4 ms·
These studies are conducted using nutrition surveys typically given out every year (or five) asking the subject about their eating habits over the last period.
by dacox 4y ago
These studies are conducted using nutrition surveys typically given out every year (or five) asking the subject about their eating habits over the last period.
They are notoriously bad data, and also do not account for the healthy user bias (or the inverse effect).
Of course that doesn't seem to stop researchers from using them to tease out extremely small effect sizes
- boltzmann-brain 4y agowell said.
- hombre_fatal 4y agoYou haven’t asserted how it’s bad data. Perhaps it’s also inconvenient data?
- brokenkebab2 4y ago"...asking the subject about their eating habits over the last period"
- hombre_fatal 4y agoWell, that sounded like a blanket dismissal of food logging rather than a specific criticism. Which study asked people to recall what they ate a year ago?
- drekk 4y agoWe know separately from this study that red meat is carcinogenic. https://www.cancer.gov/news-events/cancer-currents-blog/2021/red-meat-colorectal-cancer-genetic-signature#:~:text=Evidence%20Mounts&text=In%202015%2C%20based%20on%20data,probable%20carcinogen%20(Group%202A) https://www.cancer.gov/news-events/cancer-currents-blog/2021....
- plonk 4y agoIt’s very annoying that this article mixes “red and processed meats” like they are the same thing. The author themselves mention that there’s much better evidence against processed meat, but keeps talking about how dangerous “red and processed meats” are. The author clearly can’t be trusted. Besides, that study has the same issue GP mentions: they just asked people how much meat they ate.