3 ms·

Many of us are wearing devices with continuous optical heart rate monitors. Heart rate variability at rest is one of the best indicators of stress and these dev

Many of us are wearing devices with continuous optical heart rate monitors. Heart rate variability at rest is one of the best indicators of stress and these devices are pretty good at estimating sleep duration and sleep cycles. Regulators need to recognize that continuous tracking devices are a new tool for establishing a health baseline and they should not be expected to match the performance of single-test medical devices. Apple, Garmin, and Fitbit need to build standards to share this baseline data with medical professionals and offer analytics to determine when changes occur. Questionnaire based medicine seems backwards given the digital tools available.

gedy 6y ago
I will say that learning about and monitoring my HRV after getting a smart watch has been useful to monitor stress and trends. Like with anything though, I've had to try and not self-induce more stress watching this number continuously ha.
nitrogen 6y ago
There's a DIY-ish service that can be self hosted (not sure if it's still maintained) that tracks all kinds of data in a way that you can share with a physician. It's called Fluxtream, and might take a few extra keywords to find. Written in Java+Spring. I haven't used it myself. Putting more and more of our data, especially health data, in the hands of others is the wrong direction to be moving IMO.
- kiddico 6y ago
  Oh boy, you're not wrong. People apparently love using that to name projects. Found it though: https://github.com/fluxtream https://github.com/fluxtream
klyrs 6y ago
> Many of us are wearing devices with continuous optical heart rate monitors. I don't know how representative you folks are. I know a few people who invest time, money, and effort into tracking and measuring every aspect of their health. It looks like neuroticism to me. Especially when they look for explainations of every blip on every chart.
- Andys 6y ago
  On the other hand, its incredibly useful to establish what is a healthy baseline for your own body, so that when you do get a disease it is easier to compare your results.
  - klyrs 6y ago
    You're not wrong, but my point is that folks who are interested in knowing every little detail; who have the time, energy, and funds to invest into tracking these statistics; will provide a very biased dataset. That isn't necessarily a good baseline for everybody else.

gedy 6y ago

I will say that learning about and monitoring my HRV after getting a smart watch has been useful to monitor stress and trends. Like with anything though, I've had to try and not self-induce more stress watching this number continuously ha.

nitrogen 6y ago

There's a DIY-ish service that can be self hosted (not sure if it's still maintained) that tracks all kinds of data in a way that you can share with a physician. It's called Fluxtream, and might take a few extra keywords to find. Written in Java+Spring. I haven't used it myself. Putting more and more of our data, especially health data, in the hands of others is the wrong direction to be moving IMO.

kiddico 6y ago
Oh boy, you're not wrong. People apparently love using that to name projects. Found it though: https://github.com/fluxtream https://github.com/fluxtream

klyrs 6y ago

> Many of us are wearing devices with continuous optical heart rate monitors. I don't know how representative you folks are. I know a few people who invest time, money, and effort into tracking and measuring every aspect of their health. It looks like neuroticism to me. Especially when they look for explainations of every blip on every chart.

Andys 6y ago
On the other hand, its incredibly useful to establish what is a healthy baseline for your own body, so that when you do get a disease it is easier to compare your results.
- klyrs 6y ago
  You're not wrong, but my point is that folks who are interested in knowing every little detail; who have the time, energy, and funds to invest into tracking these statistics; will provide a very biased dataset. That isn't necessarily a good baseline for everybody else.