In this episode we review Bayesian Reasoning in general, and Nic get’s the opportunity to geek out on talking stats. We cover likelihood ratios, positive and negative predictive values, sensitivity and specificity, and pre- and post-test probabilities. Hopefully, we used examples that will help you understand and remember these concepts.
Episode 57: Bayesian Reasoning
February 09, 2020
Hey guys, really enjoying your podcast. I was wondering if you had the link to the website that gave positive and negative predictive values for different signs/findings?
Nick and Art- I enjoyed your podcast on Bayesian Reasoning (just like I enjoy all of your podcasts!) and the sidebar entertainment provided by your discussion of FIAS/Fatal Itchy Arm Syndrome. I was further entertained while reading MKSAP 18 last night and stumbling across a more benign form of FIAS, known as BRP/BrachioRadial Pruritus. an actual entity: “Brachioradial pruritus (BRP) is a localized neuropathic dysesthesia of the dorsolateral upper extremities. It is commonly seen in middle-aged white females with a seasonal predilection for warmer summer months. Cervical radiculopathy or neuropathy in the upper extremities in conjunction with ultraviolet radiation (UVR)… Read more »
Hi Nick and Art
Loved your Bayes episode.
It is something that I have thought about a lot and gave this talk a few years ago at SMACC:
https://broomedocs.com/2016/11/smaccdub-bayes-2016-diagnostic-odyssey/
I think you will like it – another attempt to translate Math into English!
Likelihood ratios are best thought of as a measure of “signal to noise” when you get a result. They tell you how much noise you might be seeing when you get a ‘positive’ or a ‘negative’ result