Changes in sleep architecture can occur over time due to this illness, medication usage, or other factors such as physiological stress. These can have lasting effects. Over the long term, sleep complaints in ME/CFS can be the hardest area to address.
I wanted to look more closely at my sleep habits to see if I could pinpoint clues that could then be influenced to achieve better sleep. Since October 2014, I began to track aspects of my sleep using an inexpensive ($4.99) app for iOS called Sleep Cycle (also for Android).
I’ve written about Sleep Cycle previously in a post about tools to biohack chronic Illness. It provides a simple way to look at sleep wave cycles throughout the night by using the phone’s accelerometer. It also has an updated membership feature ($1.99) that allows data cloud backup, web support, and additional metrics.
I want to show you some of my data.
More Coffee Please
Sleep Cycle allows you to take notes about your habits before setting the alarm clock. These include by default caffeine usage, meal time, exercise, and stress level. You can customize the sleep notes to add whatever metric you’d like. This is a great way to track the effectiveness of a new sleep aid or treatment intervention.
Prior 23andMe data analysis told me the good news that I’m genetically a fast caffeine metabolizer! Regardless, I found the data here quite surprising. I added the “drink coffee” sleep note only on days when I drank a cup of coffee after noon. Perhaps I should have afternoon coffee more often.
Eating dinner early is a key factor for me as well. I defined “eating late” as a meal time after 8pm. In general, I try to eat an early dinner before the sun goes down so as not to disrupt circadian rhythms, and as a means to intermittent fast.
Post-Exertional Malaise Quantified
In addition to tracking sleep, Sleep Cycle also collects pedometer data from your device. This provides a nice means to quantify pacing strategies. How many steps does it take to trigger a bad night of sleep? Does that bad night of sleep also correlate positively with inducing PEM?
Based on my data, activity between 7,500+ steps, but less than 10,000 steps is best for sleep. The effect of greater than 12,500 steps is of substantial detriment to my sleep quality!
Best Nights, Worst Nights
A number of studies in ME/CFS patients have reported reduced time in slow wave sleep relative to controls. It is thought to be that cytokine signaling plays a role in the altered sleep homeostasis in many patients.
Looking at my data over time, and comparing good nights with bad, there is apparent slow wave sleep cycle disturbances.
Take a look at one of my better nights. The data indicates about 3-4 cycles in which I reach slow wave sleep. Slow wave cycles should comprise about 1/5th of total sleep.
Compare that to one of my worst nights—few cycles in slow wave sleep with most of the time spent in deep sleep. Another consistent finding on my poor nights, is a long onset to sleep and waking at least once, which is nicely shown here.
Still other data from nights of bad sleep, show the opposite, periods without reaching deep sleep. This is consistent with some ME/CFS research, where REM sleep has been shown to be reduced.
I plan to continue to use this excellent tool to find clues about how various factors affect my sleep. To strengthen this data, I plan to incorporate specific notes on supplements, dietary trends, as well as hormonal cycle data. Another major finding in ME/CFS sleep research is low heart rate variability (HRV) during sleep, which can also be assessed using Sleep Cycle in conjunction with HRV apps like Heart Math. Stay tuned for more insights.
Jackson ML & Bruck D. Sleep Abnormalities in Chronic Fatigue Syndrome/Myalgic Encephalomyelitis: A Review. J Clin Sleep Med. 2012 Dec 15; 8(6): 719–728. http://www.ncbi.nlm.nih.gov/pubmed/23243408
Burton AR, Rahman K, Kadota Y, Lloyd A, Vollmer-Conna U. (2010) Reduced heart rate variability predicts poor sleep quality in a case-control study of chronic fatigue syndrome. Exp Brain Res. 204(1):71-8. http://www.ncbi.nlm.nih.gov/pubmed/20502886