Educational Environment and Mental Health of Moroccan Medical Students: A Study in the Faculty of Medicine of Marrakech

Introduction: Studies concur that an optimal learning environment is a vital aspect for effective learning and for enhancing students’ well-being. Conversely, medical training is reported to be a suboptimal environment, thereby compromising students’ learning and well-being. Purpose: To evaluate the relationship between students’ perceptions of the educational environment and their mental health. Methods: This was a cross-sectional descriptive and analytical study, using the GHQ-12 and the DREEM questionnaires, with 380 students from the Faculty of Medicine and Pharmacy of Marrakech (3rd, 4th, 5th, 6th, 7th and 8th year), during the academic year 2017-2018. Results: We recruited 358 students in a period of 2 weeks, achieving a response rate of 94.2%. There was a predominance of females (66.48%). The mean age was 22.20 ± 2.149 years. The average GHQ score was 6.37 ± 3.484, with a psychological distress rate of 66.76%. The mean total score of DREEM was 86.5 ± 29.194 which indicates the existence of several significant problems. There was a statistically significant association between the poor perception of the educational environment and psychological distress. Conclusion: Improving the educational environment and promoting deep learning approaches for medical students will improve their psychological health during medical training.


Instruments
The Dundee Ready Education Environment Measure (DREEM) is a questionnaire developed by Roff et al (Roff et al., 1997) to measure the educational environment in health professional education programs.
The questionnaire was developed using a Delphi approach involving a range of professional health educators in different settings and countries.
However nine of the 50 items (4, 8, 9, 17, 25, 35, 39, 48 and 50) are negative statements and are reverse scored. The 50-item DREEM has a maximum score of 200 indicating the ideal educational environment. It is also divided into five subscales: 1) Students' perceptions of learning (12 items, max score 48) 2) Students' perceptions of teachers (11 items, max score 44) 3) Students' academic self-perceptions (8 items, max score 32) 4) Students' perception of atmosphere (12 items, max score 48) 5) Students' social self-perceptions (7 items, max score 28) The DREEM can be used to pinpoint more specific strengths and weaknesses. Items with mean scores ≥ 3.5 are considered as highly positive points. Items with mean scores between 2 and 3 indicate aspects of the environment that could be improved. While items with a mean of 2 or less should be examined more closely as they indicate problem areas.
The effect of distress and the consequences to mental health were estimated by the General Health Questionnaire (GHQ), offering four different categories of answers (graded from "better than usual", "as usual", "less than usual", to "much less than usual") for measuring the feeling of tension, depression, inability to defend, disturbed sleep based on anxiety, lack of self-confidence and self-esteem and other symptoms of mental health disturbance. The GHQ has been translated to and been validated in more than 40 languages, and exists in five versions that vary on the number of items (12,20,28,30,60). The 12-item version was chosen in the present study. From seven validation studies of the GHQ 12, Goldberg and Williams found a median sensitivity of 87% and a median specificity of 82%. (Goldberg et al., 1991) Two different scoring systems exist: Likert system (0, 1, 2, 3) for which the scoring range for the GHQ 12 goes from 0 to 36; and the standard method of scoring (0-0-1-1), which allows a maximum score of 12.
The cut-off scores for detecting cases vary in the many studies conducted to find the best threshold. For the purpose of comparison the 4+ was chosen to be the threshold in the present study. Students marking in the last 2 weeks will be classified as having a clinically significant problem and belong to the case group.

Subjects and Settings
This cross-sectional study was conducted in the Faculty of Medicine and Pharmacy of Marrakech in January 2018. The questionnaire was distributed to clinical stage medical students (3rd, 4th, 5th and 6th years) in the University Hospital Mohammed VI of Marrakech, several sessions were organized within many hospital departments to explain the interest and purpose of the study. Then, the students, having accepted to participate in the survey, received the questionnaires to fill, while respecting their anonymity. These questionnaires, once completed, were given to the secretaries of each department.
The students of the 7th and the 8th year received for their part the questionnaire directly within the faculty of Medicine and Pharmacy of Marrakech.
Given the personal nature of certain questions in this survey, the questionnaire was also made available in electronic format in order to respect the privacy of students and to promote their sincerity.

Statistical Analysis
The Data were analyzed using Excel 2010 and the statistical analyzes were performed by the ANOVA test. A P-value < 0.05 served as the cut-off value for statistical significance.

Result
The response rate was 94.2% (total 358 out of 380 students). Among the 358 students, 120 (33.52%) were male and 238 (66.48%) were female. The mean age of participants was 22.2 (SD 2.149) years.

Educational Environment
The mean DREEM total score was 86.5 (SD 29.194). Total DREEM scores ranged from 11 to 185.
The descriptive statistics for each of the five DREEM subscales are presented in Table 1. The highest score was found in the subscale of students' perceptions of teachers (21.71/44 (49.3%)), and the lowest score was found in the subscale of students' perceptions of learning (17.38/48 (36.2%)). Table 1 shows the individual item analysis of DREEM according to the five different subscales. 35 items scored less than two. Among them, 12 items were from the students' perceptions of learning subscale, 4 items were from the students' perceptions of teachers subscale, 4 items were from the students' academic self-perceptions subscale, 11 items were from the students' perceptions of atmosphere subscale and 4 items were from the students' social self-perceptions.
The remaining 15 items scored between 2 and 3 and there was no area of excellence (Item score ≥ 3.5).
The lowest score was 0.57 for Item 3 "There is a good support system for students who get stressed".

Mental Health Status
The mean score of psychological distress (GHQ) was 6.37 (SD: 3.484) ranging between 0 and 12, and 66.76% scored above the threshold (4 points) indicating notable mental problems. wanted medical treatment and/or psychotherapy.

Correlation Analysis
Correlation analysis revealed significant correlation between psychological distress and the educational environment; it also showed a significant association between the educational environment and the need for psychiatric help. Table 2 Table 3 We also found a very significant correlation between psychological distress and 23 individual items of DREEM, since item scores were very high among students who had a GHQ score less than or equal to 4 compared to students with GHQ scores greater than 4. Table 4 The need for psychiatric help is also a factor associated with 23 individual items, since the average item scores were higher for students who reported that they did not require psychiatric help.    The students irritate the teachers 2,28 2,55 0,007

Discussion
The high response rate (94.2%) obtained in our study was due to the brief introduction given to students about the aim of this study, which convinces them that the results of such a study would lead to significant changes. The students also perceived it as an ideal opportunity to express their opinions.
The response rate in other studies ranged from 44.6% to 96.9%. This showed that our response rate was among the highest, indicating that our students were keen to participate in such study to improve their school and their mental health status.
This response rate is comparable to that obtained in Canada (91%) (Till, 2004) and in Australia (90%) (Vaughan et al., 2014). On the other hand, the lowest response rate obtained in King Saud University (44.6%) was explained by students' fears of participation in their study and its impact on their exam results (Al Ayed & Sheik, 2008).
The mean age of the respondents was consistent with similar studies carried out in other medical schools (Backović et al., 2013;Bíró et al., 2011;Oku et al., 2015;Sherina et al., 2004), in which the mean age was 22.2±2.149 and the majority of the students were between the ages 19 and 31 years.
Overall, two out of three respondents were female, which is quite different from what was obtained in a study conducted in Japan, with 66.3% male and 33.7% female (Ohtsu et al., 2014).
A study in Singapore reported that 79 studies showed total DREEM scores ranging from 100 to 150, and only 3 studies reported excellent scores between 150 and 200. (Chan et al., 2018) The global DREEM score of 86.5/200 indicated the existence of many significant problems in the educational environment of our faculty. As far as we can verify, our study had the lowest score Among the subscale scores, students' perception of learning was lowest in our study (36.2%). This is fairly close to the score of 38.3% reported by Andalib (Andalib et al., 2015) and of 39.58% reported by Till (Till, 2004), but lower than the score of 71.7% (34.42/48) reported by Vaughan (Vaughan et al., 2014). Table 6 The perception of learning atmosphere, where other studies showed to have significant impact on students' behavior, academic progress and sense of well-being, scored low in the present study. The Medical students around the globe seemed to share similar concerns as reported in studies that utilized the DREEM instrument (Mayya & Roff, 2004;R.M. Harden, 2000). It is interesting that most areas of concern are related to what is taught rather than how it is taught and allude to the curriculum content rather than its delivery.
There were 35 items that scored below 2, which indicated problematic areas of the learning environment. Item 3 (There is a good support system for students who get stressed) had the lowest score (0.57) in the questionnaire. This item also scored the lowest in other studies. (Aghamolaei & Fazel, 2010;Al Sheikh, 2014;Al-Hazimi et al., 2004;Al-Qahtani, 2015;Dimoliatis et al., 2010;Herrera & Oslando Padilla, 2015;Rotthoff et al., 2011) A study in Greece (Dimoliatis et al., 2010) found 19 problem areas, another study in Germany (Rotthoff et al., 2011) reported 18 items with scores below 2, while a study in Iran (Aghamolaei & Fazel, 2010) objectified the existence of 22 problem areas.
In our study, no area of excellence (Score ≥ 3.5) was reported, which is in agreement with many studies. (Bennett et al., 2010;Rotthoff et al., 2011;Shehnaz & Sreedharan, 2011;Tontus, 2010;Veerapen & McAleer, 2010) The psychological morbidity in our study was significant and a cause of concern for the faculty and administrators. Previous literature suggested consistent evidence of higher prevalence of anxiety, depression, and burnout (Dyrbye et al., 2008(Dyrbye et al., , 2006 and psychological distress in medical students than in the general population and age-matched peers (Dyrbye et al., 2008;Grant et al., 2004).
Assessment of psychological morbidity or mental health status of the respondents using the GHQ12 was a key finding in this study. The prevalence of psychological morbidity was 66.76%. This was found to be high compared to other studies (Dendle et al., 2018;Farahangiz et al., 2016;Ohtsu et al., 2014;Oku et al., 2015). However, studies conducted among medical students in Australia, England and Malaysia, which used the same cut off as the present study, showed similar results (Firth, 1986;Willcock et al., 2004;Yusoff, 2011). Table 7 Our highly stressful educational environment, the personal characteristics of our students and possible previous mental health problems may be considered as the reasons for our high levels of psychological morbidity. These variations in mental health status of medical students shows that effective supportive and mental health services still need to be instituted as a necessary part of the under graduate medical training both in developed and developing countries.
The significant findings from our study are that the educational environment has a direct effect on the students' mental health and interestingly, psychological distress also has a direct effect on the educational environment. Several studies have demonstrated that an unfavorable medical training atmosphere leads to a high prevalence of psychological distress (Wolf, 1994;Yusoff et al., 2010) and eventually leads to unwanted consequences either at the personal level or professional level (Dyrbye et al., 2008(Dyrbye et al., , 2005. A Malaysian study reported similar results, supporting that a favorable educational environment directly improves the psychological distress of medical students. (Yusoff & Arifin, 2015) One of the important implications of this finding is that our faculty should make the effort of conducting a regular evaluation of the educational environment to detect potential areas of concern; it should also be aware of the high prevalence of psychological distress among their students, as it could be the sign of an unfavorable educational environment.

Conclusion
This study extends the evidence of the important relationship between the educational environment and the students' psychological health, and thus the faculty, the medical educators and the students should work hard together to create an optimal environment.