UKH

I acquired a Garmin Fenix 6X a few months ago and one of the new metrics it provides is a pulse oximeter reading. Mine consistently reads in the range 91-96%. I live in London (at sea level) so this seems quite suboptimal. Question for other Garmin users with this feature- do you trust the result? do you measure consistently boringly high (as you should expect)? have you used at altitude (at the least I would expect it work relative to itself)? should I pay any attention to this at all?

I have tried measuring two other people- one measured what I would consider normal at 98%, and one who had fairly recently recovered from a confirmed case of COVID at 91%. I have not knowingly had COVID but did have a mystery respiratory illness in early January 2020. Lastly a few years ago know I did have a lowish haematocrit reading (~37ish I think) and was recommended to supplement with iron which I have done since. 

I feel that my running performance is down a considerable amount compared to previous years- but in the year before buying the new Garmin, I only ran untracked, so don't have any training metrics to show I wasn't just lazier than ever before, or that the drop off was specifically after January illness (I definitely took a good few weeks off doing any aerobic work at all). My "I'm not lazy, but my physiology knowledge isnt great" hypothesis says my lungs were somewhat damaged by the illness and have never fully recovered, or I am still somewhat anaemic. Any thoughts from the runners of UKC? I don't really feel I can ask a GP about this kind of thing right now, it doesnt seem that important, but maybe should stretch to paying for a blood test.

I don't trust it. The hospital one will show 100%, while the garmin shows between 90-95%

Limited at best. 

So you feel you cope with aerobic excercise less well than you did.

You measure low on blood oxygen content.

Other people measure normal on the same piece of equipment.

Sounds like you may have a problem.

Just a question though. Does the watch measure oxygen content by shining IR light into your wrist? If so do you have black skin and did the people you tested it on have the same colour skin?

Another thought, have you tried pushing the watch into you wrist to ensure it is in intimate contact with your skin? 

I wouldn't trust the pulse oximeter function on a watch. A cheap amazon pulse oximeter will probably be far more accurate.

IR is not absorbed by melanin.

To be honest, I did some work on pulse oximetry during my PhD and I would not trust these readings at all. Even hospital pulse oximeters have rather large error margins, what they measure really well is change in SpO2, not absolute values.

To be honest I wouldn't make any health decisions based on Garmin metrics.

I just measured my pulse ox and got 100%, tried it again a couple of minutes later and got 96%.

It tells me my training is 'unproductive', then I go for a gentle couple of miles with the dog and I'm 'overtraining'. I also have a 'fitness age of 20', which Mrs Ridge seems to find hilarious for some reason....

If you're really concerned speak to a GP.

When I got my Fenix 6 last Xmas I had just about recovered from a virus which showed most of the Covid symptoms, only lasted 3 days but left me with a cough which is still hanging about. Initial experiments with oximeter were hopeful, usually 96 to 100% at 95m ASL. Trying the continuous nighttime option had me convinced I was still at death's door with readings as low as 88% whilst asleep. Stopped paying any attention to it, a majority of spot readings fail first time anyway, and running is going well despite the odd cough (had an X-ray and antibiotic just in case). Like Ridge, I am also overtrained, unproductive and have a fitness age of 20 (at 69.97). Doesn't say much for the kids of today, I reckon...

Thanks for responses. Yeah I'm not making any decisions based off this really, I've just been musing about it and wondered if others had similar experience.

The garmin unproductive/overtraining status thing is very simply based on 7d load score, I understand it can be weird but its understandable! (it's annoying that sometimes you do the workout it recommends, and it THEN says your training is unproductive, but it's not rating that one workout but the last week, based on HR I think)

I have white skin though apparently not relavent, and i've been taking readings for a couple of months and it's never been over 96% (except on other people) so I don't think it's random, but agreed it is hardly a medical instrument and probably not calibrated. The reading does often fail but it tries many times per day so you get a few good ones each day.

When I got my watch (same model) I sat at my desk with a proper pulse OX and compared the two. 92-96% on the watch was always 100% on the medical device. If you stay very still and push down lightly on the watch face you might get a better (higher) reading.

I too go from needing to do a hard session to overtraining and back again and on some occasions one screen is telling me that I have had some hard weeks and need to rein it in and another telling me to do VO2 max intervals - completely bizarre.

I am pleased to hear you too are in that amazing top 5% of athletic 50 somethings - I feel like Mo Farah (not). We must be the fittest people here! Its amazing what you can do on 25 miles a week.

The other thing that might be different between you and the other people that you tried it with is the shape of your wrist.

Oximeters normally use your finger tip because there's lots of soft tissue and lots of blood vessels their that means the light gets a nice clear path through lots of blood. The trouble with your wrist is it's nothing like as good a place to measure absorption by blood (I'm assuming that's how this device works). It's possible your wrist is slightly different to other peoples, wrists and thus giving a lower reading. Have you tried pushing the watch against your thigh or forearm to see if you get a higher reading?

Just tried it on... 

Thumb, finger, palm, forearm and calf.

Calf got the highest reading. Finger the lowest. 

Only once though, so not scientific at all.. 

Load score over 7 days is one component, more important is change in estimated VO2max - if this score reduces (and you only get to see whole numbers so the change is imperceptible) then your training is Unproductive. I look at the 5k estimated time under Training Status.

Back in 2011 when I started doing parkrun my PB was 22:03. Currently I can get close to 25 mins (25:09 best since March) but the watch still thinks I can do 22:12 with a VO2 of 47. Until 3 weeks ago it was saying my 5k potential was 21:11 with a VO2 of 49. In the interim I have run 1 mile, 5 mile, 6 mile and 10 mile "standard" times for Silver award (county standards) and each hard run has increased my 5k estimate and lowered my VO2. I got to the dizzy heights of 49 by running slowy or steadily every day in October and not racing. Weird. On my indoor bike however I get 51 or 52 almost regardless of the effort I put in.

Ah interesting- since getting this device my VO2 max has only changed from 50 down to 49 despite training and definitely feeling a lot better for it, so I haven't observed it adjusting the training status due to that I don't think. I don't see how it can even attempt to estimate V02 from less than hard effort runs though except by using a very generalised performance model.

I think it's interesting that there is a motley collection of ages and training levels on this thread, all hovering around the same VO2 max according to Garmin.

I'm down from 50, must train more/less but do it slower/faster over a shorter/longer distance depending on how the planets align.

I have a cheap one, finger type, I bought a few years ago off Amazon, it constantly reads around 94%-96% for me, my missus is higher 99%. I can get mine up another 3% by breathing quickly and deeply. Swapping fingers doesn't change the reading so repeatability is good.

I don't know the answer to your question though I'm afraid, but if it's reading "better" on other people it's not likely to be the fault of the unit.

Can you not compare with another meter?

HaHa! I got my comeuppance for complaining on this thread - this morning's 5k jog (with a faster finish) has resulted in Maintaining (rather than Unproductive - hooray), a new estimated 5k time of 22:02 (9 seconds improvement - hooray), VO2 max still 47 but worst of all - fitness age is now 26. I got 6 fitness years older overnight...

Yes I could just buy a finger based one. I feel like I will just use it once and never use it again so will avoid that until curiosity gets the better of me!

so the garmin attempts to calculate what your expected time at a distance could be, based on your recent running?

Yes. They used to buy the heartrate variability algorithms from Firstbeat Technologies, and now they have bought the company. If you search online there are White Papers about how the data they can pick up from an appropriate HR strap can allow estimates for exercise load and predicted times for future events (5 & 10k, half and full marathon). My 5k estimate has fluctuated by about a minute over the last year, when I am training slowly it predicts I could race fast, when I try really hard it says I will be slower (but still about 3 minutes faster than I actually can manage). There seems to be no logic in the calculation to detect that when I try really hard, that is all I can actually do! I get the impression that it does not compensate for age properly, I could have managed to match their estimates 10 years ago...