← Back to context

Comment by Traubenfuchs

3 months ago

The real shocking information I gained from this paper is that the AIS goes from 0-24 (0 = perfect sleep, 24 = total insomnia) and the study participants had an amazing average AIS score of 4.3 (SD 3.3)! Wow, how well all those people must sleep!

As someone scoring 12, it's pretty bad and I am suffering a lot while trying to sleep and during day time because I did not sleep well.

If my understanding of statistics, standard deviations and the standardized partial regression coefficient are correct, potassium supplementation in the evening only DECREASES this score by about 0.2178 (Beta −0.066, multiplied with SD of 3.3), which is kinda worthless.

Please correct me if I am wrong.

4 comments

Traubenfuchs

Reply
OutOfHere  3 months ago

It decreases the score. It doesn't increase the score. An increase would be harmful. The beta is negative. The abstract is wrong.

It is not worthless. For good sleep, potassium levels have to be adequate. Once one improves the level, one can move on to other factors.

As for what works for me, avoiding caffeine after 12 pm helps, as does sunlight exposure in the daytime.

With regard to a supplement stack, these help: collagen hydrolysate 12g, magnesium citrate, calcium, B6 as P5P, melatonin 4 mg, L-theanine 100-200 mg, and various sleep promoting herbs.

Ensure your BP is optimal, well below 120/80 for most people under 70.

Uncorrected acid reflux too worsens sleep, but avoiding consumption in the last three hours and also famotidine help.

  • Traubenfuchs  3 months ago

    > It decreases the score.

    Yes DECREASE instead of INCREASE. What I was going for in my head was "improve" I guess. Thanks for pointing that out.

    > B6 as P5P

    Don't forget it's one of the few nutrients that accumulates and that you can get too much of. It causes nerve damage and mystery sores.

    > melatonin 4 mg

    That's probably too much:

    https://www.lesswrong.com/posts/E4cKD9iTWHaE7f3AJ/melatonin-...

    • OutOfHere  3 months ago

      The P5P form harms less if in excess. This is in contrast to the default form which is pyridoxine. I have found 20 mg of P5P twice daily to be quite useful for keeping stress spikes in check.

      Regarding melatonin, newer data up to 2024 in https://pubmed.ncbi.nlm.nih.gov/38888087/ confirms that 4 mg is an optimal dose for sleep. For a discussion, search for this article on r/FoodNerds.