Medical students experience mistreatment, with a focus on gender discrimination. Cross-sectional study at one Swedish medical school

Abstract The aim was to study discrimination, with a focus on gender and sexual harassment among medical students at Uppsala University in Sweden. A survey was sent via email to all registered medical students in the spring semester of 2020. Data were compared with two previous studies. Questions about gender and sexual harassment were the same as in the study conducted in 2002 and 2013. In addition, the 2020 survey included a question about other grounds of discrimination. Forty percent, that is, 453 out of 1,130 medical students, participated. The proportion of students reporting gender-based discrimination during the preclinical semesters was similar for females and males. During the clinical semesters, significantly more females than males reported gender discrimination (41% vs. 21.5%, p = 0.005) and sexual harassment (22% vs. 5%, p = 0.001). Physicians were the most commonly reported perpetrator. Reports about not being respected had increased from 2% to 20% between 2013 and 2020 among female clinical students. The prevalence of those who experienced several sexually harassing behaviors increased for the female and male clinical students and the female preclinical students. Receiving an unwelcome touch increased from 1% to 7% for the female clinical students. Discrimination due to ethnicity was reported by 36% of the students born in a country other than Sweden compared to 3% of those born in Sweden. Our findings confirm that experiences of different forms of discrimination exist in this medical school, and females and minorities are particularly affected.


Introduction
Experiences of discrimination among medical students remain common, despite systematic work on equal rights and opportunities in academic education worldwide (Sheehan et al., 1990;UNESCO, 2021).Most studies on discrimination at medical schools report gender mistreatment, ranging from abusive behaviors and gender harassment to sexual harassment (Jendretzky et al., 2020;McClain et al., 2021;Ng et al., 2020).In the latest meta-analysis (2011), 59.4% of 38,353 medical students, mainly female, had experienced gender discrimination at least once during their training (Fnais et al., 2014).Moreover, medical students may also be mistreated due to their ethnicity/race and sexual orientation (Hill et al., 2020).Discrimination can affect confidence, participation and performance, and the selection of future careers (Kristoffersson et al., 2018).It can cause adverse mental health outcomes, including burnout, substance use, depression, and alcohol-related disorders (Cook et al., 2014;Goto et al., 2013).In addition, previously mistreated medical doctors are more likely to perpetuate discrimination against medical students in the future (Ng et al., 2020).Fear of a negative impact on grades and career opportunities may discourage peers from reporting discrimination (Grinshteyn et al., 2022).
We investigated gender discrimination of medical students at Uppsala University in Sweden in two previous studies in 2002 and 2013.Specifically, we found that gender discrimination was experienced in particular during clinical clerkship training, and that the prevalence had decreased in men but increased in women.However, in the same studies, a decrease was seen in the prevalence of sexual harassment (Kisiel et al., 2020).A study from another group found that medical students from racial/ethnical minorities were more likely to encounter discriminatory treatment from supervisors, peers, and patients (Kristoffersson et al., 2021).
Since our last follow-up study (2013) on gender-based mistreatment, a global movement against gender discrimination has occurred.In 2017, the #MeToo movement raised the importance of awareness of gender discrimination in the general society in many countries (Choo et al., 2019;Kaiser, 2018).Furthermore, a recent qualitative Swedish study on medical students and a report from the Swedish Association of Medical Students also highlighted an urgent need for an increased awareness of other forms of mistreatment, especially regarding ethnicity and race (Kristoffersson et al., 2021;Ström, 2021).
Although there are several reports on mistreatment of medical students, few studies have monitored gender discrimination over time, in addition to other forms of inequality.Our current study aimed to close those research gaps and evaluate whether medical students experience less or more gender discrimination over time.We also aimed to determine whether medical students encounter other forms of inequality, such as discrimination due to age, ethnicity, religion, functional disability, sexual orientation, and sexual identity.

The study designs
A cross-sectional study was conducted at the Medical School of Uppsala University.All registered medical students at the university medical program in the spring semester of 2020 were invited to participate in a survey on all forms of discrimination.There were no exclusion criteria.In May 2020, all medical students received an email linked to the anonymous survey system, the Research Electronic Data Capture (REDCap) (Harris et al., 2009).The students received three monthly reminders until the survey closed in August 2020.The part of the survey about gender discrimination was a follow-up of previously conducted questionnaire studies in 2002 and 2013 (Kisiel et al., 2020).Additionally, the 2020 survey included questions on other inequalities.
All participants were requested to complete an online written informed consent form before initiating the survey.No individual identifiers were recorded to ensure students' anonymity.The study was approved by the Ethical Review Board in Uppsala No. 2020/00311.The study was carried out in accordance with the Declaration of Helsinki.

Study population
The survey was sent to the university email accounts of 1,130 registered medical students.As in the study in 2002 and 2013, those who consented to participate in the survey were stratified into gender (female/male) and preclinical and clinical semesters.The Swedish medical school comprises 330 high education credits (five and a half years of study) that lead to a medical degree.The medical education at Uppsala University is divided into preclinical semesters (one to four), where peers have only limited contact with the clinical environment.In the clinical semesters (five to nine and eleven), the students participate in several clinical internship periods per semester in general hospitals and primary care centers.The 10 th semester is devoted to a scientific project.

The survey
The same survey questions on specific examples of behaviors representing gender discrimination and sexual harassment were used in the surveys conducted in 2002 and 2013, respectively (Kisiel et al., 2020).The questions were formulated to assess the personal experience of different types of gender discrimination, and the student could indicate one or several types (Diskrimineringsombudsmannen, 2022).In accordance with the Swedish Discrimination Ombudsman, gender discrimination includes gender harassment (an act that violates someone's gender dignity) and sexual harassment (an act of a sexual nature that ranges from the subtle offence of someone by words, comments through unwanted sexual attention or touch to sexual assault) (DO, T.E.O., 2022).The medical students were asked to report the perpetrator and where the sexual harassment occurred in the medical school.Perpetrators included other medical students, supervisors/teachers (at the preclinical moments), medical doctors or nurses, and other healthcare personnel categories, such as psychologists, occupational therapists, physical therapists, biomedical analytics, or patients (at the clinical moments).A student could indicate one or more perpetrators.
Moreover, the 2020 survey included questions on other grounds of discrimination, transgender identity, ethnicity, religion, disability, and sexual orientation.Information about age groups, gender, country of birth, semester of studies, and sexual orientation was also collected.In this study, we asked about legal sex, defined as the biological sex listed in the Swedish population register (female/male) (Socialstyrelsen, 2020).We also asked about self-reported gender as women/men/ non-binary (Broughton et al., 2017).We considered the country of birth as a proxy of ethnicity.Ethnicity is a broad and complex term and can be defined as a group with a shared origin and culture, tradition, identity, language or religion, a link to a particular geographic area (Wikipedia, 2023).In many countries, including Sweden, the country of birth is used as the indicator of ethnicity, for example, in population registers and research on the labor market and health (Careja & Bevelander, 2018;Gibson-Helm et al., 2014).The advantages of using country of birth as ethnicity are that it is an objective, reliable, stable character and allows for uniform identification of groups that can be compared over time and between the studies (Stronks et al., 2009).All survey questions used are presented in the Appendix, supplementary materials.

Statistics
The students were divided into study stages (preclinical and clinical) and sex (female and male).In those groups, categorical variables were reported as frequencies with proportions and continuous variables as means with standard deviation.The difference between males and females in the preclinical and clinical semester and between the underrepresented minorities and no minorities was assessed by Chi-2-test.Furthermore, we calculated the proportion of students who reported experiences of different discriminating behaviors and their frequency.In this analysis, as in the previous studies from 2002 and 2013, medical students from the 10 th semester were excluded as the students were not doing clinical practice during that semester.The differences between 2002-2013, 2013-2020 and 2002-2020 were assessed by a risk ratio (relative risk) of mean values with 95% confidence intervals (95% CI) (Wikipedia, 2022).Risk ration was defined as: RR = {n11/(n10 + n11)}/ {n01/(n00 + n01)}, based on a 2 × 2 table : Mean slider scale scores were calculated to summarize the frequency of these behaviors: never (0), one (1), or several times (2).Moreover, the proportion of students reporting different perpetrators was shown.A p-value of less than 0.05 was considered significant.R version 4.0.2 was used to perform all the statistical analyses.

Basic characteristics
There were 453 students who completed the 2020 survey (response rate 40.1%) (Supplementary Table S1).The respondents' characteristics are shown in Table 1.Significantly more females than males in the preclinical and clinical semesters responded to the survey.Four students reported being nonbinary or other gender (1%), and those students indicated their legal sex as female.All the other students that reported their legal sex as female reported their gender as women.All students reporting their legal sex as male reported their gender as men.We divided the students in accordance with the semester of study and legal sex.We used sex and not gender since the group non-binary or other gender was small, and we did not want to exclude any students who completed the survey from the study.The ratio between the female and male students was slightly skewed in the statistical sample (66% female/34% male in the survey and 60% female/40% male among registered students).

Discrimination due to gender
The proportion of students who reported an experience of gender discrimination, gender harassment, and sexual harassment was significantly higher in females than in males in the clinical semesters (Figure 1).In the preclinical semesters, 15% of the females and 19% of the males reported that they had observed other students being discriminated against.The corresponding prevalence among students in the clinical semesters was 31% of the females and 23% of the males.In addition, similar proportions of females and males in the preclinical semester reported hearing about other students being mistreated (46% and 42%, respectively).Significantly more females (76%) than males (59%) during the clinical semester had heard of gender discrimination against other students (p = 0.021).
Being favored due to gender was reported by 4% of the females and 6% of the males in the preclinical and 21% of the females and 33% of the males in the clinical semesters (p = 0.011).Observing other students being favored due to gender was reported by 22% of the females and 15% of the males from the preclinical and 43% of the females and 21% of the males in the clinical stage (p = 0.002).

Perpetrators of sexual harassment
The most frequently indicated perpetrators of sexual harassment during the preclinical semester were nurses and/or other healthcare personnel categories, and the most common scene where the harassment occurred was at lectures.The clinical students reported that medical doctors were the most common perpetrators (Table 2).

Gender discrimination and sexual harassment over time
Experience of gender harassing behaviors, such as "experiencing stereotypic statements," 'being ignored,"and "being not respected," had increased over time for females in the preclinical and clinical settings and males in the preclinical setting (Figure 2 and Suppl Table 2).The prevalence of those who experienced being ignored and not being respected has increased from 1% to 27% and from 2% to 20% between 2013 and 2020 among female students in the clinical setting.Male students in the clinical setting experienced more discrimination than male students in the preclinical setting across the study years 2002-2020.The most frequently reported discrimination by males was stereotypical statements (Figure 2).
The prevalence of sexually harassing behaviors, including "unpleasant comments on private life as well as clothing and appearance" and "receiving unwelcomed touch," had increased in female and male students in the clinical setting and female students in the preclinical setting.In contrast, the prevalence of these behaviors had decreased over time for males in the preclinical setting.The prevalence of "receiving unwelcomed touch" had increased from 1% to 7% between 2013 and 2020 for female students in the clinical setting.Receiving requests for a sexual relationship at least twice increased from 0 to 3% for male students in the clinical setting between 2013 and 2020."Encountering sexually offensive jokes and comments" decreased for all the students over time.

Discrimination due to other grounds
Discrimination due to age, religion, function disability, and sexual orientation was reported almost exclusively by minorities, whereas the observation of these behaviors was reported by nonminorities (Table 3).Discrimination due to age, religion, function disability, or sexual orientation was similarly reported by females and males during the clinical and preclinical semesters (Table 3).Of the peers who reported not being born in Sweden, 36% reported discrimination due to ethnicity.In the survey, 48% of those with functional disabilities reported mistreatment due to functional disability.In addition, we found that discrimination due to sexual orientation was reported significantly more often by homo-and bisexual than heterosexual students (Table 3).

Discussion
The main result of this study was that during the clinical semesters, significantly more females than males experienced gender discrimination.The proportion of students reporting gender inequality in preclinical semesters was lower, with no significant difference between females and males.Physicians were the most commonly reported perpetrators of sexual harassment, and clinical settings were the most common place where these behaviors were experienced.Our study also showed an increased prevalence of gender-discriminating and harassing behaviors in females and males in the preclinical settings.Alarmingly, sexual harassment, such as "pointed out as being of sexual interest," "received intrusive touch," or "receiving a request for a sexual relationship," has increased among students in clinical settings over time.In addition, a substantial number of medical students from minority backgrounds reported mistreatment due to other grounds of discrimination.Discrimination against students born in a country other than Sweden and those with functional disabilities was particularly high.
In our study, discrimination due to gender was most frequently reported, followed by mistreatment due to country of birth, age, religion, sexual orientation, and functional disability.This is congruent with a recently published Swedish study on discrimination in final-year medical students in seven medical schools (Kristoffersson et al., 2018) and several international studies addressing different mistreatment grounds (Fnais et al., 2014;Hill et al., 2020;Makowska & Wylezalek, 2021;Najjar et al., 2022).
In line with several previous studies conducted before and after the #Metoo movement, we demonstrated that female medical students experienced gender discrimination (Hill et al., 2020;Najjar et al., 2022), including sexual harassment (Jendretzky et al., 2020;Schoenefeld et al., 2021), more often than males.However, a direct comparison of the proportion of affected medical students is difficult because of the diversity in study design.In a previous study from the U.S., based on 27,504 surveys of graduating medical students, a significantly higher percentage of females than males reported facing discrimination due to gender (Hill et al., 2020).Another study from Germany, based on an online survey, reported that 59% of medical students experienced sexual harassment (Schoenefeld et al., 2021).In a Swiss study, it was shown that 22.5% of the female and 9.8% of the male medical students reported having been victims of gender discrimination and sexism (Najjar et al., 2022).In contrast, a comparable percentage of both genders reported having witnessed discrimination due to gender (Najjar et al., 2022).Consistent with this result, our study found no significant differences in the proportion of females and males who reported having observed other students being discriminated against.Interestingly, significantly more women than men in the clinical semesters reported hearing about other students being discriminated against.One explanation could be that women discuss gender issues with other female colleagues more often than men.
We also found that a higher proportion of female students than males observed other students being favored.Consistently with this, we also found that more male students experienced being favored compared to females, which was significant during the clinical semesters.Congruent with our result, a previous study showed that males more often received funding, scholarship, and grants that finance their medical school or MD/PhD education (Snyder et al., 2021).In the Swiss study, female students expressed the need to receive support in their professional career more often than their male colleagues (Najjar et al., 2022).The reason why males were more favored might be that they were much more likely than female colleagues to express interest in leadership or/and academic/scientific career on the side of their clinical career (Laurent et al., 2017;Snyder et al., 2021).As a result, males have been overrepresented in leadership and academic roles in academic medicine (Laurent et al., 2017).The reason for this inequality included stereotypes and societal pressure on women to fulfill the role of primary caregiver for the family and children, in addition to gender discrimination.In contrast, males tend to opt for higher income and social prestige (Edmunds et al., 2016).
Consistent with previous studies (Schoenefeld et al., 2021), we saw that significantly more female than male students experienced discrimination during clinical clerkships compared (Frank et al., 2006;Fried et al., 2012) to other parts of training.The students in the clinical settings pointed out that physicians were the most common perpetrators.Similar results were shown in our previous studies conducted in 2002and 2013(Kisiel et al., 2020) ) and in reports from several other countries (Colenbrander et al., 2020;Mavis et al., 2014;Roling et al., 2020).The reason why physicians might be perceived as behaving inappropriately while supervising students has been explained by complex organizational psychology as "see one, do one, teach one" (Rees & Monrouxe, 2011).As many medical doctors have experienced discrimination, similar expressions might form their collegial expression and hidden curriculum (Billings et al., 2011).This might lead to cynicism and also the use of inappropriate teaching methods while supervising students.The reason women are more affected by gender discrimination during their clerkship has been broadly addressed and claimed to be a combination of hierarchical and dependent relationships between students and supervisors in an often maledominated and stressful environment (Dzau & Johnson, 2018).
Since the 1970s, Sweden has continuously worked to provide equal rights and opportunities in the labor market and education system (Carlson, 2021).This has resulted in a universal act called the Discrimination Act (SFS 2008:567), which took legal effect in 2009 (Carlson, 2021).This act forbids unlawful mistreatment based on all grounds of discrimination.Since 2015, Uppsala University has followed an action plan based on the Discrimination Act and the Action Plan of Equal Opportunity (UFV 2015/766) (Uppsala University, A.b.t.V.-C, 2017).One of the aims was to monitor equal opportunities by the newly formed research groups (Univeristy, 2023) and another to recruit and appoint 48 percent female professors (Uppsala University, A.b.t.V.-C, 2017).In the medical program, these aims are communicated and reflected through a mandatory course on gender medicine for all students during their final semester of medical studies.The course goals are to educate students on gender and other grounds of mistreatment in the context of their forthcoming profession.
We found that several forms of gender discrimination and harassment of females during the clinical semesters increased over time.In addition, sexual mistreatment has increased among students in clinical settings.An explanation for why the reported prevalence of gender discrimination has increased between 2013 to 2020 might be due to a greater awareness as a result of interventions from the university and the #Metoo movement.An increased reporting of these experiences after #Metoo was also found in the survey on working environments conducted in Sweden among randomly selected women employed in academic settings.In this report, there were 400% more reported cases of sexual harassment in the year after #Metoo (2018) compared to the year before (2017) (SCB, 2019).
There have been a few longitudinal studies on mistreatment among medical students.Fried et al. examined discrimination among American medical students after third-year clerkships over 13 years between 1996 and 2008.This study used comparable definitions of sexual harassment to our studies.They showed that these experiences remained at the same level throughout the study period and that females were significantly more affected than men (Fried et al., 2012).
More than one-third of the students born in a country other than Sweden reported experiencing discrimination due to ethnicity, and there was no difference between females and males.Our study used the country of birth as a proxy for ethnicity.A recent national report by the Swedish Medical Association indicated that 5% of graduated students reported being mistreated due to ethnicity (Läkarförbund, 2020).A recent study from the U.S. found that discrimination due to race/ ethnicity was significantly more commonly experienced among females than males (Hill et al., 2020).Interestingly, another study highlighted that ethnic mistreatment was the form of discrimination that medical students were least likely to report formally (Fried et al., 2012).
In our study, we found that discrimination due to higher age at the medical school was reported by between 14 to 23% of the students.The reason for that might be that traditional medical school applicants in Sweden are younger and graduate at 25-30 years of age and become residents in their early 30s (Socialstyrelse, 2021).Older students might be discriminated against because becoming a medical doctor takes many years of education and further professional development (L, 2017).On the other hand, a previous study showed that older females had the best academic performance at medical school, followed by younger males and then younger females (Haist et al., 2000).
We found that some students were discriminated against due to religion, with no differences between females and males.In line with our study, 6.5% of medical students experienced discrimination due to religion in the recent Swedish study (Kristoffersson et al., 2018).Another study found that 20% of medical doctors from other countries than Sweden perceived they had been disfavored due to another religion or belief.In that study, Muslims, in particular, were accused of having a negative view of women (Sturesson et al., 2019).Religion might be closely connected to a person's cultural background and ethnicity.In a previous review study, religious freedom was related to medical students' well-being and helping them to deal with stress (Ray & Wyatt, 2018).Medical doctors identifying with a religion agreed that faith influences their medical practice positively (Ray & Wyatt, 2018).
Moreover, in our study, a few students (n = 25) reported functional disability, such as dyslexia.Half of them reported being discriminated against due to their disability.Only a few previous studies have addressed learning disabilities among medical students.Students with functional disabilities usually receive lower grades, which may lead to lower self-esteem, stress, and anxiety (Majumder et al., 2010).Previous studies have shown that most healthcare teachers were unaware of the student's learning disabilities and did not know how to support these students, whose needs might be ignored, even unwelcomed, or not understood (Majumder et al., 2010).
Discrimination due to sexual orientation was also assessed in the present study.For example, we showed that LGBT+ students experienced discrimination due to sexual orientation significantly more than heterosexual students (8% vs. 2%, p < 0.001).In the study from the U.S., there was a much higher proportion of students who reported mistreatment due to sexual orientation (43% in LGBT+ and 23.6% in heterosexual, p < 0.001).Females reported significantly higher discrimination because of sexual orientation than their male counterparts (Hill et al., 2020).

Strengths and limitations
The primary strength of this study was that we evaluated discrimination due to gender and sexual harassment directly compared to data from 2002 and 2013 covering 18 years.We also addressed the frequency of discriminating behaviors and who was the main perpetrator.In addition, we assessed several other grounds of mistreatment.
The limitation of the study, in addition to the known constraints with a cross-sectional design, was the low response frequency that poses the same methodological challenges faced in several similar studies (Jendretzky et al., 2020;McClain et al., 2021;Schoenefeld et al., 2021).A potential selection bias can be caused by the lack of time and interest in the study participants and recruitment procedures (Rothman & Lash TL, 2008).The previous studies found that people are much more willing to participate in a survey that has a personal significance (Galea & Tracy, 2007).Therefore, those who did not experience discrimination can be less prone to answering the questionnaire.We do not know whether all the students who were sent an online invitation received it in their emails.The disadvantage was that the study coincided with the start of the COVID-19 pandemic.Thus, we could not distribute the survey during lectures as we initially had planned to do, and as we did in the previous studies, 2002 and 2013, where the response frequency was higher.Therefore, the generalizability of our current results should be considered with caution.In this study, we also could not determine in which department or clinical setting the mistreatment occurred.Another potential weakness of our study was that one medical student could indicate several types of gender discrimination, several types of discrimination, and more than one perpetrator.

Importance and implications
Since discrimination seems to be common in medical school, those responsible for medical education must be made aware of the problem.Discrimination has negative consequences on the student's academic motivation and can lead to poor self-esteem and negatively affect one's physical and mental health (Oancia et al., 2000).Encountering discrimination can lead to lower job satisfaction and productivity and a decline in empathy toward patients (Samuels et al., 2021).This, in turn, may lead to the inability to care for patients, avoidance of certain specialties, or those particularly affected might quit their job, reducing workforce diversity in healthcare (Thakkar et al., 2020).
The burden of mistreatment in the present and several other studies (Fnais et al., 2014;Hill et al., 2020;Najjar et al., 2022) should motivate those responsible for academic education to evaluate the existing strategies (Mavis et al., 2014).We suggest that the mandatory course regarding gender medicine which medical students attend in their last semester could be moved to the beginning of the clinical part of medical students' training.This would improve students' understanding of which behaviors represent discriminatory practices and how to recognize, prevent, and respond to them before they enter their clinical clerkships (Mavis et al., 2014).We also think that medical schools should improve the anonymous reporting system.The academic institution should be responsible for educating their faculty and staff in teaching medical students regarding discrimination (Oser et al., 2014).Moreover, hospital staff in teaching hospitals need education about discrimination.
A future direction might be to perform further follow-up studies among medical students at Uppsala University.We might also determine which specialty the students perceive there to be the highest risk of mistreatment.In addition, a third part could be to invite health professionals to answer questions about their attitudes toward students and their experiences of discrimination.This might trigger reflection and understanding of inappropriate behaviors.

Conclusion
Our study showed that self-perceived gender discrimination is common and has increased among female students and students enrolled in the clinical semesters over the 18 years of follow-up.We also found that many medical students experienced other forms of discrimination due to ethnicity, age, religion, sexual orientation, and functional disability.Both improving awareness of discrimination and its eradication continue to be a challenge for the medical program.

Questions
Possible answers • being ignored • not being respected • being ridiculed • being denied rightful help/services • stereotypical statements • condescending comments • unwelcome comments on clothes/appearance • sexually offensive comments • being pointed out as a sexual object • unwelcomed touch • sexually offensive jokes • request for sexual services Perpetrator of sexual harassment in the medical school was . . ..(you can indicate one or several perpetrators) • other students • lecturer/teacher • medical doctor • nurse or other medical personnel

Figure 1 .
Figure 1.The proportion of students who reported discrimination due to gender, gender harassment, or sexual harassment stratified by legal gender and preclinical or clinical semester of studies.The difference between females and males was analyzed with the Chi2 test.

Figure
Figure 2. The proportion of students reporting experiences of different gender and sexually harassing behaviors over time (2002, 2013, 2020).Mean slider scale scores were calculated to summarize the frequency of these behaviors: never (0), one (1), or several times (2).

Table 3
was . . . . . . . . .due to my gender (indicate if you have experienced it once or at least twice at the medical school; you can select one or several types of experiences)