Vitamin D destroyed my deep sleep

I want to tell the whole story of how I figured this out, but I will just cut to the data.  Here is a graph of the amount of deep sleep.  Its almost a year of data (Jul 30, 2011-Jul 8 2012).



You'll notice that the first night I had 54 minutes of deep sleep which is the highest amount of deep sleep in the time period.  There was one other 50+ night  and 9 40+ nights of deep sleep and they were all before Aug 27 2011 (the red line).

For perspective, according to Zeo, the average amount of deep sleep for males aged 30-39 is 59 minutes.

After the red line, my deep sleep cratered. It hovered in the 10-20 minute range. Except for one period from the middle of January to February, it stayed that way for almost a year.

I have done all kinds of tests in the year trying to improve my deep sleep and haven't experienced much success. So what was the red line?  It was when I switched from 4000IU to 5000IU Vitamin D every day.

There has been research suggesting that a Vitamin D deficiency can cause sleep issues. In March of 2011 (before I switched to 5000IU) I was tested at 44ng/mL...so no reason to think I was deficient at any point in the sleep tracking.

In the alternative health blog world, there is a lot of discussion that Vitamin D timing can affect sleep quality.  In most of 2011, I took it at night.  Most of 2012, I took it in the morning.  You can see from the graph it didn't make much of a difference.

I haven't seen anything before that suggests too much Vitamin D (at any time) can cause sleep issues. Also, to be clear, I had my Vitamin D level tested in December (after 4 months of 5000IU).  It was 89.5 ng/mL - well under the level considered excessive (100 ng/mL).

But I still think too much Vitamin D is what caused the decrease in deep sleep.

I made this determination on July 8 (the other red line).  I stopped taking any Vitamin D and put on suntan lotion when I went outside.  The halflife of Vitamin D in the body is a month.  So what happened a month after I stopped?



My deep sleep went way up.  That's what happened.  I have had multiple nights of 50+ minutes of deep sleep and my first ever 60+ minutes of deep sleep. I have gotten more deep sleep  in the past 11 nights than the last 33 nights while I was taking Vitamin D. Pretty awesome.

I went and had my blood tested the first time my sleep broke 50 minutes.  It was 55.5 ng/mL.

I'd love to hear any comments/questions. I think a lot of people in the paleo/alternative health world are taking a lot of Vitamin D so I think it would be great

EDIT:  I updated the post to go into a little more detail about what time of day I took Vitamin D.

Magnesium helps me sleep (Potassium may help even more)

After 20 days of taking a bath in Epsom Salts (MgSO4), I definitely went into deeper sleep than I have.  This has been one of the more clear results.

Column1	nothing	mask/earplugs/magnesium pills	magnesium bath	magnesium bath And Potassium
z score	75.6	71	67.2	69.3
deep	37.4	32.5	30.2	32
awake	17.4	20	13.5	10.8
awoken	3.6	4	2.8	2.5

Clear reductions in # of times awoken and time awake.  It includes a recent night that had 0 wake ups.  Unprecendented.

You can also see that when I started adding potassium to my diet the time awake and times woken were even better.  I have been high potassium for 13 of the 20 bath taking days.  I don't know if the addition of Potassium was truly a help or whether that was just a result of taking the bath consistently.  I am planning on doing both for the time being.  I figure pretty soon there will be a time where I can't do both and I can see then if the effect is synergistic or not.  Not going to do an experiment just for this.

What is also interesting is the lack of improvement in deep sleep.  The two (awake time and deep time) are clearly not perfectly correllated.  A subjective analysis shows that sometimes I would go into deep more than once (Something I rarely did before) and rarely woke up in the middle of the big initial deep sleep (Something I did often before).

This makes some sense since deep sleep is the physical recuperation portion of sleep and I haven't really increased my physical exertions.  Perhaps it would increase if I exercised more.  Staying asleep certainly isn't enough.

I plan to continue with Magnesium and Potassium.  The next big sleep thing will be using the emWave2 that arrived today.  I am skeptical on this one for sleep.  I generally have no problem falling asleep.  If my theory on deep sleep is correct it seems like it wouldn't affect that either.

I may expand to starting Dual N-Back training as well,  I have enough cognitive performance data now and there are reports it increases REM sleep.  Starting emWave and N-Back Training at the same time may be a bit too much right now.

Back to the Drawing Board on sleep experiments

Upon the collection of further data, I have basically made no progress on improving my sleep.  In fact, in general, I am worse than I was during the control week.  Check it out.

Column1	nothing	just mask	any mask	mask/ear plugs	mask/earplugs/magnesium
z score	75.6	72.2	74.4	74	71
deep	37.4	36.8	35	34	32.5
awake	17.4	22.8	17.8	17	20
awoken	3.6	3.2	3.4	3.4	4

Its not the proudest I have ever been of a table, but the results are pretty clear if you take a second.

Lets talk about the magnesium a second.  I noticed one thing....I pooped more.  Turns out the poop part is very common and means it isn't really getting into your system.

In follow on research, I have become convinced that I still am not getting enough magnesium (most people aren't).  And if you are deficient, in a cruel twist of fate, it can be hard to get it orally (thus the squirts). 

So now the next step...ditch the sleep mask, earplugs and take epsom salt baths to get magnesium topically. 

Sleep updates - Sleep Mask/Earplugs = good, caffeine = bad

So wearing a sleep mask and earplugs definitely reduces the number of times I wake up (3.6 to 2.5) and the amount of time awake (17.4 to 12.8).  However, it didn't seem to do much for the amount of deep sleep I got (37.4 to 38.3).  I rarely felt tired during this time (didn't really want to take a nap), but I also didn't feel particularly energetic.  I don't have a better way of describing it.  There was also no real increase in productivity during the time of this improved sleep.

These results are confounded a bit by the fact that I also cut out caffeine for all but one day of this trial.  I don't think the results are largely due to that though.  One day, I intentionally drank an above average amount of caffeine and that was the night I only woke up once for two minutes and had the second highest amount of deep sleep ever.


I did notice that I had a headache for three days after quitting.  Which seems like withdrawal...that isn't cool.  I also noticed I didn't have the afternoon dip after I got through the headaches.  So I probably will continue to avoid caffeine even though it doesn't appear to directly affect sleep.

Bottom line....a lot of waking/wake time is bad for deep sleep, but there is clearly more to increasing deep sleep than just sleeping through the night.

I am going to try experimenting with 800mg of Magnesium now.  There is a lot of anecdotal evidence in forums and from Tim Ferriss that it improves sleep.

Orange Glasses don't work....but Sleep Mask might

There is no connection between the amount of deep sleep I get and the Orange Goggles (37.4 vs 37.7 min) The theory that I need to wind down more to get more deep sleep is not true. No matter how I sliced up the data it took about 50 minutes before I hit deep sleep.

In retrospect, this makes a little sense. If there were any problems with me and light...It is probably that I am not getting enough rather than too much. May explore more time in the sun in a future experiment.

So it must be interruptions that are taking me out of deep sleep.

There is clearly a connection between the number of wake ups, time awake, and deep sleep. The x-axis below is number of times to wake up. You can see amount of deep sleep decreasing as the number of wakes.

The trouble is the Sleep Mask doesn't appear to be doing the trick. It decreases the average number of wake ups (4.3 to 3.2 per night), but doesn't significantly change the amount of deep sleep (36.8 to 38 min night). It actually appears to have increased the amount of wake time (18 to 22.8 min).

My only guess is sound. I live in an urban area next to a train track. I haven't personally noticed noise in the night but it is all I got.

I'll start tonight wearing earplugs along with the sleep mask and see what happens.

Orange Glasses sure feel like they work....

But I'll be damned if I can prove it.

Time to sleep didn't really change.  I am not the type to go to bed till I am ready so that seems unlikely.  My time to go to bed didn't change much either though.  I notice I yawn when I wear them, but I am not looking to increase my yawn count.

All that being said, I do think it gives me a drowsy feeling that increases the likelihood that I go to bed at a decent hour.  It may be placebo, but that is good enough for now.

However, there is no doubt that there isn't any increase in deep sleep.  In that respect, the experiment was a failure.

Still learned quite a bit.  The amount of deep sleep directly affects how I feel in the morning/throughout the day.  The only day I felt noticeably better was the day I got almost twice as much deep sleep as I did any other day (even though it was the lowest total amount of sleep since I started tracking). 

The current hypothesis then has to be that something (light/noise) is interrupting my deep sleep.  Since I am more comfortable with the idea of wearing a sleep mask rather than earplugs/white noise machine....we'll start with that.

For the next three nights (may have to be four since I will probably skip the night when i have houseguests), I will continue with orange glasses an hour or so before bed and a sleep mask before I go to sleep.

First four nights of data and the first sleep experiment

Have completed the first four nights of data collection with the Zeo.  Initially, I had intended to get a whole weeks worth of data before experimenting began.  However, there are a few patterns that are so obvious I am going to go ahead and work on them.

First, the data with a little commentary

Friday night...note the transition from Deep to Light right before 2.  This was the result of some drunks outside my window.  Notice all the waking that happens after 6.

Saturday night (Sunday morning) I took a nap that afternoon so sleepy time didn't happen till 4.  Notice the near total lack of Deep (I'm not sure if something happened at 5 to wake me up)  Again, notice all the waking after 6.

Sunday night - Worked out during the day.  Didnt seem to cause any notable change.  41 minutes of Deep.  Again, notice all the waking after 6.

Monday Night -  Almost don't want to count this night since I woke up a little before 6 and instantly went to check the data.  Feel I may have thrown it off. 37 minutes of Deep.

So two issues instantly pop out (you see where this is going?).  The waking up around 6....I can only assume is related to sunrise.  This is a surprise.  I would not have guessed this would be an issue (which is why I didn't have a sleep mask on the things I wanted to test) Except for Monday, whenever I go to sleep I start spending a lot of time awake after 6. 

Second is the low amount of Deep.  I haven't felt especially refreshed after any of these nights.  My REM sleep is above average (on the high side in fact) and my Deep is on the low end.  That makes me think  the lack of Deep sleep is preventing me from feeling great.

The sunrise issue can probably be dealt with pretty easily.  I already have a sleep mask and if that showed great results, I have already priced out some black out curtains.  However, I dont think that is going to be too helpful.  On workdays, I generally don't sleep too much past 6 anyway and it would probably only add to my REM or light sleep which I am getting plenty off.

I am going to test the orange glasses as a way of increasing Deep sleep.  The theory being that my brain will be shutting down while I wear them so when I finally hit the sack I will drop down to Deep.  I am going to try to wear them at least an hour before going to bed. 

If the theory is correct, than the results should be pretty instantaneous.  Next entry will be after three nights of data.

Sleep Hacking

While I intend to start measuring for everything simultaneously....the first active work I am going to do is on sleep.  I think it will affect all the metrics positively, and it will also be the easiest to measure/play with thanks to the Zeo I ordered.

I will collect a week of control data.  Then I will test a number of the ideas I have heard for improving sleep.

• Orange Glasses - I already own Orange Glasses and believe they do help.  It will be nice to have data to confirm deny the effects. (Here is a way to get them cheaper)
• Standing on one leg -  I have my doubts.  Will be probably be a bit more convenient than the glasses if it has an effect
• Pzizz - This will be at the bottom of the list of things I try.  It costs money.
• Earthing - Sounds very shady. Also costs some money so will be at the bottom of the list
• Meditation - may try this, but quantifying the meditation part may prove to be a challenge

Expect the Zeo to be here any day now.  Will get started after that.