Ultra-processed junk science on ‘ultra-processed foods’

A new “study” from the Chinese-owned and operated Harvard T.H. Chan school of Public Health claims that consuming so-called “ultra-processed foods” is associated with a four percent (4%) increase in all-cause mortality. It’s a bunch of crap.

The study is junk science for any and all of these reasons:

1. Accepting the correlation at face value (a relative rise of 1.04), that is close enough to zero (i.e., a relative risk of 1.0) to be regarded as a zero correlation. Before the rise of modern junk science epidemiology, epidemiologists disregarded correlations until relative risks were at least on the order of 2 or 3.
2. But we don’t accept epidemiological results at face value.They are merely statistical exercises (not science) that prove nothing.
3. The data used is the Nurses Health Study, which is a perpetual junk science machine. It consists of self-reported dietary “data” from a cohort of nurses. The data is not verified or verifiable. No one knows for sure who ate what and how much. So the data quality is unknown.
4. The term “ultra-processed food” is a relatively new term for attacking the food industry. Data in the Nurses Health Study predates “ultra-processed food” by a lot and so the researchers have manipulated old data into new food classifications. More uncertainty and opportunity for manipulation.
5. Epidemiology is useful for studying higher rates of rare diseases. None of that is present here. This is a study of a very low rate (relative risk = 1.04) of a common endpoint (death). The data is too imprecise for such a small statistical claim.
6. Finally, there is no demonstrable biological plausibility for the notion that “ultra-processed food” increases risk of death.