is the self-determination theory.
The present study examined the relationships between various factors associated with academic stress in medical students, including motivation, personality, academic performance, and depression. The causal relationships among these factors were investigated via path analysis. Our findings may form the basis for management of student mental health in medical schools.
The present study was based on the results of a survey that focused on evaluating and promoting mental health in medical students. The survey was conducted in November 2009 at the University of Ulsan College of Medicine in Seoul, Korea. Korean students may apply to medical school, which consists of 4 years of training, after receiving an undergraduate degree. They may also apply immediately after finishing high school. In this case, students take a 2-year pre-medicine curriculum prior to 4 years of medical school. The present study was conducted at one of the latter institutions. The survey was organized by school administrators and was taken by 160 students who were entering their third year of education (i.e., first year of medical school). The curriculum of the students included basic science courses (i.e., anatomy and histology) during the second year of pre-medicine coursework. During the first 2 years of medical school, the students are enrolled in extended basic science/clinical coursework (e.g., cardiology and reproductive medicine), and the last 2 years consist of clinical ward training. The aims of the survey were explained in detail to all of the students, and questionnaires were handed out. The students were given ample time to complete the survey. During the 4-week period after the surveys were completed, the students visited the psychiatry outpatient clinic and received individual assessments conducted by a psychiatrist. The psychiatric evaluation included the Structured Clinical Interview for DSM-IV (SCID-IV) and the Mini-International Neuropsychiatric Interview (MINI). The interviews were conducted according to the principles of SCID-IV and MINI, although due to time constraints, the tests were not completed in full. Students who were determined to be at high risk for mental health problems were asked to participate in a second interview. All of the students participated in at least one interview. None of the students in the present study was diagnosed with a significant mental health problem. Informed consent was provided by all of the students, and all of them were included in the final analyses. The Institutional Review Board at the Asan Medical Center approved the present study.
The following questionnaires were used in the present study: the Medical Stress Scale (MSS),
The MSS is an 11-item questionnaire designed to assess the amount of stress medical students experience during medical school.
The AMS is a 28-item questionnaire developed to describe academic motivation
The MMPI-2 is an instrument used to identify personality structures and psychopathology.
The HAM-D-17
Academic performance was assessed with school grades, which were provided by the university with consent. Grade-point averages (GPAs; A+=4.5; A0=4.0; B+=3.5; B0=3.0; C+=2.5; C0=2.0; D+=1.5; D0=1.0, F=0) during the previous semester were used in the analysis.
Chi-square analysis and Student's t-test were used to compare the demographic and clinical characteristics of stressed and non-stressed participants based on the dichotomous categorization described above. All analyses were two-tailed. Pearson's correlation was used to identify the association between various factors and stress. A linear regression model was constructed to determine the independent contribution of each factor to academic stress. Path analysis was used to estimate the causal relationship among stress, motivation, academic performance, depression, and personality. The Statistical Package for Social Sciences (SPSS) version 15.0 was used for all statistical analyses, and AMOS version 16.0 was used for the path analysis.
Stressed students were significantly younger compared with non-stressed students (21.5±1.6 vs. 22.1±1.7 years, p=0.03). Moreover, stressed students had lower GPAs (3.1±0.7 vs. 3.4±0.6, p<0.01) and higher scores on the depression scales compared with non-stressed students (HAM-D-17: 8.6±6.0 vs. 4.0±3.9, p<0.01; BDI-II: 7.2±6.1 vs. 3.3±3.7, p<0.01), although the mean scores were within the subclinical or normal range. Among the subcategories of the AMS, stressed students showed higher scores in amotivation (8.4±2.6 vs. 6.6±2.6, p<0.01), intrinsic motivation to accomplish things (13.3±2.2 vs. 14.2±2.5, p=0.02), and extrinsic identified regulation (15.4±2.8 vs. 14.0±2.8, p<0.01) (
Separate analysis via analysis of variance (ANOVA) and post-hoc analysis using the Tukey and Bonferroni methods revealed that students in their first year of medical school displayed significantly higher levels of stress compared with students in other years (27.0 vs. 25.5 for second-year pre-medical students, 24.1 for second-year medical students, and 22.2 for third-year medical students; p<0.01 for both post-hoc tests).
The MSS scores were positively correlated with high MMPI-2 scores (r=0.27, p<0.01) and depression scale scores (r=0.45, p<0.01 for both) and negatively correlated with GPA (r=-0.29, p<0.01). Five of the AMS subcategories were significantly correlated with MSS scores, including amotivation (r=0.39, p<0.01), intrinsic motivation to know (r=-0.20, p=0.01), intrinsic motivation to accomplish things (r=-0.20, p=0.01), extrinsic external motivation (r=-0.16, p=0.04), and extrinsic identified regulation (r= 0.18, p=0.02). Amotivation and extrinsic identified regulation were positively correlated, whereas the other three subcategories were negatively correlated with MSS scores. The depression scale scores showed the strongest correlation (
The AMS and MSS results represented corresponding variables in our model. The number of high MMPI-2 scales was used to represent personality. All subcategories of the AMS, including amotivation, were entered as indicators of the unobserved variable motivation and were arranged as covariates of each other and of the observed variable personality. The HAM-D-17 and BDI-II results were arranged as indicators of the unobserved variable depression and were arranged as covariates of each other, the observed variable of personality, and the AMS subcategory of amotivation.
Previous studies have used the chi-square value, Comparative Fit Index (CFI), Normed Fit Index (NFI), and Root Mean Square Error of Approximation (RMSEA).
Medical students who were under stress displayed several differences compared with students who were not stressed, and several of the factors were significantly correlated with stress levels. The results of the depression scales as well as the scores for amotivation and extrinsic identified regulation were independently associated with MSS results. The path analysis model revealed that motivation, grades, and stress may be sequentially associated with one another, forming a feedback loop. These factors may also be associated with depression and personality.
One of the implications of the path analysis model is that stress may be associated with motivation, and motivation may have an indirect association with stress through academic performance. Amotivation is defined as the absence of intent or drive.
Several inferences may be drawn from our path analysis model. First, academic stress, motivation, and grades form a feedback loop. Therefore, if one of the three factors deteriorates, the other two factors may be stabilized as a protective psychological mechanism. For example, a decrease in grades would cause an increase in stress, which in turn may cause motivation to increase. This, in turn, would cause grades to increase, thereby correcting the initial decrease in grades. In real-life situations, the responses to each step would certainly be influenced and modified by other factors, and this protective mechanism may not function easily. From a practical viewpoint, however, our results imply that actions that promote motivation may greatly benefit from interventions against stress or depression. Moreover, appropriate stress management may increase motivation. Interestingly, the proposed psychological mechanism strikingly resembles hormonal negative feedback loops. Moreover, contradictory associations were noted with depression. Depression showed comparatively weak positive associations with academic stress, which in turn was positively associated with motivation, making the net effect positive. Depression also showed a stronger negative association with motivation. Several unsuccessful path models were constructed, so our results are inconclusive at this time.
Previous studies have not investigated the associations among the three factors (i.e., academic performance, academic stress, and academic motivation) in medical students. Two reports, however, focused on the relationship between two of the three factors. Specifically, one study examined the influence of perceived stress factors on academic performance in dental students and concluded that the association was non-significant.
Several previous studies have revealed that stressed students were more likely to be in their first year of medical school compared with non-stressed students. This is, however, controversial.
The present study has several limitations. First, some of the assessment tools that were used, especially the MMPI-2 clinical scales and the depression scales, are designed for pathological assessments. Most of results from these tests were within the normal range and may not be clinically meaningful. The present study was also unable to identify the aspects of personality that have the strongest association with other factors. The tools that were used to measure stress and motivation were designed for use in a normal population, and thus the overall influence was probably minor. Next, many of the assessments were based on self-reports. Nevertheless, the self-report forms used in this study have all been validated, and therefore, this probably had very little effect. Moreover, unauthorized absences, dropout rates, and clinically significant levels of depression were not included in the final analyses because the number of students falling into these categories was too small (n≤10) and the results were not significant. Finally, the present study was a cross-sectional survey and not a longitudinal follow-up study. Thus, it is not possible to infer causal associations based on the results of our path analysis. Future longitudinal studies that focus on stress, motivation, and academic performance in medical students are warranted to confirm our results.
Taken together, the results of the present study indicate that academic stress in medical students may be influenced by an interaction among motivation, school grades, depression, and personality. Therefore, strategies to reduce academic pressures in medical students should take into consideration the importance of these factors. Increasing motivation may occur through interventions against stress and depression. Moreover, appropriate stress management may help students to become more motivated. Screening students once a year with self-report measures that evaluate stress, depression, and motivation may be helpful. This would identify students in need of individual counseling for stress management. Furthermore, as the number of students suffering from significant levels of stress is expected to be high, universal interventions may also be beneficial. Future studies are warranted regarding the interaction between motivation and depression.
The structure of motivation. Based on Deci and Ryan's self-determination theory.
A path-analysis model of the relationship among academic stress, academic motivation, personality, and grades in medical students. χ2=20.28, df=20, p=0.44, Comparative Fit Index (CFI)=1.00, Normed Fit Index (NFI)=0.97, Root Mean Square Error of Approximation (RMSEA) <0.01 Rectangles and ovals represent observed and unobserved variables, respectively. Numbers printed next to single-headed arrows correspond to standardized regression weights. *p<0.01. The Academic Motivation Scale, the Medical Stress Scale, and the Minnesota Multiphasic Personality Inventory (MMPI-2) were used to evaluate the corresponding variables in this model. Personality was defined as the number of MMPI clinical scales with a T-score above 65. The following details were not included in this diagram: 1) all subscales of the Academic Motivation Scale were included as indicators of the unobserved variable motivation and were arranged as covariates of each other and of and personality and 2) the Hamilton Depression Scale (HAM-D-17) and Beck Depression Inventory (BDI-II) results were arranged as indicators of the unobserved variable depression and were arranged as covariates of each other, of the observed variable personality, and of the Academic Motivation subscale amotivation. The data used in the present study were cross-sectional, and the associations implied by our path analysis model are hypothetical.
Clinical characteristics of stressed and non-stressed
'Stressed' and 'Non-stressed' participants had Medical Stress Scale scores ≥28 and <28, respectively. 'School Year' was re-coded as follows; the second year of pre-medicine as 2, the first year of medicine as 3, the second year of medicine as 4, and the third year of medicine as 5. For 'Grades', grade-point averages (A+=4.5; A0=4.0; B+=3.5; B0=3.0; C+=2.5; C0=2.0; D+=1.5; D0=1.0, F=0) of the previous semester were used. MMPI-2: Minnesota Multiphasic Personality Inventory, HAM-D-17: Hamilton Depression Scale, BDI-II: Beck's Depression Inventory
Correlation of demographic and psychological characteristics with stress
For 'Grades', grade-point averages (A+=4.5; A0=4.0; B+=3.5; B0=3.0; C+=2.5; C0=2.0; D+=1.5; D0=1.0, F=0) of the previous semester were used. MMPI-2: Minnesota Multiphasic Personality Inventory, HAM-D-17: Hamilton Depression Scale, BDI-II: Beck's Depression Inventory