Experiments with sleep measurements
Experiments with sleep measurements
data, information, presentation, feeling, intuition... What you feel is how you are!
Each day, I wear my Fitbit Versa 2 to bed to monitor my sleep patterns. This is my third Fitbit and my seventh fitness tracker overall, over the years. During the daytime, I use an Apple Watch SE, and prior to the Fitbit’s I had used Jawbone UP, UP 2, and UP 3 trackers. Jawbone of course innovated themselves into oblivion. That’s typically one tracker for about 15-18 months over the last 9 years
Recently, I did an experiment with wearing three of my trackers for a few nights - my Fitbit Versa, my Apple Watch, and a beta version of the Amazon Halo band. When I reviewed the data collected, analyzed, and presented, these three devices often told a completely different story on my deep sleep, REM sleep, light sleep, and not sleeping states. Both the patterns of the duration of logging as well as the times at which these states were logged looked significantly different. And, the margin of variance was more than 20-25% between the devices. One could treat these devices only as experimental at best.
I also repeated an experiment where my wife wore two Fitbits to sleep - her Fitbit Charge 3/HR and my Fitbit Versa 2. And guess what, there was a huge variation in the sleep log with over 15% difference in times logged and sleep state totals between Fitbit’s own two models.
In addition, before I review my sleep patterns, I determine how I felt my night’s sleep was. Often, there was no concurrence between what the devices reported and how I felt. The closest pattern that I could hypothesize was “the more the deep sleep that was logged (1 hour or greater) the better I felt that my sleep was”. There was no correlation to good sleep and for eg. overall sleep duration. In fact, some nights I slept for less duration and had a deeper sleep, I felt better than the nights I slept longer.
This aspect of sleep data logging and tracking started in 2012 when I was found to have Obstructive Sleep apnea. At that time, my breathing stopped over 30-35 times a night. I actually reversed this problem (for the most part) with a 50-60 pound weight loss since 2013 and also kept this weight off over the years. My apnea is still peripheral, but I have learnt to combat it through other techniques like sleeping on my sides rather than on my back and sleeping in an extremely elevated position.
The moral of the story is that these devices are just a data point. You don’t want to treat them as just another data point to verify against how you feel. If some patterns emerge for you - good or bad, then potentially you could use it to investigate potential issues that one might surmise out of this.
A final experiment that I do sometimes is not to wear a tracker to bed. These happen when the devices run out of charge and I’m too lazy to charge them before going to bed. And when I get up in those mornings, it mostly feels fantastic - there is no reference point to compare the feelings. Those days, it is “What you feel is how you are!”