Hacker News Reader

I think in general the advice is "preprints are not yet holy gospel". Don't start preaching on it, don't start basing your life on it.

View on HN

Look, even peer reviewed papers are not supposed to be treated as any kind of unquestionable truth, especially in medical sciences. It is unfortunately very common to have papers with repeatability issues or even outright wrong. Peer review just means that experienced scientists read a paper and concluded that nothing they saw was outright wrong.

Neither are published papers in peer reviewed journals, we need to always treat everything with healthy skepticism. That obesity and bad covid outcomes happen is beyond any doubt, the interesting part for me in this paper is that they try to explain the mechanism for obesity itself being a problem, rather than say diabetes or atherosclerosis.

I guess my whole point was if they're trying to explain a mechanism, their sample size doesn't generalize to the population; even limiting it to the obese population. If anything they've shown, in their small cohort, that the hypothesis is worth further study.

Having spent a lot of time in academia before moving into biotech, I think this is a pilot study and they're fishing for more research funding. The way those funding mechanisms work, I tend to view manuscripts like this with a critical eye.

Reply on news.ycombinator.com