Are Pulse Oximeters "Intrinsically Biased Against People of Color?"

A Wall Street Journal article suggests that pulse oximeters mismeasure blood oxygen levels in American minority populations, but the article fails to ask some important questions.

I am a huge advocate of using pulse oximeters, which measure blood oxygen saturation levels, to help monitor how Covid is affecting one's lungs. A pulse oximeter may have even saved my life; when I caught Covid in fall 2020, I had no trouble breathing but my handy pulse oximeter told me that my measurements were at a level where I should seek medical attention, and physicians in my family told me to go immediately to Georgetown Hospital. Once there, they found significant pneumonia in my lungs, admitted me, treated me, and released me good as new (after a few weeks recovery).

One concern about pulse oximeters, however, has been that they may work less well on people with dark skin. Specifically, that they slightly overestimate blood saturation levels; that slight overestimation, however, can have significant consequences regarding medical attention and treatment.

Today, the Wall Street Journal breathlessly reports on a study that seems to confirm that the concern is warranted.

The recent study looked at more than 7,000 patients diagnosed with Covid-19 between March 2020 and November 2021 at five hospitals in the Johns Hopkins Health System.

Pulse oximeter measurements were taken for all 7,000 patients. More-accurate but invasive arterial blood gas measurements also were taken from about 1,200 of the patients.

When the researchers compared measurements from the arterial blood gas tests with pulse oximeter readings, they found discrepancies among patients of all races but a more pronounced difference for patients of color. Compared with white patients, pulse oximeters overestimated oxygen saturation by an average of 1.2 percentage points among Black patients, 1.1 percentage points among Hispanic patients and 1.7 percentage points among Asian patients.

Some doctors, not quoted, expressed reservations to the Journal, but Dr. Martin Tobin stated that the study shows that pulse oximeters are "intrinsically biased against people of color." That's an oddly unscientific way of putting it; at most, we can say that the device gives less accurate results for people with darker skin pigmentation. Many "white" people have darker skin than many Hispanic, Asian, and some Black Americans.

Which brings me to some concerns about the study. First, you want to measure how accurate pulse oximeters are for people with dark skin, why not directly study that? Self-identified race and ethnicity is a crude substitute for skin tone.

This is particularly true given the internal diversity within each category. The study in question used data from the Johns Hopkins health system. Is the Hispanic and Asian population of the mid-Atlantic representative of the national Hispanic and Asian population in terms of skin tone, given that they have populations with different national origins (e.g., way fewer Mexican Americans on the East Coast than in the nation as a whole)? If so, how dark-complexioned are they compared to whites and African Americans? I doubt anyone has even considered those questions.

And then the results themselves are odd, if skin tone is really the decisive factor. Recall that the study finds that "pulse oximeters overestimated oxygen saturation by an average of 1.2 percentage points among Black patients, 1.1 percentage points among Hispanic patients and 1.7 percentage points among Asian patients." I don't need a scientific study to tell me that the average African American is darker-complexioned than the average Hispanic or Asian American (though if you want scientific studies, we do know from DNA studies that Hispanic Americans have a much higher percentage of European ancestry than do African Americans, which would be reflected in average skin tones.) If dark complexion were the issue, we would expect the overestimation to be greatest among Black patients, not Asians, and we would expect the gap between Black and Hispanic patients to be greater.

The reporter, apparently caught up in the "people of color" paradigm, does not seem to even notice these questions.

The article concludes:

The FDA recommends accuracy studies include 10 or more healthy subjects who vary in age and gender. At least two participants, or 15% of the participant pool, should be "darkly pigmented." The FDA said it was evaluating whether its guidance should be modified.

It strikes me that what needs to be done is a wide scale study that ignores America's unscientific official racial and ethnic categories and directly measures subjects' skin pigmentation, and measures whether darker skin pigmentation correlates with less accurate results. And if the government wants to measure whether pulse oximeters indeed work less well on, e.g., "Asian Americans" for reasons other than complexion, it needs to break the categories down by subgroup; there is no particular genetic or sociological commonality between South Asians like Indians and Pakistanis and East Asians like Chinese and Vietnamese.

Comments

[Joe_dallas]

Pulse Oximeters are relatively crude devises for measurement. The delta between races (skin color) is less than the normal error range

[Joe_dallas]

the accuracy range is 3%-5%. The delta for “asians” ( a very wide and diverse skin color ) is 1.7% which implies the study is reaching a conclusion based data that doesnt support the finding.

[jimc5499]

At least someone gets it. A person who has one of these should be able to establish a baseline. Let’s face it, if you read 98 when healthy and one day you are reading 78, you might want to see a doctor.
It’s sad that they call this “science”. All they did for the study was to gather data that was already there and played the “race card”. They didn’t examine anybody. A true study would have measured the skin pigmentation set a value for that and then compared the readings, with no identification of race. Do that enough times and the pulse oximeter manufacturers could make one that was adjustable for a pigmentation value.

[Brett Bellmore]

” Let’s face it, if you read 98 when healthy and one day you are reading 78, you might want to see a doctor.”

If you read 98 when healthy, and one day you’re reading 78, it’s probably somebody else making that call, you’ll be unconscious. The need for urgent medical attention kicks in at about 92%.

I agree that the pulse oxiimeter manufacturers should probably add the capability of measuring and compensating for pigmentation.

[Rossami]

I think you’re missing the point, Brett. Pulse oximeters are a crude measure of absolute blood oxygen level but a more reliable measure of relative blood oxygen level – and that remains true largely regardless of your skin color.

Consider an albino and someone with melanism (I’m not sure what the noun is for someone with this condition – melano?), both with a true blood oxygen level of 95. The albino takes multiple pulse ox readings that average to 96. The melano’s average shows as 90. That difference doesn’t matter. What matters is that tomorrow, they both start feeling poorly and take new measurements. The albino shows 86 and the melano shows 80. In both cases, that drop of 10 is significant and concerning. That drop of ten is also reliable even though the absolute measurements are not.

This is consistent with the FDA advice on evaluation of pulse oximeters – “When taking pulse oximeter measurements, pay attention to whether the oxygen level is lower than earlier measurements, or is decreasing over time. Changes or trends in measurements may be more meaningful than one single measurement.”

[Frank Drackman]

Have you ever seen blood? Everybody’s is Red (either bright or dark depending on Oxygen Saturation)
thats how the Friggin Pulse Oximeter works!!! ( I could tell you exactly how, but then I’d have to kill you(not really, but can’t wait to see “Top Gun/Maverick”

[Rossami]

Which has nothing to do with the discussion above about the skewing effect of skin pigmentation because the pulse oximeter measures the passage of light through both blood and skin. Even if everybody’s blood was exactly the same red (because their oxygen saturation was identical), more light will pass through the albino’s finger.

[jimc5499]

Just using numbers for an example. I’m an engineer not a medic.

[Brett Bellmore]

Me, too, but I was studying to design medical instrumentation, so human biology was my second major.

[jdgalt1]

That adjustment is the perfect solution. So long as people using the device know to make the adjustment, no race issue should arise.

Of course that will not be acceptable to the DEI crowd, who want a race issue to arise. So they will rule that pulse oximeters are illegal under the ADA and take them away from everybody.

[Davedave]

“So long as people using the device know to make the adjustment, no race issue should arise.”

But they don’t. The research hasn’t been done. That’s what this is about.

What actually happens is that the pulse oximeter readings are given far too much weight, and patients who are obviously struggling to breathe are told they’re fine because the reading isn’t low enough.

[DaivdBehar]

I% is not clinically meaningful. Take a deep breath. You can change the reading more than 1% on yourself. The cut off is 92%. However, repeat readings after deep breathing in the 80s requite action. Study is nitpicky,misleading and politucal. Did you say, Hopkins, as from Baltimore? Dismissed.

[DaivdBehar]

The color of skin was not measured nor was dryness. Race was self identified. Any difference may come from education and from following instructions. Those racial differences were far smaller than the official error range of 6%. Trash study from woke Hopkins.

[DaivdBehar]

Yes, Hopkins, you woke dipshits. The device engineers colluded to have dark skin people show 1% lower oxygen blood saturation.

[Flight-ER-Doc]

In 40 years as an Emergency Physician, a 2-point difference would not be of clinical concern. 97% vs 95%? Thats OK. 89% vs. 91%? Doesn’t matter, both would be noteable and worth considering the cause, and correction.

Pulses oximetry is a valid, if crude, tool to help, not define, treatments.

[Joe_dallas]

Flight doc – I agree with your point – ie that oximeter are a valid tool, My point is the measurement error for each reading is greater than the delta with the various races. The point being is that it is hard to reach any conclusion when the error is less than the margin of measurement error.

See my comment on the Bell McDermott study on premature mortality and increases in ground level ozone

[Flight-ER-Doc]

That too.

[Frank Drackman]

It’s not the Hypoxia, it’s the Hypercarbia

[Dr. Ed 2]

Yes, but when you realize how invasive the arterial blood test is, they are damn nice to have. And HOW accurate a reading do you need, particularly if you norm the instrument against the person when healthy?

As long as the variance is consistent, and as long as the medical people are bright enough to understand consistent variance, it becomes a moot point. If the person’s dark skin causes it to read 1.5% low, it will always be 1.5% low.

You COULD calibrate it for the person, and back when these things cost a few thousand dollars, it might have been worthwhile doing so.

[Brett Bellmore]

“If dark complexion were the issue, we would expect the overestimation to be greatest among Black patients, not Asians”

I’m guessing you’re unaware that “Asians” are a very diverse group indeed, varying from quite light skinned to as dark as some Africans?

[Mr. Bumble]

I don’t believe that most Americans consider that people from the Indian sub-continent are Asian.

[Brett Bellmore]

Are Thais not considered Asians? How about Filipinos?

There’s a curious dynamic here, that celebrities from this region tend to be quite a bit lighter skinned than the general population, I think this causes people to not realize how common dark skin is in Asia, even outside of India.

[Bob from Ohio]

“Are Thais not considered Asians? How about Filipinos?”

Rough determination is Pacific versus Indian oceans.

So both Thais and Filipinos are probably Asian.

[Squirrelloid]

I mean, I wouldn’t consider anyone ‘asian’ racially, because its a nonsense term. (Grouping Tatar tribes with Hmong or Bamar in one group doesn’t make any sense).

But I’m not sure Filipinos are asian, even in the US government’s weird classification. Wouldn’t they be Pacific Islanders?

[David Bernstein]

Asian, not Pacific Islanders according to the US government.

[Flight-ER-Doc]

What about Maori’s? Australian Bushmen? Are they Asian or Black?

A ‘Hispanic’ (according to federal standards) like me (while I look Dutch, mostly), vs an Afro-Carrib?

A stupid study, designed and conducted to create more racial foment rather than add to the body of knowledge. But, easy to do, rather than actually classify skin tone – which is hard to do: Which part of the skin do you classify? The dorsum of the hand? The buttocks? Buttocks laterally, or deep in the gluteal cleft? And most importantly, WHY?

[Bob from Ohio]

“Maori’s? Australian Bushmen? Are they Asian or Black?”

Bushmen look black so are. Maori’s look vaguely Hawaiian so are. This isn’t science, its not consistent at all.

US racial classifications are bizarre. Arabs include Iranians to many, many Americans.

[Lee Moore]

Maori and Hawaiian are both Polynesian, originating fairly recently – ie within the last 1,000 years, from somewhere round the Marquesas. So they’re about as different as Germans and English, ie not very.

Whereas Bushmen are VERY different from say West Africans, and not closely related at all. Nor are Bushmen particularly black. Browner than whities but much less so than Nigerians or Tamils or Melanesians.

[Bob from Ohio]

I took “Bushmen” to mean Aboriginal. I’d disagree they are not dark skinned like an African but of course they are not closely genetically related.

It doesn’t matter, we are talking perception, not science.

[Lee Moore]

“Maoris” prefer to be called “Maori”

[jimc5499]

I have some Filipino friends who are darker complected than some Blacks.

[Brett Bellmore]

Heck, my half Filipino son is darker than half the blacks around here. Probably a consequence of that silly ‘one drop’ rule.

[Frank Drackman]

What color was your Postman?

[Gandydancer]

All that is necessary for the quoted sentence to be true is that blacks on average have darker skins than Asians. And he observes, correctly, “I don’t need a scientific study to tell me that the average African American is darker-complexioned than the average Hispanic or Asian American.” Your sentence anyway contradict nothing I noticed him say.

[Á àß äẞç ãþÇđ âÞ¢Đæ ǎB€Ðëf ảhf]

I’m guessing you’re unaware that “Asians” are a very diverse group indeed, varying from quite light skinned to as dark as some Africans?

I’m guessing you didn’t read this fine article, nor are you familiar with his other fine articles about the same subject.

[Brett Bellmore]

Yeah, he’s apparently aware that Asians vary in coloration, but none the less reasons as though coloration couldn’t be the cause of their readings being systematically different on average. He treats it as a confounding variable.

The point here, and I agree with him, is that these political “racial” categories are too arbitrary to be of any particular scientific/medical use. They need to look at the actual biological reality that correlates very poorly with political “race” categories.

[John F. Carr]

If all these devices use the same basic technology somebody should fund a test to determine correction curves based on facts that can be measured.

[Rossami]

You could but it would be largely pointless. As the FDA and everyone except, apparently, the author of this “study” already says, pulse oximeters are not especially accurate for point measurements but they are reliable for identifying trends.

In other words, I don’t much care about the number you test today. We already knew that it can be skewed by skin pigmentation, skin thickness, and tobacco use to name just a few. But I care a great deal about the difference between today’s number and yesterday’s number because all those confounding factors are constant (in the short term) and therefore cancel out. The difference is the useful measure.

[Davedave]

But that is not how they are used. >92% reading, get sent home. <92% reading, get admitted to hospital. It might be stupid, but that is what has actually been happening.

[Rossami]

It is both stupid and contrary to the manufacturers’ instructions, the FDA’s approval documentation and all competent medical advice. If it is being done by a medical professional, you should consider reporting their incompetence to the appropriate board. If it’s being done by a non-medical-professional … why would you be listening to their advice in the first place?

[rsteinmetz]

For some time I’ve been concerned that many such “studies” stop at race, as crudely defined, and don’t account for other factors which could be more significant and often repeat conventional wisdom, which is not verified.

For example were the results of this finding correlated to age, obesity or other health conditions?

There seems to be false assumptions baked into a lot of this. It is for example well known that blacks on average are poorer that whites. However if you do some math you will quickly find that most blacks aren’t poor and there are more poor whites that poor blacks.

In my home city with a high black population and a high poverty rate, even is every single poor person was black more than half of all blacks couldn’t be poor.

I haven’t researched enough to see if the actual researchers did the necessary work and the journalist failed to comprehend the results.

[ReaderY]

Professor Bernstein’s consistent rhetorical strategy, a common one, has been to be demand that racial classifications be perfect. If any deviation from perfection is an unmitigated evil, then the world as a whole is pretty much unmitigated evil. This is no different.

As in any classification, or pretty much anything in this world, this one isn’t perfect. But imperfect classifications, like imperfect things generally, can nonetheless be highly useful. If we start from a premise that we can accept imperfections if the overall classification scheme provides value, as we pretty much have to if we want to accomplish anything at all, Professor Bernstein’s argument falls to the ground. As Aristotle put it, “the perfect is the enemy of the good.” This is no different.

It would be a mistake to characterize the development of pulse oximeters as reglecting intentional racial discriminatation. The existence of natural biological differences is not the result of human design. But the current ideological rhetoric that racial differences have no biological implications is also false.

So if there are small differences in pulse ox normal ranges by race, it’s perfectly reasonable to take race into account. There may perhaps be better ways of arriving at a more accurate individual normal range. But race is a perfectly reasonable start. If it gets you a better result than starting with the population as a whole, it adds value.

And the possibility Professor Bernstein raises that with additional work a different indicator yet to be proven might add more value in no way detracts from the value of using race as an indicator and is in no way a legitimate argument against using it.

[Joe_dallas]

While I agree that are biological differences in the races (genes do make differences), the difference in measurements between the “races as caterogized ” is significantly less than the overall measurement error range.

Similar errors in the Bell mcdermott study of ground level ozone and premature mortality of 96 us cities which is considered the “gold standard” of attribution studies. Measurement error range far exceeding the study’s conclusion.

[David Bernstein]

“Professor Bernstein’s consistent rhetorical strategy, a common one, has been to be demand that racial classifications be perfect.” My complaint is no that they aren’t perfect, but that they are arbitrary. In this particular case, they are using being a racial minority as a proxy for an average having darker skin to study the effects of darker skin re a medical device. That much is ok. Then it turns out that in terms of skin tone, we have something like “whites lightest, then Asians and Hispanics, then blacks darkest.” But then the results are “whites most accurate, blacks and Hispanics basically tied, Asian least accurate.” So the results don’t correlate with average skin tone. The gist of the article though, is that they do, because the reporter (and doctor quoted) ignore the more granular, albeit still crude, data re racial group in favor of looking as “people of color” as if they are an undifferetiated mass. Which is arbitrary and absurd, not merely “imperfect.” And all completely unnecessary, because skin tone can be measured directly.

[Sarcastr0]

So doctors ignore it. Reporters ignore it. Everyone including most of the commenters here seem to know what’s being communicated here.

Is it such a big deal that race and skin color are being used synonymously, as they often are in common conversation?

[Stephen Lathrop]

Standard racial classifications are not arbitrary in cases—maybe like this one—where standard racial classifications will predictably be the basis on which the study results get applied. However much additional accuracy might be provided by some other research modality, if in practice the results were applied on the basis of standard racial characteristics—clinicians estimating racial background, or self-reporting—some of that extra accuracy might then become a wild card which disrupts treatment, however slightly.

[Ridgeway]

Don’t the results also imply that skin “darkness” is in fact not the relevant factor, and that the factor(s) that causes the measurement inaccuracies is only partially correlated with skin color?

[Sarcastr0]

Isn’t partially correlated all you need here?

[Ridgeway]

Not necessarily. The cause could be something like tobacco use (to make up an example) where the “races” sampled happen to use that product in greater percentages than do whites.

In that instance, looking at skin color as a proxy for some sort of relevant genetic difference would be a red herring.

[Sarcastr0]

But until we know the cause, any non-swamped correlation is relevant to diagnosis and treatment, no?

[Ridgeway]

Not necessarily.

Assume for the sake of argument that tobacco use causes the oximeter to overcount oxygen levels by 5%. Telling an Asian non-smoker to adjust his oximeter to read 1.6% higher will actually make his readings less accurate.

My point is that skin darkness as the determinative factor is pretty clearly wrong given this data, so you need to look at the data more granularly (or collect different data) to try to find a better potential cause.

[Sarcastr0]

You assume no other variables will be looked at???

[Brett Bellmore]

Hard to say. Could, given the irrational basis of the ‘racial’ groups being used, just mean that skin darkness was only partially correlated with the ‘race’.

[Sarcastr0]

Crying about the irrational basis of normal language is like those who yell how gay people aren’t really happy.

You’re taking an interaction whose function is communication, and complaining it does not create a rock-solid consistent system.

[David Bernstein]

Would you be ok with the Skin Cancer Foundation stating that “white people need to be especially careful to put on sunscreen before going out in daytime during the summer, but people of color are less likely to get skin cancer from sun exposure,” as opposed to “people with fair complexions need to be especially careful… but people with dark complexions ….” Are these statements about equal in the accuracy of the information they convey?

[Sarcastr0]

…yes. I know exactly what they are talking about.
I’m so white I burn even in cloudy days is a thing I heard literally this weekend,

[njanetos]

“…Dr. Martin Tobin stated that the study shows that pulse oximeters are “intrinsically biased against people of color.” That’s an oddly unscientific way of putting it…”

To the contrary, although it may not be how he meant it, that’s the statistically correct term for the finding (the measurement is biased).

Some comments note that the bias is less than the range of error. Bias is a little different though and means some % of the population with a bias will be wrongly categorized even if the bias is less than the error.

The rest of your points are well taken.

[David Bernstein]

Bias is the right scientific term. “Biased against” is not.

[David Bernstein]

And even if “biased against” was correct scientific terminology, “biased against people of color” is not. If the presumed problem is pigmentation, the device won’t be “biased against” a light-skinned Hispanice, black, or Asian, but will be “biased” against a dark-complexioned “white,” say an Armenian or Greek with dark skin. So if the device is “biased against” a definable group, it’s “people with dark skin,” not (what we in the US call) “people of color.”

[Sarcastr0]

Everybody conflates race with skin color. This is normal parlance; you’re railing against common speech as though it were legal writing.

[VinniUSMC]

Yeah, who cares about precision in language? Doctors and scientists should just use normal parlance.

[Sarcastr0]

This situation is not a clinical study. Doctors and scientists use what language they need for the moment. It’s pretty easy to switch from common parlance to scientific jargon when you need to!

People call all black people African American even if they’re neither black nor African American. Somehow we survive.

[Bored Lawyer]

No, it’s in a scientific paper that should know better. Using “race” when one should refer to skin color is politically motivated stupidity. Everyone knows that “blacks,” “whites,” “asians,” and “hispanics” each have a wide variety of skin tones, even if on average some races are darker than others.

[Sarcastr0]

The OP is about a Wall Street Journal article.

The scientific paper seems to be about how skin color discrepancies will have a disparate racial/ethnic impact. Do you think that’s innacurate?

[David Bernstein]

“People of color” in common parlance means people of certain ethnic heritages, excluding European heritage and sometimes others. The term has been politicized so that, e.g., the very pale Linda Sarsour has persuaded the media to call her a “person of color” because she wears a hijab. Regardless, people of the relevant ethnic heritages have a wide range of skin colors, including Sarsour paleness and ever fairer. OTOH, people not deemed people of color can be quite dark. A medical researcher commenting on a study should know better than to conflate “person of color” with “people of dark complexion.”

[Mr. Bumble]

When will we be seeing the study on flesh colored band aids?

[John F. Carr]

You reminded me of https://www.gocomics.com/bloomcounty/1989/01/05

But see the followup https://www.gocomics.com/bloomcounty/1989/03/14

[jimc5499]

You can buy band-aids in various skin tone colors including what would be called Black.

[Brett Bellmore]

I need some in a freckle camo pattern. No solid color isn’t going to stand out against MY skin.

[Craig Johnston]

David, I also had COVID in the fall of 2020 and I agree about the value of the pulse oximeter. For me, it triggered a trip to the VA ER which diagnosed pneumonia and COVID and sent me home with antibiotics (the COVID ward was full). You should be sure your lungs are clear now. The pneumonia that comes with COVID can leave behind fibrotic tissue that show up in x-rays as “ground glass opacities”. Happily, these changes don’t progress like pulmonary fibrosis, but better to know than not.

[David Bernstein]

They sent me for a followup lung x-ray which was clear, but my understanding is that this isn’t a guarantee of lack of lung damage.

In my case, remedisvir had just been approved and at the time to do the five-day full course, you had to be admitted to the hospital. Also was given steroids.

[docduracoat]

As a board certified anesthesiologist with 30 years of clinical experience, I can tell you that pulse oximetry has been the biggest advance in my career.

I will say a change of 2 in the saturation is insignificant.
That is within the margin of error of the device.

[Flight-ER-Doc]

Thank you.

And it’s of more clinical significance in showing trends in oxygenation, than absolute levels.

[Brett Bellmore]

Unfortunately, I’ve found that mine is pretty unreliable, on account of chemo IVs having really wrecked the circulation in my hands. It’s probably accurately reporting the oxygen saturation in my finger, but I don’t need to know the O2 saturation in my fingers, I need to know it in the rest of my body, and due to impaired circulation they don’t have a lot in common.

[Frank Drackman]

ET CO2’s more important,
Just ask whatever sap Ologist was “Supervising” Bill Clinton’s CRNA Mom when she “Tubed the Goose”
Jeez, Shysters talking medicine is even worse than Sawbones talking Law,

Frank

[jimc5499]

Nice to hear from an expert.

[Krayt]

The FDA recommends accuracy studies include 10 or more healthy subjects who vary in age and gender. At least two participants, or 15% of the participant pool, should be “darkly pigmented.

That would be 1-2 dark pigment testees, which isn’t enough statistical certainty to stick in your eye. 10 isn’t enough. They are barely proof of concept early test suggestions things might be good enough to warrant further testing.

[PeteRR]

The reading is nice to have, but the changes in the reading are more important. At the beginning of the pandemic, I used a pulseox to monitor my GF and I. She consistently read 96% and I read 93%.

The readings never varied and I never worried as a consequence.

[M L]

The oximeters should be charged with a hate crime.

[Longtobefree]

Of course, the downside of a pulse oximeter is that it shows numerically how much the cloth mask that doesn’t stop the Communist Chinese Virus does stop air.

(and yes, the study was bullshit as far as scientific rigor is concerned)

[jorgyusa]

There is just too much junk science being reported as real science. Pulse oximeters are not accurate devices and the tests in this were not controlled. It was just lazy crappy reporting which we see too much of these days.

[Frank Drackman]

Pulse Oximeters overrated. With Cyanide Poisoning you’ll stay at 100%, right until you go into Asystole from the Histotoxic Hypoxia.

[Brett Bellmore]

Eh, you just need a higher end pulse oximeter. Uses multiple frequency bands so that it can discriminate between hemoglobin in different states.

[Frank Drackman]

Co-oximeters? been around for centuries (well 20th, and 21st), never seen one

[Frank Drackman]

Want to talk Medical Discrimination,
Why does Jay-Hey give Afro-Amuricans Colon Cancer 5 years before every one else? (Just another example of a Surpreme Being keepin a Brother down) At least that’s what the American College of Gastroenterology’s Colonoscopy recommendations were until last year, when they recommended everybody get one at 45 (too complicated the other way, does Obama get one at age 47.5?)
And “Reverend” Kirkland, I didn’t say anything about Watermelon/Fried Chicken (love both of them)

Frank

[Stephen Lathrop]

. . . there is no particular genetic or sociological commonality between South Asians like Indians and Pakistanis and East Asians like Chinese and Vietnamese.

I am fairly certain Bernstein does not have basis for that. It is a complicated question, with factors easy to overlook, including sociologically- and historically-mediated factors.

I am surprised no one has yet mentioned the possible confounding factors of either personal history, or genetic selection, associated with living at various elevations. Without speculating on what oximetry results might show, I would guess random samples of U.S. Europeans, Asians, Hispanics, and Africans would sort roughly in that order on a variability-of-elevation-in-personal-background criterion—whether considering genetic backgrounds, or personal-experience backgrounds.

My unscientific take on geography and world population is that Europeans and Asians live at lower elevations in population fractions higher than do Hispanics and Africans. My sense is also that Africans and Hispanics encompass historically greater variations in residence elevation. High plateaus in Africa, Central America, and South America support populations which are notable fractions of the entire populations from those regions. High plateaus in Asia are extensive, but not populous compared to the lowlands. Populated high plateaus in Europe barely exist.

There is no reasonable doubt that extended personal experience at higher elevation alters oxygen metabolism, compared to sea level. It seems plausible that multi-generation selection among populations living at especially high elevations might confer genetic differences in oxygen metabolism. That may already have been observed and measured.

Anyone seeking more scientifically-based insight into oxygen metabolism among populations could hardly ignore that elevation question. Sociological and historical factors are linked to it. For instance, might we suppose that residence elevation played a selective role to define population subgroups which chose to immigrate to the U.S. from various regions? Might we further suppose that any such role might play a part, but look different from place to place, based on historical interactions among population subgroups involved?

Bernstein seems to be looking for a simplicity which he is unlikely to find.

[David Bernstein]

“. . . there is no particular genetic or sociological commonality between South Asians like Indians and Pakistanis and East Asians like Chinese and Vietnamese.

I am fairly certain Bernstein does not have basis for that. It is a complicated question, with factors easy to overlook, including sociologically- and historically-mediated factors.”

There are studies of the genetic origins of South Asian and East Asian people. There is very little overlap. There is actually significant diversity in origin even within the South Asian subcategory, with some South Asian groups being primarily European in origin, and others have origins in Asia (but not East Asia).
Sociologically speaking, if you have a good theory why, e.g., Pakistani American are “like” Hmong, or Bangledashis “like” Japanese Americans in any meaningful way, I’d love to hear it.

[David Bernstein]

(And fwiw, they weren’t put in the same category because anyone in the government thought they were alike in some way. In fact, South Asians were originally designated as “white”, but an Indian American group lobbied successfully to change them to Asian, and the government’s official reason for agreeing was that they got a lot of letters in support and very little opposition. Not exactly a scientific or social scientific rationale.

[Stephen Lathrop]

Did that just fly over your head? My suggestion was that the various Asian groups you mention may indeed share genetic similarity, in that most could be from populations genetically adapted to lower-elevation oxygen metabolism. In principle (I am not qualified to speak to the science) that could be true of all of them, without affecting other markers of diversity. Even without genetic diversity, it may be that lifetime experience, or decades-long experience could condition oxygen metabolism findings in a subsequent study.

To illustrate, without claiming more than anecdotal authority, I was raised for 25+ years at near-sea-level elevation, where for many of those years I practiced aerobic-intensive athletics, including swimming and rowing. Thereafter, I moved to a location at 6,000 foot elevation, and frequently went higher, to nearly 12,000 feet. I enjoyed that life for about 15 years. During my time living and exerting at higher elevation, I acclimated rapidly at first, and more gradually thereafter, but never thought I quite reached parity with many less-trained people who had been born at those elevations. The higher the elevations got, the more the natives pulled ahead. I am under no illusions that my oxygen efficiency could ever have been a match for that of an athlete born and trained on the Ethiopian plateau.

Now I am back at sea-level, and for whatever reason my resting pulse oximeter readings tend to be about 92, while I remain in excellent cardio-pulmonary health. Invariably, a few deep breaths returns the number to 97 or higher.

I do not know what to make of any of that, except to suggest it looks like a complication a pulse oximeter study ought to investigate. I do not share your confidence about knowing what is going on.

[David Bernstein]

“My suggestion was that the various Asian groups you mention may indeed share genetic similarity, in that most could be from populations genetically adapted to lower-elevation oxygen metabolism.” They may, but so may any other two random groups, there is no particular reason to think so of these groups.

[Just Eric]

This “oddly unscientific way of putting it” applies all too often these days in public debate and efforts to quash public debate.

[CJColucci]

I am a huge advocate of using pulse oximeters

That’s all very well, but I prefer to get my medical advice from doctors.

[abhudd47]

Prof. Bernstein, didn’t you fall into the trap a little bit too? It seems you came to the conversation with the conclusion that it was skin tone that was the factor determining these discrepancies, but the results of the study seem not to support that. If the lighter skinned race-group is the farthest off, then shouldn’t we be looking at something other than skin-color that is common within that racial group? Or perhaps, something other than race that those patients have in common? Sounds like the methodology here was seriously lacking anyway. For example, I don’t see any mention of cross referencing other points of commonality between test subjects. Also, it’s based on how people “self-identify,” so a couple of Rachel Dolezals in there could throw off the whole thing…

[David Bernstein]

Did you get to the final paragraph, last sentence?

[abhudd47]

I did, but I missed the words “other than.” My apologies!

[Frank Drackman]

Jeez, listening to Mouthpieces talk about medical stuff is enough to make me rend my Scrubs, THIS is how Pulse Oximeter’s work, and blood all looks the same to the nekkid eye, (Maybe Superman can tell if someone has Sickle Cell Trait with his Super Electrophoretic Vision)
A typical pulse oximeter uses an electronic processor and a pair of small light-emitting diodes (LEDs) facing a photodiode through a translucent part of the patient’s body, usually a fingertip or an earlobe. One LED is red, with wavelength of 660 nm, and the other is infrared with a wavelength of 940 nm. Absorption of light at these wavelengths differs significantly between blood loaded with oxygen and blood lacking oxygen. Oxygenated hemoglobin absorbs more infrared light and allows more red light to pass through. Deoxygenated hemoglobin allows more infrared light to pass through and absorbs more red light. The LEDs sequence through their cycle of one on, then the other, then both off about thirty times per second which allows the photodiode to respond to the red and infrared light separately and also adjust for the ambient light baseline.[29]

The amount of light that is transmitted (in other words, that is not absorbed) is measured, and separate normalized signals are produced for each wavelength. These signals fluctuate in time because the amount of arterial blood that is present increases (literally pulses) with each heartbeat. By subtracting the minimum transmitted light from the transmitted light in each wavelength, the effects of other tissues are corrected for, generating a continuous signal for pulsatile arterial blood.[30] The ratio of the red light measurement to the infrared light measurement is then calculated by the processor (which represents the ratio of oxygenated hemoglobin to deoxygenated hemoglobin), and this ratio is then converted to SpO2 by the processor via a lookup table[30] based on the Beer–Lambert law.[29] The signal separation also serves other purposes: a plethysmograph waveform (“pleth wave”) representing the pulsatile signal is usually displayed for a visual indication of the pulses as well as signal quality,[4] and a numeric ratio between the pulsatile and baseline absorbance (“perfusion index”) can be used to evaluate perfusion.[31]

[Rossami]

Your mistake is in the statement that “the effects of other tissues are corrected for”. That is simply untrue as both this study and the many prior disclosures by the FDA and the device manufacturers have already said.

More precisely, the effects of other tissues (principally the skin) are not entirely corrected for – and that difference is … significant in the eyes of folks like Dr Tobin and insignificant to everyone else who notices that the margin of error for the device is considerably greater than the alleged “racial” bias.

[EscherEnigma]

Ah yes, because nothing bad has ever happened from letting lawyers tell doctors and researchers how to do their jobs.

Nothing bad. Ever.