Predictors of self-reported research engagement and academic-career interest amongst medical students in the United Kingdom: a national cross-sectional survey.

BACKGROUND: We hypothesised that the gender/ethnic disparities and reductions in the UK academic-clinician workforce stem from research experience in medical school. This study investigated the factors influencing research engagement and academic-career interests among UK medical students. METHODS: Using a 42-item online questionnaire, a national multicentre cross-sectional survey of UK medical students was conducted over 9 weeks in the 2020/21 academic year. Multiple binary logistic and zero-inflated negative binomial regressions were used to evaluate associations between the predictor variables and research engagement (yes/no), number of research projects conducted, and academic-career interest (yes/no). P < 0.05 was considered statistically significant. RESULTS: In total, 1573 students participated from 36 medical schools. No ethnic/gender differences in research engagement were observed. However, compared to men, women had a 31% decrease in the odds of being interested in an academic-clinician career [odds ratio (OR): 0.69; 95% confidence interval (CI): 0.52, 0.92]. Positive predictors of interest in academia were being a PubMed-indexed author (OR: 2.19; 95% CI: 1.38, 3.47) and having at least one national/international presentation (OR: 1.40; 95% CI: 1.04, 1.88). Career progression was the primary motivating factor (67.1%) for pursuing research, whereas limited awareness of opportunities (68.0%) and time constraints (67.5%) were the most common barriers. CONCLUSION: There were no ethnic differences in research engagement or academic-career intent. Although there were no gender differences in research engagement, female students were less likely to be interested in an academic career. This could be tackled by providing targeted opportunities to increase research productivity and self-efficacy in medical schools. Key messages: What is already known on this topic: There has been a decline in the number of academic clinicians, with a disproportionate gender and ethnic representation in the academic workforce. Engaging medical students in research activities during their medical training could mitigate the declining number of academic clinicians. Differential attainment occurs in medical school and persists after graduation. What this study adds: Although there were no gender/ethnic differences in research engagement amongst UK medical students, our study suggests that female students were less likely to be interested in pursuing an academic career. Time constraints, a lack of awareness of opportunities, and difficulty in finding research supervisors/mentors were the most common barriers to research engagement, whereas PubMed-indexed authorship was the strongest positive predictor of interest in an academic career. How this study might affect research, practice, or policy: Medical schools should facilitate the selection of good-quality research mentors that would provide adequate support to ensure that their students' works are published in peer-reviewed journals. Medical schools should employ local research officers to increase students' awareness of research opportunities.

Predictors of self-reported research self-efficacy and perception of research amongst medical students in the United Kingdom: a national cross-sectional survey.

PURPOSE: There has been a decline in the number of academic clinicians in the UK, and there are ethnic/gender disparities in the academic workforce. Higher research self-efficacy (RSE) and a positive perception of research (PoR) amongst students are associated with a higher motivation to engage in academic medicine. Hence, this study aimed to determine the factors that influence RSE and PoR amongst UK medical students. METHODS: This is a multicentre cross-sectional survey of medical students in 36 UK medical schools in the 2020/21 academic year. Multiple linear regression was used to investigate the association between students' demographics and RSE/PoR. P-values less than a Bonferroni-corrected significance level of .05/28 = .0018 were considered statistically significant. RESULTS: In total, 1573 individuals participated from 36 medical schools. There were no ethnic differences in PoR or RSE scores. Although there were no gender differences in PoR, female students had lower RSE scores than male students (adjusted ß = -1.75; 95% CI: -2.62, -0.89). Research experience before medical school (adjusted ß = 3.02; 95% CI: 2.11, 3.93), being in the clinical training phase (adjusted ß = 1.99; 95% CI: 1.09, 2.90), and completing a degree before medical school (adjusted ß = 3.66; 95% CI: 2.23, 5.09) were associated with higher RSE. CONCLUSION: There were no associations between the predictor variables and PoR. Female students had lower self-reported RSE scores. Future studies should investigate the role of targeted research mentorship in improving RSE amongst female medical students.

Predictors of self-reported research productivity amongst medical students in the United Kingdom: a national cross-sectional survey.

BACKGROUND: The number of academic clinicians in the UK is declining and there are demographic inequalities in the clinical-academic workforce. Increased research productivity by medical students is believed to reduce future attrition in the clinical-academic workforce. Thus, this study investigated the association between student demographics and research productivity amongst UK medical students. METHODS: This is a national multicentre cross-sectional study of UK medical students in the 2020/21 academic year. We appointed one student representative per medical school, and they disseminated a 42-item online questionnaire over nine weeks, through departmental emails and social media advertisements. The outcome measures were: (i) publications (yes/no) (ii) number of publications (iii) number of first-authored publications (iv) abstract presentation (yes/no). We utilised multiple logistic and zero-inflated Poisson regression analyses to test for associations between the outcome measures and predictor variables at a 5% significance level. RESULTS: There are 41 medical schools in the UK. We received 1573 responses from 36 UK medical schools. We failed to recruit student representatives from three newly formed medical schools, whilst two medical schools prohibited us from sending the survey to their students. Women had lower odds of having a publication (OR: 0.53, 95% CI: 0.33-0.85) and on average had fewer first-author publications than men (IRR: 0.57, 95% CI: 0.37-0.89). Compared to white students, mixed-ethnicity students had greater odds of having a publication (OR: 3.06, 95% CI: 1.67-5.59), an abstract presentation (OR: 2.12, 95% CI: 1.37-3.26), and on average had a greater number of publications (IRR: 1.87, 95% CI: 1.02-3.43). On average, students who attended independent UK secondary schools had a higher rate of first-author publications compared to those that attended state secondary schools (IRR: 1.97, 95% CI: 1.23-3.15). CONCLUSION: Our data suggest that there are gender, ethnic and socioeconomic inequalities in research productivity among UK medical students. To tackle this, and potentially improve diversity in clinical academia, we recommend that medical schools should facilitate targeted high quality research mentorship, funding and training, especially for under-represented-in-medicine students.

Machine learning model to predict mental health crises from electronic health records.

The timely identification of patients who are at risk of a mental health crisis can lead to improved outcomes and to the mitigation of burdens and costs. However, the high prevalence of mental health problems means that the manual review of complex patient records to make proactive care decisions is not feasible in practice. Therefore, we developed a machine learning model that uses electronic health records to continuously monitor patients for risk of a mental health crisis over a period of 28 days. The model achieves an area under the receiver operating characteristic curve of 0.797 and an area under the precision-recall curve of 0.159, predicting crises with a sensitivity of 58% at a specificity of 85%. A follow-up 6-month prospective study evaluated our algorithm's use in clinical practice and observed predictions to be clinically valuable in terms of either managing caseloads or mitigating the risk of crisis in 64% of cases. To our knowledge, this study is the first to continuously predict the risk of a wide range of mental health crises and to explore the added value of such predictions in clinical practice.

Regional Distributions of Iron, Copper and Zinc and Their Relationships With Glia in a Normal Aging Mouse Model.

Microglia and astrocytes can quench metal toxicity to maintain tissue homeostasis, but with age, increasing glial dystrophy alongside metal dyshomeostasis may predispose the aged brain to acquire neurodegenerative diseases. The aim of the present study was to investigate age-related changes in brain metal deposition along with glial distribution in normal C57Bl/6J mice aged 2-, 6-, 19- and 27-months (n = 4/age). Using synchrotron-based X-ray fluorescence elemental mapping, we demonstrated age-related increases in iron, copper, and zinc in the basal ganglia (p < 0.05). Qualitative assessments revealed age-associated increases in iron, particularly in the basal ganglia and zinc in the white matter tracts, while copper showed overt enrichment in the choroid plexus/ventricles. Immunohistochemical staining showed augmented numbers of microglia and astrocytes, as a function of aging, in the basal ganglia (p < 0.05). Moreover, qualitative analysis of the glial immunostaining at the level of the fimbria and ventral commissure, revealed increments in the number of microglia but decrements in astroglia, in older aged mice. Upon morphological evaluation, aged microglia and astroglia displayed enlarged soma and thickened processes, reminiscent of dystrophy. Since glial cells have major roles in metal metabolism, we performed linear regression analysis and found a positive association between iron (R 2 = 0.57, p = 0.0008), copper (R 2 = 0.43, p = 0.0057), and zinc (R 2 = 0.37, p = 0.0132) with microglia in the basal ganglia. Also, higher levels of iron (R 2 = 0.49, p = 0.0025) and zinc (R 2 = 0.27, p = 0.040) were correlated to higher astroglia numbers. Aging was accompanied by a dissociation between metal and glial levels, as we found through the formulation of metal to glia ratios, with regions of basal ganglia being differentially affected. For example, iron to astroglia ratio showed age-related increases in the substantia nigra and globus pallidus, while the ratio was decreased in the striatum. Meanwhile, copper and zinc to astroglia ratios showed a similar regional decline. Our findings suggest that inflammation at the choroid plexus, part of the blood-cerebrospinal-fluid barrier, prompts accumulation of, particularly, copper and iron in the ventricles, implying a compromised barrier system. Moreover, age-related glial dystrophy/senescence appears to disrupt metal homeostasis, likely due to induced oxidative stress, and hence increase the risk of neurodegenerative diseases.