2.1 Study Design and Participants
An electronic survey composed of four sections: demographics, stressors and their impact on residency and career choice, World Health Organization (WHO) well-being index and stress management and resources was distributed to clerks (third- and fourth-year medical students) in all Canadian medical schools, with 10 of the 17 faculties participating and forwarding the survey to their clerks. The survey was developed and delivered in French and English. The study protocol, consent form and recruitment documents were approved by the institutional review board of the Université de Montréal (CERSES-20-060-D).
Eligible participants were clerks attending one of the 17 Canadian medical faculties, and whose rotations were suspended due to the COVID-19 pandemic. Clerks whose clerkship had been suspended for any other reason were excluded.
2.2 Survey Development
The survey (Appendix 1) contained four sections: (1) Demographics, (2) Stressors (3) WHO well-being index, and (4) Stress management and resources. These four sections of the survey were developed by either adapting existing sources(16) or by conducting short focus group sessions with a small sample of junior and senior clerks from a single Canadian medical school.
Demographics: The first section of the survey consisted of questions on demographics; age, gender, level of education, university, implications and occupations, COVID-19 infection status and preferred residency. Senior clerks were asked about the specialty to which they had already matched.
Stressors: Stressors included in the survey were developed with a focus group of Canadian medical clerks. This focus group was composed of four third year medical students and two fourth year medical students whose clinical rotations were suspended due to the COVID-19 pandemic. A meeting was held in order to select the main stressors by reflecting on their own experience as clerks and the experience of their classmates.
This section included an initial question allowing an assessment of the respondent's stress level in the context of the COVID-19 pandemic. The three subsequent questions assessed the impact of the pandemic on their choice of residency and their decision to pursue a medical career. If the respondent reconsidered his residency choice, an additional question was asked to see the type of residency change they considered. This section also contained a question allowing the respondent to rate, on a Likert scale, the level of stress associated with potential stressors they may have experienced. Most of these stressors were directly linked to future educational and professional pathways.
WHO well-being index: The third section of the survey assessed the respondent's state of well-being by using the WHO well-being index.(16) This measurement tool consists of five statements that respondents apply to their own lives. It has high clinical validity and is among the most widely used questionnaires assessing subjective psychological well-being.(17) It has been translated into over 30 languages, including French.(17) The maximum score of this index is 25 and a score of 13 or lower indicates a poor state of well-being.(18)
Stress management and resources: The fourth section of the survey contained five questions evaluating whether the respondent used university psychosocial resources during the COVID-19 pandemic, as well as the perceived usefulness of the resource. This section also included a question exploring the level of agreement regarding certain academic solutions. The level of agreement with each solution was assessed using a 5-point Likert scale ranging from "Strongly agree" to "Strongly disagree". A final open-ended question allowed respondents to offer other potential solutions.
In order to ensure validity of the survey, its contents were reviewed by a group of experts composed of members of the medical stimulation and education research group at the Université of Montréal and the Ontario Tech University, consisting of physicians, researchers in medical education, residents, and medical students.
2.3 Recruitment and distribution of the survey
The survey was developed and distributed using the Qualtrics XM platform.(11) The request to participate was sent to the deans and curriculum directors of all 17 Canadian medical faculties. They were asked to review the summary of the study and the survey and decide if the faculty would participate. Because the aim of this study was to capture the well-being, stressors, and management within the acute stage of the COVID-19 pandemic, we requested a response within a month. For the faculties that agreed to participate, the clerkship directors were asked to send the survey to all eligible clerks by distributing an email containing a digital link. For faculties that chose not to participate, the clerkship directors were asked for a short explanation and reasoning for not participating in the study.
Participation in this study was voluntary and consent was obtained with a consent form at the beginning of the survey. The survey was anonymous and took 10 to 15 minutes to complete.
2.4 Data collection and statistical analysis
The study was conducted between April 9th, 2020 and June 3rd, 2020.
Data collected for each of the four sections of the survey were analyzed separately. The data was first cleaned and analyzed in Microsoft Excel. Except for participants’ characteristics, missing data from participants who elected not to answer certain questions were excluded from all calculations.
Demographics: Categorical variables (age, level of study, home university, preferred residency match, COVID status, and family COVID status) were described as proportions of the study population.
Stressors: Descriptive statistics were performed. Proportions were used to assess the number of clerks who reconsidered medicine as a career and their residency choice, and to assess levels of stress during the pandemic.
WHO well-being index: First, we assessed the frequency distribution of the index and performed descriptive statistics for all participants with the mean and standard deviation. Second, we compared the mean WHO well-being index between two groups of participants: participants who reconsidered medicine as a career and participants who did not. This data was obtained from part 2 of the survey (Stressors). We used unpaired t-tests to compare these two means. For each analysis, confidence intervals were computed for a 95%-degree confidence. P-values less than 0.05 were interpreted as indicating a statistical difference. All inferential statistics were performed in GraphPad Prism. (19)
Stress management and resources: Descriptive statistics were performed using proportions. For the final open-ended question, responses were grouped by theme. The themes that came up more than ten times by different students were considered valid solutions.