Estimated Oxygen Variation graph is broken

[AndreaVersa2can]

Hi, I just noticed that the green and yellow lines have changed and now are all over the place on the graph. All graphs for different days start with a 0:00 hrs. The fluctuations are not at all in line with what I was reading up until today.  Anyone experienced this issue?? Thanks

Moderator Edit: Clarified subject

[Jim-in-AZ]

I see several references here to a green and yellow EOV chart. On my Galaxy tablet and phone the non-green color (of which there is a lot with the current glitch) looks orange, not yellow.

[SunsetRunner]

Yes!  And since I have asthma and was suffering an allergic reaction to something at the time I thought the data was credible.  It's extremely comforting to know that it's not just me seeing the crazy graph and I don't need to rush off to immediate care or something like that.  The timing was really bad and I did freak out a bit.  I guess we'll just have to sit tight for an update to fix this.  

[Blake.S]

Miko82's screenshots show perfectly what's going on. The new graphs show every single oxygen desaturation event. These are normal for everyone, but showing them in this style accidentally exaggerates the severity. The old graph shows a little high only when he had a cluster of desaturation events, and it smoothed out the ones that occurred at a normal frequency, which is a better way to show it to people who haven't studied sleep apnea. Consistently less than 5 yellow spikes per hour is probably good, but good luck counting more than that on these squished graphs! I kind of wonder if the old graphs needed FDA approval...

[dazmlc]

As of about 1 week ago, the display of the nighttime estimated Oxygen Variation drastically changed, even for past  nights data. I used to always be low variation at night. Now even for past nights long ago I am showing high variations. Obviously the calculations have changed. Was this done purposefully and why?  

[LZeeW]

Hello @dazmlc and welcome to the Community.  You don't say what type of phone you use, but there is a new issue with the Android app and Fitbit ia aware.

[dazmlc]

it's a Sense watch.

[BillJ99x]

Thank you for the update. It's crazy that a bug this bad made it into production and is taking a week to resolve. 

[BillJ99x]

Interesting explanation - thanks! Obviously events that are in the normal range should not be orange. 

[Scarlett1982]

Yes, the same has happened to me and I am concerned too. I think that it is a bug of some sort. 

[Wcjh]

Hopefully a tech in Fitbit will fix this glitch very soon. Not sure how to notify Fitbit of this problem.



Sent from my Galaxy

[KathieP]

Hello, 

Same issue. I have sleep apnea. I tested with and without my cpap machine. The green line was steady with my machine and erratic without. I wonder if new charts are more accurate?

[Debbie29469]

And,  it's doubtful that a reset will fix the issue because it's in their programming,  not our watches. At one time,  the SpO2 information wasn't transferring from my watch to the app.  Fitbit repeatedly told me to reset to fix it.   I refused,  for the same reason you state.  It took over 2 weeks,  but when they got it fixed,  everything went back into place.   That specific issue happened to me twice.  I'll just try to be patient 🤷‍:female_sign:🥺🙂.

[Debbie29469]

I use Android with my Versa 3.  

[SunsetRunner]

Huh, that's interesting.  I was screened for sleep apnea as part of my asthma work-uo and don't have it.  Every historical graph has changed to include high variations even those where my SPo2 only fluctuated between 96/97 and 98%.  If the graphs were more accurate, that difference would be ignored.  Something's definitely wrong.

[tboll01]

I agree with you, Blake.  The more I see postings that include BOTh old and new EOVs together, the more I think that the New EOVs are just "over-sensitive" or just too RAW of representation.  The old graphs may haven not have done a good enough job of alerting us to breath-pauses, but the NEW graphs are shocking and UNINTERPRETABLE for lay people.  Fitbit would probably be more helpful if they just SKIPPED the EOV graph and instead gave some scores like average desaturations(ie. large O2 variations)/hour so that we just had one number/score to deal with and we could see a trend if the number got higher or lower over time.

[BillJ99x]

Fitbit will probably charge for that information. Maybe this isn't a bug but a feature intended to drive more Premium subscriptions 🙄

[SunsetRunner]

I was shocked when I checked oxygen variation, all of a sudden it was whackO. I would like an answer and a fix if needed.

[Wcjh]

Not sure how to contact Fitbit regarding this matter, do you?



Sent from my Galaxy

[Blake.S]

The FDA would probably ding them if they started to give results that look too much like a sleep study. The latest I can find is that they still need to convince the FDA that wrist only analysis can do the job. I got an at-home sleep study kit when I got diagnosed, and it was a wrist device with an attached finger pulse oximeter and a thing that sticks to your chest for sleep position and breath rate. Getting these numbers at just your wrist probably takes some novel techniques. They clearly have to do some crazy math to get your average O2 for the night, given that it takes an hour to show up, even when EOV shows up immediately. Although, I like to imagine they have some intern running the punch cards over to the 1950s UNIVAC.

[Yu123]

Exact same thing happening on my Sense.

All my past normal data look all abnormal and broken with lots of yellow.

Hope this is a bug.