How do I interpret the blood oxygen variation graph?

[Emmy_M]

Good morning,

I use a Charge 3 and recently noticed the addition of a blood oxygen variation graph in my sleep data. I've been googling a bit but am unsure how to interpret it. Can someone help?

Unfortunately, I'm unable to include a picture, but I'll try to explain as best I can.

The x-axis is clearly for time, but what about the y-axis? There's a line about halfway up where labels say "large" above it and "small" below it (my translation; I use the app in Swedish). The graph legend also states that a green line indicates small variations, while yellow marks bigger variations, so that's also pretty self-explanatory. My graph is almost entirely green every night, but it's fairly low on the chart. Does this mean my Charge 3 estimates I have a fairly low level of blood oxygen, but with small variations? The few times the graph has been yellow, it's been to form a peak rather than a trough.

Thanks in advance.

[netmann]

@Chris_uk35  Not a faulty watch, rather an incapable watch that cannot record minute O2 variation so it defaults to flat line or lines...! See my previous post...

Chris, I never had a total straight line, like yours, without any spikes before. Mine always had combination of spikes and flat lines which means my Versa 2 correctly records the larger variations as spikes and records the minute variations as flat lines because of its reduced resolution manipulation...!

[Chris_uk35]

Another night and another flat line.. That's 3 out of the last 4 nights it's done this now. 

Surely I have a faulty watch? 

Also I have noticed that when the red light is on its a solid red light., it does not flicker like the green light is it supposed to flicker?

Moderator Edit: Merged posts

[WilsonFitbit]

@Chris_uk35, 

I believe it is not a fault, regarding your concern about the lights, your Fitbit device tracks your EOV using red and infrared sensors on the back of the device. The sensors shine red and infrared light onto your skin and blood vessels, using the color of the light that bounces back to approximate how much oxygen is in your blood:

Richly oxygenated blood is red and reflects more red light than infrared light.
Poorly oxygenated blood is bluish red and reflects more infrared light than red light.

Note that you might see the red and infrared sensors continue to blink for around 30 minutes after you wake up. On a side note, please note that 

this feature and the metrics within this feature are not intended to diagnose or treat any medical condition and should not be relied on for any medical purposes. It is intended to provide information that can help you manage your well-being. If you have any concerns about your health, talk to a healthcare provider. If you believe you are experiencing a medical emergency, call emergency services.

@netmann, thank you so much for your input and feedback! 

See you around. 

[Chris_uk35]

Except my red light does not blink.. Its just constant on 

[WilsonFitbit]

@Chris_uk35, thank you for your reply! 

Similar to the green LED light to track your Heart Rate, the red light is constantly on while tracking your estimated oxygen variability during your sleep. But if you see a weird behavior of your Fitbit device and its lights, you may try to restart it or contact our Support Team for further assistance. You can contact them through chat or over the phone. 

Hope this helps!

[n8vz]

No, I’ve had no direct responses to answer my questions.  Now the last few weeks, I’ve been getting a few nights a week with spikes that just touch the line.  These seem to correlate often with periods of REM sleep.  Also, at night I have a very low pulse — often in the low 40s, with occasional spikes up upper 50s; not sure if this is related or not. This whole topic is quite confusing, and I had a career as a medical writer. 

I’ve noticed a strange correlation between spikes into the orange and SpO2 readings.  Usually my SpO2 is about 93 when I’m asleep and about 98 when up and awake.  However, this week I had two nights with spikes and on both of those nights my SpO2 average was reported as higher — 95 and 96.  One would expect the opposite — less average oxygenation of blood when the variability is greater.  Something is not right here.

The first graph below is from last night, SpO2 = 96, the second graph is from Wednesday, SpO2 = 95.  Previous to Wednesday had over a week straight with no orange spikes, and SpO2 average was consistently 93 and 94, while asleep. 

I’d like to revive this thread.  This whole EOV thing seems a mystery to me.  Most nights I’m still all in the green, but a few nights a week I’m getting a single spike into the orange, usually not very far.  Here’s a recent example, from last night.  Often the spike is even less than this, barely touching the line.  

Let add that on some of these spike nights there seems to be a correlation with REM sleep — not all the time, maybe half the time there’s a spike.  I also noticed last night that at about the time of the spike my resting heart rate jumped from 49 to 65.  A bad dream raising anxiety and, therefore, heart rate?  Is so, this would likely be an significant increase in SPO2, not a decrease. Strange stuff.

Moderator Edit: Merged posts

[WilsonFitbit]

Hello @n8vz. It's great to see you around. 

Thank you very much for the detailed information and for taking the time to provide your feedback. I'd like to let you know that:  

Small variations show that your estimated blood oxygen saturation levels remained consistent.

Big 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.

Frequent big variations on your estimated oxygen variation graph may indicate that you experience breathing disturbances during sleep. Breathing disturbances can lead to higher than expected variations in blood oxygen levels, which can indicate that your body is not getting enough oxygen while you sleep. If you frequently see big variations on your graph, consider talking with another health profesional that can give you a second opinion, particularly if you experience symptoms such as excessive tiredness, loud snoring, or gasping during sleep.

Other factors, including your arm position and certain physical characteristics, can affect your estimated oxygen variation readings, making them appear to be more variable. Make sure your Fitbit device is snug and secure on your wrist. For more information, see How do I wear my Fitbit device?

Note that your Fitbit device is not intended for medical purposes. As mentioned above, it is recommended to have a second opinion and consult with a health care professional about any questions or health issues you may have.

See you around. 

[n8vz]

Thank you for your response that didn’t specifically address anything in my posts on this subject, or those of other folks who are puzzled about the interpretation of this EOV graph.  There must be some interpretive difference between barely hitting the line between high and low variation, one spike above the line, multiple spikes above the line, spikes correlated with increased pulse rate, spikes correlated with REM sleep, etc.  If you aren’t going to supply detailed information on graph interpretation, it would be best if you discontinued this feature.  You are doing a great disservice to your user community as it now stands.  — Carl

P.S.  Also thanks for merging my posts in a manner that takes them out of context and makes them less understandable. 

Moderator edit: format

[MBGrape]

n8vz I feel your pain but me thinks it's self induced. The moderators have answered every one of your questions bur you just refuse to accept them. Nothing you question is a problem with your tracker or the algorithm. It's doing exactly what it was designed to do, provide a glimpse into your O2Sat levels. It's not an absolute but a variation. There is not an O2 sat level, only a change. You need to figure out why it changes on your own, that's not the aim of the feature. Change is small (green zone) or large (yellow zone). Period. You can correlate that to whatever you want, Doc. I get it, all my friends who use FBs get it, stop being a PITA. I've also had a rare flatline on the graph. On rare occasions I get a generic sleep pattern. I readjust my band or accept the fact that I may move around a little to aggressively (on rare occasions). My bad, not the tracker's. If you want accurate O2 sat numbers go buy a good co oximeter. You have a good tracker at a good price that does what it says it does. Learn how to use is or change models. Luv ya.

[n8vz]

If you think "these responses” are engaging the customer in real dialogue, then go for it. Some of us would prefer more. And, in point of fact, yesterday’s response was the first one from any FitBit employee that I’ve received. Previous responses were from other users. I stand by my point that if FitBit is not more forthcoming on the meaning of this EOV feature it should be discontinued as it probably causes more angst than any real knowledge that it imparts. I would suspect that if a complaint was lodged with the FDA that they might agree. At this point I’m not going to do that. I’m happy with the rest of the features of my Versa 3. — Carl

=============================
Carl Jón Denbow, Ph.D.
=============================

Moderator edit: personal info removed/word choice 

[Tinkerb3ll94]

Estimated oxygen variation what does that mean 

[audree_woods]

What does it mean when my graph is in a straight line? It’s all well in the green but did not budge all night. 

[AnnaMccoy]

Sleep apnea is the question for me, and the blood oxygen level connected. If there is a big variation (and also I think you're correct that there is not a specific dimension on the y-axis-- only a relative variant) I check heart price to see if its raised. At high variance, there could be less oxygen absorption.

[juliefouc]

Agreed. Exactly what you said. If they can't provide an explanation (or even just a darned legend), then they should discontinue the feature. People without a background in this field will just blindly accept the chart, and Fitbit is banking on that. I don't even care about the readings at this point. I have an spo2 meter at home that I use when I need something very accurate. This is a matter of principle at this point...

What is low variation as opposed to medium variation? How do you get a line that starts in the middle of 'low' variation, and then goes up, and then LOWER than the starting point? It makes absolutely no sense. How can you get less variation than your starting point?!?

[MizladyP90]

You want to see a “low”  oxygen variation on this graph. That would mean you have very little or very few fluctuations (changes, differences etc) in you Oxygen levels while you sleep. This means your oxygen levels stay consistent and have a low occurrence of changes while you sleep. 😉

[Alex_h12]

Hi Carol, i was reading this thread and noticed this. I have experienced getting up to use the bathroom at night (nocturia) for years, and it turns out it is actually a stronger indicator of sleep apnea than snoring. Perhaps you are experiencing apneas, hence the high variations, which are causing you to get up to use the bathroom, rather than the opposite.

Maybe consider getting a sleep study? I am 34 and fairly fit. Anyone can have sleep apnea.

[n8vz]

My Fitbit odyssey is over.  Now that Apple will be adding complete sleep analysis to Watch OS7, due out in the fall, I’m officially switching to an Apple Watch that I got a good deal on today at the T-Mobile store. The erratic behavior of my Versa 3, and my previous Fitbit, requiring bi-weekly complete resets was just too much of a hassle. Signing off! 

[SunsetRunner]

Thanks for commenting, @Alex_h12. I actually had a sleep study done several years ago and sleep with a CPAP. Your theory is right. But in my case, the action of getting up, removing the CPAP, and walking to the bathroom … then reversing it all … that is causing my spike. The spike(s) correspond exactly to the times I get get up. 

[Peggy9970]

My wife is on Cpap (actually Apap) for mild sleep apnea. So I've a particular interest in the Fitbit (versa 2) observation of O2 levels.

We have shown these EOV graphs to her Dr, and HE asks the following question: " Variation from what? Is the current O2 level compared with the previous measurement, or the average over the past few minutes, or over the past hour? " 

On the other hand he is impressed with the pulse rate graphs, as those are clearly marked, and can be scanned as 5 minute averages.  Pulse Rate variations (for my wife indicative of AFib episodes) are very useful and actually provide a record of Afib events.  Unfortunately the Estimated Oxygen Variation graphic is useless without a full explanation of the Y-Axis. 

That is really unfortunate, because the watch actually displays the estimated O2 levels, and bluetooths all that data to the app, the programming of the app, apparently decides to be "cute" and processes the individual readings into a graphic that neither I, my wife's PCP, nor her sleep doctor and understand. 

I recently talked with Fitbit rep, where my question had been "escalated" and waited a week to get the "more advanced team" response. What they told me was that the 'line" between high variation and low variation was actually at 95% Oxygen Saturation.  The "high zone" is between 95 to 98%, the "low zone" is between 92 to  95%. That makes no sense at all.

Suggestion to Fitbit: Just report the data that was collected, average it over some period of time (like pulse rate), and graph the O2 values (like is done with pulse rate) and let us and our medical providers decide what is high or low variability, and if it is medically significant.

Please do not tell us to take our EOV graphic to our medical consultant, if he can't interpret that graphic or the magnitude. 

[n8vz]

I need to update this a bit.  After trying the Apple Watch for a month, I’m back to using my Fitbit.  The AW experience was horrible.  The sleep tracking, even using auxiliary apps was horrible — said I was asleep when I was watching TV and wide awake.  It didn’t count steps well and was hard to navigate.  Needs to charged daily!  The Versa 3, with all of its quirks, is much better.