Prof. John Ioannidis is probably the most capable doctor in the world nowadays, also extremely good in Mathematics&Statistics and absolutely integer and human. We need people like him urgently. Thank you Professor!
When the worldwide event began in late 2019/early 2020 I looked for this man's opinion immediately. I am an average guy, no degree on the wall and in 2021 many may call me a bum. Call me what you will but I am without any debt, no children by choice as I was aware of the world not being a place in my opinion to raise a child. I am one that surround myself with my books of old and I am selective when reading anything new. Corruption in medicine due to greed is disgusting. I was studying to be a biomedical engineer and walked out as in 2000 the writing was on the wall that money was the most important aspect for those in that career so I couldn't do it and would not throw my soul out for a couple hundred grand a year. This man is a true doctor and a good man. No one is perfect and that is why evidence based medicine is the topic at hand right here. His videos from early 2020 regarding covid were buried and shadow banned. Why? He is not a sell out like most in medicine. Pay attention and you will learn from a wonderful human being. Thank You Doctor.
@@Fomites manipulating data during clinical trials for medicine.Publishing only positive trials and ignoring the negative trials.Selection bias like cherry picking data due to conflict of interest.This applies to science but very specifically applies to medical science and pharmacy industry published clinical trials
Nutrition Epidemiologists were criticized for using models and not testing them properly by Gary Taubes when the low fat high carb diet was exposed as pure BS. Today we are seeing what happens again when we follow Imperial College’s modellers down the drain.
Wonderful how a mainstream prof at Stanford is questioning the whole mainstream paradigm and vested interest culture of mainstream medicine and medical practices. I'm surprised though he hasn't yet got the boot.
on the metric main page there is an entry on exactly why we should go slow w/effect sizes bc psych is already too wrapped up now, and they are becoming exaggerated & independent replication has brought up some novel problems (new is old again re sampling)~5min clip: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-CltcG-wKrV4.html
Professor Ioannidis is at the same one of the most hot proponents of the NHST and goes against "ditching statistical significance" strongly, so it's not that easy. He proposes other measures to address it, but stays within the NHST. BTW effect size has little or no connection to the MCID - the practical significance (minimal clinically important difference), which is the major filter of clinical relevance. Moreover, different methods of calculating of the effect size (for parametric and non-parametric methods) may (and usually do) yield very different outcomes. People overrate effect sizes and quarrel on their "thresholds" no less fervently than about any other measure.
18:10 Conflict declarations: Are you declaring you are bowing out of this field (& any influence on guidelines youve had should to be removed). Announcing you are "conflicted" doesnt allow judges to rule on their own cases. And science best have an even higher standard.
@@Dan-iy8ig In a 2012 meta-analysis paper co-authored by Ioannidis, entitled "Effectiveness and harms of seasonal and pandemic influenza vaccines in children, adults and elderly", the primary conclusion was (Ioannidis's own words): "Most influenza vaccines have been shown to confer some protection against naturally acquired infection and no evidence for major harms has emerged. In adults and children, the efficacy/effectiveness of current seasonal vaccines was generally high for laboratory-confirmed cases (especially for LAV in children aged 2-17 y), and modest for clinically-confirmed cases and for the elderly." - Dr. John Ioannidis
So true.. actually 90% of socalled "research" is NOT out of necessity but for personal gain, promotion, recognition or monetary compensation by pharmaceuticals.
Discovery is a nuisance 96% of clinical papers since 1990 documented “significant” p-values and claimed to discover something new... Most “discoveries” don’t end up with any new “solutions” to help change clinical outcomes There’s a difference between what is “statistically significant” and what is “clinically significant” but we don’t have a standard way to differentiate the two yet...
In Engineering, we have to consider a factor clearly missing from Professor Ioannidis' claim of overreach in the reaction to the Coronavirus pandemic, which is the "significance of failure." I applaud your contributions to society, I deplore your dissemination of erroneous accusations of overreach, which I also question as a politicization of a tremendously horrific occurrence in the already limited existence of mankind.
Incredibly poor math for an engineer. The virus has run it's course in China. 3,200 dead. That is less than 1/10th of the death toll of the flu in America during the same time interval, 34,000 plus. Wuhan has na population of 12,000,000 and is the transportation hub of China, nicknamed the Chicago of China. Taking into account the housing, sanitation, food handling, and medical fields of China, this pandemic is the definition of hype and propaganda.
