Mental Health Disturbances and Related Problems in Italian University Medical Students from 2000 to 2020: An Integrative Review of Qualitative and Quantitative Studies

Background and objectives: The presence of mental health problems in the population of medical students in Italy has been evaluated in several cross-sectional studies, which have used different methodologies and study designs. However, a global overview of the prevalence of mental health problems in Italian medical students is not available, although this would be essential for promoting preventive strategies and supportive treatments. Materials and Methods: An integrative review aiming to describe the prevalence of mental health problems in Italian medical students has been performed. Results: The most relevant findings are the high prevalence of substance use, in particular alcohol and nicotine, and of depressive and anxiety disorders in Italian medical students. In particular, substance use ranges from 13 to 86%, which is higher compared to Italian students coming from other faculties. Italian medical students show a high rate of smoking and of depressive symptoms of about 20%. Conclusions: Our findings highlight the need to develop appropriate supportive interventions for the medical student population, which are rarely provided and implemented among the routine activities of Italian medical universities. A relevant aspect to be considered is the stigma and anticipated discrimination attached to mental disorders, which reduce the help-seeking process in medical students.


Introduction
Up to 35% of university students suffer from common mental disorders or related health problems [1]. The transition from high school to university is a critical period in terms of biological, psychological, and social development, with the creation of new relationships, a new identity and an increase in autonomy and responsibility [2]. Moreover, most mental illnesses arise and develop during late adolescence.
The admission into medical school and the period of graduation are very competitive and highly stressful. In fact, compared with their peers, students from medical school show higher rates of mental health problems, including depressive and anxiety symptoms [3], and are at higher risk of using illicit substances [4] or of developing full-blown mental disorders [5]. Students reporting these symptoms have reduced academic performances and a consequent poor quality of provided healthcare and increased medical errors [6]. Risk factors for the development of psychiatric symptoms in this student population include female sex, exposure to recent stressful life events, excessive use of smartphone and low quality of sleep [7].
Tobacco consumption and smoking dependence represent a rising problem for medical students. In fact, Mas et al. [8] found that medical students' tobacco consumption increases during the years of training in medical schools. Moreover, tobacco use has been associated with the risk of developing depressive symptoms [9]. Another unhealthy lifestyle behaviour frequently adopted by medical students is the consumption of energy drinks, which are used for their stimulatory effect and for claims of giving benefits, such as physical endurance reaction, concentration, and reduced need for sleep. Smoking represents a risk factor for the use of energy drinks in university students [10][11][12]. Therefore, the use of energy drink, smoking habits and alcohol consumption are considered "proxy" measures of the global mental health of medical students.
Despite medical students frequently have mental health problems, their levels of help-seeking are quite low [13]. It may be that their access to specialist care is delayed by the presence of stigma still attached to mental disorders, as recently documented by the American Psychiatric Association that more than 50% of medical students meeting the diagnostic criteria for a mental disorder are reluctant to seek professional help because of fear of exposure to stigma [14]. Moreover, after graduation, the fear of stigma and of financial and professional consequences are a significant barrier to seeking help among doctors [15][16][17]. In Italy, the levels of stigma attached to mental disorders are still quite high, particularly in specific socio-cultural contexts impacting negatively on the help-seeking. In particular, a survey carried out in the general population indicates that depression is usually seen as a reaction to significant life events. Moreover, the levels of stigma towards depression are quite high, as highlighted by the fact that participants report that it is embarrassing to report depressive symptoms to general practitioners and there are several concerns regarding the use of antidepressants [18].
The prevalence of mental health problems in Italian medical students has been evaluated in several studies, adopting different methodologies and study designs. However, a global overview of the prevalence of mental health problems in Italian medical students is not available, although this would be essential for promoting preventive strategies and supportive treatments. Therefore, we decided to perform an integrative review of qualitative and quantitative studies aiming to describe the prevalence of mental health problems in medical students in Italy, where we expect that the levels of mental health problems in medical students are higher compared to other countries [19], which could be due to the high levels of stigma, the prestigious role of medical doctors [20] and the lack of a structured organization of psychological support during medical training [21].

Materials and Methods
This review was performed in five stages: definition of the problem, literature search, data evaluation, data analysis and presentation of findings [22]. Our goal was to define the prevalence rates of mental health problems such as depression, anxiety or suicidal behavior or abuse/dependence of drugs among Italian medical students.
The following search terms, "students", "medical", "mental health", "Italy", "mental disorders", "depression", "anxiety", "suicidal ideation", "suicide attempted", "suicide", "alcoholism (alcohol and drug use)", "street drugs (abuse, dependence, binge)", "prevalence" and "risk factors" were entered in ERIC, MEDLINE, PsycARTICLES, PsycINFO, SCOPUS and PUBMED (Table 1). Terms and databases were combined using the Boolean search technique, which consists of a logical information retrieval system (two or more terms combined to make search more restrictive or detailed). Students, medical AND "mental disorders" AND Italy AND "risk factors" 0 1 0 13 12 4 Three authors from Italy (GS, FZ and VG) and three authors from Brazil (EGLS, RJGU and GML) independently searched papers, which had been published between 1 January 2000 and 30 June 2020. Authors (GS, FZ, VG, EGLS, RJGU and GML) searched the same databases. The selected articles were saved in an Excel shared document.
Qualitative and quantitative studies on the prevalence of mental health problems (depression, anxiety, suicidal behavior and drug abuse/dependence) among Italian medical students, which had been published in peer-reviewed English or Italian journals (with abstracts written in English), have been included in the review. When studies included mixed student populations (i.e., medical and non-medical students) and it was not possible to extrapolate data only on medical students, these studies were excluded.

Selection of Study and Data Extraction
Two members of the review team (one in Italy and one in Brazil) independently screened titles, removed duplicated papers, evaluated eligible studies, and performed data extraction. Any discrepancies in the study selection were resolved through discussion with a senior expert reviewer. Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) was used to demonstrate the selection of study [23] (Figure 1). Data from each study included: study design, sample size, prevalence of depression, anxiety or suicidal behavior or drug abuse/dependence among Italian medical students, mean age, sex, main aims, data collection, risk factors, main conclusions and GRADE (Grading of Recommendations, Assessment, Development and Evaluations) criteria [24].

Quality Assessment According to the GRADE Criteria
The quality of quantitative studies was evaluated according to the GRADE criteria by two authors from Italy (GS and FZ) and two from Brazil (RJGU and GML), independently. Four levels of quality are identified, from "high" to "very low". The quality of evidence is considered "high" when the study is a randomized controlled trial; and "low" in the case of observational studies. The attribution of a lower level is based on the presence of limitations, bias, inconsistencies, indirectness, and impreciseness in the study.
The authors independently assessed the studies against these criteria and a discrepancy rate of about 10% was found, which was solved through online discussion.

Results
According to the search strategy, 873 records were identified. Based on title and abstract's review, 296 were duplicates and were excluded. For the remaining 577 papers, the main reasons for exclusion were "topic not relevant" (N = 411) and "not the target population" (N = 146). The remaining 20 articles were retained, and seven additional eligible papers were identified by searching the references' lists of the papers. Therefore, sixteen studies have been included in the review ( Figure 1). In the majority of cases [25][26][27][28][29][30][31][32][33][34][35], the articles' title was related to the use of cigarette, energy drink or alcohol. In 12 out 16 studies, the main aim was the evaluation of the consumption rate among Italian medical students and their knowledge and attitudes about substance use. Four studies [34,[36][37][38] investigated the prevalence of alcohol, depressive symptoms and suicidal behavior, and evaluated the factors associated with these syndromes ( Table 2).
The sample size of included studies ranged from 44 to 794 participants. Almost all studies adopted a cross-sectional design, while only three studies [29,32,35] used a multicenter cross-sectional design.
The prevalence rates were 15.3-31.4% for cigarette use, 22-75% for energy drink and 13-86% for alcohol use. The percentage of smokers was significantly higher in male than female students [26]. Moreover, male students were at higher risk of using multiple substances (drink, alcohol, nicotine and coffee) and of being heavy smokers compared to female students [33]. However, women showed a higher prevalence of depressive and anxiety symptoms [36].
On the global sample, the prevalence of depressive symptoms ranged between 2.5 and 21.4%, while severe suicidal thoughts were found in 17% of participants [37]. Depressive symptoms and deficit/hyperactivity symptoms were more likely to be present among students seeking counseling [38]. Students with severe suicidal behavior had higher rates of anxious and cyclothymic temperaments compared with non-severe suicidal group [36].
Only one study [38] with a small sample size (N = 44) found a higher prevalence rate of adaptation disorder (36%) and of psychotic disorder (14.3%), compared to dysthymia and panic disorder (7.1%).
The levels of knowledge regarding the risks of tobacco use, its epidemiological aspects and possible treatments were relatively low among Italian medical students.

Summary of Design Quality
The GRADE criteria were adopted to evaluate the quality of included studies. The majority of the studies (N = 10, 62.5%) were of moderate quality, being cross-sectional studies and conducted in one or two cities only. The remaining six papers were rated of low quality due to the small sample size and the lack of limitations or to the risk of reported bias. Since the overall quality of the included studies is moderate, we recommend that multicenter longitudinal studies involving students attending different years of training in the medical schools should be carried out.

Discussion
To our knowledge, this is the first integrative review aiming to evaluate the prevalence rate of mental health problems in Italian medical students. While many healthcare professionals report high rates of burnout and poor mental health and wellbeing, it is likely that the student population is even more vulnerable to the development of mental health problems [40][41][42][43]. Several studies have reported that medical students have a poor mental health and a low access to adequate treatments [44][45][46][47][48][49][50][51]. Furthermore, the mean age of access to a medical school corresponds with the mean age of onset of many mental disorders. The highly competitive environment of university settings could represent a trigger factor for vulnerable students, thus precipitating the onset of any mental disorder [52]. It is likely that mental health can worsen when the medical school starts and can remain poor during the whole training period.
Based on the included studies, the most striking findings are the high prevalence of substance use, in particular alcohol and nicotine, and of depressive and anxiety disorders. In particular, substance use in Italian medical students is very high, ranging between 13-86%, which is higher compared to Italian students coming from other faculties. In a study by Jackson et al. [52] in the US, alcohol abuse and dependency among medical students was associated with higher levels of burn out. This finding can be due to the fact that expectations at medical schools are very high, since students are trained to become future scientists and health professionals.
Compared with other European students, Italian medical students show the second highest rate of smoking [53]. A systematic review by Niu [54] on a Chinese population of medical students found some improvements in the status of tobacco use, following an educational campaign against tobacco targeted to medical students. Moreover, while the prevalence of smoking is about 20% in the Italian general population, it is slightly higher (ranging between 15-31%) in Italian medical student population [55]. However, we found that the prevalence rates of smoking of Italian medical students is lower compared with those of Italian students from other university settings. This finding may be due to the increased knowledge on the detrimental effects of tobacco during medical training, which can contribute to reduce the consumption of nicotine along the years [56].
In line with previous studies carried out in Europe, the prevalence rate of depressive symptoms was about 20% in Italian medical students. However, Rotenstein et al. in their metanalytic study [3] found a wide variability of the prevalence of depression in medical students worldwide. We found an association between the presence of depressive symptoms and cyclothymic temperament, which highlights the importance of screening programmes in order to evaluate personality traits in students [57][58][59][60][61][62][63]. Furthermore, depressive symptoms are more frequently reported by female medical students compared with their male peers, thus confirming the higher prevalence of mood disorders in females. This finding needs a careful evaluation, considering that the number of female students starting a career in the medical field has doubled since 1980s. Therefore, it may be useful to develop preventive and supportive interventions for depressive disorders taking into account biological, social and psychological gender differences [64][65][66].
According to the reviewed studies, the prevalence of suicidal thoughts in Italian medical students is about 17%, and it is higher in males. This finding, which is consistent with previous evidence showing high rates of suicidal thoughts and behaviours in medical students [3,[67][68][69], underlines the need to develop appropriate preventive strategies to tackle suicide risk in medical students [70,71].
The main limitations of included studies were the small sample sizes, the crosssectional designs, and the use of self-reported instruments. In particular, when using self-reported instruments, the findings may be biased by the so-called "social desirability". Moreover, these self-reported instruments should not be considered valid to formulate a diagnosis of a mental disorder, but they are rather screening tools. In order to evaluate the prevalence of mental health and related problems in medical students, there is the need to carry on longitudinal, multicenter, population-based studies with sound methodologies. Therefore, our findings must be interpreted cautiously, and further, longitudinal studies may be carried out in order to fill in this gap and obtain conclusive data.
The present integrative review has some limitations which should be acknowledged. Firstly, the decision to select and include only studies carried out in Italy makes our findings not representative of the global condition of mental health problems in medical students worldwide. However, this decision is due to the fact that the Italian context has some specific socio-cultural features which should be taken into consideration in order to develop ad hoc preventive interventions targeting medical students. Another limitation is due to the narrow time frame considered, but previous years have been covered by international systematic reviews [3].

Conclusions
Our findings confirm the high prevalence rates of depressive-anxiety symptoms and of substance abuse in the medical student population. Therefore, these data highlight the need to develop appropriate supportive interventions for this specific population [72,73], which are rarely provided and implemented among the routine activities of Italian medical universities. A relevant aspect to be considered is the stigma and anticipated discrimination attached to mental disorders, which reduce the help-seeking process in medical students. In particular, medical students feel concerned by being treated for a mental health disturbance, further delaying the first contact with psychiatric services [74][75][76][77][78]. On the other hand, treatment programmes for medical students with substance abuse should be developed, as well as training about smoking cessation [79,80].  Institutional Review Board Statement: Ethical review and approval were waived for this study, due to the adopted methodology (i.e., review of existing studies).