What’s Data Quality Got To Do With It?

Carol Macumber, MS, PMP, FAMIA (00:03):
Welcome. Thank you for joining us this morning on our inaugural very first presentation of HIMSS 2024 in the Clinical Architecture Theater. My name is Carol Macumber. I’m the EVP of Client Services at Clinical Architecture, and I have the pleasure of being joined today by some friends here to talk about data quality and what data quality has to do with it. And we’ll find out what it means as we get going here. So from my right here, I’m going to start with Angie. Angie Glotstein is a thought leader with more than two decades of diverse experience in healthcare. In her current role as Vice President at GEHA, she oversees several programs including care management, wellness coaching, clinical quality and medical plan accreditation. Her initiatives are focused on outcome improvements, and related reporting. She achieves results by enhancing solutions in their continual effort to evolve member experience.

(00:55):
Previously, Angie was an Executive Director at Cerner Corporation, where she served as a leader in the emerging space of cloud-based health IT founding a fast growing business unit focused on the development of data intelligence supporting Cerner’s platform. Next to her is Ms. Michelle Dardis, who is currently a director in the department of Quality Measurement at The Joint Commission. In her role, she oversees activities related to the development, testing, implementation and maintenance of quality measures, performance measurement systems, and quality measure analytics. Her federal and other advisory experience includes serving as a track lead in CMS and ONC, Kaizen events pertaining to interoperability and clinical quality and membership on measured development technical advisory panels. Prior to her role at The Joint Commission, she served as a health IT standards expert to multiple clients, including the Centers for Medicare and Medicaid Services, and as a principal and investigator on projects to develop and maintain chart abstracted claims-based and ECQM measures used in national reporting.

(02:02):
Last but certainly not least, Dr. Michael Barr is the President and Founder of MEDIS, a veteran owned independent healthcare consulting firm. Before founding MEDIS, Dr. Barr served as executive vice president at the National Committee for Quality Assurance. In his role, he oversaw the development of HEDIS performance measures, research and analysis projects, and the contract and grant portfolio. Dr. Barr contributed to NCQA’s strategic initiatives including the digital measure strategy. Prior to NCQA, he held prominent positions at the American College of Physicians, including senior vice president of the Division of Medical Practice and Vice President of Practice Advocacy and Improvement. Needless to say, experts in their own right and many others. So welcome should have clicked the slide. The focus of the presentation here is to talk about the interdependencies of interoperability and data quality, for these guys to kind of share the limitations and opportunities of interoperability and data quality, and steps that your organizations can take to improve interoperability and data quality within your organizations and with others.

(03:15):
We’re going to do this through a series of discussion topics. There’s four of them. We’re going to do the very best we can to open it up to questions as soon as we’re done. And the first being, I’d like to start with what is their definition of clinical data quality, right? Everybody’s data quality. You say that it’s kind of like implementation. One person’s implementation is another person’s configuration and installation. So I thought it best that we start with how do these guys define clinical quality and how does it impact their organizations today, both positive and negative. Let’s get to the nitty gritty. So Angie, if you don’t mind, we’ll start with you.

Angie Glotstein, RN, BSN (03:49):
Absolutely. Thank you. So where I’m coming from, the lens that I’m taking is an FEHB payer. So data quality to us is very different when we think about data quality from someone else. So my definition of data quality is using the true intention of the data from the source where it originated, and then being able to apply it against an appropriate use case, like quality reporting in a way that can be backed by an audit. So understanding that true intention, applying it, backing it by an audit even when it’s outside of its original source.

Carol Macumber, MS, PMP, FAMIA (04:28):
Michelle.

Michelle Dardis, MSN, MBA, RN (04:29):
We’re working in order here. So from my context at The Joint Commission, we receive clinical data from over 2,600 hospitals around the country at the patient level. And so that audit is so important, but outside this purview of what we do. So our focus on data quality really focuses on conformance with HL7 standards and the values that’s used in our measures.

Dr. Michael S. Barr (04:51):
So we’re going to continue to widen the spectrum and agree with everything you said, and my bias is how does it improve quality of care for individuals and populations? Because as a use case to Angie’s point, thinking about that we need the utmost best of data to improve care at the point of care and different use cases require different fidelity and completeness of data. So the quality can be different for each different use case. So defining, as you said, Angie, with the end in mind what we need and then working towards that.

Carol Macumber, MS, PMP, FAMIA (05:22):
Right? I mean, arguably we can’t achieve the goal of semantic interoperability without good data. And poor data quality things like duplicates and having to sift through them or completeness have a direct impact to that. The ability for us to achieve that, I’ve heard it been called it’s an ocean of data with a desert, desert of insight. So I mean is the answer to that and how it impacts your organization’s more data for simply just better data?

Dr. Michael S. Barr (05:52):
I’ll jump in real quick. Moving data is not the same as using it. And more data doesn’t mean better data. So I think that’s, if we start with that as sort of the two parables, axioms or whatever you want to call what we need to improve and recognize that just simply…

Carol Macumber, MS, PMP, FAMIA (06:11):
I’m not sure.

Dr. Michael S. Barr (06:13):
Anyway, moving data is not the same as getting it to the point of care where you need to use it again, back to the different use cases. So yeah, too much data. It doesn’t mean more data doesn’t mean you’re going to get the data you need.

(06:25):
Right. Any other thoughts?

Michelle Dardis, MSN, MBA, RN (06:28):
I think we do need some more data though, but very specific data that’s been curated and tested and grown and nurtured. So we’re looking at using social determinants of health or health related social needs, expanding the definition of gender identity and sexual identity. These are areas where the standards community is working and implementers are working and we need to kind of build that continuum so that those data are useful.

Carol Macumber, MS, PMP, FAMIA (06:55):
I would go on to number two, what is that relationship between data quality and interoperability? So we kind of started going down…

Dr. Michael S. Barr (07:01):
That, I apologize, I jumped the gun, but…

Carol Macumber, MS, PMP, FAMIA (07:04):
I prompted it though.

Dr. Michael S. Barr (07:05):
I do want to build on what Michelle said, just the last comment. I think we also need to be sensitive of that as more data become available, how it’s reflected to clinicians and clinical teams. Because yeah, we may need it for certain purposes, but if it doesn’t get to them in a usable way, that other end of the usability spectrum, not just the data itself but usable in clinical practice to improve care, then we’re failing. So yes, we need all the data you described, but it needs to be capturable in electronic health records to be reflected the data that we use and also usable by clinicians at the point of care and shareable.

Carol Macumber, MS, PMP, FAMIA (07:41):
I did a little research last night before showing up today, and I came across NCH’S annual EHR survey, and in that survey 40% of the providers that answered it said they rarely, if ever, get information about the patient from an external organization. Not so surprising, or maybe surprisingly, they said 35% of the time they wouldn’t use it. Okay. So I mean what does that say about the trust also in the exchange of this? So there’s interoperability and there’s the quality that you’re talking about, and from that survey, which was January, 2023, I mean they’re reporting back a year or two. I mean it just feels like the providers and the folks that are providing data for you guys don’t really trust the information even if they get it.

Angie Glotstein, RN, BSN (08:34):
So many thoughts there.

Carol Macumber, MS, PMP, FAMIA (08:35):
Let’s hear ’em.

Angie Glotstein, RN, BSN (08:37):
So I think that it’s a slower uptake on the provider side than where we’re sitting, what we expect for them to even document in such a way that it’s interoperable or usable across venues. So for their own reporting as well as sending it out to anyone trusting us enough to send it and sending it in a way that it’s usable as it was intended. I think that’s a difficult hill to climb. And I think you brought up social determinants of health and race and ethnicity and gender. Those are things that we absolutely need to understand in greater depth, but capturing it in a way that is interoperable outside of the EHR itself where it was documented is not occurring today as we expect at a regulatory level.

Michelle Dardis, MSN, MBA, RN (09:30):
If I can comment on that too, I think we have a people and process challenge on the front end of interoperability right now, maybe more so than we have even in past years when it comes to entering structured data due to workforce challenges. I was talking to a health system earlier this week who are trying to have the surgical techs enter diagnoses on the behalf of surgeons because there’s just no time currently and the surgical schedule’s too tight due to surgical staffing shortages. So data quality on the front end I think has some unique challenges right now that we maybe didn’t see in past years to the same detail.

Dr. Michael S. Barr (10:07):
Michael, it’s two quick thoughts. Trust in data is incredibly important. Timeliness of data or the lack thereof undermines trust. So by the time the information gets to the clinicians that they can act upon it, it may be too late. So it may not be that they reject the data, but they’re not sure it’s timely enough for them to act upon for the individual. That’s different than in trends or retrospective analyses where the data don’t need to be as timely. The other point I just want to make is I think we’re at danger of over structuring the clinical record. And in this age of augmented intelligence, artificial intelligence that can build cases. I wonder whether we need to dial back on how much we ask clinicians to structure in their clinical record. I mean, there’s a historical reflection I did in an editorial years ago when records were totally unstructured doctors wrote these wonderful prose. Captured the story about everything and it was great. And then there was an effort to structure data so they could do some population health and statistics. And that worked for a while, but then they overstructured, then doctors started writing around the margins of the paper and it was really hard for them to do anything with it. I think we’re in danger of over structuring the record and therefore creating more work than is necessary, especially with the new tools and AI and generative AI that can help us pull out relevant information from unstructured data and make it structured so that we can use it for the purposes that we’re talking about.

Angie Glotstein, RN, BSN (11:30):
So I absolutely agree with you, but we’re in that gray space right now where it’s not happening. So the unstructured data, actually, I was talking to one of our EHR aggregators, Data Link, and I was talking to them about this last night and this morning they are taking out 50% or more of the data because it’s unusable because it’s in that unstructured format. And so we’re losing all of that goodness. But I believe that over structuring it or trusting the AI to translate it to structured and then backing it with an audit, we’re not there yet.

Dr. Michael S. Barr (12:04):
No, of course not. But what we need to start talking about is so we can get there because what we’re doing now ain’t working, right.

Carol Macumber, MS, PMP, FAMIA (12:14):
I mean, listening to you talk, I mean it’s kind of trying to find that sweet spot between standards and policy and actual point of care and delivering clinical quality upwards. Don’t have an answer, but…

Dr. Michael S. Barr (12:28):
Well, I mean until the electronic, I mean, is there a pathway to continue using structured if we redesigned electronic health record, so it’s actually usable on the point of care back to a similar survey over 40, 50% almost of clinicians find their EHRs not usable enough.

Carol Macumber, MS, PMP, FAMIA (12:47):
Discussion topic three. So we keep jumping ahead, but that’s how natural this conversation is. Describe the limitations and opportunities associated with data quality and its relation to your organizations or to those organizations you serve or perhaps even just exchange data with.

Angie Glotstein, RN, BSN (13:06):
So our number one challenge is access to data. It doesn’t matter even that it’s interoperable or not, we just do not have access to data. I think interoperability will help eventually, but the access is the difficult part for us.

Carol Macumber, MS, PMP, FAMIA (13:23):
Right. Now, that’s a policy thing. Yes. I mean meaning they can still say no.

Angie Glotstein, RN, BSN (13:30):
They can still say no.

Carol Macumber, MS, PMP, FAMIA (13:31):
Yes, they can still say no. Even though you are required to have that content, and as the quality measurement worlds evolves and they go completely digital, well then now what?

Angie Glotstein, RN, BSN (13:44):
Well, and then you would think because payer perspective will then require the data, and you can do that to a point. So if you think about a pay for performance with an ACO and you’re incentivizing for gap closure, for example, that would then contribute towards HEDIS. There’s ways to prove that individually in each EHR, but then getting that data out of each individual EHR at a provider level on a nationwide perspective is impossible at this point.

Carol Macumber, MS, PMP, FAMIA (14:18):
Impossible and improbable. Yes, Michelle.

Michelle Dardis, MSN, MBA, RN (14:22):
But it can happen for some. So at The Joint Commission, like I said, we have about 2,600 hospitals that submit EHR extracted data to us for the purpose of quality measurement. I think the biggest challenge there is the administrative and documentation burden facing the organizations who collect the data. It is about two thirds of the hospitals who are now required by CMS and others, and that requirement is growing through HRSA is starting to require electronic data measures and some states like the state of Maryland are starting to require hospitals to report EHR extracted data. So the administrative and clinical burden piece of data quality is becoming a central issue for those folks. Michael?

Dr. Michael S. Barr (15:09):
Sure. I come in for a slightly different angle As a consultant, I have health plans as clients, as well as technology companies. So just speak from those two perspectives. Health plans are struggling to figure out what they have in terms of data and where to get the data they need and what they can expect as a lift from that data to improve their outcomes or their HEDIS, Medicare Advantage and so on. And then the technology companies have some incredible solutions. A lot of ’em are here going to health plans and trying to convince them they ain’t doing what they think they’re doing because there’s a way to do it better, simpler, easier. So it’s that nexus between what I need to know, what I don’t know and where to get it, and what can I use to help me improve the data quality, data access for the use cases that are most important. Translated to delivery systems, accountable care organizations, so on, they are looking for data to support their population health and where to get it. They look at their own clinical records, but they recognize because of the fragmentation of the healthcare system, they don’t have a complete picture of the individuals they’re serving or the populations they’re serving. So they’re looking for the data improvement they can get from other sources and the connectivity, leveraging interoperability to support their use cases. So working backwards for the use cases. There are a lot of different challenges along that chain.

Michelle Dardis, MSN, MBA, RN (16:26):
Is there an opportunity to harmonize on the use cases? What are the top use cases for interoperable data? I do think at The Joint Commission we have looked at data quality across organizations to understand can organizations report the same measure the same way they would from chart abstraction through an electronic clinical quality measure? And the answer is yes. We are seeing quite a bit of maturity with electronic clinical quality measures where we do really feel they reflect care. And so that’s where we feel a lot of focus has been on the hospital side. But are the use cases we’re pursuing the same that the payers and other stakeholders are interested in?

Dr. Michael S. Barr (17:04):
Great question, Michelle, and great point. I think on the ambulatory side, and with all due respect to my colleagues who are measured developers out there, my NCQA former colleagues, the measures that we’re attuning to report are not the ones we need for quality improvement in patient care. Because those are accountability retrospective reflective measures that tell you what happened, not what should happen, not what happened during the entire course of the year. We need different measures that linked to those accountability measures that drive improvements. And for that, you need the high degree of trust, high degree of completeness, high degree of fidelity, and high degree of, I’ll say trust because it’s worth mentioning twice and I lost count of the ones I was going to say. But I think that’s the critical aspect. So I think there’s an opportunity to align, but I don’t think that exists currently. I think we’re frustrating the delivery systems with measures that don’t really help them deliver care.

(17:59):
Absolutely agree. That’s a solid point.

Carol Macumber, MS, PMP, FAMIA (18:01):
Great. I mean, what other advice do you guys have or realities of trying to shift from the retrospective, how well did we treat a disease? To prescribing more preventive measures, looking at wellness, trying to reduce the costly inpatient admissions and such?

Dr. Michael S. Barr (18:22):
So let’s understand that the accountability measures are going to be slow to change because they have to go through several bars of evaluation and testing and to change those on the fly ain’t going to happen. So let’s accept them from what they are for the near term. We need better measures that reflect care at the point of care. An example of diabetes, the current measures in diabetes look at was the hemoglobin A1C below a certain number? Do they have eye exam? Do they have the blood pressure control and so on and so forth. Again, on the blood pressure control one, it’s the last measure of the year, not was the average blood pressure over the course of the year managed appropriately, was improvement in measure increasing intensification of or deintensification medication based upon a response? Or was the hemoglobin A1C adjusted during the year adequate or? All those measures are approximate to the accountability measures and more usable.

(19:18):
We should be developing those in digital format, distributing them, let the practices improve. You might even say what digital measures or measures are being used in the field crowdsource it from the way up. There are a lot of innovative delivery systems that are looking for some of these improvement measures. Can we crowdsource those and sort of making them more available in digital format for distribution and link them to the accountability measures. If you’re improving blood pressure average over the year, likely the last blood pressure of the year is going to be fine. If your hemoglobin A1C and glucose glycemic index is adequate during the year, you’re going to likely do okay on the Medicare Advantage measure and so on and so forth.

Michelle Dardis, MSN, MBA, RN (19:56):
I agree with that. I think a portfolio of measures where we have the accountability measures, but these derivative measures that are electronic that help with actual care management would be a great step forward. We can use the same value sets and define the condition in the same way, but use those in real time for improvement.

Dr. Michael S. Barr (20:14):
Absolutely. Thank you.

Carol Macumber, MS, PMP, FAMIA (20:15):
I mean and realistically the healthcare IT environment is monolithic. I mean it moves very slowly. So realistically, how long would you think it would take to implement change like that?

Dr. Michael S. Barr (20:28):
I’ve known people can create a digital measures in a week, right? It is not going to meet the same level as a HEDIS measure by intent. And you could write a measure in a week if you wanted to.

Carol Macumber, MS, PMP, FAMIA (20:40):
You probably wrote one standing here.

Dr. Michael S. Barr (20:43):
And I mean a really good one, leveraging existing value sets. If you’re dealing with the same, I think that’s a great idea. Why create a new value set? The measure should be reflective and connected to the accountability measure. So the question is how quickly can you get them out and distribute it in a way that’s usable, that work in all the health systems and so on, and leverage the data back to the better data. You still need better data for better measures. A lot of these measures cannot be populated. A lot of the ones I would envision, we get challenged to populate with the current data infrastructure, but we’re getting there. So asking, I don’t know, I don’t have a crystal ball, three to five years.

Carol Macumber, MS, PMP, FAMIA (21:17):
Get it.

Angie Glotstein, RN, BSN (21:18):
I want to mid step to that, that’s the right place to go, but I want to go in the middle to where you were. If we could just agree on the definition of things with a value set or even numerator denominator criteria, if we could all agree on that, it would make it a little bit easier for us to understand the data that we’re getting, right? Because we’re all, when you’re working on an ECQM at an EHR level, that’s a silo of translating data into a use case for that ECQM. But when you move that data out of the EHR, it might be that we’re looking for the same thing, but the meaning isn’t there because it was normalized to an ECQM and not NCQA value set. So I will actually, it’ll appear that the measure was missed, if you will, for lack of a better way of saying that, even though it wasn’t.

(22:12):
So if we could agree on normalizing that data in one way, I think that would be a great first step. And then maybe a baby step towards where we need to be for making it usable at the clinician’s hands would be just changing the HEDIS measure. I’m going to speak HEDIS specifically, but any, where I’m using every day is day one, so that I’m running value of the year and not just a retrospective view. So that I can understand performance of where I am right now and where I’m heading a little more proactively and then get to perfect state.

Carol Macumber, MS, PMP, FAMIA (22:49):
Be still my heart. You start talking terminology and I’m like, okay, let’s do this. So I mean you’re looking for standardized value sets, reusable, harmonized. Not go to VSAC, I find seven different value sets for diabetes, right? I But I think that at the core of that is still this perception that the terminology used in those value sets are not really indicative or they don’t capture the clinical meaning that the physicians are trying to express. I mean, so how do we get over that perception, right? If you want common value sets, then there has to be a way for an intermediate between the clinician and that terminology to ensure that the intent, that clinical intent that you were talking about is maintained, right? We’re up. We’re currently at the point where what we do is we take the words, the natural language, the way that they describe things and we try to find a semantic equivalent. Versus it just natively being what they’re trying to say. So what do you think about that?

Dr. Michael S. Barr (23:51):
I’m not a value set expert by any stretch of the magic, but I like the idea of using language clinicians are used to.

Carol Macumber, MS, PMP, FAMIA (23:57):
I do too. If only. Are my SNOMED friends in the room? No? Okay. Alright.

Michelle Dardis, MSN, MBA, RN (24:04):
Can I comment on?

Carol Macumber, MS, PMP, FAMIA (24:06):
Please, please.

Michelle Dardis, MSN, MBA, RN (24:06):
A problem with the value sets we have today going in the value set direction is that there are seven for the same concept. Because, and the developers do work to harmonize. There was a value set harmonization project, maybe eight years ago. Yeah. It was a journey, but I think we are looking at the patient at slightly different points in time. And so it’s the same condition, but sometimes different diagnoses or different procedures and the title is the same. And we lack the metadata or the transparency to help consumers understand what the intent of which value set was. And at the end of the day, in your electronic health record, maybe you need to map to one or two of the terms that exist across all of the value sets and you don’t need all seven, but there’s a lack of transparency around that.

Carol Macumber, MS, PMP, FAMIA (24:55):
Right? There’s an HL7 standard on about the characteristics of a value set definition. It is near and dear, but it’s in HL7 speak, right? But the whole purpose of that was to standardize those characteristics so people could discern themselves what the difference is between all of those. But the reality is that takes a lot of effort and it takes a lot of familiarity with value sets and HL7 standards even to really glean from it as a layman the differences between those things that they’re finding. So what happens? They just get frustrated and create their own. So now there’s an eight and then a nine, right?

Dr. Michael S. Barr (25:30):
Is there a role for generative AI in doing some of the mapping back to the clinical terms?

Carol Macumber, MS, PMP, FAMIA (25:35):
Perhaps.

Dr. Michael S. Barr (25:37):
I mean at the point of care.

Carol Macumber, MS, PMP, FAMIA (25:38):
Right. Okay, I’m going to be of time and get to four. Identify steps your organization can and should take to improve data quality, access to structured real-time, clinical data and performance and value-based payment models. Anybody want to start?

Angie Glotstein, RN, BSN (26:00):
We just need the data.

Carol Macumber, MS, PMP, FAMIA (26:01):
We need it. Please give it to us.

Angie Glotstein, RN, BSN (26:03):
We need the data, but then we have the responsibility ourselves. Once we receive that data, we need to, I mean, so we’re moving to, in our EDW, we’re moving to FHIR, so we need to, what is, keep the data, I can’t think of the right word. But we need to house the data in a way that remains true to its intention from the source in which we received it for an audit. But in a way that is also interoperable out of our own systems as needed. So when we bring it in, even if it’s not in FHIR, and oftentimes it is not, we’re trying to translate it into that model so that we’re interoperable on the other end.

Carol Macumber, MS, PMP, FAMIA (26:40):
Thanks. Michelle,

Michelle Dardis, MSN, MBA, RN (26:42):
Yeah, I think maybe I’ll skip the data quality piece a little bit here to talk too about security. The more real time data we get access to and in the more ways we want to use it, we need the security and performance layer in place to be able to use it effectively along with the semantic layer. So just figuring out the cloud computing environment is a challenge that we’re currently looking up.

Dr. Michael S. Barr (27:02):
And I’m going to go back to something we said earlier. Start with the use cases. What use cases do you need data for and how complete and high quality do you need for each of those use cases? Whether your health plan, delivery system, an ACO, and look backwards. Okay, what data am I currently getting? What are the gaps between what I need and what I have? Where is the data coming from? Is it being transmitted into FHIR, API? What do I need to, what use cases does it support? What does it not support? How can I fill in those gaps and from where? And then look for opportunities to improve the data quality in that way because they’re not going to be good to start with. You need to improve it to some degree, mix and match and so on. But look for clinical data in particular to match the claims data.

Carol Macumber, MS, PMP, FAMIA (27:47):
Great. So we reached the part where I get to be Vanna White. I do have a cane. It’ll take me around to get a while, but we would love to get some questions from the audience. All right. Nice and close. Yes.

Audience Speaker 1 (27:59):
Thank you, Carol. This has been fantastic. It’s really held together extremely well. And Angie, your comments opening and closing on getting access to the data prompted me to have the question. Do Providers have a natural advantage in that space? And I’ve got another question, but go ahead with that.

Angie Glotstein, RN, BSN (28:21):
Yes.

Audience Speaker 1 (28:21):
And so I think, Michael, your comments, especially about time and therapeutic range, you didn’t use that term, but absolutely right on. And Michelle, your comments about harmonization and value sets and the history of how that’s gone and your validation of that is near and dear to my heart. I think based on the fact that Providers seem to have an advantage and there’s either 30 of them or 200 of them depending on how you count it and VA and Kaiser Permanente and others are demonstrating it, what can we do to create that so we can get over the, they can still say no problem. I love that term.

Angie Glotstein, RN, BSN (29:06):
I don’t have a good answer for you. It’s something that we’re working towards all the time. The best answer we have so far is contracting with providers via the network contractor to require data share. Slow road, especially from a nationwide perspective. But that’s the best answer that we’ve come up with so far.

Carol Macumber, MS, PMP, FAMIA (29:30):
Do you think it’ll be the carrot or the stick that gets you over the hump?

Angie Glotstein, RN, BSN (29:34):
I’d rather it not be either. We’re using both, right? So the carrot comes from…

Carol Macumber, MS, PMP, FAMIA (29:39):
Good answer. Very nice answer. Alright, anybody else?

Dr. Michael S. Barr (29:41):
Just curious, Angie, we’ve had this conversation before at the NCQA meeting, right? Yeah. What value, when they share, if they were to share the data better, what value could they get out of doing so that we can highlight as a way to break through some of the challenges you’re describing?

Angie Glotstein, RN, BSN (30:05):
We have discussed this before and I don’t know. So what value can providers receive for sharing data back with clients?

Dr. Michael S. Barr (30:14):
In other words, can it be a part of the loop of data completeness back to them so they emulate a payvider situation? Because you’re getting data from your network. Their patients may or may not be out of network or outside their, so basically getting more complete data back because they’re all sharing with you as the central hub.

Angie Glotstein, RN, BSN (30:34):
Yeah, we are not even in the position to say that we would be able to provide that value because of the lack at this point. Is there…? My brain goes towards the longitudinal record of the patient, right? And so if you have a central collection point where you can create an entire picture of someone outside of just the EHR silo that provider sees, I think that’s exceptionally valuable. I’m years from that.

Dr. Michael S. Barr (30:59):
Understood. Understood.

Audience Speaker 2 (31:03):
Hi, Julia Skapik, National Association of Community Health Centers. Can you say something as a provider, quality measures aren’t that exciting to me because what I need is assistance from the technology tool to do the right thing for the right patient at the right time, i.e. clinical decision support. So can you say something about flipping quality measures on their head and implementing them as decision support artifacts first and then feeding that sort of back to the clinician afterwards?

Angie Glotstein, RN, BSN (31:36):
Hi Julia.

Michelle Dardis, MSN, MBA, RN (31:38):
So as a measure developer, I would love engagement from clinical decision support vendors with a feedback loop. The measures often lead currently with the value sets that are required because of the stick that exists with quality reporting. But could there be a feedback loop on the quality of the data model and value sets that we’re providing so that they can be useful for clinical decision support? I would love to also be a clinical decision support vendor so that you had the measures and the ECQMs together, but I think they’re in various separate spaces in practice. And so I think team building is the way to get it done.

Dr. Michael S. Barr (32:18):
So I’ll go back to something I said earlier. If we build based upon the measures we have in the accountability framework, you’re never going to be satisfied. We need to start with building measures in tandem with the clinical distance support and the type of data we need so that we can generate improvements in the accountability measures to take care of themselves. So yes, I think that’s part of what I should have said earlier. If we develop those better measures at the point of care to drive improvements, they should be embedded in clinical decision support or vice versa. But don’t take the HEDIS accountability measures and drive CDS associated with that.

Carol Macumber, MS, PMP, FAMIA (32:56):
Yeah. What about HL7? That great organization … who is a new chairman of the board? Dr. Skapik just happens to be. Okay. It’s not clicking, but the next slide was questions. So any others? Otherwise, thank you. So yes, please. Yes.

Audience Speaker 3 (33:22):
Okay, we continue to talk about data.

Carol Macumber, MS, PMP, FAMIA (33:25):
Unfortunately, you got to go right out here.

Audience Speaker 3 (33:27):
Oh, we continue to keep talking about the inputting of the data, but in my world, a provider has 15 to 20 minutes to see a patient. And if you’re not in the IT world, if you’re not checking that box, you’re not going to get that data filled to be able to transfer the data. What do we do? So the provider can spend their time with the patient and still collect the data? Because you know, as a provider, we’re telling the patient one thing. There’s not a box there to click. Yeah, you got free text, but who’s really going to do that?

Dr. Michael S. Barr (34:04):
I think there’s a big issue. And it has a lot to do with burnout and frustration, clinical practices driving clinicians out of practice retirement. So I think part of it is the design of the EHRs and collecting the information. I think there are a lot of interesting workarounds available now technology wise to take data that’s coming in reflected better, to make it more actionable. But we do really need to redesign that entire engagement. Whether the generative AI, ambient AI is going to be helpful- some of those pilots are going on. Whether there’s the slide in sort of a reflection of the data that you can click through and get the ribbon technologies. Those are all opportunities to improve the record because the records are going to be very slow to change themselves. And also to make it more meaningful so that we are doing that work. You’re actually improving the care of the person in front of you and appealing to the professionalism of the clinicians and the clinical team and not just for the sake of clicking through a measure requirement.

Michelle Dardis, MSN, MBA, RN (35:01):
There’s also a group convened called the National Burden Reduction Collaborative that’s co-convened by AMIA, HIMSS, AMDIS and a few other stakeholder organizations to identify the priorities related to documentation burden. Again, it’s a measures reference, but develop a measurement of documentation burden, but also put forward those solutions that are going to matter for providers. So one of them is advice on how to restructure clinical documentation for providers. And they’re distributing that with EHR vendors and EHR vendors are participating.

Carol Macumber, MS, PMP, FAMIA (35:34):
Right. I think we are at the end of our journey. Thank you so much. You’ve been great.