Multiple N=1 experimentation

On a separate thread there is an appetite for coordinating some more structured N=1 experimentation. On the assumption that @RapAdmin is happy for that to happen on Rapamycin News then I have started this topic.

There are two obvious areas of experimentation to me which seem to be useful if we can obtain structured results.

The first is one I have always thought was needed, but have not really seen although I have done it

That is the impact of Vitamin D supplementation on Serum levels. What this needs is a blood test for 25OHD levels then a record of supplementation (cholecalciferol or 25OHD or a combination of the two) and then a closing blood test of 25OHD levels.

Hence the variables are the number of days between blood tests, the total cholecalciferol taken in units (and the type of supplementation and manufacturer of supplement), similar for 25OHD, and the levels of 25OHD at start and end.

This enables people to have an idea of the range of outcomes that can occur with levels of supplementation.

The second obvious area is Rapamycin Dosing, Timing and blood levels. There are quite a few variables here. Those include whether taking boosters or inhibitors. This is something which is much harder to structure and would certainly require more than one set of data as there are so many ways of looking at this.

A subset of that may be linking Rapamycin to outcomes (again multiple areas). This could be glucose levels (variation from normal glucose levels), sleeping problems, blood pressure, heart rates.

Whereas with the vitamin D experimentation it is relatively easy to get structured results with some meaning this I think will be harder.

However, that starts the ball rolling.


Absolutely - we need to do a lot more of this. I’ve tried to get more focus on this issue a year ago… but did not see much interest.

See this former thread: The n-of-1 Clinical Trial: For Longevity Drugs?


I joined Rapamycin News in September 2022.

Why is the Vitamin D supplementation on serum levels important?
I can do that, I am very certain I am deficient because I haven’t been taking vitamin D regularly. So it is interesting for myself to know what level it is at and how it moves with supplementation.

I think it is important because there are lots of recommendations about how much to supplement. However, I am not aware of any experimentation that looks for individual results on the follow through from supplementation to serum levels. That really has to be the first thing to guide people.

In the end we can have a list of experiments people can try and results they can provide. However, each individual experiment is entirely optional. I don’t think we should pressurise people to try something that they don’t want to do.

My view is that the results for different people are likely to vary. I actually have a formula I can use to predict my serum levels based upon quantities of cholecalciferol and 25OHD.

I’ll do it. I just ordered some high dose vit d for 1x/week supplementation. I was getting lots of sunshine but … winter is coming. What do I need to do for the group?

The main thing to do is to have a starting blood test for vitamin D, and then take a known amount of supplementation and then have an ending blood test. You need to record the amount of supplementation taken and what you supplement with.

What we can build up is a data set which indicates what the likely range of outcomes are for particular supplementation levels.

1 Like

I think it’s important to record every day you take it, in case someone forgets a day when taking it daily. Rather forget a day and see it in a record than guess about it.

1 Like

Yes, it is important to record it every time you take it. Then you can work out the impact of forgetting to take it on a particular day.

1 Like

I would think each topic should have its own thread or this thread will get unmanageable. I would assume there would be discussion of topic before testing starts?
Obviously n of 1, data can be of debatable value.
(I couldn’t really tell if you understand the common meaning of n of 1.)


Now that I think about it, it’s probably important to not overly complicate things so more people would be interested. Instead there should be expected to be a margin of error. At the last test or so people can estimate how many days they had supplemented, etc, between the tests.

I tend to disagree here.

If people don’t actually record what they are doing as they go along their results are much less reliable. I would prefer only reliable data rather than more data some of which is unreliable. Hence personally I would ask that people keep contemperaneous notes.

1 Like

We could have both. A more regulated and stringent category and a more relaxed one. We could then compare results.

1 Like

Well there will still be a margin of error like if you take vitamin D with fat or not, somewhere a line has to be drawn. Other than vitamin D and Rapamycin, what else is on the table by the way for healthspan increase?

Okay the minimum can be to use an excel spreadsheet with date for rows, then columns for taking vitamin d or not? should there be column with fat?


I agree that there can be a more stringent category.

1 Like

Other than vitamin D and Rapamycin, what else is on the table by the way for healthspan increase?

There are other similar tests that could be done. For example Lithium supplementation vs Lithium serum levels, Citrate supplementation can be tested against lots of things, more recently I have looked at PSA (Prostate-Specific Antigen). I have noticed that as Citrate levels go up PSA goes down. That could be tested more widely as a quantitative test. Copper is perhaps less of a longevity thing, but people might find it useful to know what levels of supplementation result in what serum levels.


I certainly would be interested in participating in something like this. I already track a lot of details about my lifestyle and biomarkers. It would be great if this information could be made more useful by aggregating with other people’s data to test various hypotheses and discover non-obvious correlations. I would be happy to adhere to a specific testing protocol to make the results more robust.


Looking at the recent discussions I am thinking that starting with a Rapamycin orientated MN=1 experiment might be best.

What I am thinking is that looking at the lowest sleeping heart rate on the days before taking rapamycin and the week afterwards might be quite a comparable data set. If people are using a fitness tracker then finding the lowest heart rate is quite easy. Because people are asleep when this is measured there is less potential variation. There are things other than Rapamycin that impact on this (such as alcohol), but it strike me that if Rapamycin increases serum glucose, thereby the heart rate and thereby blood pressure the easiest way to measure this would be with the heart rate.

Any thoughts on this?

1 Like

If you press the cogwheel/settings while writing a post you can build polls, it would be easier to find out what people want that way.



Thank you for this. However, at this stage a discussion is needed rather than a poll.