This national survey cross-sectional study investigated the understanding and attitudes of medical students and doctors towards EBM, principally research concepts and critical appraisal. This is a necessary issue as adequate training in EBM is required for clinical decision making and good conduct of clinical research.
True to our initial hypothesis, only a limited majority of respondents claimed they received teaching in CAS as part of their medical curriculum. This was as low as 48% in medical students with an increase to 65% for clinical academics. As our results show that less than 50% of medical students received training it indicates a sizeable neglect in this area by the medical curricula. Furthermore, given the significance of CAS as a core skill for clinical academics [16] it is unsurprising to see that 40% of clinical academics attended extracurricular courses on CAS outside of formal learning, most likely to bridge the gap between limited medical school teaching and academic requirements. When this is compared to the 15% of clinicians who attended external courses on CAS this further highlights the necessity of CAS as an academic skill.
Given this, it is expected to see that the clinical academics were consistently the most confident in all the research concepts featured in the survey. Generally, clinicians were more confident than medical students in understanding concepts except for the PICOS framework and null hypothesis. A potential reason for these marginal differences in confidence could be due to certain concepts such as biases having more meaning in a practical setting, whereas PICOS and hypotheses are routed in theoretical learning which students may have familiarity towards. However, given the heterogeneity of results for these two groups a stronger study would be required to assess any statistically significant differences. Moreover, it is not unexpected that in the more ubiquitous concepts such as study types and P-values all three groups were ‘confident’ in these areas whereas, the more technical areas of error and bias type had a net score of ‘neutral’. This shows that whilst some concepts in research are common knowledge, higher level areas which are seldom discussed but perhaps as equally important in critical appraisal require further education on.
Despite a widely varied education in EBM and differing levels of confidence in research concepts, all three groups agreed that CAS for practicing clinicians was an ‘important’ skill. The Likert score given to this was greater in the clinicians than medical students, perhaps due to the day-to-day relevance critical appraisal has in clinical practice.
The percentage of respondents in each group who agreed with this seemed to increase alongside their overall confidence of research concepts. This supports the idea that only through teaching of EBM does its importance to clinical practice become more apparent. Encouragingly, most respondents expressed interest in participating in research during their medical careers. As expected, this was highest in the clinical academic group (94%) followed by 78% of medical students and 66% of clinicians. Students who intercalated or completed a previous degree were more likely to be interested in performing research. However, as the Foundation Programme no longer awards extra credit for students who have intercalated it may be less appealing for students to continue with intercalation [14]. As intercalation was the predominant source of CAS it can be expected that there will now be a decline in EBM knowledge of future medical students. 66% of medical students also expressed interest in attending extracurricular teaching of critical appraisal demonstrating the interest in learning in this field.
Our study benefited from a robust design by expanding on previously published surveys to evaluate our outcomes. Given the heterogeneity of utilized survey tools and blinding of participants (medical students and doctors) in-depth statistical analyses could not be performed. Despite the plausibility of our results and homogeneity with the conclusions of previous studies, the results of these surveys are subject to response and recall bias. Additionally due to the relatively small sample size and medical student responses from a limited number of medical schools there may be some institutional bias. Furthermore, respondent bias, is also likely present as those with some interest in EBM were most likely to have completed the survey. To mitigate for this, we incentivized as many eligible respondents from diverse backgrounds as possible with a £25 gift voucher. As one of the primary aims of this study was to establish confidence in research concepts the objective research knowledge of each group may be different and provide a stronger or different argument. Future studies in this field may consider evaluating knowledge using objectives measures between groups.