Stress
First-generation college students
Compared to their CGCS peers, the FGCS in our study had significantly higher perceived stress at the end of third year clerkships and no significant reductions in academic performance (table 2, figure 1). One interpretation of this finding may relate to imposter syndrome, which is characterized by an overwhelming belief that one does not belong in a certain setting despite evidence to the contrary and fear about being discovered as a “fraud” [34]. People experiencing imposter syndrome have chronic self-doubt and are unable to internalize personal achievements [34]. Levant, et. al. (2020) found significant correlations between stress (measured with the 10-item perceived stress scale [PSS-10]) and imposter feelings (measured with the Clance Impostor Phenomenon Scale), and that PSS-10 scores were 28-31% higher in those experiencing imposter phenomenon [35]. Studies on FGCS at the undergraduate level have shown that these students more frequently report difficulty fitting into campus culture and often doubt their abilities to succeed, feelings that are directly related to imposter syndrome [36, 37]. Canning, et. al. (2019) looked at associations between peer competition, generational status, and imposter feelings in students taking science, technology, engineering, and mathematics (STEM) undergraduate courses [38]. They found that FGCS were significantly more likely to experience imposter feelings in settings where perceived competition was increased (e.g., STEM courses) [38].
It is possible that the first-generation students in our study may have experienced imposter feelings related to the highly competitive learning environment during the third year of medical school. Imposter syndrome is further supported by the absence of significant reductions in exam performance, as this provides evidence that these students are not academically inferior to their peers. Although the increase in perceived stress in the FGCS was isolated to a single educational milestone in our study, sources of stress in these students should be investigated, as should other measures of well-being.
Academic performance
Underrepresented in medicine & the older medical students
URM students, on average, scored lower than their non-URM peers on all academic performance measures, with statistically significant lower average performance on the pre-clinical exams (table 5, figure 2). This is important because it echoes historical trends in assessment performance disparities impacting racial and ethnic minority students [39-43]. Research on the achievement gap in medical education highlights how structural inequities in education stemming from poorly funded K-12 schools that primarily serve low-income and minority children leads to disparities in performance on standardized exams, including the Medical College Admission Test (MCAT) and United States Medical Licensing Examinations (USMLE) [39-41]. The most recent report on medical school applicants and matriculants data from the AAMC showed that URM students (Black/African American, Hispanic/Latino/Spanish, and American Indian/Alaskan Native/Native Hawaiian/Pacific Islander) had lower average GPAs and MCAT scores than non-URM students [44]. A 2019 study by a team at the National Board of Medical Examiners (NBME), found that Hispanic and Black students were significantly more likely to score lower on any of the Step exams compared to White students, and that nearly half of all students who initially fail Step 1 are racial/ethnic minority students [42].
Students who were 30 years of age or older at matriculation had significantly lower average exam scores across all academic performance measures compared to their younger counterparts (table 9, figures 3 and 4). These findings indicate that a need is not being met for these students, which may relate to curricular design and/or assessment formats being poorly suited for the adult learner. Previous research suggests that older medical students may have different learning strategies than their younger counterparts. In a 1998 study at McGill University, Feil et. al. found that older medical students approached learning more abstractly, with concern for the thought processes involved in basic and clinical science, while younger students were more inclined to study by memorizing facts for a test for the sake of getting good grades [45]. In 2000, Kick, et. al., reported that older students perceived medical school to be more intrusive upon their deeply developed personal lives, and that responsibilities at home made it difficult to study [19]. Further investigation to identify factors that are undermining academic success for URM and older medical students is warranted.
Negative findings
In none of the study groups did we find both higher stress and lower academic performance. The first-generation students had higher perceived stress (table 2) and no significant differences in academic performance (table 3), while the URM and older students had significantly lower academic performance in the absence of differences in stress (tables 4, 5, 8, and 9; figures 2, 3, and 4). In fact, our results show that URM and age 30+ students arrived at medical school with the lowest reported stress of any group, including the control groups (tables 4 and 8, respectively). Based on socioeconomic status, no significant differences in stress or academic performance were found. These findings, or lack thereof, yield multiple interpretations.
The collection of program evaluation data by Kirk Kerkorian School of Medicine necessitates that the surveys be a required component of the curriculum. Because of this, it is possible that our results are confounded by individual differences in attention and reflection on the survey questions. While the PSS-4 was designed to be better suited for settings in which respondents may not have the time or desire to complete the longer versions of the PSS (e.g., the 10-item and 14-item questionnaires) [30], and indeed this is the rationale for using the Short Form PSS, it is possible for students to simply click though the survey because it is required to do so without responding thoughtfully to the survey questions. If this is the case, then the survey results would not reflect what we are trying to measure.
Another possible explanation for our results showing no differences in stress, with a single exception between FGCS and CGCS, may reflect individual differences in stress appraisal and resilience. Research on stress theory holds that the effects of stress depend, at least in part, on whether the stress is perceived as enhancing or debilitating [46]. For those who perceive stress as enhancing, it can improve performance and enhance motivation to overcome a challenge [46]. Perhaps the FGCSin our study, while they have significantly higher stress than CGCS at the end of third-year clerkships, may be less inclined to appraise stress as a negative factor, and thus academic performance is not impacted.
Moreover, by virtue of their nontraditional sociodemographic backgrounds, these students may possess greater resilience. The older students under-performed on all academic measures yet report some of the lowest stress levels of anyone until they approach the residency match (table 8). Studies on older medical students indicate that they are more likely to hold an internal locus of control, demonstrate greater critical thinking abilities, and have an increased propensity for self-reflection [47, 48]. Having led full adult lives, with prior careers and other life experience, it is possible the older students approach the burden of medical school differently and may be more accepting of their personal limitations. Similarly, studies have shown that racial and ethnic minority groups may be less likely to report stress than their White counterparts [49]. Dr. Assari posits that having faced more adversity leads to greater resilience, and thus an increased threshold for stress tolerance [49]. If these students have lived experiences overcoming greater adversity, then they may be less inclined to perceive, or report, increases in stress.
Implications
Expanding the sociodemographic diversity among physicians, and by extension, medical students, has long been recognized as an important avenue to address healthcare inequities for marginalized populations in the US [50]. In 2009, the Liaison Committee on Medical Education (LCME) began implementing accreditation standards for the structure and function of medical schools regarding the benefits of diversity [51, 52]. These standards have evolved over the years and now also include policies on anti-discrimination, cultural competency, and addressing disparities in social determinants of health [53]. Today, considerable effort is put towards increasing the matriculation of students from nontraditional sociodemographic backgrounds. Medical schools have universally adopted holistic admissions policies that consider socioeconomic status, demographic characteristics, and life experiences of applicants to encourage the matriculation of nontraditional students [24, 54-56]. Pipeline programs have been implemented in some areas to recruit students to medicine from community colleges, where many nontraditional students begin their post-secondary education [22-24, 41, 57, 58].
Despite these efforts, however, current evidence indicates that medical education continues to greatly favor young applicants from wealthy families, and that medicine has yet to meet the needs of the population. A 2017 analysis of socioeconomic diversity among US medical students found that roughly a quarter of medical students report parental income in the top 5% of all US households (greater than $225,251), and over three-quarters were from households with annual income greater than $74,870 [28]. In 2022, 82.9% of incoming medical students were 25 years of age or younger, and only 5.7% were 28 years of age or older [32].With regards to racial and ethnic diversity among physicians, the majority of physicians currently practicing in the US identify as White (56.2%), as do the majority of recent medical school graduates (61.2%) [59, 60]. Finally, and perhaps most importantly, disparities in the availability and quality of healthcare resources, burden of cost, health insurance coverage, patient outcomes, general health status, and overall life expectancy continue to exist for racial/ethnic minority, low-income, and inner city, and rural communities [61]. Together, this implies that there is a “leak in the pipeline” for students from nontraditional sociodemographic backgrounds.
Limitations
Our study has several limitations. Our study population is small and involves students from only one institution, which limits generalizability. Additionally, statistical analysis of differences in Step 2 CK performance in the “age 30+” group violated the assumption of equal variances, which remains unexplained by our data set. Results from the PSS-4 only capture stress at a particular moment in time and tell us nothing about the quality of stress. Employing qualitative methods could elucidate sources of stress and factors undermining academic performance. While exam performance is a critical component of a student’s competitiveness for residency programs, it is closely followed by narrative evaluations. To fully understand the experience of nontraditional medical students, exploring the outcomes of these evaluations is necessary. We did not include gender-based subgroups in our analyses. Finally, the next logical step in this research would be to compare professional outcomes between traditional and nontraditional students.