I stopped eating, I slept better
I fasted for 30 days, here is how my sleep changed
Some weeks ago, I made a promise. For thirty days, I quit food and water, for 22 hours each day.
I had done this in the past as well. But this time, I collected everything I can from heart-rate and sleep data to glucose levels and workouts.
I pricked my fingers over and over to measure my blood sugar. And jumped out of pain every single time, as if it was my first.
Today, I released all of this data online in this link.
While I was playing around with the sleep and glucose data, some of the results blew me away: It seemed like my heart-rate went 7% lower and how my deep sleep went 18% higher.
These changes also seem to go in the same line as some of the research done on fasting, which I will mention below.
In future posts, I will try to jot down other findings. For now, let us see how the heart-rate, as well as sleep, were affected by the cheapest trick: not eating.
Heart-Rate improves
Look at the Average resting Heart-Rate graph below provided by Oura:
The resting heart-rate drops down below baseline during the fasting period. A lower value of the resting heart-rate indicates better sleep, recovery and overall health.
At the same time, my heart-rate variability, for which higher values are “usually” better, went up during the fast.
In fact, this difference between Fasting and Baseline is so big that you can easily separate fasting days on a graph (red dots below)
Ok sure. But how much of this is actually significant?
To verify this I used the data from the Oura ring (you can learn how to download the Oura ring data in a previous post) I ran a simple permutation test. We can start from the hypothesis: Fasting and Baseline show no difference in the average resting Heart-Rate. And then, we can easily see that: it is unlikely that the difference is due to chance. The blue graph below shows how that difference is distributed, whereas the red line shows the original difference.
All of the data used here and the hypothesis testing can be found here.
Deep sleep increases
At a first glance, we can quickly see it in the Oura ring’s dashboard.
Again, this does not necessarily mean it is a significant improvement. Or does it?
So, let’s verify that deep sleep increased significantly.
In the same way we did for Heart-Rate, we can clearly see that the increase in deep sleep is significant:
All of the data used here and the hypothesis testing can be found here.
But wait, It could be that I was just sleeping more during this period, right? Let’s verify that too.
The total amount of sleep stayed the same during both periods
This increase in deep sleep was also accompanied by a decrease in REM sleep.
Similar studies where participants underwent a similar fast during the month of Ramadan have also shown an increase in deep sleep and a decrease in rem sleep. Rem sleep usually happens at the end of the night. So it is not surprising that more deep sleep will result in less rem sleep, assuming the total sleep is the same.
Conclusion
Although these results are significant, we should be careful about concluding that fasting directly improves deep sleep, or heart-rate.
Why? Because other factors can jump in the way. For example, while I was fasting I stopped taking vitamin D. Maybe this is what impacted my sleep.
This is why, as I mentioned in a previous post, we need to design a controlled experiment to confirm these results. But we should also remember that the absence of evidence is not an evidence of absence.
In a future post I will write more about what I found in the fasting data. But science can be tricky, and all of the above can be wrong. So please don’t take it too seriously.
And have a nice day!
