Wearables and HRV monitor my stress, but digital stress balls don’t work for me. Help!
The pandemic has rocked all of us in the past few years. Studies have described the ongoing pandemic and uncertainty as cumulative stress that our bodies are ill-designed to sustain over a long period of time (e.g., in this body of work at the Yale School of Medicine), leading to a similar type of symptoms found in PTSD and significant mental strain for a large population.
Our wearables monitor time-series data for our sleep, our heart rate, our heart rate variability (HRV), and many more. HRV is a proxy for stress and while individuals are different, then monitoring your own trends over time may be helpful. Yet, when we search for stress management apps on the App Store, we find comparable digital stress balls and some HRV analysis charts. Shouldn’t applications be more individualized to really suggest healthy patterns and trends, not just stress analytics?
There have been various non-invasive sensors that have been developed over the past few years. Most recently, Rockney Photonics unveiled various new capabilities at CES this year, including the ability to determine “heart rate (HR and HRV), core body temperature, hydration, blood pressure, blood oxygen, alcohol, lactate, and glucose indicators.” Others are developing new innovative ways of measuring the body’s stress levels, including “galvanic skin response or measuring the electrodermal activity of your skin ” by Monoa Health.
With all this data, shouldn’t we be able to create better models than only a digital stress ball?
For example, meditation is often given as a tool to manage stress and applications like Headspace are doing fantastic work. At the same time, they are not using the person’s stress data to gauge what mood they are in (or even asking them!) to see what type of mental workout is appropriate.
For feedback, it would also be great to see how your stress readings are impacted by different activities such as meditation, but also your sleep and exercise. Not just on a daily level, but maybe also the time when you meditate and the place where you exercise impacts your stress and stress recovery? You can quite quickly start to map out a complex series of different types of data points that have an impact, some more material than others.
The reality is that it is not so simple to build tailored experiences. Individualized applications require a lot of data, but they also require a lot of application training and learning on this individual data. In the end, it is almost like creating personal applications for each individual user, where the application not only responds to their data, it is actually built using their data. Bespoke applications, if you will.
And it’s not just the physical data we collect, it’s about combining it with behavioral and activity data, that is contextual and complementary data. A 78 bpm heart rate reading will tell you one thing if the person is sleeping, another thing if they are meditating, and a third if they are jogging. That same 78 bpm will be impacted by what the person is watching, reading or listening to. How then can you determine what the 78 bpm reading is telling you? This is the problem we face in spot-readings, where the highest utility and value is focused on time series data and trends, and spot-readings often contain significant risk for error. Looking at a spot reading you might conclude the person is in trouble!
Building individualized applications requires a lot of data in both quantity and quality, so much so that we simply cannot keep moving all the data between places, such as between the user and developers servers. It also raises privacy concerns. With such in-depth data, privacy is simply impossible unless the data is processed in a decentralized way.
That does not mean that the contextual data is irrelevant; quite the opposite. In fact, it does mean that the way of utilizing it is missing. With Prifina, we focus on building individualized applications that run on user-held data, but they are also built in a user-centered environment where they can be trained and constructed to be truly individualized. This guarantees full privacy for the user, but also a rich training ground for the application for the developers to take advantage of. How you train this application in a decentralized manner then, is a question of the use case and developer ingenuity to serve their customers.
For now, it is of great interest to have more data driven mental health applications that can go one step beyond ‘how you feel’. By utilizing personal data in these types of applications, we can get more results, better suggestions and even outside the box ideas that your data suggests might work better for you, than a stress ball.
Connect With Us and Stay in Touch
Prifina is building resources for developers to help create new apps that run on top of user-held data. No back-end is needed. Individual users can connect their data sources to their personal data cloud and get everyday value from their data. Follow us on Twitter, Medium, LinkedIn, and Facebook, or listen to our podcast. Join our Facebook group Liberty. Equality. Data. where we share notes about Prifina’s progress. You can also explore our Github channel and join us at Slack.
