HD-OPM Archive
Topics › How to read research

Absence of evidence is not evidence of absence: weighing unproven interventions

Louis Nervegna · Oct 30, 2026 · ♥ 6 reactions · 1 comments · original Facebook post
How to read research
AI summary & key facts

Louis explains that an intervention that has not been properly studied is indeterminate, not proven ineffective, and that this differs from interventions proven not to work. For speculative items such as supplements, he suggests assigning a subjective probability from limited evidence, while warning against betting your life on indeterminate interventions.

“Absence of evidence does not mean evidence of absence.”
Translation: Just because an intervention hasn’t been properly studied, does not mean that it fails to work. It simply means that it is indeterminate. Nobody knows for sure.

It is important to distinguish interventions that fall into this indeterminate category from interventions that are definitively proven not to work.

A lot of the more speculative things that we talk about in the group fall into this indeterminate category, e.g. supplements. Maybe they work. Maybe they don’t.

The best you can really do is try to assign a subjective probability to whether or not they work based on the limited evidence which is available.

This is sometimes referred to as a Bayesian approach.

The subjective nature of interpreting this limited evidence means that this assigned probability will vary from person to person. Ultimately, you have to make this assignment for yourself.

Needless to say, it is probably unwise to bet your life on things that fall into this indeterminate category.

#evidenceHD

#supplements
← newer: Green Mediterranean diet cut visceral fat about 14% in an 18older: Semaglutide extended lifespan 12% in elderly female mice, hi →