Factors inuencing Research Engagement among Medical students in the United Kingdom (REMED-UK): protocol for a prospective, observational, national, multi-centre questionnaire - study

Introduction: There has been a decline in the number of academic clinicians in the UK. Also, there is sex and ethnic inequalities in research opportunities amongst UK clinicians. The decision to get involved in research as a clinician is usually made as a medical student. Hence, this study aims to investigate the factors affecting UK medical students’ involvement in research, and to understand their perceptions of research, motivation for and the barriers to their engagement in research. Methods: This study is a prospective, observational, national, multi-centre study that will utilise an online questionnaire using the Qualtrics TM survey tool. Medical students that are at least 18 years old and enrolled in a UK medical school in the 2020/21 academic year are eligible to partake in this study. The following data will be collected: demographic, research experience, research perception, research self-ecacy, motivation, and barriers to engaging in research. The outcome measures are: number of research projects, publications, presentations, and successful grant applications. Conclusion: This study will produce evidence-based reports that will guide UK medical schools and relevant organisations in developing resources and programmes that will increase the number of future academic clinicians and resolve inequalities in research opportunities and engagement amongst future clinicians.


Introduction
Research can contribute to the creation of new knowledge that leads to advancement in clinical care and practice. In the United Kingdom (UK), the General Medical Council 'Outcomes for graduates' states that newly quali ed doctors must be able to apply scienti c methods and approaches to medical research and be pro cient in evidence-based medicine within their clinical eld [1]. Clinicians should participate in research because compared to a purely basic scientist, they regularly encounter clinical problems that may inspire relevant questions for scienti c research. Also, academic clinicians can bridge the gap between science and practice, by translating research ndings into clinical settings [2].
Recent reports indicate a decline in the number of clinical academics in the UK in the last few decades [3][4][5]. Also, the ethnic and gender pro le of the clinical academic workforce is not re ective of the wider population of licensed doctors. A 2018 report indicates that in the UK, there is a disproportionately low number of Black, Asian, and ethnic minority (BAME) and/or female clinicians in the clinical academic workforce and the upper echelons of research leadership [5]. A recent report by the Royal College of Physicians indicates that female clinicians and BAME clinicians have reduced access to research opportunities compared to their male/white counterparts [6]. It is believed that a greater balance in terms of gender and ethnicity in the healthcare workforce may improve patient outcomes, including those of deprived populations and ethnic minorities [7].
For most clinicians, the decision of whether to pursue a research career is made before they graduate from medical school [8]. Furthermore, early engagement in research has been reported to enhance medical students' research-related skills, stimulate their research interest, and is positively associated with sustained scienti c activity and productivity after medical school graduation [9,10]. Hence, medical school provides a fertile ground to initiate programmes that will increase the number of future academic clinicians and resolve the gender and ethnic imbalance in research engagement amongst clinicians.
However, there is little empirical research conducted in the UK to guide the development of such programmes.
To stimulate medical students and keep them interested in research, it is important to understand their motivation for and the barriers preventing their participation and attainment in research. Hence, the primary aim of this study is to investigate the factors in uencing research opportunities and attainments amongst UK medical students. Also, the study will investigate the perception of research, motivation for and barriers to research engagement amongst UK medical students.

Methods
Study design REMED-UK study is a prospective, observational, national, multi-centre study that will utilise an online questionnaire using the Qualtrics TM survey tool. This study is being run by Melanin Medics, a UKregistered charitable organisation that aims to increase diversity in medicine. Exclusion criteria: Any student that is unwilling or unable to give informed consent or does not meet the inclusion criteria.
The UK National Research Collaborative Model [11] will be used to distribute the questionnaire to medical students currently enrolled in medical schools listed by the Medical Schools Council at the start of the 2020/21 academic year. The survey will be disseminated through emails and various social media platforms. Prospective data will be collected over eight weeks (22 March to 16 May 2021).
The questionnaire (Appendix 1) has the following domains: Pre-processing steps will include re-categorising free text entries into existing similar data categories and grouping free text entries into new data categories. Statistical analysis will be conducted on IBM SPSS Statistics 26 software. Variables with less than 80% responses will be excluded from the analysis.
Descriptive statistics will be presented as mean ± standard deviation for parametric data, and median ± interquartile range for non-parametric data. Frequencies will be presented as both absolute numbers and percentages.
Binary logistic regression analysis will be utilised to test for associations between categorical outcomes (binary) and the independent variables. Multinomial logistic regression analysis will be utilised if the categorical outcome has more than two categories. Multiple linear regression analysis will be utilised if the outcome variable is continuous. Generalised linear models (binomial distribution and logit link) will be utilised if the outcome variable is a proportion but with a binary outcome. The assumptions for these statistical tests will be investigated before running the tests. For all analysis, a p-value < 0.05 will be considered statistically signi cant. See Appendix 2 for the hypotheses that will be tested in this study.

Independent variables:
Demographic: Age, Sex, ethnicity, Russell group vs non-Russell group university, stage of training, previous degree, part-time job, type of secondary school, parent's highest level of academic quali cation, previous research experience, medical school decile Desire to get involved in various aspects of research (e.g. data collection, analysis, writing manuscript.): Categorical (yes/no). These components will be analysed separately. PubMed-indexed publications, number of rst-authored publications, and collaborator status Presentations: analysed at categorical and continuous levels. Secondary analysis will consider oral presentations, and level of presentation (international or national: yes vs no).
Grants: Proportion of successful grant applications (whilst still accounting for the number of applications) Con dence in various aspects of research: Analyse the individual Likert items as continuous variables in isolation Involvement in various aspects of components of research (e.g. data collection, analysis, writing manuscript.): Categorical (yes/no). These components will be analysed separately.

Anticipated Outputs
On completion of the study, the data will be analysed, and reports will be prepared. The study report(s) will be submitted for publication in peer-reviewed journal(s) and presented at national/ international meetings.

Declarations
Con ict of interest: None