Sajid Javid Orders Review Of Medical Device Racial Bias

[naomi24]

//A review into whether medical devices are equally effective regardless of the patient's ethnicity has been ordered by Health Secretary Sajid Javid.

Research suggests oximeters, which are clipped to a person's finger, can overstate the level of oxygen in the blood of people from ethnic minorities.

Ministers want to know whether bias could have prevented patients receiving appropriate Covid treatment.

Mr Javid said any bias was "totally unacceptable". //

https://www.bbc.co.uk/news/uk-59363544

The word bias smacks of something quite deliberate which I'm certain isn't the case. Should Mr Javid et al choose their words more carefully? I think so.

[bobbinwales]

Maybe but do's biased have to mean deliberately. Unconscious bias?. A biased dice?
Bit yes words should be thought through but it depends on whether it was a prepared speech or off the cuff answer

[calmck]

If in fact the oximeter does record higher levels of oxygen in people of colour this is something the manufacturers need to address. A machine cannot be accused of "bias". It would be ludicrous to suggest NHS staff, if the machines do record wrongly, know this and by using the machine on coloured patients do this deliberately

[Ken4155]

Perhaps, instead of using the word 'bias' (which is open to misinterpretation), they ought to have said that oximeters are proving to be less effective on dark skin.

[mushroom25]

since the devices work by beaming light through the blood, it does seem a little incredible that it never occurred to the makers that skin pigmentation might affect the result?

of course it could be that the makers did take that into account, but the recalibration process is such a faff that staff using them can't be bothered.

[THECORBYLOON]

He said, " "It is easy to look at a machine and assume that everyone's getting the same experience.

But technologies are created and developed by people, and so bias, however inadvertent, can be an issue here too."

By including, "however inadvertent" it is clear (to me) that he HAS been careful with his words.

[barry1010]

These machines don't work properly for people with Reynaud's Syndrome but very few manufacturers mention that in their blurb. It can cause a great deal of anxiety as it shows very low oxygen in the blood when there really isn't a problem

[pastafreak]

Maybe the "bias" here has more to do with statistics than any kind of prejudice.

[pat2604]

My answer would have reflected what Ken has written.

But this indicates that racialism is involved according to Javid and I think he uses the word bias with that intention.

Mr Javid said: "It is easy to look at a machine and assume that everyone's getting the same experience.

"But technologies are created and developed by people, and so bias, however inadvertent, can be an issue here too."

It has come to light because of Covid. It's important. But the media will make it a racist issue as usual. The media will thrive on it.

[dannyk13]

I have an Oximeter and the notes that came with it did state that under certain circumstances the reading could be affected eg sweat or grease on the finger.

[rowanwitch]

We have known about this for years, it's why in the old days we were taught to look at the patient as well as the device and take into account their underlying condition. If they have enough breath to walk round the ward and chat to you but the device is giving a low reading first check the device but it might be they have adjusted to a lower oxygen percentage. I.e it's normal for them. . If it gives a normal reading and the patient is breathless deal with the patient if they aren't improving, suggest the nice junior doctor practices taking arterial blood for gases. About the time I was leaving the NHS it seemed the rule was nurse the numbers, not all the new nurses but a lot of them. Had a few discussions with them when Dave was in hospital along the lines of " of course his sats are rubbish, he's got lung cancer and end stage COPD, and no you can't turn the oxygen up as he will stop breathing! '

[pat2604]

I had two oximeter readings after my fall. They varied considerably. I was told by the nurse this often happens even when a patient changes position or on sweaty hands because of being anxious.

[bednobs]

it's been anecdotally known for years that o2 Sat's monitors do not work as reliably on black skin as on white. I am amazed that it hasn't been addressed previously

[bednobs]

They are also unconsciously biased against people with fake nails

[bednobs]

Ps rowan YES YES YES!

[Spicerack]

Ever since Sajid lied about receiving that threatening (racist) letter, I don't put too much credence in anything he says.

[THECORBYLOON]

What letter did he lie about?

[youngmafbog]

//I am amazed that it hasn't been addressed previously//

Perhaps its not easily fixed, at least in one machine. And can you imagine the problems if they had two machines (what about the mixed races etc).

As with all things like this they are a guide.

[youngmafbog]

CORBYLOON, it was the "Punish a Muslim" abusive letter.

[rowanwitch]

All you need is an adjustment chart based on pigment density. Match the shades are to the patient and add a set percentage to compensate. A decent sized study would give you fairly accurate ranges.

No one should have pulse oximetry on false nails, even nail polish distorts the reading. Use an earlobe adaptor or try a toe.

[Peter Pedant]

bin drinking
here is an email I sent the Beeb last night
do I really send insulting and abusive emails to our hallowed and respected broadcaster? - yes I do
You will see that 1) ABers get off lightly when it comes to vulgarity, ridicule and abuse for moronic and stupid views.
2) Sjoding was interviewed tonight and unfortunately kept to the view " it is all true and I need very large grant and many many coal face workers to show what I said was true true true and then I will be even more famous"

you are talking about 10% errors ( peanuts ) and are they errors which are constant - always 10% under
or 10% under only 10% of the time ?

The other thing they should show is how to get an arterial sample ( gold standard )out of a kid. Putting a drip up is bad enough.

Dear Sir

The Beeb bits of fluff doubling for hacks have a truly appalling history of garbling science news articles. I note for some reason pulse oximetry is to be indicted for racial bias.

For the hacks' education, the relevant paper is here

https://www.bmj.com/content/372/bmj.n298

which is on the old side - Feb 2021 I think

Sjoding who started it all off - here Dec 2020

https://www.nejm.org/doi/10.1056/NEJMc2029240

and a comment from a London doctor here

https://www.nejm.org/doi/full/10.1056/NEJMc2101321?query=recirc_curatedRelated_article

The doctor says it is not race but how dark your skin is


I write this email to head off - or perhaps give a head start to the hack ( fluffy, obliging, gullible, easily led, out to lunch, marxist, tree hugging, -ology graduate ) delegated with this task ( if there is one of course) and prevent the usual awful pig's ear the Beeb makes of straightforward science subjects

Luv PP

Sjoding ( hi mum! ) rather dropped the ball, I thought.
In summary Sjoding said, "It reads low
it does not consistently read low"
but has it led to increased deaths? no..
Increased deaths is not good enough. You have to show the increased deaths are a result of using these duff oximeters, which would not occur if you just sat and looked at them

and you can tell THAT by showing that the death rate where oximeters were used was higher than those where you used a thermometer and a bottle of white mixture only

it isnt by the way - the death rate is NOT lower if you dont measure saturation

This is a standard witch hunt social construct