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

[SunsetRunner]

Then I wonder why to even have it on the watch? If things cannot be
accurate, then why bother? Apple watches don't have this issue. The other
smart watches don't have this issue. Seems to be only fitbit

--
*This is not a dress rehearsal*

[glenj]

Received an update this a.m. All Oxygen graphs appear as they should.

Looks like issue has been resolved. Yay!

[OKRider]

I log out and log back in again when this happens. It's annoying, and I
have to do it a lot, but it fixes the problem. Maybe the latest updates
will resolve this, too.

[MarkFolse]

Okay this thing is hosed. Got the 3.65 update last night. About 30 minutes after I got up it showed appropriate looking sleep data for last night on the top screen of sleep. The drill down screen didn't show any analysis. I just went to recheck it and the data on the top level screen has changed and is incorrect. It's entirely possible they made it worse.

[Bubbabighatt]

Mark, How can you tell the data is incorrect? My graphs don't have any labels on the axis so I don't have a clue what is "high" or what is "low".  

[RodrigoMFitbit]

Hello everyone. 

Thanks for your feedback and informing about the current situation regarding the EOV graph. The latest Fitbit app update shows now the corrected graphs. If you have not received the latest update yet, I appreciate your patience. The rollout is done gradually and soon all users will get it. 

Please note that the graph before only showed "high/low" on the Y axis. Now it seems that only the colors of the graphs will show you high and low variations. If it stays green, the variation were low. I recommend you take a look at this article: How do I read the estimated oxygen variation graph in the Fitbit app?. 

• Low variations show that your estimated blood oxygen saturation levels remained consistent.
• High variations show that your estimated blood oxygen saturation level fluctuated. A high variation on the graph may indicate you had variations or disturbances in your breathing during sleep.

Note that your Fitbit device is not intended for medical purposes. Consult your health care professional about any questions or health issues you may have.

I would also like to confirm that the minimum requirement for the Fitbit app to be installed in an Android device is actually Android 8. You can see this here: Android Fitbit app 3.65 is now available - 08/10/22

[MarkFolse]

That based on reviews of my CPAP data my doctor believes its
working correctly. And it shows a display of how many detected breathing
interrupts I have per hour each night. And the graphs before version 3.65
are clearly inconsistent with my CPAP's nightly result.

[Peggy9970]

The pulse rate graph is a plot of 5 minute averages, and (at least on my app) is interactive. (that is the averages and the time of that average is revealed). 

Why Fitbit chose to handle SpO2 data differently is a mystery, but their current display of SpO2 recorded during sleep appears to be undocumented and largely creates more confusion (and apprehension).

On my IOS app, the Y axis is undefined; there is no "zero".

If the graph were a first derivative (aka slope) of  the SpO2 v. time curve, then a unchanging SpO2 would have a slope of zero. 

Similarly, an increasing serial SpO2 would have a positive slope (and be represented by a upward spike), while a decreasing  serial SpO2 would have a negative slope (and be represented by a downward spike).  That does not seem to be the way the graph appears,

To be useful to users (and potentially to their medical caretakers), the graph needs to explained by the Fitbit engineers. 

As best I can tell, members in this community have been asking for such an explanation/documentation for several years - without any satisfaction

[DannyFB]

The explanation is vague, rendering the graph too vague to be of any use.

It is presented as though it is discrete data points. A variation of a
single data point is nonsense. Hence discussions degenerating into wild
guesses of first derivatives and such.

End user should never be left having to take wild guesses as to what the
device is displaying.

"High variation" has no real meaning. No value. Does it mean a discrete
data point was more than 10% above or below the average? 20%? 90%?.

THAT is the normal and clearly general interpretation of this otherwise
useless graph. It is being interpreted as and presented as a specific ( has
a value) variation around some level.

[home4all]

I just checked my SPO2 (or EOV) graphs and they appear to be working the way they did before the mid-July 'update.' They at least show a similar curve to the way I remembered it. Since that is what I was used to I am satisfied with the update. The app says I now have version 3.65 instead of 3.64.5. All of my historic EOV graphs seem to be displaying the same type of info in a similar way that they did before mid-July as well. The line thickness.is much thicker and the resolution seems to be changed. The graphs day to day may have different scales but since there is not the familiar xy axis with values (and never was) I am only going by what I think I see. I am using a Charge 4 and Samsung S22+. The 3.65 version stated it was new as of Aug 9 but I received the update today, Aug 17. I am not using Fitbit Premium.There are types of displayed info that is new, some of which could be useful to me. I will be checking out all the changes soon.Sent from my Verizon, Samsung Galaxy smartphone

[SunsetRunner]

There are no wild guesses here. A graph of a change in a set of values is by definition the first derivative of that set of values. You would see this more clearly if you were familiar with dimensional analysis.

[Nancwood]

I was able to update to 3.65, oxygen variation is still off, not right. AND now I cannot set UK through WiFi.

Will these issues correct themselves? Getting frustrated.

[DERIEL]

I just called,  and then tried to cancel appointment with my doctor requesting a  sleep study for apnea. They said a lot of people have been requesting  this starting last month....

Your bar chart Oops is going to raise everyones insurance rates next year if a lot of people run to the doctor.  

If,  on the other hand your chatting is now accurate and was formerly inaccurate I'd  like to know why.  

[SueChem]

So yes, I got the update and the graph looks much better and thankfully now has the time on the x‐axis.  HOWEVER, I typically would have one or two  high variations in the past per night.  When I go back now and look at my historical data there are almost no days showing any orange (found one fron November of last year with a tiny orange blip).  Somehow they have changed the criteria for the variation.  That's pretty odd.  I also wish they had the option for ALL the graphs to go full screen.  I have one of the biggest phone screens out there, and even with my glasses these charts are pretty tiny and hard to read.

[lkmtxla]

As of Wed morning, 8/17/22, the oxygen stats on my Fitbit seem to be accurate.  Fitbit was aware of the glitch and it seems the issue has now been fixed.  However, in the interim between the time in July when I notified Fitbit of this issue and when it was corrected this week, I purchased an o2 ring to be able to track my oxygen levels since I have sleep apnea and use a CPAP.  Of course, that new device didn't arrive until a day after Fitbit corrected its  o2 problem!  Glad that Fitbit addressed this issue and its o2 readings are now more reliable.

[Audreybrooks056]

The problem with that support is always going to refer to 2 things talk with your doctor if you have concerns and second it is not a medical device not meant to treat diagnosis or etc. So in return this is to watch your trends I remember my first fitbit boy do I miss the days if simple technology and motivation in my own opinion just me I wish since there is no y axis the oxygen variation was just deleted off the program just saying we all have been so frustrated but a heads up it's a new day POSITIVE IS A MINDSET....you all have a wonderful Thursday 😊 

I to wear a CPAP.....want to invest in spo2 ring it is a medical device with no guess work 

[tboll01]

@Audreybrooks056 I agree.  The feature that I want most from Fitbit is the ability to SELECTIVE what I want monitored.  If I turn it OFF, I don't just want the graph to disappear, I want the raw to NOT be COLLECTED (and stored on the Fitbit server).  Sometimes Too Much Information does more harm (anxiety, stress) than good.

[Audreybrooks056]

Tboll01 exactly could not have been said any better!!!! Technology sometimes can be the greatest or worst...I have been loyal because in the beginning as you said had choices selective.....but I will say I hope FITBIT sees thru these threads all the input of how fitbit is not the same the SUPPORT has basically last 2 years here us a case number reboot!!!! On a positive note I just hope they turn all this around I think they are headed in the right direction 

[artype]

The Fitbit Android update this week fixed my graphs. Thank you!!

[DannyFB]

A graph with no axis labels and no scale is a graph of nothing.