It has been recognised that bullying and sexual harassment have negative consequences for students on an individual level, as well as for the universities with increased risk of students dropping out of their education.³¹ A Swedish study investigating sexual harassment and sexual violence found associations between increased risk of depression and anxiety for female students,³² and two reviews covering consequences of bullying and sexual harassment of nursing students show an increased risk of mental health problems, including burnout symptoms,^{6, 33} For medical students, sexual harassment¹⁸ or other mistreatment³⁴ has been found to be associated with increased rates of feelings of burnout. For instance, medical students who said that they had been sexual harassed by patients reported higher levels of feelings of burnout (21% for women and 5% for men) compared to students who did not report sexual harassment.¹⁸According to a systematic review focused on higher education among female students,³¹ university dropout, was identified as a consequence of sexual assault.

Until now, there has been a lack of national data covering Swedish medical and nursing students’ vulnerability to bullying, sexual harassment, stress and burnout, meaning that there is a knowledge gap regarding students within the Swedish context. The aim of this study was therefore to examine the prevalence of bullying and sexual harassment of nursing and medical students and to investigate the relationship between these risks and perceived stress, burnout symptoms and intention to drop out of education.

This study is part of a collaboration programme between four universities in Sweden (Karolinska Institutet, Royal Institute of Technology, Malmö University and University of Gothenburg). The questionnaire was developed by researchers from these universities in collaboration with Statistics Sweden. The questionnaire mainly consists of validated scales. Survey data on students, PhD students, and employees from 38 universities that are part of the Association of Swedish Higher Education Institutions (SUHF) were collected by Statistics Sweden between March-July 2021. Participants were randomly selected and extracted from the Swedish higher education register.³⁵ The survey was pilot tested with a subgroup of the target population, confirming that the questions were clear and suitable. Cronbach’s alpha showed high reliability in both the pilot and the main study. As no changes were needed, the final questionnaire was used in its original form. The study was approved by the Swedish Ethical Review Authority and with adherence to the Declaration of Helsinki’s ethical guidelines. The web-based questionnaire was distributed by Statistic Sweden including information that it was voluntary to participate, and that it was possible to withdraw from participation at any time. Informed consent was given by participants by sending in the questionnaire. Statistic Sweden processed and depersonalised the data before delivering it to Karolinska Institutet for analysis which guaranteed anonymity for the participants. A total of 372 396 students were eligible for inclusion, of which 75 927 students were randomly selected to participate. An invitation letter was sent by post to the selected participants, which provided information about the study and login details to the online survey. Three reminder letters were sent to non-responders during the data collection period.A total of 18 582 participants filled in the questionnaire and the response rate of 25.2 percent.

In the current study, a subgroup of the data from the main study was used, consisting of individuals who at time of data collection were “students in the nursing or the medical programme”. Descriptives are presented in Table 1. Data consist of 431 medical students (men 162/women 269) from all seven medical faculties in Sweden that provide medical education as well as 1083 nursing students (men 157/women 926) from 25 universities or university colleges.

All included variables are validated scales from the Copenhagen Psychosocial Questionnaire (COPSOQ).³⁶ The scales have been extensively validated in different work-settings and occupational groups in Sweden and elsewhere.

The first outcome variable measured burnout symptoms. Participants were asked: "During the last 4 weeks, how often have you": (1) been emotionally exhausted, (2) been physically exhausted, or (3) felt worn out.  Response alternatives were: (1) All the time, (2) A large part of the time, (3) Part of the time, (4) A small part of the time, and (5) Not at all.  An index denoting burnout symptoms was created by taking the mean value of the above three items. Reliability statistics of the burnout index, Cronbach’s alpha, was .86, indicating good reliability.

The second outcome variable measured stress. Participants were asked: “During the last 4 weeks, how often have you": (1) found it difficult to relax, (2) been irritable, (3) been tense. Response alternatives were: (1) All the time, (2) A large part of the time, (3) Part of the time, (4) A small part of the time, and (5) Not at all.  An index describing stress symptoms was created by taking the mean value of these three items. Reliability statistics of the stress index, Cronbach’s alpha, was .82, indicating good reliability.

The third outcome variable assessed intention to quit the studies. Participants were asked: “Have you considered resigning from your current studies?”. Response alternatives were: (1) Never, (2) A few times, (3) Sometimes, (4) Often, and (5) Many times. 

Workplace bullying was measured with one question, “Have you been exposed to bullying at your place of study during the last 12 months?”, with the accompanying explanation: “Bullying is when a person is repeatedly exposed to unpleasant or degrading treatment and finds it difficult to defend themselves”. Response alternatives were: (1) Yes, daily, (2) Yes, every week, (3) Yes, every month, (4) Yes, a few times, and (5) No.  Sexual harassment was measured with one question: “During the last 12 months, have you been exposed to undesired sexual attention at your workplace”. Response alternatives were: (1) Yes, daily, (2) Yes, every week, (3) Yes, every month, (4) Yes, a few times, and (5) No. Questions regarding the relationship to the aggressor were measured with two questions regarding the aggressor’s gender and position with the response alternatives teacher/supervisor, student, patient or other.

Bullying and sexual harassment variables were dichotomised. The response alternatives 1 to 4 were recoded into ‘Yes’ and 5 into ‘No’ as recommended in COPSOQ³⁶. The response alternatives for stress and burnout were transformed to 0 to 100, as recommended in COPSOQ³⁶: (1) ‘All the time’ was transformed to 100, (2) ‘A large part of the time’ was transformed to 75, (3) ‘Part of the time’ was transformed to 50 (4), ‘A small part of the time’ was transformed to 25, and (5) ‘Not at all’ was transformed to 0.

A series of statistical analyses were performed using version 27.0 of the IBM SPSS program. First, descriptive statistical analyses were performed, including means and standard deviations for continuous variables, percentages for categorical variables. Cross-tabulations were performed to demonstrate how the variables differ across various subgroups, such as for e.g. gender. Two-sample t-tests were performed to compare stress levels, burnout symptoms and intention to quit in the bullied vs. non-bullied groups of students as well as the harassed vs. non-harassed groups of students. Results were considered statistically significant at p<.05.

Participant demographic characteristics in medical and nursing programmes as well as reported bullying/sexual harassment are presented in Table 1.

As can be observed, around 10% of women in the medical programme were subjected to bullying compared to around 7% of men. With regards to sexual harassment, the proportion of women subjected was around 13% while the proportion of men sexually harassed was 6%. The highest prevalence of both bullying and sexual harassment was reported by female students in medical education (10.4/13.4%). For male students, the prevalence was highest in nursing education with 8.9 and 10.3% for bullying sexual harassment, respectively. We also conducted additional analyses to compare these results with the total cohort of students from the national data collection of which this study was a part. Among students in all higher education programmes, except for those in nursing- and medical programmes, women reported 9.1% for bullying and 4.5% for sexual harassment and the equivalent numbers for men were 6.0% for bullying and 2.3% for sexual harassment demonstrating that the prevalence is higher in medical- and nursing programmes compared to other higher education programmes.

Overall, the bullied students reported higher levels of stress and burnout symptoms, as well as expressed a stronger intention to quit studies.

Among the female students, t-test results indicated that those subjected to bullying reported statistically significantly higher levels of stress symptoms t₍₁₁₈₈₎=4.91, p<.001, mean difference 11.47, 95% CI[6.88, 16.05], burnout symptoms t₍₁₁₈₈₎=5.83, p<.001, mean difference 15.07, 95% CI[10.00, 20.14], as well as expressed a stronger intention to quit studies t₍₁₁₈₆₎=4.30, p<.001, mean difference 0.46, 95% CI [0.25, 0.68].  Among the male students, t-test results indicated that those subjected to bullying reported statistically significantly higher levels of stress compared to non-bullied ones t₍₃₁₇₎=3.15, p=.002, mean difference 14.25, 95% CI[5.36, 23.14]. Male bullied students also reported statistically significant burnout symptoms t₍₃₁₇₎=3.49, p <.001, mean difference 17.26, 95 CI[7.52, 26.99]. Finally, male bullied students expressed a stronger intention to quit studies: t₍₃₁₆₎=3.67, p<.001, mean difference 0.76, 95% CI[0.35, 1.17].

Women in the medial programme reported higher levels of stress and burnout than men, as well as somewhat stronger intention to drop out of their education. Bullying in the nursing programme was reported by nearly equal proportions of women (about 8%) and men (nearly 9%), while the share of women subjected to sexual harassment was around 7% compared to about 10% of men. Overall, women in the nursing programme reported higher levels of stress and burnout, together with a weaker intention to quit studies, compared to men.

Data collection was conducted during the Covid-19 pandemic, which may have impacted the outcome of sexual harassment, bullying as well as mental health data. Caution should therefore be taken when comparing the results with other studies. Even if education during the pandemic became, to a large extent, digital, students from the nursing and medical programmes were not greatly affected since they still had clinical rotations in health care settings. In addition, the Swedish model, with no regular lockdown, meant that during the Covid-19 pandemic Swedish medical- and nursing education was subject to fewer changes in daily life compared with students from other settings.⁴²

In this study, we used self-reported data, with single direct questions on bullying and sexual harassment. To increase the validity and comparability, we only used scales validated in a Swedish context.³⁶ Still, single direct questions are known to give lower prevalence numbers than questionnaires with multiple questions to measure a phenomenon.⁴³ In addition a cross-sectional study does not allow conclusions about causality or the direction of effects, we therefore strongly recommend longitudinal studies in the future to be able to establish causality between sexual harassment, burnout, and dropout intentions, overcoming the limitations of cross-sectional design.

To reduce the risk of bullying and sexual harassment, several steps are needed. First, these behaviours should be recognised and addressed even if they are not severe because when bullying and sexual harassment can pass under the radar there is a risk that these inappropriate behaviours are seen as an inherent part of the educational milieu and become normalised.⁴⁴ To create a positive learning environment, students need to feel safe on campus and in clinical settings.

Second, universities and clinical settings need to collaborate regarding the prevention of bullying and sexual harassment, this mean that both university staff and clinical staff need to have knowledge regarding bullying and sexual harassment. To train the staff is one way to make them prepared to take action when needed. A large proportion of the education takes place in clinical settings, and sometimes the perpetrators are patient or relatives. This means that clinical supervisors need to have knowledge of how to prevent harassment during clinical practice. When patients or relatives pass the line of appropriate behaviour, a supervisor should be able to support the student and react to inappropriate behaviour. Since they are role models, their response to different forms of mistreatment will be part of the student’s professional socialisation. In a qualitative study where medical students were asked to bring solutions to the problem of sexual harassment in medical education, they emphasised  the need for staff training.⁴⁴ Incorporating a gender perspective is crucial in preventing bullying and sexual harassment, as it helps to better understand how gender influences the risk of these behaviours and its consequences for both men and women.

Third, Russel and colleagues conclude that reporting systems must be easy to follow and they pinpoint that it is crucial to be able to report harassment behaviours anonymously ⁴⁵ as victims often fear that reporting can have negative consequences for them especially in medical settings.

Fourth, universities must inform their students about how to and the importance of reporting bullying and sexual harassment when it occurs, and secure that no reprisals will be made to those who disclose. In a previous study among university staff, we found that social support from managers buffered negative effects (burn out symptoms) in relation to bullying.⁴⁶ In accordance to those results, universities could develop guidelines in order to support students that have encountered sexual harassment or bullying. Universities could include seminars in the curriculum to increase knowledge for students concerning what is considered acceptable behaviour from staff, patients and relatives and how to report problems. Increased levels of drop out for students and intention to leave health care professions are serious issues, especially in light of the current situation, where the WHO has estimated a shortfall of 10 million health workers by 2030 globally.⁴⁷

The authors of the manuscript would like to thank the participants.

Funding: Forskningsrådet om Hälsa, Arbetsliv och Välfärd, STY-2021/0001

The authors declare that there is no conflict of interest.