Mental health differences in medical students based on curriculum and gender | BMC Medical Education

Our data shows that anxiety was twice as common as depression and that over a third of students screened positive for moderate or severe mood disorder. This finding parallels previous studies that show higher levels of psychological distress among medical students compared to the general population, suggesting that medical training is a time of great personal stress / emotional distress for many students. Notably, our findings showed higher rates of both depression and anxiety than previous studies which may reflect the implementation of our study during the height of the COVID-19 pandemic [12].

Further evaluation by gender revealed higher levels of anxiety, depression, and burnout in female students, a finding supported by previous studies [13, 14]. Among female respondents, almost half had high PHQ-4 scores (> 6), a finding that correlates strongly with functional impairment, disability days, and healthcare use [27]. In comparison, only a quarter of male respondents had similarly high scores. This discrepancy is reflected in the general population where women are twice as likely as men to have anxiety and depression [23].

Notably, there was no variation in anxiety or depression between students in different class years; this did not change even when grouped by preclinical (1st /2nd ) and clinical (3rd /4th ) years, suggesting the transition from an academic to clinical environment has a negligible impact on mood. However, there was a noticeable increase in the level of EE, a strong predictor of burnout, as students transitioned from first to second year [26]. This may be due to the increased pressure to study in preparation for the first board exam in addition to previous obligations from first year. Interestingly, the increased rates of EE did not continue into third year despite the need for preparation for the second board exam. This might suggest that as students advance in training, they acquire the experience and resilience necessary to better manage the increased stress that comes with increased expectations. From third to fourth year, the incidence of EE did not change significantly. However, it should be noted that the survey for fourth year students was distributed at the end of the academic year, after the residency match, a notoriously stressful process, had concluded.

We found that the PBL curriculum, which is based on working in small groups, may be less stressful for medical students than the traditional LBL curriculum. Although outside the cutoff for statistical significance, anxiety and depression tended to be lower in PBL students than LBL students. Anxiety and depression are strongly associated with burnout which when represented as EE, was significantly lower among PBL students [28]. While novel, this finding may not be entirely surprising since previous studies have suggested that PBL provides an improved educational environment compared to LBL [22]. Compared to LBL, which is characterized by large class sizes and one way dissemination of information by the instructor, PBL offers small group sizes and collaboration led by a facilitator. These factors may positively influence students’ perceptions of their learning environment and provide regular interpersonal engagement which may translate to an overall better mindset as they progress through medical school. However, these observations must be viewed in light of an inherent selection bias as students generally self-select into the two curriculum tracks. However, PBL spots are limited, therefore not all accepted students may get their preference. It is possible that students who are uncomfortable working in more closely interacting small groups may have selected LBL; conversely, students with greater confidence in their abilities may have chosen the PBL curriculum. These personal characteristics could very well contribute to the curriculum differences observed in this study. Therefore, this confounding factor makes it difficult to assign an improved educational environment in PBL as the sole factor for the differences observed.

When evaluating anxiety and depression by both curriculum and gender, female students had similar rates of anxiety and depression regardless of whether they were in the PBL or LBL curriculum. The finding suggests that the influence of gender on mood outcomes is stronger than any curriculum difference or potential selection bias.

Another interesting finding in this study was the inverse relationship between EE and sleep and exercise. This raises the question of whether EE, as a component of burnout, leads to decreased sleep and exercise or vice-versa. One study identified too little sleep as a predominant factor in the development of burnout, supporting the idea that adequate sleep is important in preventing burnout [29]. However, there is likely a bidirectional relationship between sleep and burnout. For medical students, establishing good sleep habits early on can help reduce burnout, potentially reducing the risk of making detrimental medical mistakes in the future [30].

This study supports prior studies documenting the prevalence of anxiety, depression, and burnout among medical students [13] which, as Dyrbye et al. pointed out previously, can contribute to suicidal ideation [9]. The potential consequences warrant investigation into reducing such prevalence. We highlight new evidence suggesting that students completing their preclinical years of medical school under the PBL curriculum have less EE than students in the traditional LBL curriculum. This previously undescribed result may represent an additional benefit of the PBL curriculum model which applicants might consider when making their medical school choices. Schools with only an LBL curriculum may seek to offer more PBL-like structure or an entire PBL curricular track. However, the true benefit of PBL cannot be fully assessed without controlling for self-selection.

Limitations

It is recognized that the study has an inherent selection bias as a major limitation since the LBL and PBL samples are nonrandom. Another limitation of this study is the subjective nature of some of the survey questions and corresponding answers. A further limitation is the smaller sample number of PBL respondents compared to LBL respondents. This is a consequence of the already smaller PBL class sizes. Despite these limitations, the study is the first of its kind where a comparison between two curricula is conducted within a single institution and lays the foundation for future analyses.