[FRIAM] COVID SaO2 at 7k feet

[uǝlƃ ☤>$]

Steelman which one? I proposed 3:

1) SpO2 may not be accurate.
2) Health apps are only used by an rareified demographic.
3) Composition of individual effect to group effect is hard.

It sounds like you're targeting (3), given your mention of NOVID and R0. But it's related to both (1) and (2) by test/symptom recognition and attention/follow-through, respectively. Were I to combine them and try to steelman that combination, I would simply say that the more every individual thinks hard about such things and, importantly, realizes that they are very hard, the more they'll give people like Fauci a break. Everything's hard. Everything sucks. But sometimes we see little nudges in progress due to the huge amounts of work done by hard-working people. Or, in other words, look deep inside your most hated villain and you'll find some heroic components. Obamacare is a candidate example.

Is that the steelman you're looking for? It seems a bit too vague for me. It's certainly not a use *case*. It's not really even a usage pattern. But it smells a bit like Pinker's neoliberal conception of progress. BTW, I found this, this morning: https://rootsofprogress.org/a-new-philosophy-of-progress It reminded me of Pieter's technological optimism, which I often share ... but am leery of.

On 9/28/21 9:46 AM, Steve Smith wrote:
> Yours is a pretty good strawman use case, do you have a steelman one?
> 
> My tin man ideation was more like:   If you have or can get an SPO2
> monitor (especially a continuous read like the Ouros rings),   then add
> a simple app that gets a baseline from your readings and then alerts you
> to anomolies that *might* be early warnings of *some kind* of infection
> effecting your lungs.
> 
> The idea was only really to help those who were A) trying to be
> responsible and B) unable to simply self-isolate all the time to have an
> early warning that they *might* be infected and thereby A) be (yet) more
> careful in their protection of others (more rigorous social distancing,
> better use of better quality PPE, etc) and B)be more attentive to other
> symptoms.
> 
> In other words, not wait until you are shedding viral particles wildly
> to take yourself out of the network of possible spreaders and verify
> your status more concisely before putting yourself back into the mix.
> 
> At the time, it was no more than a "silver bullet ideation", looking for
> some simple, straightforward (though maybe not so much I suppose) way to
> cut the likely R0 for an individual drastically and thereby maybe shave
> a small percentage off of R0 for the larger population.
> 
> I believe now that it was yet one more of many dumb ideas, given the
> context of the time.   I would not have considered it as a possibility
> if I had not already had a modest amount of experience with
> self-monitoring of SP02.
> 
> I've been running the NOVID app since it was young (well over a year)
> and have only a small handful of "hits", meaning over that whole time, I
> only came across a couple other people running the app whose bluetooth
> got close enough to mine to note we "might" have exposed one another.  
> I don't believe I simply never encountered another person who was
> infected in this 18 months enough *to have triggered* the NOVID app, it
> was simply that the penetration of the App into the population was not
> significant (at least in my circles).
> 
> Apparently none of us ever reported an infection (in a time window?)
> because I was never notified I was at risk of having been infected (by
> another NOVID app runner).   I haven't checked but I am guessing that
> well motivated effort was for naught, but I'm glad they (and those of us
> who tried it out) tried...   I do believe that my own social
> distancing/masking/etc likely lowered *my* chances of having caught
> COVID this last 18 months significantly and by extension the probability
> of me contributing to the spread.   When R0 is near 1.0, a small delta
> makes a big difference.
> 
> 
> On 9/28/21 8:45 AM, uǝlƃ ☤>$ wrote:
>> Well, it's worth noting that SpO2 is *not* a good approximation to SaO2 under a pretty large variety of conditions. So, given that most of our electronic health record data is seriously dirty already, adding even more, dirtier, data doesn't seem all that useful to me.
>>
>> Add to that the fact that normal people don't even use any health apps at all (e.g. <https://www.healthit.gov/data/data-briefs/individuals-access-and-use-patient-portals-and-smartphone-health-apps-2020>) and whatever data we do get from such devices will probably be too biased to be statistically useful, at least until we de-bias it.
>>
>> But de-biasing data isn't very easy. Crossing threads back to the inverse map from phenomena to generator: https://www.thelancet.com/journals/landig/article/PIIS2589-7500(21)00137-0/fulltext
>> "Accuracy was greater in White (90%, 89–91) and Hispanic populations (91%, 88–94) than in African (84%, 81–87) and Asian populations (82%, 78–86; table)."
>>
>> On 9/28/21 7:31 AM, Steve Smith wrote:
>>> When COVID hit, I was using a fitness band that was *capable* of reading
>>> SPO2 but it required a deliberate "stop and hold the band tight against
>>> the wrist and trigger and take several readings but accept the
>>> highest"...  there were one or more new "fitness rings" out which
>>> presumably did better (better contact with capillary rich skin.   
>>>
>>> It seemed to me that were such devices *widely* available, they could be
>>> used to predict COVID before other symptoms (it seemed like low SPO2 was
>>> an early symptom) grew noticeable.   At the time, we were most worried
>>> (and rightly I think) about RO numbers and fast self-quarantine would
>>> cut the contact-component down fast (for anyone out and about) down
>>> better than anything.
>>> If wishes were fishes.
>>>> Ah, I see. No,this infection is run of the mill. Just snort some salt water and wait it out. I monitored my SpO2 as a signal whether to get a covid test. Since it never dropped very low, I had no fever, no loss of smell/taste, etc., I didn't bother to get a covid test. Had any one of those obtained, I would have gotten a pcr.
>>>>
>>>> Thanks for the idea that low SpO2 might require more heartbeats. I hadn't thought of that either.
>>>>
>>>> On September 27, 2021 6:05:23 PM PDT, Steve Smith <sasmyth at swcp.com> wrote:
>>>>> On 9/27/21 4:11 PM, uǝlƃ ☤>$ wrote:
>>>>>> What am I struggling with?
>>>>> "But while fighting my infection"   I took this to mean you were "struggling" with an infection.  I understand/appreciate that your SPO2 numbers weren't necessarily causing you any symptoms... I assume you were measuring them for some reason though?   Curiousity I get... I used mine as crude biofeedback to (re)learn how to breath properly, but most of the time I was taking readings out of curiosity...  trying to understand correlations between what felt like a good, hard measure (SPO2) and various activities and symptoms.
>>>>>
>>>>>>  Thanks for the stories about SpO2. They nicely demonstrate that variation is normal. To be clear, when I talk about SpO2, I'm not talking about symptoms at all. I'm simply talking about the number that comes from the little machine. I've never had any symptoms that correlate with a low SpO2 measurement. And I think your (and Nick's) stories indicate that there's little, if any, correlation between the two (symptoms and low SpO2).
>>>>> I'd say that the effects of low SPO2 are less obvious (to a point) than
>>>>> one would imagine...  I can't say that when I was down in the 70s, there
>>>>> was no correlation with my fatigue, chills, blue lips and fingernails,
>>>>> etc...
>>>>>> However, what was interesting to me during this very normal cold was my elevated heart rate. Even though I quit running seriously about 5 years ago, my resting heart rate is ~63. I've never really monitored it through other infections. But because I happen to have that number along with SpO2, now, I noticed that at the nadir/height of the infection, my resting heart rate was ~100 or ~90 bpm. It's about 80 now, on day 10 since symptoms started. It just never crossed my mind that infections like the rhino would raise your heart rate. But I guess it's common.
>>>>> One might guess that low SPO2 might raise your heart rate to deliver the
>>>>> same amount of O2 per unit time?
>>


-- 
"Better to be slapped with the truth than kissed with a lie."
☤>$ uǝlƃ