[atomical]

> "...eating 12 hazelnuts daily (1 oz) would prolong life by 12 years (ie, 1 year per hazelnut), drinking 3 cups of coffee daily would achieve a similar gain of 12 extra years, and eating a single mandarin orange daily (80 g) would add 5 years of life. Conversely, consuming 1 egg daily would reduce life expectancy by 6 years, and eating 2 slices of bacon (30 g) daily would shorten life by a decade, an effect worse than smoking. Could these results possibly be true?"

Why would the author expect all research to be useful and life changing? There is a lot of noise but there is noise in every aspect of life (investing, programming, etc). Do Charlie Munger and Warren Buffett complain about the news and say it's useless because there are a few bad articles?

[Err_Eek]

I think the point was that these studies aim to make a serious statement about serious topics, but they rarely get to be properly reproduced and tested. And when they're not challenged they can creep into our official medical guidelines, where they become doctrine, and it can take decades to convince people that not all fat is bad for you or that children with mild allergies should be exposed to allergens, and that a coffee per day will not cause high blood pressure.

[DataWorker]

The scientific process as it stands at present is greatly overestimated in its ability to reach firm conclusions. It comes down to trust. Do you trust the studies that say fat and salt and alcohol have no deleterious effects? Do you trust meta-analysis because “more must better”? These issues are coming to a head and I think we’re in a transition right now but I don’t know what’s next. Some want to abandon significance testing, some think frequentist approaches should be discarded in favor of Bayesian. Meanwhile people cling to the dogma of scientific rigor because without that, what do we have. Whether the flaws in our current efforts are endemic or the result of economic factors is unclear. Publication bias is clearly an economic problem, but the problems I worry most about are things like measurement error and clerical mistakes like this https://www.bloomberg.com/news/articles/2013-04-18/faq-reinh...