Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The effect sizes are implausibly large. Breastfeeding would not be a live debate if the causal effect was that large. For example, a 36x risk of SIDS is really just too much to believe. This lit review: https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/apa.13...

Doesn't mention any studies with effect sizes larger than 4x. And it is concerning habitual breastfeeding or other feeding treatments. If breastfeeding a single time provided a 36x safety factor, these kinds of results would be impossible. A 14x reduction in infant mortality is similar far beyond the reasonable belief.

Personally I wouldn't toss out a claim like that without strong evidence to cite.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: