No One Knows You Like Your Wearables — Are They On Your Side?
What we might learn about ourselves by combining data from multiple wearables
Who knows your personal patterns best? If you thought of your friends and family — consider someone who never leaves your side, your wearable device(s). What could they tell you?
By combining information from all your wearables, including your phone, into the same place, you can look at your holistic data points and answer interesting questions. Well, at least start interesting arguments with and pose more questions. With Prifina, that “one place” is actually your personal data cloud. Having data on your side allows you to really utilize it and build on top of it.
During the previous pandemic year, wearables have shown that they are not just an entertainment tool, but also can play a role in early detection of symptoms. Even without going into life-altering situations, there are several interesting questions to pose.
- What’s the optimal regimen according to your data for a good night’s sleep?
- Where outside your home, relaxes you the most?
- How many intense exercise minutes do you need for the lowest resting heart rate, short term and how about long term impact?
- How does your inactivity impact your sleep, how about your stress?
In this short post we explore and discuss the relationships between exercise and sleep, exercise and stress (as a function of HRV) and inactivity based on a naive interplay of data we can find in common wearables.
We also discuss what other data points we could combine at the user level, to even better understand and build agency around the choices we make once we become empowered by our own data. To do this we use an example app built on the Prifina platform that allows you to navigate and compare data points from different wearable and personal devices.
Does high intensity exercises result in deeper sleep? Being able to combine different wearable data sources on the Prifina platform, we can explore these in the same place. We could plot a graph to explore intense exercise minutes and deep sleep.
Studies show that exercise minutes can improve deep sleep, but they are also clear that other exercise parameters such as intensity, time of day (in relation to bedtime) are also particularly important factors. In our individual data set it is hard to draw a direct link between activity and deep sleep.
Drawing inferences and interpretations from a few graphs is of course futile and glancing at the data shows it goes deeper than just two data points, the biggest peaks you can discern correlate with weekends more than exercise and sleep. In order to look at exercise’s correlation with sleep, we have to look at data points such as resting heart rate and heart rate variability (HRV), to see exercise recovery and associated stress, as well as when that plateaus.
One could make the hypothesis that exercise, particularly high intensity exercise, can help resolve stress and therefore may raise one’s HRV values (the heart beats less variability between beats, less HRV, when stressed). As the previously-referenced Harvard article states, a lower HRV “is even associated with an increased risk of death and cardiovascular disease”.
We could plot high intensity exercise minutes against average HRV readings to explore whether or not all that sweat working out actually lowered our stress. Let’s take a look.
Hmm. That didn’t really tell us anything. And the other countless plots we ran to explore this were just as trivial. The conclusion is that we are missing out on other data, surely stress is more linked to other factors than exercise minutes alone. And there’s a huge question of causation here too, which we cannot ignore (running like crazy when stressed).
How about sleep? Good or bad, would high intensity exercise cause more REM sleep in minutes? We can plot this to find out.
From plotting active minutes (high, medium and just total active exercise), we can see some positive correlation. Not strong, but it seems to exist.
What about body temperature? The Oura Ring tracks your temperature’s deviation from your baseline too.
In all honesty the only thing temperature deviation seems to tell us is a positive correlation with muscular and cardiovascular recovery. See the following chart for correlation between active calories and temperature. The same was true for the countless other metrics we tested, like inactivity which correlated negatively with a positive temperature deviation.
Genetics, environment and all those other factors
It’s evidently clear that we can only start a discussion when it comes to individual data points. In order to actually be able to draw any real conclusions, we’d need more subjects’ data to explore and it would also be interesting to understand more a context around the individual subjects, their environment, habits and even genetics. This is certainly something rigorously already explored in countless other discussions.
Utilizing something as private as genetics would also require a truly privacy-protected data processing model, such as retaining all your data in your data cloud and using it without sharing.
Being able to see my own patterns of behavior, my recovery trends peaking on weekends (most efficient sleep catchup Friday/Saturday and Sunday a definite early morning run day, with the energy to match), I can pose a lot of interesting questions for myself. At the end of the day that’s what I also care about more, my own activity and personal data, not just statistics.
There’s countless studies pointing to the incredible growth in the wearable market, to nearly 1 billion devices sold by 2030.
We as individuals have access to increasing data, but it’s equally important that data is actually moving us forward in practice, with small nudges to help us navigate our lives and make informed decisions.
