Study design

The present study was conducted during the months of January to March 2021, a period in which the second wave of infections began in the country and clinical rotations of medical students in hospitals were still restricted in Peru [3].

Study population

The population consisted of students from the 3rd to 7th year (clinical years) from 24 different universities in Peru; from the three natural regions (coast, mountains, and jungle) that have a medical school [14,15]. The sample size was calculated with a prevalence of mental health disorders of 50.0%, a confidence level of 95%, a precision of 5%, a design effect factor of 1, and considering a infinite population which gave us a number of participants of 384, which was exceeded in our study (808 participants) [16]. The type of sampling was the snowball type, which happened when “one subject gives the researcher the name of another, who in turn gives the name of a third, and so on” [17].

We considered students who belonged to courses superior to the third year, regular students (those with more than 12 credits) and who accepted the informed consent. We excluded students who belonged to years below the third because they do not study in the clinical field according to the curriculum in our country. Also, because the instrument Dundee Ready Education Environment Measure (DREEM) that was used to evaluate the educational environment focuses on asking about characteristics of the clinical environment.

Instrument and data collection

The online survey was divided into four sections: 1) Informed consent; 2) sociodemographic data, and educational data; 3) DREEM questionnaire that evaluates the EE perceived by the students; 4) 7-item Generalized Anxiety Disorder Scale (GAD-7) that evaluates symptoms of generalized anxiety disorder (GAD); and 9-item Patient Health Questionnaire (PHQ-9) that evaluates depressive symptoms. To collect the data, we created a survey on the Google Forms platform that was sent to the study collaborators and through them we proceeded to share the survey with all groups of students, in addition to sharing it on social networks (Facebook, WhatsApp, Telegram, Instagram) by posting on the class pages of medical schools. Consent was obtained from the participants at the beginning of the survey. Prior acceptance triggered the subsequent sections. Otherwise, the survey was terminated. Participants were asked to complete the survey only once and as honestly as possible, as we limited the survey to one response per participant, which ensured that no multiple responses were given with consequent overestimation of the results.

Anxiety and depressive symptoms

The GAD-7 scale was used to assess anxiety symptoms. It consists of 7 items, with scores ranging from 0 (not at all) to 3 (almost every day), so that the total score ranges go from 0 to 21 and, in turn, can be classified into 4 severity groups: minimal (0–4), mild (5–9), moderate (10–14) and severe (14–20) [18]. The instrument has a sensitivity (89%) and specificity (82%) [19]. The GAD-7 was validated in Peru with good reliability (Cronbach’s alpha = 0.89) [20]. We defined the presence of anxiety with a cut-off point > 4.

Depression was assessed using with the PHQ-9. The Spanish version shows a sensitivity of 92% and a specificity of 89% [21]. This instrument was validated in Peru and was found with an adequate internal consistency (Cronbach’s alpha = 0.903). It consists of 9 items and each item is scored according to a Likert scale that ranges from 0 (no day) to 3 (almost every day). The PHQ-9 scores reflect 5 categories of depressive disorder severity: none (0–4 points), mild (5–9 points), moderate (10–14 points), moderately severe (15–19 points), and severe (20–27 points) [22]. We defined the presence of depression with a cut-off point > 4.

Educational factors

The perception of the EE was assessed with the DREEM instrument as this is considered the most optimal instrument for its evaluation. The instrument was validated and developed in Peru, with high reliability (internal consistency); Cronbach’s alpha = 0.93 [23]. It involves 50 items. It is answered with a 5-level Likert-type scale: completely agree (4 points), agree (3 points), not sure or no opinion (2 points), disagree (1 point), and completely disagree (0 points). Negative perception of the EE was defined as a score between 0 and 50, with many problems between 51–100, more positive than negative between 101–150, and excellent between 151–200 [24,25]. Among the items of this instrument, we will find questions on the identification of areas of EE strengths and weaknesses, academic performance, self-perceived competence to work, aspects of the work environment, and others [26].

Biostatistical methods

Categorical variables were described as frequencies and percentages, and continuous variables as mean with SD (standard deviation) or media with IQR (interquartile range) depending of the distribution of the data. We performed chi-square to determine the association between the sociodemographic and educational factors with anxiety and depression. In addition, simple and multiple regression models were created using generalized linear models (GLM) with Poisson distribution, robust variance, logarithmic link function and grouping by type of university to obtain prevalence ratios (PR) with 95% confidence intervals. The multivariate multiple model analysis was adjusted for all variables entered in the simple regression. Data analysis was carried out in STATA version 16.1 (College Station, TX: StataCorp LL).

Ethical aspects

The study was approved by the institutional ethics committee of the Universidad Peruana Unión (Code: 2020-CEUPeU-00047). Informed consent was obtained from each participant and the data were anonymous and confidential. Only the researchers had and will have access to the database.

High prevalence of depression and anxiety

We found a higher prevalence of depression compared to other studies. In Shanghai universities, a prevalence of 46.2% was found, with males being the least affected; in Czech and Slovak university students, 52% and 47%, respectively [27–29]. However, in relation to the prevalence of mental health symptoms, our results are similar to those reported in a study carried out in Morocco that showed high levels of mental health symptoms especially women, who are in the preclinical stage and live in regions with a high prevalence of COVID-19 cases, where the prevalence of anxiety and depression was 77.6% and 67.3%, respectively [30]. Similarly, another study reported an anxiety prevalence of 75.4% and a significant association with female sex, but this study was conducted on first-year university students [31]. However, they are contrary to a study that revealed a higher prevalence of anxiety and depression in face-to-face classes (42.3% and 49.3%, respectively) than in virtual classes (15.5% and 27.6%) [1]. A study in Peru during the pandemic that reported a prevalence of anxiety and depression of 47.6% and 47.3%, respectively, in young university students [32]. Denoting that in the pandemic context, the development of mental health symptoms was conditioned by the dissatisfaction associated with the maladjustment of universities to the virtual environment of education, which in our country occurred in 65.0% of students [12]. Another issue that could contribute to the high prevalence of mental health problems is the interruption of clinical rotations, which raises concerns about the potential impact on physicians’ professional development [33].

Sociodemographic factors associated with mental health disorders

Being male has presented a higher probability of presenting depression compared to women, a result similar to a study conducted in Asia [34], but, different from what was found in another where the most affected population were women [35], similarly, another study found no differences in relation to gender [36]. It has been reported that women are more likely to develop depressive symptoms [28,37] as opposed to men, because they have a lower threshold for developing these symptoms [38], however, our results add to the current literature regarding this association. Our result could be explained due to the fact that that men have been seen to have greater loneliness, greater economic distress and lower levels of resilience that predispose the development of these symptoms [39], a result that should be taken with caution because men with symptoms of health disorders patients do not seek treatment due to stigmatization, a situation that leads to the chronification of symptoms and the risk of committing suicide [40,41].

Having a history of illness is associated up to 10 and 12 times more with developing anxiety and depression, respectively. Similar to what was found in the Peruvian population, which was higher in women and young people [35], it is also supported by the result of another study carried out in Malaysian university students [42], showing that taking or living with health comorbidities predisposes to the condition mental [43].

People with a diagnosis of COVID-19 were 7 times more likely to present anxiety. This was different from a study of French university students who were 45 times more likely [44], a result similar to that evidenced in the Canadian population up to 55 times more [45] highlighting that this study was conducted in the general population. Our result could be explained by the fact that presenting a disease with high stress load such as COVID-19 conditions the increased production of cortisol as a result of a dysfunction of the hypothalamic-pituitary-adrenal axis, which keeps the organism in a constant development of anxiety in addition to other mechanisms that have not yet been elucidated [46].

Students who lived in regions such as the mountains and the jungle presented up to 11 times more likely to develop mental health symptoms, where 41.8% of students came from these regions characterized by high poverty (44.9% and 34.4%) of the Sierra and Selva, respectively [47], the same result evidenced at the educational level where approximately 25.0% of the residents present low pedagogical level [48], associated with situations that condition the development of mental health symptoms due to the little knowledge that has been able to be developed about COVID-19 [49,50], which has possibly generated greater uncertainty.

Educational factors associated with mental health disorders

Students with a negative perception of the EE presented more anxiety and depression symptoms. University EE is known to have a significant impact on education. It is considered a determining factor in the academic development of the medical student. As well as the expression of the administrative management of the university and the quality of the study plan of the career [51]. Similarly, a study from Malaysia evaluated the association between EE and psychological distress found that a positive environment has direct influences on the psychological health of medical students [52]. Therefore, negative EE could lead to emotional instability or burnout resulting in lower academic performance [53], affecting their emotional stability and motivation to continue with their careers [54].

Our study found a lower score in the perception of the level of EE compared to a study carried out in Saudi Arabia, which obtained a mean score of 122.4, a result that could be due to the context prior to the pandemic [55]. In another study conducted in India, a total score greater than 130.6 was found, indicating a higher quality of the EE. This difference may be due to the pre-pandemic context, having a smaller sample and conducting the study in every academic year [56]. However, it should be mentioned that the DREEM instrument has questions about the clinical setting. Therefore, it cannot be applied to students entering medical school to take basic and general science courses. Our results are also lower than those of other Latin American medical schools. In Chile they found a result that ranged between 103.1 and 126.9 points. Similarly, a study conducted in Argentina found total scores for the 1st, 3rd, and 5th academic years ranging between 149.6 and 136.6 [57]. In Colombia, the total mean was 152.0, but it should be noted that this study was only conducted in one course [58]. This suggests that these countries have better curriculum plans than ours. However, it is worth mentioning that all the studies, including ours, presented a more positive than negative perception.

Students who came from a national university are 10 times more likely to develop some of the mental health symptoms. The national universities in our country correspond to more than 52.5% of the 40 faculties that exist [14]. National universities have fewer student health support programs and less connectivity to Internet networks [59] which leads to maladaptation to virtual education, a situation that has led to greater stress in university students in general [12] and even greater anxiety in medical students [2]. The medical schools of private universities, on the other hand, have better resources and were able to present a better adaptation to EE [16]. However, we have not found studies that evaluate this variable and have demonstrated this association, therefore, our results add to the current literature.

Recommendations

The current context of COVID-19 affects the mental health of the student population; due to increased loneliness, mourning, and hopelessness. Also, there is concern about academic overload, ineffective communication with teachers, the future of the career, the delay in clinical rotations, and the fear of the infection spreading [30]. For this reason, it is imperative to improve EE in medical schools, since a constant and effective evaluation of online learning in universities is considered a potential protective factor for symptoms of anxiety and depression. During the pandemic, increased stress due to virtual classes, thoughts of abandonment, and decreased productivity were also perceived; these are considered potential risk factors for increased anxiety and depression [60]. For all of the above, it is recommended to improve EE through the implementation of curricular restructuring programs, mental health counseling, and well-being to reverse these consequences that affect academic performance.

It is suggested to carry out EE evaluations per academic year in the universities to estimate it’s the perceptions of students and thus intensify psychological support in universities [61]. At the same time, generate a culture of acceptance, concern and destigmatization of mental disorders in university students in crucial in every university [62]. Psychological counseling is indispensable, especially during the pandemic period or high-stress events, and measures must be taken to address mental health issues, such as providing remote counseling for students.

The experience shown in the study is valuable to be able to periodically monitor curricular changes, as well as to evaluate the pedagogical innovation that can be implemented. Annual studies should be conducted, by academic cycle, region of origin and type of university, focusing on factors associated with mental health symptoms.

Limitations and strengths

Some of the limitations were the cross-sectional design of our study, which does not allow us to identify causal relationships between the study variables. The symptoms of anxiety and depression in this study were evidenced through the analysis of the universal self-assessment scale, which differed from the DSM-5 diagnostic criteria; therefore, the results of this study are for reference only and cannot be used as a guide for clinical diagnosis and treatment; however, it provides us with updated information on the burden of mental health in a particular population, medical students. Our study may present biases due to the lack of representativeness and the fact that the sampling was convenience sampling; however, we included students from more than half of the licensed medical schools (24 out of 40). Finally, our study may present measurement biases, since the questionnaires are self-administered, but as a strength, the instrument used to measure EE was applied to a larger sample compared to other published studies on the subject, and was applied to students in the clinical stage, therefore it can provide the information requested by the instrument.

On the other hand, to our knowledge, our study is the first study that evaluates the perception of EE and the prevalence of mental health disorders during the COVID-19 pandemic, and in a large and varied population, Peruvian medical students from 24 of the 40 medical schools in Peru, including the three regions of Peru, different by their economic status and quality of education, important in understanding the results. This study serves as a reference for implementing new educational environment strategies, and is further evidence that medical students deserve special attention to their mental health. This study also serves as a basis for further studies that wish to assess the mental health of students, in order to convince those in charge of education to improve the educational environment and reach conventional agreements with students to reverse the alarming numbers of mental illnesses they may have, which can alter the course of their lives and careers.