Medical graduate views on statistical learning needs for clinical practice: a comprehensive survey

Abstract

BackgroundThis paper seeks to contribute to a reputable evidence base for required competencies across different topics in statistics and probability (statistical topics) in preparing medical graduates for clinical practice. This is in order to inform the prioritization of statistical topics within future undergraduate medical curricula, while exploring the need for preparing tomorrow’s doctors to be producers, and not merely consumers, of statistics.MethodsWe conducted a comprehensive online survey from July 2013 to August 2014 for a target group of 462 medical graduates with current or prior experience of teaching undergraduate medical students of the University of Edinburgh of whom 278 (60.2%) responded. Statistical topics were ranked by proportion of respondents who identified the practice of statistics, performing statistical procedures or calculations using appropriate data, as a required competency for medical schools to provide in preparing undergraduate medical students for clinical practice. Mixed effects analyses were used to identify potential predictors for selection of the above competency and to compare the likelihood of this selection for a range of statistical topics versus critical appraisal.ResultsEvidence was gleaned from medical graduates’ experiences of clinical practice for the need for, not only a theoretical understanding of statistics and probability but also, the ability to practice statistics. Nature of employment and statistical topic were highly significant predictors of choice of the practice of statistics as a required competency ((F = 3.777, p < 0.0005) and (F = 45.834, p < 0.0005), respectively). The most popular topic for this competency was graphical presentation of data (84.3% of respondents) in contrast to cross-over trials for the competency understanding the theory only (70.5% of respondents). Several topics were found to be more popular than critical appraisal for competency in the practice of statistics.ConclusionsThe model of medical graduates as mere consumers of statistics is oversimplified. Contrary to what has been suggested elsewhere, statistical learning opportunities in undergraduate medicine should not be restricted to development of critical appraisal skills. Indeed, our findings support development of learning opportunities for undergraduate medical students as producers of statistics across a wide range of statistical topics.

Impact of COVID-19 outbreak on the mental health status of undergraduate medical students in a COVID-19 treating medical college: a prospective longitudinal study

“…Unfortunately, only limited studies were done on the impact of epidemic/ pandemic on the mental health of medical students. Studies on the impact of the COVID-19 pandemic on medical students are limited to cross-sectional surveys assessing attitude, awareness, knowledge, precautionary measures, concerns, risk perceptions, impact on education and confidence, and fear of COVID-19 (Agarwal et al, 2020;Ahmed et al, 2020b;Choi et al, 2020;Khasawneh et al, 2020;Nguyen et al, 2020;Taghrir, Borazjani & Shiraly, 2020). Literature search showed only a single study about the psychological impact of COVID-19 on medical students, which cross-sectionally assessed their anxiety levels (Cao et al, 2020).…”

Methodologic Progress Note: A Clinician's Guide to Logistic Regression

“…Fitting the model is the process by which statistical software (eg, SAS, Stata, R, SPSS) estimates the relationships among independent variables in the model and the outcome within a specific dataset. In equation (2), this essentially means that the software will evaluate the data and provide us with the best estimates for β 0 (the y-intercept) and β 1 (the slope) that describe the relationship between the variable x and p ln ( ) .…”

“…We can build more complex models using multivariable logistic regression by adding more independent variables to the right side of equation (2). Essentially, this is what Schaffer et al 4 did when, for example, they described clinical factors associated with indemnity payments (Schaffer et al, Table 3).…”

“…T he ability to read and correctly interpret research is an essential skill, but most hospitalists-and physicians in general-do not receive formal training in biostatistics during their medical education. [1][2][3] In addition to straightforward statistical tests that compare a single exposure and outcome, researchers commonly use statistical models to identify and quantify complex relationships among many exposures (eg, demographics, clinical characteristics, interventions, or other variables) and an outcome. Understanding statistical models can be challenging.…”

Effects of problem-based learning modules within blended learning courses in medical statistics – A randomized controlled pilot study

“…The study of Swift et al [ 2 ] showed that medical doctors considered medical statistics useful for “accessing clinical guidelines and evidence summaries, explaining risk levels to patients, assessing medical marketing and advertising material, interpreting the results of a screening test, reading research publications for general professional interest, and using research publications to explore non-standard treatment and management options”. Future physicians also thought that there was a need for a practical application of knowledge in medical statistics, not only its’ theoretical basis [ 3 ]. Lack of knowledge in medical statistics can lead to misinterpretation of clinical findings [ 4 ].…”