I needed to disable self-signups because I’ve been getting too many spam-type sign-ups lately. Please contact me directly if you want membership on this forum. Thanks.

Forum Navigation
Please to create posts and topics.

Serum RA and Metabolic Syndrome

This paper stumped me: https://academic.oup.com/jcem/article/101/4/1686/2804551

Higher serum RA predicts better metabolic health:

Subjects with lower RA levels had a progressively worse cardiometabolic risk profile at baseline

It's not the most rigorous study, but the association is too strong to be ignored. Figure 1 shows a neat progression of progressively lower serum RA along side with metabolic syndrome components.

Anyone care to interpret this?

A measure is not a target - Goodharts law

Obesity is a measure of bad health but starving yourself excessively will wreck your endocrine system and put you an even worse state than before despite being a lower BMI

A higher serum retinoic acid level might just signify a higher metabolism with the body turning over more VA than if it were slower, and a higher metabolism is usually associative with good metabolic health

A study looking at long term retinoic acid supplementation on metabolic syndrome or specific events like heart attack/liver failure would be worth looking at because it would be closer to causation

Thanks for bringing the study up Rudi! Otherwise I wouldn't have looked up the 2002 study mentioned later in this comment.

I think there is nothing to interpret in that 2016 study, published 1:st of April since it is a joke.

Look at this quote from the discussion part of the study:

"Finally, serum RA concentration was measured only once at baseline of the study cohort. Therefore, we were not able to adjust for possible moderate fluctuations of serum RA during the follow-up, for example, due to increased intake of vitamin A-rich food."

They only measure serum all-trans retinoic acid concentration once which means they can't use it to predict anything since a point is just a single point. You need at least two points to even be able to do the simplest prediction. It is like trying to predict in which direction someone is traveling, or not, from a single GPS coordinate, it is just impossible.

I don't even think they tried to take into account the variability of serum all-trans retinoic acid between men and women with women seeming to generally have a higher serum level. See this 2002 study called: "Biological variation of retinoids in man" by Söderlund et. al. You can find the PDF on sci-hub using the DOI:


"We also found that all-trans retinoic acid  was signi®cantly lower in men and that 13-cis  retinoic acid was signi®cantly higher in men  than in women. To our knowledge, retinoic  acids have so far not been investigated with  regard to gender differences."

The same 2002 study also says that the variability between humans indicates in regards to the population indicates that repeated individual measurements are need:

"All I_i [Index of individuality] were less than 1, which indicates that traditional reference intervals are of limited value for diagnostic purposes. Therefore, it is important to stratify populations in order to obtain separate reference intervals for sub-populations and to cumulate data from samples from the same individual [18]."

PS. There are many awful studies out there and it can be difficult to see why they are awful in many cases.

Curious Observer and kathy55wood have reacted to this post.
Curious Observerkathy55wood