Professor David Spiegelhalter from Cambridge University discusses effective risk communication, challenges with p-values, and evaluating medical AI. They explore heart age, patient empowerment in health communication, and the impact of scientific reporting on misinformation. The podcast advocates for rethinking statistical significance and emphasizing evidence quality in medical research.
Heart age helps convey cardiovascular risk effectively by comparing individual risk to healthy average.
Rethinking use of p-values in medical research, advocating for nuanced interpretation beyond significance criteria.
Deep dives
Communication of Risk and Heart Age Concept
The podcast delves into the importance of effectively communicating medical risk to the public. Professor David Spiegelhalter discusses the concept of heart age as a tool to convey cardiovascular risk to individuals. Heart age aims to illustrate a person's risk level compared to an average healthy individual, potentially motivating behavior changes. Despite its popularity, the heart age calculator has faced criticism for not incorporating fitness data, highlighting the need for comprehensive risk communication tools.
Challenges with Statistical Significance
The episode addresses the limitations of relying solely on the p-value and statistical significance in medical research. David emphasizes the dichotomization of p-values as significant or non-significant, advocating for a more nuanced approach to interpreting results. He critiques the flawed practices of overemphasizing statistical significance and the misuse of p-values in medical publications. The discussion highlights the importance of focusing on the quality of evidence rather than solely relying on p-values.
Evaluation of Machine Learning in Medicine
David expresses skepticism towards the exaggerated claims surrounding artificial intelligence and machine learning in healthcare. He emphasizes the need for a critical evaluation of AI systems' trustworthiness and ability to explain their decisions. While acknowledging the potential of machine learning in image analysis, David urges caution in overstating its benefits in diagnosis and prognosis. He advocates for rigorous evaluation processes akin to clinical trials to assess the true impact and reliability of AI technologies in healthcare settings.
In this episode of the Heart podcast, Digital Media Editor, Dr James Rudd, is joined by eminent statistician Professor David Spiegelhalter from Cambridge University. They discuss how we can better communicate medical risks to our patients, why the p-value may have had its day and how we should evaluate medical AI before using it in daily practice.
If you enjoy the show, please subscribe to the podcast to get episodes automatically downloaded to your phone or computer. Also, please consider leaving us a review at https://itunes.apple.com/gb/podcast/heart-podcast/id445358212?mt=2
Link to David’s website and his book on the art of statistics :
https://wintoncentre.maths.cam.ac.uk/
https://amzn.to/2TQyhSb
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode