Rees, Charlotte E. PhD, CPsychol; Monrouxe, Lynn V. PhD, CPsychol
Unhappily, medical student abuse is not a new phenomenon. With the publication of Silver's1JAMA article nearly three decades ago, scholars around the world have repeatedly reported the prevalence of medical student abuse. It is now believed to be “institutionalized,”2 with students experiencing a multiplicity of abuses, including verbal abuse, sexual and racial discrimination and harassment, and physical abuse.3 Abuse typically causes emotional upset for students, sometimes long after the event(s), and can result in various negative outcomes such as depressive symptoms, escapist drinking, cynicism about the profession, and low self-confidence.4,5 Despite increasing awareness of and research about medical student abuse since Silver's article, nothing much has changed in terms of the abusive culture of medical education—Medical students are still being abused in the medical workplace. So why has the research on medical student abuse, often published in high-impact journals like JAMA and Academic Medicine, had such little impact on changing this characteristic of medical education?
One problem might be that by using fairly unsophisticated questionnaires, the studies have simply catalogued an already-known problem, thereby providing little information to facilitate cultural change. As suggested by Hinze,6 “estimates of prevalence do not reveal much about the context within which harassment unfolds.” Our study aimed to address this gap by taking a qualitative approach. By listening to students' experiences of abuse, their perceptions about factors contributing to abuse, and their actions at the time of abuse, we believe we can help answer all-important “why” questions essential for cultural change.7
Abuse typically consists of unwanted harmful, injurious, or offensive acts directed at someone by another.8,9 The related concept of bullying has a narrower definition, based on characteristics including frequency, intensity, duration longer than six months, and power disparity.10 Because medical students typically have fleeting relationships with individual health care professionals, their experience of abuse rarely escalates to bullying, so we employ abuse in this report.
Zapf and Einarsen11 differentiate between open and subtle acts contributing to workplace abuse. Such covert abuses include perpetrators' withholding information from recipients, giving recipients unpleasant tasks or unmanageable workloads, and excluding recipients. More direct acts include perpetrators' giving recipients unwanted sexual attention, making insulting remarks, and shouting at, physically intimidating, and threatening recipients with physical violence.11
In the largest study of medical student abuse to date (13,168 students across 125 accredited U.S. schools), 38.3% of students reported being publicly humiliated at least once during medical school, 22.4% of female students reported sexual harassment, and 9.4% of black men and 16.8% of black women reported racial harassment.3 This and other questionnaire surveys on medical student abuse9,12 typically focus on direct negative acts (verbal, sexual, racial, physical) rather than covert abuses commonly described in organizational psychology literature.2,13 Within our qualitative approach for the present study, participants were free to discuss a broader array of abuses (including more subtle acts) that caused them to experience a professionalism dilemma.
What Factors Contribute to Abuse?
Medical student abuse research is typically framed within an individualist perspective: Commentators focus on the personality style and abusive past experiences of the perpetrator.8,14–16 Although some authors have proposed work and organizational factors for medical student abuse, these explanations are mostly grounded in opinion rather than data.8,17 However, research from organizational psychology increasingly focuses on the complex interaction between person and environment, including factors focusing on the individual, on work, and on the organization (see Figure 1).6,7,18–21 Through the qualitative approach used in the study reported here, we were able to examine multiple contributory factors for each case of abuse reported by students.
How Do Recipients Respond to Abuse and Why?
Few medical students resist at the time of abuse by challenging or reporting the perpetrator.3,8,9,15,22,23 Resistance can be defined as “any discursive or nondiscursive act of commission or omission that counters, disrupts, or defies the bully or erodes the bully's material or symbolic base of influence.”18 Resistance strategies to bullying previously outlined in social science literature include quitting, employees' discussing how to stop the abuse, and reporting, avoiding, or confronting the perpetrator.18 Although some medical student abuse research has identified reasons why students do not report abuse,3,22 none has explored how and why students resist in the face of abuse. An exploration of resistance and inaction is essential to combat the tendency of medical students to normalize experiences of abuse by seeing them as routine or a rite of passage.6
Study aims and research questions
The findings reported here are based on data that we gathered as part of a larger qualitative study, carried out between 2007 and 2009, that elicited students' narratives of professionalism dilemmas: events during their learning in which they observed or participated in something they thought was unethical, unprofessional, or wrong.24 Through a preliminary framework analysis of the larger study's data, it was clear that students were constructing abuse experiences as professionalism dilemmas. We therefore initiated the present study to explore students' personal narratives of abuse dilemmas. For this study, we formulated the following research questions after the data from the larger study had been collected and the preliminary framework analysis was complete:
* What types of abuse experiences did the students report?
* What factors did students cite as contributing to the abuse?
* What factors did they cite as contributing to their responses (inaction or resistance) at the time of the abuse?
Because of the sensitivities around this topic, we gave students the choice of participating in individual or group interviews. For the same reason, we have not given the names of the students' medical schools in this report.
In the larger study, we studied students at three medical schools located in different countries (England, Wales, and Australia), representing different entry requirements (undergraduate versus graduate entry) and curricula (problem-based learning versus lecture-based). Following ethics approval from each school, students across all years at each school were invited to participate. A total of 200 students participated in 32 group and 22 individual interviews (see Table 1 for participants' demographic and education-related characteristics).
Students were introduced to the study through announcements during teaching sessions, e-mail invitations, information posted on noticeboards, and snowballing (i.e., students already participating in the study inviting other students to participate). All students were required to read an information sheet and sign a consent form before participating.
All group and individual interviews were audio-recorded. The interviews began with an orienting question: What is your understanding of professionalism? Once students had defined professionalism, they were then asked whether they had ever been in a situation which caused them to experience a “professionalism dilemma.” Narrative interviewing techniques were employed to elicit students' personal incident narratives, and open-ended questions were used to explore the widest range of dilemmas experienced as defined by the students. Students were asked to explain what they did and why within the dilemmas recounted.
The audio recordings were transcribed without identifying the names of the students whose statements had been recorded. Transcripts and audio recordings were first analyzed using framework analysis,25 as mentioned earlier. This first-order thematic analysis conducted by three researchers (C.E.R., L.V.M., and L.R.D.—see Acknowledgments) resulted in a coding framework comprising five higher-order themes relating to what participants said (e.g., definitions of professionalism) and seven relating to how they spoke (e.g., metaphoric talk; see Monrouxe et al26 for further details about the coding framework).
The idea for the further research that resulted in the present report began when we looked at the data in the second theme, “Medical students' dilemma situations.” Here, 833 narratives were coded using Atlas-Ti. Included in this main theme was a subtheme about student abuse, comprising 86 narratives. One of us (C.E.R.) conducted a second-order analysis of these 86 narratives to explore the additional research questions (stated earlier in this section) that both of us had developed after the preliminary framework analysis. Using Atlas-Ti, she open-coded types of abuse, factors cited by students as contributing to abuse, and their responses (inaction or resistance) in the face of abuse.
There were 86 abuse narratives (22 from School 1 from 18 students, 28 from School 2 from 17 students, and 36 from School 3 from 23 students). The abuse recipient was typically the narrator (in at least 68 narratives). The gender of the recipient was typically female (in at least 58 narratives); the perpetrator was typically male (in at least 59 narratives), a doctor (in at least 57 narratives), and a senior doctor (in at least 40 narratives). The most common setting for abuse was the hospital (in at least 68 narratives). We say “in at least X narratives” because not all narratives included details about gender, setting, etc.
Types of abuse
Narratives were generally coded to more than one type of abuse and often included multiple examples of abuse (see edited personal incident narrative in the Appendix). In decreasing order of frequency, 42 abuse narratives were coded as covert, status related; 34 as direct verbal; 22 as sexual discrimination and harassment; 9 as witnessing the abuse of others; 8 as other discrimination (e.g., racial); and 4 as direct physical abuse. We included narratives where the student was the witness because witnesses commonly experience similar emotional reactions as targets and, thus, are “secondary victims.”10,18
In the paragraphs that follow, we have included short quotes from some of the narratives to illustrate the various types of abuse described by the students. We have lightly edited the quotes to be more readable, but we have not changed their style or meaning. Several of these quotes include expletives, which we retain in the article to illustrate students' emotional talk.
Covert, status-related abuse.
Participants reported being ignored by physician teachers and nurses (e.g., teachers not turning up for teaching, arriving late, or ignoring students' conversational contributions), being asked repeated questions in an intimidating way by physician teachers, and receiving questions from physicians that were beyond their level of training (see the Appendix). Students also reported being excluded by physician tutors and nurses from learning opportunities, receiving negative feedback (both formative and summative) unconstructively and destructively (resulting in students feeling humiliated in front of patients), having information withheld from them by physician tutors and nurses, being “used” by physician tutors for their own gains, particularly in the domain of research, being forced to participate in procedures (either by watching or doing) against their wishes, and being given unpleasant tasks (e.g., intimate procedures) as a punishment or joke:
She'd [a consultant, equivalent of attending physician] hand-picked a couple of real classic patients for me to go and examine. One was like a perianal abscess and one was an STDs-UTI-everything-under-the-sun bloke. She said, “What were your examination findings?” I said, “Well, you know, I'm first year, we haven't been shown examination [procedures].” She's, like, “Well, sorry, this history is useless without an examination; you better get back in there.” (First-year male student, School 1)
Direct verbal abuse.
Direct verbal abuse included (1) getting a “good bollocking” from physician teachers and nurses, often perceived by students to be due to their lack of competence, and (2) being called derogatory names by their physician teachers emphasizing their lowly status (e.g., “grasshopper,” “the medical student”) and perceived inadequacies (e.g., “plank” and “lazy”; see the Appendix). Students also reported receiving verbal threats from physician teachers and nurses to make their lives difficult by punishing them in some way (e.g., poor grades, fines, or unpleasant tasks):
He'd [consultant, who is like an attending physician] try and idly threaten for you to do a PR or catheterize some old gent or old lady, and he was always, like, “Oh, maybe I'll embarrass you in front of everybody or maybe I won't … live in fear.” (Third-year male student, School 3)
Sexual discrimination and harassment.
Both men and women reported being sexually discriminated against by physician teachers (students of both genders thought teachers asked them trickier questions and gave them harsher grades compared with opposite-gender counterparts). Students thought male physician teachers patronized female students, men and women described being offended by unwanted sexual talk by physician teachers, and female participants reported unwanted sexual attention (verbal and physical) by patients and physician teachers.
I found the cardiothoracic surgeon, and I said to him, “Oh, you know, my name's [gives her name]. I'm a third-year medical student; can I just scrub in today?” He looked at my breasts for about two minutes and said, “Well, you may as well while you're still young and pretty enough to get away with it.” (Fourth-year female student, School 2)
Witnessing the abuse of others.
Witnessing the abuse of others included students observing physician tutors verbally abusing other health care professionals.
Had an orthopedic surgeon throw his toys out of the pram once … it was literally just over one knife that they [the nurses] couldn't find, so he just knocked something over and shouted, “Why haven't you got it? This is an effing theater.” (Second-year female student, School 3)
Other discrimination included students' personal appearance (e.g., hair, attire) being criticized by physician teachers, students being labeled derogatorily in terms of their demographics (e.g., “young and naive”; “foreigner”), and racist talk being directed at students by patients and clinical teachers.
It came up that I was [the student names her nationality]. Five minutes go by, and it's, like, “So do you drink margaritas and eat burritos all the time?” I'm holding the aorta like this, and he's making these very patronizing, all slightly racist remarks. He said again, “Oh, did you do that after eating your burritos?” or something; he kept going on with the burritos. (Fourth-year female student, School 2)
Direct physical abuse.
Direct physical abuse included actual physical violence (e.g., a student witnessing a colleague being beaten up by a patient in the hospital) and threats of physical violence to students.
[The] anesthetics nurse looked at him [his male peer] and said, “Touch any of that and I'll break your arm.” She wasn't being sarcastic or facetious or anything—She was actually making a threat or purposely being intimidating. (Third-year male student, School 2)
Factors contributing to abuse
Students cited numerous factors contributing to their abuse experiences (see the Appendix). In decreasing order of frequency, they cited characteristics of the perpetrator (153 citations), work (56), the recipient of the abuse (50), the organization (31), and the perpetrator–recipient relationship (26).
Participants reported demographic characteristics of the perpetrators (e.g., older age); a range of negative traits (e.g., perpetrator's inflexibility, intolerance, and aggressiveness; see the Appendix); perpetrators' emotional immaturity, negative affect, and inability to control their own emotions (resulting in emotional outbursts toward recipients); mental health problems (e.g., depression, alcoholism); and both social (e.g., rudeness) and teaching incompetence.
Perpetrators were described as people with reputations for abuse—emphasized by students' derogatory names for them like “psycho,” “idiot,” “prat,” “sleazy old weirdo,” “arsehole,” and “dickhead” (see the Appendix).
One work factor was consistently cited across different narratives as accounting for abuse—time pressures. However, work factors cited were largely specific to the abuse narratives recounted.
I got bawled at by my reg. once … she sent me to do a history, and this guy had a hernia. She's, like, “Go and take a history and do an examination on him,” so I went, took a history, and said, “Is it okay if I examine your hernia?” and he said, “No,” so I came back and she said, “What about the examination?” And I was, like, “Well, he said it wasn't okay for me to do it so I didn't do it,” and so she then shouted at me. (Third-year female student, School 3)
Contributory factors relating to recipients included their demographic characteristics (e.g., age, gender, ethnicity), perceived shortcomings in terms of their personality (e.g., meek, extravagant), their negative behaviors (e.g., noncompliance, rebelliousness, tardiness), lowly status, and associated incompetence within the medical workplace: “Basically we took fairly pitiful histories from a couple of patients” (First-year female student, School 2).
Participants referred to the dog-eat-dog culture of the medical workplace, particularly in the surgical specialty. The hierarchical culture (e.g., “You're the lowest of the low”) was thought to contribute to abuse, to recipients' being unable to challenge abuse, and therefore to the ongoing cycle of abuse.
The rights of a junior medical officer are absolutely ridiculous compared to what most other people out in the workforce have … it's because of this attitude that doctors have: “If it was good enough for us, it's good enough for you.” (Fourth-year female student, School 2)
Relationship factors included perceived personality clashes between perpetrator and recipient and interpersonal conflict, including communication difficulties and social discomfort. Female students also cited “female–female” gender relations as accounting for female nurses' abuse of female medical students: “They hate us—like fire coming out of them.”
Students' responses to abuse
Despite the fact that students typically narrated their abuse dilemmas with negative emotion (found in 67 narratives), students mostly reported inaction in the face of abuse. There was little evidence of student resistance at the time of abuse—challenging (in 16 narratives) or reporting the perpetrator (in 7) or debriefing (in 15). Participants gave a myriad of reasons for their inaction and resistance (see Appendix for resistance).
Factors contributing to students' inaction
Participants cited numerous factors relating to their inaction. In decreasing order of frequency, students cited the recipient (34 mentions), the perpetrator–recipient relationship (25), the perpetrator (20), the organization (15), and work (3).
Participants reported not challenging perpetrators because they felt a lack of ability and confidence. They revealed their “helpless” beliefs—that no benefits would come from challenging—and described not challenging because they forgot or were not bothered enough by the abuse to challenge. They also reported not challenging because they felt the abuse was their fault:
I thought, you know, “it's me, it's me,” and so for the whole year I never fed back to the medical school.” (Fifth-year female student, School 1)
Participants reported doing nothing in order to maintain politeness in their interactions with the perpetrator, to not appear rude, or because they would never see the perpetrator again. They also reported not challenging for fear of gaining a bad reputation affecting their long-term career prospects, especially for fear of receiving negative feedback and poor marks in imminent assessments from the perpetrator.
I know that he's doing my competency and filling in my feedback form in about the next half hour so I just stand there and take it. (Fourth-year male student, School 1)
Participants reported doing nothing because of the perceived characteristics of the perpetrator. Some students reported doing nothing because they thought the perpetrators were good teachers. Others reported not challenging because the perpetrator was “scary” and had a foul mood.
Well, I'm not going to question him because he's already in a bad mood … so I'm not going to provoke him anymore. (Fourth-year male student, School 1)
Participants reported doing nothing because of the organizational hierarchy. They reported their own sense of inferiority and disempowerment and described social etiquette about not undermining the consultant's authority.
I guess it's just the structure; you've got your pecking order … you've got doctors and consultants [equivalent of attending physicians], [who] are the sort of people you don't challenge what they do or say. Yeah, I think it's just basically the culture of the medical profession, sort of like you respect and don't question the authority of the super, like the more senior doctors, and so I think that's pretty much what stops most of us from saying anything or doing anything. (First-year female student, School 2)
Work-related factors that participants reported for not resisting abuse included their lacking the opportunity and their being advised not to by a confidante.
Factors contributing to students' resistance
Some students did resist in the face of abuse. In decreasing order of frequency, students cited factors relating to the perpetrator–recipient relationship (24 instances), the recipient (18), work (6), and the perpetrator (2).
Participants reported that they challenged perpetrators because they felt pushed to their limits by them and did not think about (nor no longer cared about) the consequences. One student reported challenging his perpetrator because he was physically bigger than the perpetrator (see the Appendix). Other students described challenging because they feared the abuse would continue if they did not—particularly in the case of students worried about their learning:
When it starts to affect our education, we have to do something, so we actually complained to the dean about that. (Fourth-year male student, School 2)
Some students said that they resisted because of their strong moral beliefs that the abuse was wrong. Participants also described challenging/reporting perpetrators because they did not want other students to experience what they had gone through. One student explained that she was not a “shrinking violet.”
Work-related factors cited by participants for their resisting included their being advised to do so by colleagues. One student reported challenging his perpetrator because they were outside the typical clinical environment at the time of the challenge (see the Appendix).
Some students explained that they reported the abuse because they knew that other students had experienced similar problems with a perpetrator (see the Appendix). One recipient of racist abuse explained that she challenged her perpetrator because of his ignorance about her country of birth.
Discussion and Conclusions
Participants across the three schools reported abusive experiences as part of their narratives about professionalism dilemmas. As illustrated in the Appendix, each narrative typically included multiple types of abuse, although the most common types across the narratives were covert, status related, direct verbal, and sexual harassment and discrimination. Participants illustrated multiple factors contributing to abuse and their responses in the face of abuse (i.e., factors relating to perpetrators, recipients, perpetrator–recipient relationships, work and the organization).
Our findings extend those originating from surveys by illustrating a broader array of abuses (e.g., covert, status related) than previously documented.3,9 Our findings also extend the current student abuse literature by revealing the context in which abuse unfolds, with students reporting multiple factors contributing to abuse. Our participants cite a breadth of factors that are consistent with those outlined in the social science literature.6,7,18–21 In terms of factors contributing to students' inaction in the face of abuse, our findings again extend those from current surveys.3,9,22,23 We present a broader array of contributory factors for inaction than previously reported, and we also present contributory factors for students' resistance.
What is striking about these findings is that students cited predominantly individual factors (i.e., factors dealing with individuals) as contributing both to abuse and also to their responses to it. This focus on individual factors is understandable, given that medical students are embedded within an individualist culture.27 In terms of contributory factors for abuse, students mostly cited factors relating to perpetrators—like those commentators mentioned in the introduction.16 However, in terms of their inaction in the face of abuse, they mostly cite recipient factors (themselves), and in terms of their resistance at the time of abuse, they mostly cite perpetrator–recipient factors. Interestingly, by focusing on the individual in their narratives, our study's students do “political work,” removing organizational responsibility for abuse and effectively blaming the perpetrator for the abuse and themselves for not resisting.18
Methodological challenges and strengths
Our study has methodological challenges that should be taken into account when interpreting the results.
First, we collected data relating to medical student abuse through students' narratives about professionalism dilemmas. Narratives can be defined as stories, with a narrator, listener(s), a time sequence, a plot, characters, and a purpose.28,29 Narratives are rhetorical: When telling a story, narrators want to portray themselves in a positive light, entertain, persuade, and rally their audience into action.29 Given these characteristics of narratives, it is unsurprising that our students tended to attribute abuse to the perpetrator.21
Second, the coding of types of abuse and contributory factors is a subjective process. Although examples of covert, status-related abuse, such as being given an unrealistic deadline, would typically be constructed as a neutral feature of work if it was an isolated act,7,30 given that our participants recounted numerous negative acts together as part of professionalism dilemmas, we interpreted them as abuse.
Finally, although we present frequencies as part of our qualitative approach in order to illustrate patterns,31 these figures do not equate with prevalence. For example, it is likely that the students witnessed more abuse than they experienced directly as targets, and yet “witnessing the abuse of others” was an abuse type less common across our narratives. Rather than reflecting the prevalence of witnessing abuse, this simply reflects the number of participants who constructed the witnessing of abuse as a professionalism dilemma.
However, despite these challenges we maintain that our study is important because there has been little research that explored medical students' experiences of abuse qualitatively.32 To our knowledge, our study is the first of its kind to examine multiple factors thought to contribute to abuse and students' responses in the face of abuse. We conducted a large number of interviews with a large number of students for a qualitative study. We also collected a reasonable sample of student abuse narratives and a large number of contributory factors across three different schools in three different countries. The fact that abuse was experienced by students at each school strongly suggests the transferability of our findings to other contexts.
Implications for educational practice and research
Students mostly talked about contributory factors at the level of the individual. Although some participants talked about multiple individual, work, and/or organizational factors within narratives, they did not typically talk about the dynamic interplay between the individual and the organization.7 For example, an aggressive individual can influence the organizational culture adversely, and a hostile organizational culture can foster aggression within the individual.7 We think this dynamic interplay between the individual and the organization is central to the abusive culture of the medical workplace and central to combating it.
Boddewyn33 states that “something … usually does not happen unless it is possible, beneficial and triggered.”7 Listening to students' narratives, we think the medical workplace makes abuse possible, beneficial, and triggered. Enabling structures and processes such as perceived power imbalances and low perceived costs of abuse for the perpetrator provide fertile ground in which abuse can grow.7 Motivating structures and processes, such as high internal competition and expected benefits of abuse to the perpetrator (e.g., gaining a higher ranking in the organization relative to peers), also provide favorable conditions. Finally, precipitating processes such as organizational change can easily trigger abuse.7 Although some of these organizational factors (e.g., hierarchy) may be impossible to change, other organizational factors (e.g., low costs of abuse) are amenable to change.
Kleinerman34 recommended that medical student abuse should be changed from the bottom up. We agree that students should be afforded learning opportunities to help them manage abuse. Students should be provided with diversity and equality training and introduced to social sciences topics such as power, obedience, and conformity to help them resist in the face of abuse.35 They should also be provided with a safe forum in which to discuss their abuse dilemmas so that they can consider future strategies for dealing with abuse.36
However, we know from our results and those of others3,9 that medical students mostly respond to abuse with inaction because of their low position in the medical school hierarchy. So, we do not think our doctors of tomorrow have the chance to break the cycle of abuse unaided. We think a top-down approach is equally essential in tackling abuse. After all, organizational leaders are more likely to be abuse perpetrators and also more likely to encourage abuse through failure to intervene in abuses they witness.7 The Medical Leadership Competency Framework suggests that competent doctors show leadership through valuing, respecting, and promoting equality and diversity.37 Faculty development initiatives around equality and diversity are therefore essential for organizational leaders.
In terms of further quantitative research, we are currently conducting a United Kingdom–wide online survey of medical students to determine the prevalence of abuse, how students respond, and the level of moral distress they experience. We have already sent anonymized study results by school to each school, but each school has been told which school it is so that it can compare its results with those of all the other medical schools studied. Those schools with high prevalence rates for abuse may be more inclined to tackle this issue if they know they compare unfavorably with other medical schools. Repeated use of this survey could demonstrate whether faculty development initiatives such as mandatory equality and diversity training change the abusive culture of the medical workplace. In terms of further qualitative research, there is a need to develop a richer picture of medical workplace abuse by additional collection and analysis of abuse narratives from the perspectives of perpetrators.21 We are also examining our qualitative data through different lenses. For example, using the lens-of-behavior explanation, we have already identified how medical students explain their behaviors of compliance with or resistance to the instructions of senior clinical teachers to conduct intimate examinations on patients without valid consent.38 Furthermore, using a narrative lens, our students' sharing of their abuse narratives with researchers can be seen as their attempts to resist abuse and enact organizational change.18,39
The authors thank their study participants for sharing their stories with us and with each other so candidly. The authors would also like to acknowledge the contribution of their late colleague, professor Kieran Sweeney, to the present study: for helping secure funding and for his thoughtful comments on the study before his death. The authors thank their research assistant, Laura Rees-Davies, for helping collect data and contributing to the preliminary framework analysis. The authors also thank their colleague professor Frank Sullivan, University of Dundee, for his helpful peer-review comments on an earlier draft of this report.
This research was supported by the British Academy (LRG-45505).
None. This research was carried out independently of the study sponsor, who had no input to the study design; collection, analysis, and interpretation of data; writing the report; or the decision to submit the report for publication. Both authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
This study was approved by the ethics committees of the three medical schools. Informed consent was obtained from all participants, along with their right to withdraw from the study at any time without penalty.