This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.
Today on This Week Health.
Ensure that patients data, when it's put in the hands of third parties is done. So in a responsible and ethical manner. And That patient population that's used in AI machine learning is representative of entirety of the population of the United States. an algorithm that works great in Southern California may not work that well within one of the metropolitan hospitals within New York city because of the diversity of the patient population.
It's Newsday. My name is Bill Russell. I'm a former CIO for a 16 hospital system and creator of This Week Health, a channel dedicated to keeping health IT staff current and engaged. Special thanks to CrowdStrike, Proofpoint, Clearsense, MEDITECH, Cedars-Sinai Accelerator, Talkdesk and DrFirst who are our Newsday show sponsors for investing in our mission to develop the next generation of health 📍 leaders.
All right. I am back from vacation. And today I'm joined by Charles Boise. The chief are technology officer. Is that right? For clear sense? Is that accurate?
Actually, actually I'm the, I'm the chief innovation officer. We now have a chief technology officer, so,
oh, really? So chief innovation officer for clear sense. Well, Charles, it is, it's great to be back from vacation, which means I'm not as prepared as usually am after spending two weeks on the, but have amount has happened since the last. We were together. The the Amazon foray into one medical, I think is interesting.
Interesting move. We have a bunch of stuff happening with regard to interoperability. We have I always wanna talk to you about AI and machine learning. Where would you like to start? You tell me what you wanna talk about and we will go in that direction.
Let's do the Amazon one real quick Amazon for 200 bill
Amazon for 200. Let me see which article should I take? Let's Harvard business review is interesting, Amazon for a into primary care. Won't be easy. Sorry. So Amazon makes the purchase for one medical they're heading in that direct. I posted a bunch of things on LinkedIn on this very topic of Amazon and one medical. And it got a lot of activity. This is what I did on my vacation, by the way I was cruising the Dan. And I would get up six hours ahead of time being in Europe.
And I would, I would throw these things out there and just watch people go back and forth on of so one of the first things I put out there is let, let me ask a few questions. See if we can figure out if Amazon one medical makes sense as a strategy. I put these five questions out there.
What percentage of Americans have a primary care physician, which we know the number it's it's essentially about 75% have a primary care physician. The next one is what percentage of 'em actually use a primary care physician. And that number is even lower than that. Somewhere in the sixties, low sixties.
Are us employers happy with health experience? They offer their employees, are employers happy with the cost they pay for healthcare. And what is the net promoter score for employees with their health experience? It's interesting. And that was really a response to another post that I did that I get the normal responses.
If you're in healthcare today, if you're in the healthcare delivery world today, you look at this and say, Everybody and their brother has tried to get into healthcare. And the story goes back for decades of people going, Hey, new company coming into healthcare, we're gonna solve it. And the most recent that we have with the Oracle announcement buying Cerner was ridiculous.
Not that it's a bad move. It's just how. Came, they came out and their press release and whatnot, and just essentially said, we're here to save the day after IBM had done it after and you can really literally go back about 20 years and have these kinds of stories. And so the healthcare providers typically look at this and go, yeah, it's just another one.
And then you have the other players who are sort of out maybe outside of the healthcare provider world, maybe in digital health startup world, maybe consumers of healthcare. And they look at it and. I like Amazon. I like the way I deal with Amazon. I like how stuff should I like the convenience? I like the transparency.
If they're getting into healthcare, that could be a good thing. So generally the responses sort of fall in both sides on this thing. And as you would imagine, I put these posts out here and they Got a lot of back and forth 44 comments on one 30 comments on another, just a lot of a lot of different back and forth.
What are your thoughts? What is, what does Amazon let's start with? What does Amazon have to do in order to be successful in this move into primary care?
Sure. I think they have to pay attention. It made an acquisition. They've made a purchase. Now they gotta take a look and I would take it first from an observation.
They need to really observe the environment. They really need to understand what does Amazon do, do well. They do supply chain management extremely well. That really is their. Their niche. That's what they've excelled at. And how can you apply supply chain principles to healthcare and even at this very early stage of primary care, there's a lot that they can do.
And then think about it from a future perspective. Amazon would like to do nothing more than be supply chain for healthcare. And this is a really good place to, to start that from an ambulatory perspective, as opposed to diving right into a integrated delivery network with under hospitals and trying to sell them an EMR tied to supply supply chain.
I think initially it'll be a. Easier for them to progress, taking it from this place forward. They have the requisite technologies. There's no issue there and how to, best to take this acquisition and really craft it so that it can deliver what needs to be delivered at this level.
Then expand from that. That's kind of how I see their play. They put a group together, bill, you remember several years back, they were gonna kind of attack it from the hospital side. And that kinda really didn't materialize. So I think it's smart attacking it from this side.
it's interesting. I mean, reasons to be optimistic about it. I believe Aaron Martin going over there, I think adds credibility to it. In that he understands that side. I think they're not trying to bite off more than they can chew in that they are looking at that primary care space and they're starting with their employees and a handful of.
Employers. And so they're very focused in an area they're gonna leverage their core strengths. They obviously have scale. Now with the one medical acquisition one medical was losing things to be concerned about. One medical was losing a fair amount of money. On this I'm not sure people are looking for healthcare to be added to their prime account just yet.
I think there's a cultural change that needs to happen before you're buying your healthcare from Amazon. There's a convenience factor that they have for sure, but there's a cultural change that I think that needs to happen. and competitors, let's not forget that United healthcare Optum is out there and.
This very space that they're looking to go after Optum, which is a, I forget the last I looked at their quarterly numbers. I just don't remember the numbers, but you know, they're a hundred they're in the top five companies in the country in terms of revenue. I think it's a couple hundred billion a year that they're, they're bringing in.
they have the data and people always talk about the fact that Amazon has all this unique data and interesting data. Well, Optum has. Just eight. Just not, just a to, I mean, they're the, their payer, they're the provider. I mean, they just have a ton of data and if they wanted to get that consumer data and whatnot, they could get that as well.
Cause that's available on the open market. So. Again, they're not, they're not going up against a small player. I think people feel like, oh, they're going up against my local hospital here or whatever. They're not really going up against them. They're looking to direct that care that happens.
And if they're able to do that, then they put themselves in a really strong. Physician of influence in terms of directing me and saying, Hey I'm gonna be your partner for care. See our primary care physicians first. And by the way, we will tell you which health system to go to, because I don't have transparency into the costs, but someone like Amazon with their data capabilities might be able to go, you know what?
Build there's four hospitals. You could go to, let me tell you what the outcomes are and the costs are. We don't shop that way in healthcare today. But if we had an intermediary who could put that information,
if we had the tools to do it, we would, we just don't have the tools. I know we have pricing transparency and, but from the patient it's pretty difficult navigate that.
Did you see this one? The, the hospitals being SU. For the data that went over via pixel went over to Facebook. Did you see that one? U C S F and dignity, health being sued over data sharing. Yep.
You know what bill, this goes along the lines of the ethical, responsible use of use of data. And we've talked about that before. There was another news article from Beckers that kind of addressed this as well. And there actually is a new organization out there that Sanford health partnered up with, sharp healthcare.
And that's really what they're attacking is to ensure that data patients data, when it's put in the hands of third parties is done. So in a responsible and ethical manner and whatnot, and that's dandelion health. And the other side of this, which is really interesting, we talked about this too, before bill is how important it is.
That patient population that's used in AI machine learning is representative of entirety of the population of the United States. And we've talked an algorithm that works great in Southern California may not work that well within one of the metropolitan hospitals within New York city because of the diversity of the patient population.
So I. Those kinds of things are extremely important, but you know, getting back to this there's also articles out the last couple of weeks about ransomware which is even more prevalent. we only hear about it every so often, but there's at least three to five cases going on and as we speak 📍 📍 We'll get back to our show in just a moment. I wanted to take this opportunity to invite you to our next webinar challenges and solutions to unmanaged devices in healthcare. This is where we're gonna take a look at the tools that are integral to keeping patients healthy in what we're doing to secure those tools and find them.
In some cases, guests will be leaders from children's hospital of Los Angeles Intermountain. And we're also gonna have representatives from mitigate by clarity on the call as well. And they're gonna share their experiences in maintaining these devices. And just some of the success stories, some of the challenges that they've had as well.
We're gonna do all that on September 8th at 1:00 PM. Eastern time, you could register on our website this week. health.com top right hand corner has our, upcoming webinars. Just go ahead and click on that love to have you register for that. You could also give us your questions ahead of time.
I can give them to the guests and we can make sure that we talk about that. On the webinar. So your topics get addressed before the webinar, we're going to be having a briefing campaign, five short episodes on the channel about this important topic of securing your unmanaged devices in in the hospital setting. You wanna check those out as well. You can also check out those on this week, health.com. So look forward to having that conversation. Love to have you join us now back to our show. 📍 📍
so let me ask you this. So one of the things that kept coming up on the negative of the Amazon. One medical deal was my patient information flowing down to Amazon, right? So if I'm a one medical client now, all of a sudden Amazon has access to my medical data. Now, every time I read that, I'm like, I don't think just because they bought the company, that they can just take the data and start moving it throughout their company.
There are protections around this. In fact, I went back and forth with someone who I, I trust on this. Who said, look, Amazon is HIPAA compliant and whatnot, but they were really talking about AWS and the infrastructure that you can sort your stuff on and whatnot. And that and you have those procedures where only certain technicians can touch that and it has to be monitored and all that stuff.
That's one thing. But the other thing for me was. We had a, as a CIO, I was constantly reminded of the fact that only people that were providing care or the administration of care should have access to the medical record. And they had to have a reason to have access to that medical record. So I don't think all of a sudden one medical get bought by Amazon and the marketing department over at Amazon can all of a sudden go to town on this medical data. That shouldn't be how it works. Right.
Yes. So basically. You're you know, bringing in the tenants of least privileged access, right? Those that only those that are required to have access. It's not like Amazon's gonna say, Hey, everybody in Amazon, Hey, all the data scientists and the data engineers.
Wow. we've got healthcare information now let's go crazy. That's not happening. That's not going to happen. So they will put policies in place that again, that use the least privileged principles. So that only those. Require access, get that type of,
but let me ask you this. Here's another, a different way to approach this. If I were at Amazon, this is the way I would be thinking about it. All right. One medical. I don't want the medical data to be flowing down to me. However, could I flow a bunch of the demographic in consumer data to one medical to make them more effective as a care delivery organiz.
Sure. So if I wanted to take all of the Amazon data that contains in its social determinants, right. As well as combine that with other publicly available and privately available social determinants, put all that together, wrap it up and take it down to the the actual patient level that can be done.
Absolutely can be done. But again, Responsible and ethical use of, of doing that and Amazon having the right structure in place that they do have ethics committees and so forth. So as they go through that, they they do it responsibly we've talked about people consider it kind of creepy that I can look up you and know everything there is to know about you pretty much.
from just, you. Privately available data and you are the same with me. And do we keep that type of information at a categorical level? Let's say I'm a diabetic I'm and one of a whole bunch of diabetics is it okay to have all that information at my level to help treat me better?
I still think we're struggling with that. Deserve definitive answer. I don't think there. Some organizations are taking it down to that level. Some are not. So from an Amazon perspective, having what they have as well as what also is obtain and the clinical, could you bring it to end of one and provide better treatment, better care? Absolutely.
What's interesting about this privacy thing is we don't get privacy, not you. And I mean, you and I have studied it and looked at it and had to implement it and whatnot, but I mean, Americans don't get privacy. I remember back in the day when somebody showed me. An article and within Facebook, I could go in and look at my profile.
And then there was like this other, like backdoor, it wasn't really backdoor, but it's like, Hey, do you wanna know what we know about you? Kind of thing. And I remember going into, in there and it essentially said, here's what we think your political affiliation is. Here's what we, it had a whole bunch of stuff.
And I'm like, how do they get there? I mean I've put like nothing out on Facebook, but the reality is every click tells them something Who are my friends? What do my friends post? Where do they live? Where do they go on vacation? It's like by association. You're like, if this is your friend group, then even if I didn't put anything out there on Facebook, they go, if this is your friend group, then likely you fall into this category.
You are this, you likely fall into this. Now some of that stuff is, is is voodoo. And some of that is, is actual, but it was kind of crazy when I was looking at it. I'm. I've put very little out on Facebook and that's pretty well spot on as to where I am in, in, in a lot of things. It, I don't think we recognize how much information we give away every day.
I mean, I'm sitting here on this screen. I have 10 windows open, I think at some level information is getting scraped from my email. Which is technically a Gmail account. I mean, it's a Gmail business account, but my guess is if I really read that closely, they're looking at that information.
I mean, how much information do we give away and how much is being collected on a daily basis on us?
There's third parties that are Social determinant folks that are selling this stuff. Right. And I will tell you that they're is close to about 250 different relevant features.
I can tell how many people are in the house, what the income level is of everyone in the house, political affiliations. I can tell you what cell phones they have the apps on the cell phones, how much they're using. Where they go from a from a marketing perspective in the web, what you bought over the last week what you've used your cards for and whatnot.
So at some point in time, is it gonna be okay for the provider to ask Hey what's your alcohol consumption? We say, oh, we really don't consume alcohol. Well, can you explain the the two cases of. Name your beer that you buy every week at Costco? Is that for you or other family members?
So all that detail is available. Again, it goes back to the responsible and use of that.
Yeah, I think, and I'm gonna show a little bit of where I'm coming from as well. People get up in arms when they see that. Google's getting associated with healthcare and they see that Amazon's getting associated with healthcare.
But when I was CIO at St Joe's, we were trying to pull all this data together. I mean, we were looking at it going all right, we have a thesis. And the thesis is that we, if we get to an end of one, We could actually provide them a better level of care. We should, we could know if it makes sense to discharge this patient.
Cuz one of the biggest things we knew in terms of readmissions, if someone from the family came and showed up and took them home, there was a greater, like a significantly greater likelihood that they weren't gonna readmit. right. So if we knew that ahead of time, we could make the necessary arrangements to make sure that that person was gonna be cared for when they left the four walls of the health system.
There's a whole bunch of use cases like that, that we were looking at going, Hey, if we knew. More information. We would've never discharged that patient. Or if we knew more information, we would've done X, Y, or Z, but we didn't have that information. And so we were looking at marrying that social determinants data with the medical data and figuring out ways to put that in front of the patient now or in front of the physician.
Now we were a long way from doing that because the physicians are like, look, I've got. Just gimme the relevant stuff and there's a lot of things around it, but I think people would be surprised the number of health systems that are actually working on exactly what I just.
Yeah, bill in discovery, we made from a research working with a pediatric population primarily pediatric patients diagnosed with insulin dependent or type two or type one diabetes.
We looked at outcomes one year post diagnosis and two years post-diagnosis and the social determinant that we discovered that had an. On whether those patients did well or not, was that whether they live, whether or not they live with grandparents and grandparents taking care of them, you and I would think that, oh my God, that's the best possible scenario, but it absolutely was not transformation needs.
Weren't being met the ability to actually pick up medications and even the day to day monitoring and whatnot kinda fell apart. And that's a discovery that was made. With use of social determinants that really wouldn't have been thought of. So paying attention to where the kids are actually where they're actually living and who's actually taking care of them in this case.
It allowed that organization to look at and provide more assistance to those kids that were being taken care of by grand.
All right. let's get the story. A friend of the show somebody's been on the show a couple times. Donna Roach CIO for university of Utah health system out of salt lake was I think interviewed. Yeah, cuz it's Q and a kind of back and forth. Sometimes these interviews are just email interviews. I don't know which this is, but essentially it's it's on interoperability and some really good stuff here.
So I'd like to touch on it with you. She gets questions, like what will it take to get EHR interoperability? Right? She goes directly at unique patient identifier. For making that. Right. And we know there's been a lot of movement lately in the unique patient identifier. What are your thoughts on the unique patient identifier?
I'm you thumbs up for it. I have no issue with it. We all have an identifier, we just can't use it. That being said, there are a lot of AI machine learning algorithms that we throw in and identify folks there. It is an issue, but it's not the end all be all. If you.
Yeah, I, I agree with you. I don't think it stand all, be all. And we were solving this a long time ago and the other thing I would say to people is there is a unique patient identifier within the four walls of the health system. And there's still duplicates of some percentage. Now, I think we've gotten a lot better at that because the systems have gotten better at that. But let's assume we have a unique patient identifier.
Bill. If we were all on genomically sequenced, we have a unique identifier to pick the protein. So that will, that would take care of that. Right?
Well, well, and, what you were referring to as social security number, we can't use the social security number as the identifier, and there's a lot of reasons for that. I mean, most of 'em sure, but the primary reason we couldn't use it at, St. Joe's in orange county, California was about 15% of. Population at certain hospitals didn't have a social security number or gave us a different one every time they came in. So yeah, those challenges, that's how we ended up with duplicates a lot.
She goes on, she gets asked the question, why can't individuals just merge their various patient portals into one? And she says people will say they can't. Why can't that record from Cerner? Just populate the epic system and And she goes on to talk about the different standards and how those are built and those kinds of things.
And the complexity of doing that is this still that complex to bring those patient records together from various portals.
I'll just get right into it. Member meaningful use. Yes. Whole bunch of whole bunch of things were mandated. Right. Right. And it was monetary compensation to meeting.
We mandate interoperability and compensate for reaching that, then I think we'll go a long way to, to really solving this problem. Technically could there be one national HIE? Absolutely. Could there be one national portal if you will? Absolutely. But And this is my opinion only this is gonna have to be com mandated with real teeth and then some type of compensation for reaching that you've seen the spending bills over the last couple of weeks or so.
Yeah, I have seen no reason why this can't be mandated, but you know, kind of going back to your question yeah, it's a difficult lift in an uncooperative environ. In a cooperative environment. I think we can reach it. Absolutely.
Yeah. That's we've talked about this before. The challenge with interoperability is a business problem, not a technical problem. There are some technical aspects of it that, that do create a challenge, but it is a business problem. It is moving this data around doesn't really make sense. I was standing with two CIOs. In a market.
And I think between those two CIOs, they represented 70% of the healthcare in that market. And I said how's the interoperability going? And they said, oh we're really nowhere kids were doing this. We're doing this, the HIE this, the HIE that, and I just stopped them. And I said, you realize the two of you could solve this.
Right. I mean, 70% of the health records generated in your market come from your two healthcare organizations. And you, you could create that record that has 70%. Well, once you get to a certain scale, Then other people are gonna start participating because people are gonna start demanding it and saying, look, I've all my, I went to that system and that system, all my records are hearing this one thing.
We have a common portal. Now, if you're on, one's on Cerner and one's on epic, they're gonna say how do you display those things in those various portals, but there's, there's ways to do that. it was interesting. They sort of looked at me like I was insane and I was a little bit, cause I was a little bit in that I know they have a lot of priorities.
It's not like they're being told, Hey, go solve this. But that's part of the business problem that I'm telling you, nobody's telling them, Hey, go solve this. You want to do it by Mand it from the government level. And to a certain extent, we have that coming down with 21st century, cures and health systems are pushing back because from a business standpoint we don't go to Amazon and say, Hey look, all that data you're collecting on all those individuals you have to make that available to Walmart and Walgreens and everyone else.
Yeah. Competitive. Yeah, no, I totally agree. But Hey it's gonna take something like that. Otherwise we're gonna be floundering around for God knows when you know what I.
Yeah. She gets asked the question. I want to pose this one to you. Is the relative lack of competition in the EHR market. Good or bad for interoperability?
well, it should make it technically I'll speak from a technical perspective. It should make it a little bit easier on the ambulatory side though. However, practice side, there's about 500, so right. That's and there's more coming out every, every year. It seems like, but yeah. Cooperate.
Are there any EHRs that you're looking at that you're saying? Yeah, that's, that's a different model. That's a different approach or is it just another, some of
the, yeah, some of the ones that from a practice perspective outside the hospital care cloud, there's, a bunch that are all built on modern technologies that have taken a different approach. And that's kinda one of the limiting factors of our, the big three, the big five, if you will, from. Hospital side, they're still on some older technologies, but technically, yeah, we could we could do this, we could do this, but again there's gotta be some type of pretty significant mandate.
Absolutely Charles, thank you for getting through my first interview back from vacation. always great talking to you cause I feel like I could throw anything at you. and you'll, you'll just, you'll just roll with it. Anything exciting happening in the industry that we should keep an eye on?
Well, I think you've kind of, hit on some of these, I think it's gonna be interesting to watch. The continuation of machine learning and AI, as it's applied to healthcare there's been some significant combinations of groups and whatnot. And I think that we're all getting to the point where we're the understanding is that one algorithm isn't gonna solve all our problems.
And again, they're extremely geospecific and there's a lot. Work that needs to be done from a maintenance perspective, as well as from an educational perspective. And I think that I think we're getting better and better at that. And then understanding, I think we've gotten over the hurdle that machine learning isn't going to diagnose all x-rays.
Isn't gonna diagnose MRIs, CT, all of that. I think we've gotten over The fact that AI is not gonna do all the diagnosis, all the, I think we've got it. Then we gotta the point where AI machine learning is gonna be adjunctive to our practices is actively gonna be assistive. So instead of artificial intelligence, it's intelligent assist. And I think we've come to the realization, we'll see us progress better as a result of that.
Yeah. And I think that I read that. Dandelion announcement with Sanford. And was it script? No, it wasn't scripts. It was sharp. Right? Sharp. Sharp. Yeah. And that's, part of the foundation of what they're putting together, right? The ability to,
yeah, it's significant. Yeah. It's significant because they will have healthcare organizations that are brought across the us. That a representative of the us population. And then of course with the responsible and ethical program that they put in place, those AI vendors will be very will, it'll be very rigid in what they're allowed to do and how they're allowed to do it, but they will get a excellent representation of the population so that if I build, if they build something out for diabetes, It will be good across the nation and it will be specific as we talked about earlier to the you different categories ones. And so.
Absolutely charles, thank you again for your time. Always a pleasure to, to
What a great discussion. If you know someone that might benefit from our channel, from these kinds of discussions, please forward them a note, perhaps your team, your staff. I know if I were a CIO today, I would have every one of my team members listening to show just like this one. It's conference level value every week. They can subscribe on our website thisweekhealth.com. They can also subscribe wherever they listen to podcasts. Apple, Google, Overcast. You get the picture. We are everywhere. Go ahead. Subscribe today. We want to thank our news day sponsors who are investing in our mission to develop the next generation of health leaders. Those are CrowdStrike, Proofpoint, Clearsense, MEDITECH, Cedars-Sinai Accelerator, Talkdesk and DrFirst. Thanks for listening. That's all for now.