17 research outputs found
National audit of pathways in epileptic seizure referrals (NAPIER) : a national, multicentre audit of first seizure clinics throughout the UK and Ireland
No full textAcknowledgements We would like to thank the following collaborators of the NANSIG Collaborative, who conducted local data collection and analysis: Ajitesh Anand, Alena Abraham, Alex Irving, Amogh Prabhakar, Catinca Ciuculete, Cindy Zheng, Daniel King, Declan Browne, Dipesh Kumar Barua, Dorota Duklas, Farhat Mirza, Fumilola Olaifa, Harmani Daler, Hassan Naveed, Heba Elzeky, Hedley Emsley, Honglin Zhu, Ian Morrison, Irtiza Syed, Isabel Summers, Jack Wellington, Jasmine Wall, John O'Dwyer, Jordan Ford, Karthikeyan Sivaganesh, Katja Lassak, Keara Jamison, Khalid Hamandi, Kourosh Parvi, Lareyna McMenemy, Lewis McColm, Lina Aleknaite, Maithili Srikantha, Maja Kaladjiska, Marie Jasim, Mark McCarron, Martina Mockova, Mohammad Marar, Naghme Adab, Najma Ahmed, Nye Rhys Potter, Pavithira Tharmapoopathy, Prithvi Dixit, Rajiv Mohanraj, Ravanth Baskaran, Richard Davenport, Robert Seah, Rohan Bhate, Rohan Gupta, Sahar Shams, Siddarth Kannan, Tahir Majeed, Timothy Counihan, Tomas Ferriera, Yihui Cheng, Zaib ShamshiPeer reviewe
Supporting mental health during the COVID-19 pandemic: implementation of an e-guide
No full textThe COVID-19 pandemic has caused significant disruption to activities of daily living, which in turn has had a profound impact on the mental wellbeing of the public. An e-guide was designed to provide remote support to the general public through the application of a Behavioural Activation approach. Interactive, brief evidence-based exercises were included in the e-guide, along with mood ratings after each exercise to assess any improvements observed. The e-guide was designed using the Xerte On- Line Toolkits open source software. Videos and interactive exercises were embedded within the resource, forming part of the brief intervention based on cognitivist and behaviourist principles. Information and further support was also provided for young people and parents. Videos from the public highlighting their experiences during the pandemic were also sourced and included (with consent). A pilot was launched to assess the impact of the e-guide. Participants were recruited from Cardiff University, mental health services and a local charity. The e-guide was piloted on a sample of volunteers (n = 3), who completed a brief survey after engaging with the resource. Following the results of the pilot, the e-guide was promoted by the university's marketing team and made available to the public. At the 6-month mark, the e-guide had been accessed by 3228 individuals throughout the UK. The e-guide has since been disseminated by support services for young people, places of employment and eduction institutions. The national impact of the e-guide is evidenced from the number of people accessing the resource exceeding 3000. With the long-term effects of the pandemic taking hold, it remains crucial to support the wellbeing of the general public through such initiatives that are administered remotely
Assessing ChatGPT 4.0's capabilities in the United Kingdom Medical Licensing Examination (UKMLA): A robust categorical analysis
No full textAdvances in the various applications of artificial intelligence will have important implications for medical training and practice. The advances in ChatGPT-4 alongside the introduction of the medical licensing assessment (MLA) provide an opportunity to compare GPT-4's medical competence against the expected level of a United Kingdom junior doctor and discuss its potential in clinical practice. Using 191 freely available questions in MLA style, we assessed GPT-4's accuracy with and without offering multiple-choice options. We compared single and multi-step questions, which targeted different points in the clinical process, from diagnosis to management. A chi-squared test was used to assess statistical significance. GPT-4 scored 86.3% and 89.6% in papers one-and-two respectively. Without the multiple-choice options, GPT's performance was 61.5% and 74.7% in papers one-and-two respectively. There was no significant difference between single and multistep questions, but GPT-4 answered 'management' questions significantly worse than 'diagnosis' questions with no multiple-choice options (p = 0.015). GPT-4's accuracy across categories and question structures suggest that LLMs are competently able to process clinical scenarios but remain incapable of understanding these clinical scenarios. Large-Language-Models incorporated into practice alongside a trained practitioner may balance risk and benefit as the necessary robust testing on evolving tools is conducted. [Abstract copyright: © 2025. The Author(s).
A picture paints a thousand words: perspectives of the use of images and videos of physical findings in virtual OSCE teaching sessions
No full textBackground: The COVID-19 pandemic has resulted in medical students’ education being disrupted and a new direction of learning towards the online classroom, with students frequently learning to identify clinical signs via online conferencing platforms. Given this shift in student study techniques in the pandemic and post-pandemic era, we set out to analyse the confidence levels of students recognizing clinical signs in patients and the opinions and patterns of resources students would use to build up the knowledge and confidence for their future clinical practice.
Methods: OSCEazy, a student-led medical education organisation, delivered an online teaching session and disseminated a questionnaire containing ten-point Likert scales, multiple-choice questions and free text options.
Results: The majority (over 75%) of respondents reported using digital resources to learn physical examinations. Most respondents were somewhat confident with recognising clinical signs during physical examinations (Median 7, IQR 6– 8). Kruskal–Wallis H-test on students’ impression of the quality of current OSCE materials showed a statistically significant difference between cohorts (X2(2) = 14.209, p = 0.014). A large proportion of students (98.41%) agreed that an online resources bank would be beneficial for their learning. Wilcoxon rank test showed a statistically significant preference for learning physical examinations using clinical images rather than videos (p = 0.014). A 29% of respondents mentioned the use of physical textbooks as a resource, while online platforms such as OSCE revision websites, YouTube and Google Images had become the mainstream platforms.
Conclusion: The findings reported here suggest that there needs to be a greater emphasis on providing online image-based resources to support medical students in learning clinical signs. A resource containing these clinical signs would prove to be of benefit for students to access high-quality self-directed learning to identify and verify clinical signs
Asymptomatic chronic subdural haematoma - does it need neurosurgical intervention?
No full textPurpose
Chronic subdural haematoma (CSDH) is a well-recognized consequence of head injury with a rising incidence worldwide. Whilst symptomatic CSDH warrants consideration of surgical intervention, the management of asymptomatic CSDH (AsCSDH) remains unclear. In this retrospective study, we investigate the natural history of AsCSDH, the requirement for radiological monitoring, and the role of neurosurgical input.
Methods
Head injury referrals to a tertiary neurosurgical unit over two years were screened to identify patients with AsCSDH. Clinical, radiological, and outcome parameters were collected for included patients.
Results
Of 2725 referrals, 106 (3.9%) patients were eligible for inclusion. The cohort consisted of predominantly male patients (70.8%) with an average age of 81.9 years and independent at baseline (79.3%). Radiological follow-up was recommended by neurosurgery in 4 patients (3.8%). Medical teams performed follow-up imaging for 57 patients (53.8%) culminating in a total of 116 follow-up scans, predominantly for falls or monitoring purposes. Antithrombotic agents were used by 61 patients (57.5%). Anticoagulants were held in 26/37 patients (70.3%) and antiplatelets in 12/29 patients (41.4%), ranging from 7 to 16 days when specified. Only one patient required neurosurgical intervention at 3 months from the time of initial presentation after the development of symptoms.
Conclusions
Patients with AsCSDH do not require neuroradiological follow-up or neurosurgical intervention in the majority of instances. Medical professionals should explain to patients, families, and caregivers that the isolated finding of CSDH is not necessarily a cause for concern but safety netting advice regarding AsCSDH should be provided
Assessing ChatGPT 4.0’s Capabilities in the United Kingdom Medical Licensing Examination (UKMLA): A Robust Categorical Analysis
No full textAdvances in the various applications of artificial intelligence will have important implications for medical training and practice. The advances in ChatGPT-4 alongside the introduction of the medical licensing assessment (MLA) provide an opportunity to compare GPT-4’s medical competence against the expected level of a United Kingdom junior doctor and discuss its potential in clinical practice. Using 191 freely available questions in MLA style, we assessed GPT-4’s accuracy with and without offering multiple-choice options. We compared single and multi-step questions, which targeted different points in the clinical process, from diagnosis to management. A chi-squared test was used to assess statistical significance. GPT-4 scored 86.3% and 89.6% in papers one-and-two respectively. Without the multiple-choice options, GPT’s performance was 61.5% and 74.7% in papers one-and-two respectively. There was no significant difference between single and multistep questions, but GPT-4 answered ‘management’ questions significantly worse than ‘diagnosis’ questions with no multiple-choice options (p = 0.015). GPT-4’s accuracy across categories and question structures suggest that LLMs are competently able to process clinical scenarios but remain incapable of understanding these clinical scenarios. Large-Language-Models incorporated into practice alongside a trained practitioner may balance risk and benefit as the necessary robust testing on evolving tools is conducted
Assessment of factors that students perceive to affect their virtual learning of clinical skills for OSCE
No full textPurpose: Objective Structured Clinical Examination (OSCE) is a vital examination that must be passed to graduate as a qualified doctor. The delivery of OSCE teaching was changed to an online format to accommodate COVID-19 restrictions. Therefore, this study evaluates factors that students perceive to affect their virtual learning of clinical skills for OSCE.
Methods: In this cross-sectional study, all medical students from across the world who attended “The Respiratory Station” session delivered by OSCEazy (a medical student organization providing free online medical education) in the academic year 2020– 2021 received an online questionnaire about their perceptions of this learning opportunity. The survey was created on Google™ forms and consisted of 5-point Likert scales as well as free-text boxes.
Results: A total of 556 responses were received (mean age: 24, female: 76.6%). Most students agreed that online OSCE teaching offers more flexibility and convenience (median: 5, IQR: 4– 5) but their likeliness to ask questions in either format was similar (median: 4, IQR: 3– 5 vs median: 4, IQR: 3– 4, p value: 0.94). The use of visual aids (median: 5, IQR: 4– 5) and breakout rooms (median: 3, IQR: 2– 4) were thought to enhance the quality of virtual OSCE teaching. The biggest concern about online teaching was access to a stable internet connection (69.1%).
Conclusion: The flexibility and convenience of virtual OSCE teaching enables the sharing of knowledge and skills to a wider audience and thus may be a very useful adjunct to face-to-face OSCE teaching in the future
Challenging pedagogical styles of teaching: Amalgamating blended learning with near-peer teaching for integrated structured clinical examination (ISCE) preparation
No full textIntroduction
Near-peer blended learning is a method of combining online virtual teaching with traditional face-to-face lectures delivered by senior students. In this study, we wanted to evaluate the student perceptions of this novel method of education and characterise its overall effectiveness.
Methods
A hybrid Integrated Science and Clinical Examination (ISCE) revision session was hosted simultaneously in-person and online. The session was delivered by fourth-year medical students to over 100 second-year medical students. A survey was sent to students prior to the session to deduce the topics they required most help in. Feedback forms evaluated students’ confidence, satisfaction with the format of the session, and overall student engagement using blended learning with free text responses which recorded individual opinions of students. Tutor perceptions were also obtained using feedback forms that assessed knowledge, confidence, whether they would attend, and teach a hybrid event in the future and their personal preferences on teaching styles.
Results
The results showed that students had a much better understanding of the content after the session and the format was deemed well-suited for delivering ISCE teaching (p = 0.001). Thematic analysis revealed students preferred a hybrid approach for ISCE teaching considering it catered to a wider audience. Live demonstrations that are not well executed in virtual formats are compensated for with in-person teaching. 96% of participants expressed that the material was aimed at their level due to relatability achieved through near-peer teaching. Three of the four tutors preferred a hybrid method of teaching while one preferred a sole online model. An overall increase in tutors’ knowledge and confidence was seen after the session.
Conclusion
Overall positive feedback provides evidence that blended learning using near-peer teaching should be utilised by educational institutions with the potential of revolutionising the delivery of medical education in this post-pandemic
Assessing the publicity and reach of peer-led online medical teaching: A single-event evaluation
No full textPurpose: The COVID-19 pandemic caused significant disruption to higher education, including medical studies. Online learner support became urgently necessary, and peer-support of learning was needed to supplement formal teaching. This study evaluates the impact and delivery media of OSCEazy, a student-led initiative supporting formal teaching across institutional and national boundaries.
Methods: A survey was created on Google Forms and disseminated via the Zoom chat function to all the participants at the end of our event titled, “The Cardiology Station”.
Results: A large proportion (99.5%) of the students have a Facebook account and (98.7%) use it to keep track of medical events (p=0.45 for comparison). However, a very small proportion of use their other social media to keep up with medical events (p< 0.0001 for comparison).
Conclusion: It is evident from our results, that most of our participants used Facebook to keep track of medical events. The use of their other social media platforms (Instagram, etc) to keep track of medical events was limited. Thus, it is important for any organisation, such as ours, who conduct online events to evaluate which platform is most suitable for them to use to publicise their events
Enhancing medical students` confidence and performance in integrated structured clinical examinations (ISCE) through a novel near-peer, mixed model approach during the COVID-19 pandemic
No full textBackground: Near-peer medical education serves as an important method of delivering education to junior students by senior students. Due to the reduced clinical exposure because of the COVID-19 pandemic, we developed a mentorship scheme to help medical students with their Integrated Structured Clinical Examinations (ISCEs) by providing a combination of near-peer mentorship together with lecture-based teaching on a weekly basis for a 12-week period. Students attended a specialty-focused lecture every Tuesday followed by a small group teaching session organised by their tutor. Methods: A longitudinal evaluative interventional study was undertaken by the international student led medical education organisation, OSCEazy. The teaching programme was organised and conducted by third year medical students to a recruited cohort of second year medical students. Students’ perceptions of ISCEs (confidence, anxiety, and overall performance) were evaluated using 5-point Likert scales while their knowledge of the specialty was assessed using 10 single best answer questions which were distributed via Google® forms at the start and end of each week. In addition, we assessed tutor perceptions of their teaching and learning experience. Results: Seventy-two tutees were enrolled in the programme (mean age: 24.4, female: 77.8%). 88.9% of the participants had not attended any online ISCE teaching prior to this. They preferred in-person ISCE teaching as compared to virtual sessions [median 4.5 (IQR 4–5) vs 3 (IQR 3–4), p < 0.0001), respectively]. There was a significant overall increase in knowledge when comparing pre-session and post-session performance [mean 53.7% vs 70.7%, p < 0.0001)]. There was a significant increase in student confidence [Confidence: median 3 (IQR:3–4) vs 4 (IQR 3–4), p < 0.0001] while no change was seen in the anxiety and perception of their overall performance in an ISCE. [Anxiety: median 3 (IQR 2–4) vs 3 (IQR 3–4), p = 0.37, Performance: median 3 (IQR 3–4) vs median 3 (IQR 3–4), p < 0.0001]. The tutors reported an increase in their confidence in teaching ISCEs online [median 3 (IQR 2–3.25) vs median 4 (IQR 4–5), p < 0.0001)]. Conclusion: Online near-peer teaching increases the confidence of both tutees and tutors involved while enhancing the tutees’ knowledge of the specialty. Thus, medical schools should incorporate near-peer teaching in their curriculum to enhance the student learning experience
