Earlier quoted context omitted.
Medical doctors are no more trained to digest scientific evidence than anyone else. Especially when it comes to population epidemiology. Statistical illiteracy is widespread among medical doctors.[1] There's no reason to trust a doctor to interpret a p-value. For these types of questions you're much better off listening to a data scientist because they have actual training in interpretative statistics. [1] https://ww…
> Medical doctors are no more trained to digest scientific evidence It's about understanding and processing this evidence in the broader context of their medical training. You're trying to generalize the results of a study with many shortcomings to say everyone may be better off dealing with increasing their odds of developing malignant melanoma and dealing with its associated mortality risks. My issues with using th…
One very consistent pattern with vitamin D research is that association studies will find a major relationship between serum vitamin D and some health outcome.[1] But an RCT using vitamin D supplementation will find no effect.[2]
That strongly suggests that serum vitamin D, at least as we measure it, is merely proxying for something else. The map is not the territory. There's an X-factor that's related to serum vitamin D, but is not just serum vitamin D. Artificial supplementation doesn't work. Since the vast majority of population vitamin D variance is related to sun exposure, that would strongly suggest that sun exposure is the X-factor that improves health.
[1]https://www.cambridge.org/core/journals/proceedings-of-the-n... [2]https://jamanetwork.com/journals/jamacardiology/article-abst...