That's great, yeah, so thank you. So, well, first we'll start off, then we'll start off with the pulse oximeter as a source for racial disparities. Disclosures, none other than serving on the ISO standards board or committee. Overview, we'll talk about the background and how it works, some scientific studies, patient, and then in order to really try to get closer to the root cause, look at it from a patient safety perspective, try to summarize a path ahead. So the pulse oximeter, their history was invented in 1972 by Takuo Aoyagi. So, I mean, it's an amazing invention. He used two wavelengths and to basically a basic spectrometry where he'd measure blood oxygen levels peripherally and non-invasively, and that's the SpO2 versus a blood gas analyzer of SaO2. Hewlett-Packard had developed an oximeter that was actually a little bit more accurate, we think, and they used eight wavelengths, more than just two, because this issue was recognized early on. For whatever reason, you know, they got more involved in computers and they settled on two separate wavelengths. So here's how it works, where you see it's reliant on the red and the infrared waveforms and wavelengths here, and then you have a photo detection, and it's actually requires a pulsating blood flow, and you're going through the artery as well as venous blood, and then tissue, fingernails in certain cases. So you're going through and there's a lot of calculation that's involved with it, but there's a lot of good engineering that can actually solve these issues. And so pre-COVID, we knew that there were, like, looking in hindsight, the data was out there to show that this was not as accurate, in 1986, 1998, 2007, so the data was there. And then after COVID, or AC, you know, where Michael Schoding submitted his letter to the New England Journal, and they started looking at this, they noticed it at the bedside, and they started looking at this. So he came up with a term that he felt was clinically relevant, where the pulse oximeter read 92 to 96%, and the saturation from the blood gas was less than 88%, so that would make a big difference with how, you know, how you're deciding to treat patients. And in the initial group, they had an 11.4% risk of that in African Americans, and 3.6% whites, white and Caucasians, and then Valeria Valbueno did a study on patients being evaluated for ECMO, and it shows a big difference, 21.2 to 10% difference, and then Andrus et al, in pediatric patients, there's a difference as well. And here's the data that you see, these outliers. Ian Wong did some additional work. His definition for hidden hypoxemia was a little different, where the SpO2 was greater than 88%, and his SaO2 was less than 88%, but there's 70% increased risk of mortality when you have this hidden hypoxemia, and it was higher in Hispanic and black patients. And Ashraf Fawzi showed some similar differences, although maybe not to that degree of Schroding's work, but black patients had a 29% likelihood of not having timely care, and Hispanic patients had 23% likelihood of not having timely care, and so we think that a lot of this is related. Now, those are real-world studies. This is out of Phil Bickler's group at UCSF, and they're a group that does, they do standardized testing in the lab, and what they do is they take patients from a saturation of 100% to 70%. These are normal, healthy patients, and so they tried to tease this out on whether or not this was a perfusion issue versus skin tone, and so here you can see in low perfusion patients, dark-skinned folks had a 21.1% risk of hidden hypoxemia, and so these are our very septic patients that are sick in our ICU, so this can have huge, huge implications. So overall, this is a patient safety issue, and I guess starting off with my interpretation of our oath of practice is that all humans are to be appreciated regardless of their skin tone or race, and we need to fix this, so we need to understand how things work, the systems, processes, what's actually happening up close, and that the whole system is interconnected, so there's transparency and openness to the group, and doing so in a non-blaming, and the purpose of that is that we can actually get closer to the truth as we can to try to identify the sources and overcome and help each other do the same, and to continuously improve, and I just want to highlight Donaldson et al, one of the challenges is that in Donaldson, in his book, his writings on patient safety, that oftentimes people in power, in organizations in power oftentimes try to shield themselves from being the source of the blame, so this adds to some of the challenges of solving this issue, but in summary, the sources of error can come from melanin, skin tone, perfusion, wave form, finger size, nail polish is an issue, motion, location, and so, and then how do we, so how do we at the bedside right now, you know, our interpretation and decision making, you know, do you have somebody who's trained and appreciates all these components of the pulse oximeter, and then do we think through the individual physiology and engineering and education, you know, we have to appreciate, we need understanding and transparency of the data and statistical issues, a lot of times it's really hard to actually get that data out that we can actually interpret it ourself on how these are validated and how these are analyzed, so we don't have the understanding and the detailed data to try to really understand where the discrepancies can be, the sources of error, how devices work, and the range of uncertainty, you know, one of the things that sometimes we do is, you know, do a blood gas and try to correlate, and some of these things that we assume work, they may not really work that well, so getting back, we have to add in our practice oath that we have to get, do the right thing, and so for right now, is it, do we change our decision thresholds, there've been some thoughts on like, will we try to increase saturation levels to 94% or higher, do we do more ABGs in patients or something else, but we have to kind of be mindful of this and get it right, you know, as opposed to sometimes there's protocolization and evidence-based controlled care, where like these protocols are sometimes simplified, and they don't allow us to kind of try to elevate and provide that human bedside lens to this, and it removes that human perspective, things get lost in translation, you know, and evidence is whether or not it's the research question, the study design, and in this case, it's the P value and statistical power of how these are, was really the root cause of some of this issue, and so in how these are developed, so there's a pre-marketing and manufacturing, you have innovators, people come up with, you know, developed pulse oximeters, but that there's an approval process, and we're currently undergoing an update to it, but there's an approval process to actually get FDA approval and then there's manufacturing and quality control, and so this approval process is, it's called the 510K clearance, they use these controlled desaturation studies, where they drop saturations down, and they use healthy individuals, and they use, needed 10 or more healthy subjects, ranging in age and gender, but at least two darkly pigmented individuals or greater than 15%, and what that means is non-Caucasian, so basically, it's underpowered, it's not powered at all to look across these skin tones, and so based on a P value, doesn't meet statistical, so it's not powered to look for it, doesn't meet statistical evidence, so it's not really, there's quote unquote no evidence that there's an issue until you actually look a little deeper and design the studies to actually try to get closer, you know, for all folks, so the standards are being updated to improve testing across skin tones, the next FDA hearing is on February 2nd, this should be, it's open to the public, and then the implementation process, so you hear about issues where, you know, we have healthcare administrators and purchase decisions, they use different parts sometimes, or copy parts, or different component sensors, and we have our biomedical engineering experts, and so the system is, we have this standards issue that's been part of the problem, and we have the system administrators and purchasers, and clinical biomedical engineering, and then we have our own clinical scenario of how we get this right for the individuals, what many of us have some control in, but then with patient safety, it requires us to actually get real-time understanding, identify issues, so if we have a ventilator that's now malfunctioning, right, we get rid of it and fix it, we need to be attuned to some of these issues, and recognize, report real issues, and make it easy to do, and then build transparency, I mean, reporting an issue with this is, it's kind of a challenge, and there's no, there's not a lot of, there needs to be more appreciation of this, and just needing to have that transparency, and then real-world studies to try to help us get closer to the truth on these issues, and move it in the right direction, and then professional organization advocates, and hopefully some NIH funding will help us to try to move towards that right answer, and keep this transparency, and rebuild this trust that is so important in this area, so what about the future? The standards will be updated, we're looking at a broader range of skin tones, there's one scale's the Munk scale, he's realized some of these skin tones, or skin color scales are not, they don't give us the range of folks, you know, it's much better than some of the other scales, there's also IDA, but it's not ideal, because it doesn't always measure at the site where it's being taken, and we have to appreciate that, you know, I don't think that this was Dr. Munk's intention, but it's now being used, embedded in like Google's, in Facebook's business models, to kind of digitize more for their online digitization approaches, and their other business models that aren't always as ethical, and then there's new advances in engineering and design, colorimetry, spectrophotometry at the specific site of the sensor, and updates in engineering, and just, you know, so what we can each do, we can, you know, educate each, you know, get educated, and teach each other, and to really individualize our decision making, to help uplift folks who may not have as accurate skin tones, and be an advocate for this perspective to optimize care for all, I mean, when there's a protocol that, when there's a protocol or something else that's kind of limiting it, we need to be able to fix that, or not have that hinder us from actually doing the right thing for our patients, and engage in real-time quality assessments, and, you know, appreciate that we're responsible for these issues, and then just, you know, organizing their critical care societies, collaboratives can have a role, and I wanna highlight OpenOximetry, their group, that they're actually, they're actually really trying to help solve this problem, and kind of take it out of the, take it more, get more of the clinician voice in this issue, and try to move things forward, and we need to engage the FDA about data transparency, even, like, the perfusion index is really not a, there are these kind of gray zone numbers that don't really help us, so data transparency, and advocate for more real-world studies, and I just wanna acknowledge, you know, all the folks work, there are a lot of good people who are trying to get this, move this forward in the right direction, it's not as straightforward, I just wanna acknowledge, too, that, you know, I've been practicing for 23 years, and I've been providing racially disparate care, you know, unknowingly, with this issue, but there's so much, you know, hope, because when we commit to our oath of practice, and really focus on trying to get to true patient safety, we can, the sky's the limit, Christian's shown us that with some of his work, so thank you. Thank you.

racial disparities

pulse oximeter accuracy

darker skin tones

medical practice improvements

equitable patient care