TRANSCRIPT:
[00:00:00] Jordan Messler: Good afternoon. I'm Jordan Messler, the Chief Medial Officer here at Glytec and am honored to introduce our next speaker, Dr. Maynard. Dr. Maynard is the current Chief Quality Officer and Clinical Professor of Medicine at UC Davis Medical Center in Sacramento, California. He's a national leader in the space of quality improvement and patient safety.
He ran the Society of Hospital Medicine's Center for Hospital Innovation and Improvement for several years, helping to establish that organization as a leader in quality improvement while mentoring numerous hospitals around the nation. His teams have won national awards for work in patient safety and quality, including improvement projects related to blood clot prevention and inpatient glycemic control.
I'm lucky to have known Greg since I first started out in quality improvement work years ago, and I consider him an important mentor in my career. His efforts in glycemic improvement and metrics are well-known and we're very excited to have him with us today.
If you have any questions for Dr. Maynard, make sure to add them to the Q and A, and we'll answer as many as we can at the end. Anything we don't get to today, we'll follow up with you after the event.
I'm happy to present Dr. Greg Maynard.
[00:01:05] Gregory Maynard: Thank you Jordan, and thanks for having me. Glad to speak today on this topic. Um, so we are talking, uh, about the CMS measures, but by way of grounding, we'll be talking a little bit more than that. Um, we will be talking about the eCQMs that are coming from CMS, how they might impact you and your finances, but we'll also look at, uh, some background on glycemic, uh, metrics and future, uh, venues and measures that will be coming on the horizon that I think will be very helpful to improving teams.
Um, I'm going to try to make this so that you have some takeaway, some actual items that you can prepare for these measures.
There's a lot of reasons why we do glycemic measurement. Um, we assess the baseline in our own hospital and garner support for change. We often have to, uh, convince doctors and nurses that implementing this insulin protocol is safe and, uh, can be done without undue hypoglycemia. You certainly want to track progress over time.
You want to be able to identify your strong units and your weak units and have them teach each other. Uh, and you want to prioritize efforts. For example, if you have really bad hypoglycemia, you don't want to push glycemic control really badly or really hard until you improve the hypoglycemia. Uh, it lets you really assess those trade-offs and I think one of the most important things is, that you can benchmark, um, yourself against other hospitals, if you have common measures across hospitals.
All right. So, uh, Unfortunately, a lot of hospitals don't have metrics of these types. Uh, they might have some rudimentary metrics, but often they're not like other hospital's metrics, so they can't benchmark themselves and, um, we find a lot of variability in this regard.
So how I got into this basically is tied in to the history of, uh, glycemic control and glucometrics and how that got to be ported to the Society of Hospital Medicine.
Uh, I was Hospital Division Chief at UC San Diego, starting there in 2003. And like Jordan, uh, glycemic control is my first quality improvement project. I didn't know I'd still be in it this many years later. Um, I think that happens to a lot of us. Uh, through my work with Society of Hospital Medicine, we convened a task force and I was fortunate to begin to network with all these experts in glycemic control, many of whom are speaking at this meeting today, including people like Guillermo Umpierrez and Jane Jeffrie Seley and Silvio E Inzucchi, and on and on. Uh, we got input from so many experts, uh, and it was very helpful.
Uh, that task force convened and we started to build out implementation guides and resource rooms, uh, and, uh, a supplement to the Journal of Hospital Medicine.
We started to build out our own glucometrics, um, in 2009 at UC San Diego. And as we launched these mentored implementation programs that Jordan and I have been a part of, um, at, uh, SHM, we realized they were hindered by not having common measures. So in 2012, we ported the UCSD glucometrics to Society of Hospital Medicine and allowed people to upload their data on a monthly basis and then we would interpret that and give them back the ability to go and, on demand, look at how any unit or group of units was doing across any time period.
Um, at, That kept going strong for a while. Uh, I moved up here to UC Davis in 2015. Uh, we kept benchmarking periodically. We had the fourth round of SHM VIC benchmarking in 2015 and now here we are years later at the 17th round of SHM benchmarking, it happens about twice a year. Um, and we kept putting out refined, uh, glucometrics and, uh, updated toolkit, the last one being in 2015. Probably, uh, due for a refresher that includes some new topics, like CGM for example.
In the meantime, this is available at no charge on the SH, SHM website, and it includes most of the principles I'm going to cover today around metrics and data collection. So if I don't cover something fully, that and more is available for free online.
Some of the basics are that there's different units of analysis or different ways to look at glucometrics. It seems so simple because you're just looking at glucose values and a date and timestamp and the unit where they're at. But, um, eventually, uh, it can actually get quite complicated.
Early on, people are using the actual blood glucose, blood glucose reading event itself as a unit of analysis. That will give you a really large denominator but you can imagine how much the measures would vary depending on whether or not you're, uh, whether you're checking glucose four times a day or 24 times a day.
So we began to move towards patient-day metrics, um, which basically look at hyperglycemic or hypoglycemic days. And they create things like stay weighted, uh, or excuse me, day weighted measures for glucometerics. So you'd add up the mean for each day, um, and that would be the way that you would gauge glycemic control and hypoglycemia at our hospital. We find that was probably the most useful.
Patient-stay measures are also useful, they tend to translate. So if you say we had, you know, 5,000 patient stays and this percent had hypoglycemia, people can kind of get that, get their arms around it. So it's easy to interpret.
This is an example of an on-demand report you can pull up if you participate in the SHM, uh, eQUIPS Program, which is, uh, electronic quality improvement program.
Basically you can pull up any date or any range of dates with any, uh, kind of unit or kinds of units. Um, and you can set your own date range and it gives you all these different measures by both patient-stay and patient-day units of analysis.
Uh, so in this case you can see it shows mean and median, the percent with the day weighted mean over there, a whole stay that was greater than one, greater than or equal to 180.
Uh, it's got a metric for readings and range, um, during the stay. And it's got a lot of hypoglycemia metrics. Uh, glucose less than 40, less than 54, less than 70 on another count with glucose greater than 300, uh, recurrent hyperglycemic days and some fairly unique metrics around timeliness of hypoglycemic management and recurrent hypoglycemic days.
So all those were available on demand. Uh, and you can also drill down further in these patient-day metrics where you see month by month, what happens in addition to the roll-up amount for that time period. And I'm not showing them, but, they, you can also generate, uh, run charts that show your performance over time in a graphic manner.
This is the benchmarking I was talking about. Uh, we do benchmarking for critical care and non-critical care units separately because they're such different environments. Um, and the, your hospital would be marked in red on these reports. Um, this is, uh, basically hypoglycemia on the y-axis and around these are the rankings for the hospitals in gold numbers down here. And this is the hospital ID on the top of that purple box.
So basically you can see a huge range of hypoglycemia, uh, exists in these hospitals. You know, it's a five or six fold difference between the best performing and the worst performing hospital, which is one reason why it's important to have better measures and more attention paid to this and more standardization.
We thought it would be useful to, in benchmarking, have a way to look at both hypo and hyperglycemia on the same chart, so you can put them in context. So in this case, glycemic control is on the Y axis here and a marker of hypoglycemia is on the x-axis and your goal is to get in this bottom left quarter.
Uh, quartile, where you're in the best quartile for both glycemic control and hypoglycemia. So you can imagine that if you do benchmarking, the approach of hospital, this hospital in this quadrant, would be entirely different from the hospital that's up here in the upper outer quadrant. So it gives you that context, uh, it tells you where you're at compared to a hundred other hospitals, and it's relatively low cost. I forget exactly the cost, but I think it's you know, $5,000 or $6,000 a year or something like that.
Over time in conferring with all these experts, we came to prefer the patient-day as a unit of measurement.
Um, basically repeated hypoglycemia readings during a given day, mean that it's less useful perhaps, they see how many hypoglycemic events during any given day. And it really evens out the playing field in terms of a patient, uh, having a really long, length of stay. Um, so you're looking at each individual patient-day as something that's important to value.
So transitioning to CMS E uh, uh, quality measures, uh, it is a messy process how they do this. So they propose a measure, they have a contractor that develops it, they go through a really long process. Um, and eventually, they ended up in a clearinghouse called the National Quality Forum. And that's important because it keeps, as many measures as the are, there would be multi-fold more if we didn't have a national clearinghouse. So getting into the NQF is really important.
Uh, even once you're in there, it's no guarantee, you have to go through public comment. There is work with the IT vendors and other stakeholders to make sure that it's extractable and easier, easier to do. Um, They have to go through proposed implementation with more comment, even after they release it for reporting, then they have to assess it and they decide whether or not they're going to have incentive or penalty with reporting.
Once it's out there, it can be used in multiple venues. Um, so you might find it showing up in three or four different publicly reported dashboards.
Well, back to the hypoglycemia NQF measures and eCQMs. All the way back to 2014, we thought we were almost there. We thought we finally had publicly reported measures that would garner support, uh, to help glycemic control teams, get the support they need to standardize care, get the metrics they needed and so on.
Um, Because we had this, uh, measure that went through NQF with a corresponding glycemic control measure, at a numerator of severe hypoglycemic events that were proceeded by anti-diabetes drug, denominator as the number of inpatient days with the, w, with greater than one or, uh, with one or more, uh, anti-diabetes, drugs being administered. Uh, at the time frame was adjusted for the kind of insulin and they had typical exclusions, people with really long length of stay, they tried to get rid of spurious events, events within 20 hours of a prior event to rule out just a repeat event. And, uh, it was really for adults, so patients less than 18 were out there.
So we got that endorsement, we thought we were home free and then it stalled. Guh, uh, it went nowhere, much to our, uh, basically, basically our frustration with that whole process. Then finally in 2019 and, uh, the NQF endorsed a different measure, they withdrew their endorsement of the prior measure. So NQF 3503e is now the current endorsed NQF measure that had a final proposed rule come out in October, 2021, which is a very important step, one step further than the last one ever got.
Uh, basically this one looks at patient stays or inpatient admissions. So technically speaking, it may be easier, um, metric quality-wise, may be not as good as the prior measure, but again, any public measure is probably better than none. And, uh, it would definitely give people some directional, uh, value about where they're at.
Uh, this is the corresponding hyperglycemia measure. Uh, events in the first 24 hour period are excluded. The day of discharge is excluded. The numerator is the number of inpatient days with a blood glucose greater than 300. And the denominator includes those qualifying inpatient hospital days for patients 18 years of age older.
Here we are back at day. So, uh, um, you know, they kind of mix and match their unit of analysis here. Uh, the other, uh, population measures are relatively straightforward. They're trying to make sure they have diabetes or have an ADD administered, or have a blood glucose greater then equal to 200 to qualify somebody with hyperglycemia or diabetes.
So that is coming. It sounds like, um, I'm going to try to give it a little bit more granularity around that, about the timeframe and how that's coming. The eCQMs are really part of the inpatient quality reporting program, and that is a P4R reporting quality program that reduces payments to hospitals who fail to meet program requirements of reporting.
It's a big deal, cause it would subject you to 25% reduction of annual CMS payment for all your Medicare patients for failure to report. So virtually everyone, uh, uh, participates in this program.
CMS. There's currently removing five measures that are more or less now routinely met and they're replacing them with five new measures that include these two new, uh, NQF measures, these eCQMs for hypo and hyperglycemia. That reporting begins in calendar year 2023. Um, and, in, uh, fiscal year 2025, they'll start making payment determinations. So start deciding that there'll be penalties for not meeting a certain benchmark or, uh, or other ways to monetize that, uh, those measures.
So the measures will be there, there'll be available to pick from, but they will not be mandatory. So hospitals need to pick three measures to report on, in addition to a mandatory opioid metric. They have 7-9 to pick from so they don't all I have to pick those two that I just showed you. Many hospitals will, I suspect, and it's, they're designed to act as a pair so that if you pick one, you should report on the other as a balancing measure,
Uh, financial impact will depend, I mean, your results may vary. Uh, certainly you had to participate in the IQR program, that doesn't mean you have to report on this right away. Uh, but the impact could run into the millions, particularly if these measures get integrated into other reports like a HAC Reduction Program, or Hospital Acquired Condition Reduction Program, or their incentives or penalties, um, depending on your performance.
You can expect, I'm sure, for the eCQMs to garner more interest and support for glycemic control teams and the tools that encourage more appropriate use of insulin. I think that's a big win we've been waiting for for a long time. It should be fairly straightforward for your hospital to follow all eCQMs at your website, even if you don't report to CMS, because if you've got an EHR, and you would adhere to the standard bill from your EHR vendor, your EHR vendor is working with you to make sure it's fairly easy and straightforward to report the eCQMs.
It shows that a bit more graphically, a new proposal goes into effect October, 2021, hospitals will work with the EHR providers to implement reporting packages, uh, in the next few years. Uh, the eCQM rules for data collection period is January-December, 2023, it's first full year. 2024 is when you'll start deciding whether or not you're going to report on that, on those particular measures.
If it works in your hospital like it works in mine, you probably have a group that is looking at all these measures and deciding which ones are most strategic for them, are the ones who will be able to most likely, uh, improve on and reach the goals. Um, so, uh, again, not everyone will pick these, but even if you don't pick them, you'll likely have greater visibility of these metrics.
Uh, and then the data become available publicly. Uh, you won't be able to change 2023 data once it comes out. Uh, it'll be old data by then, but that's when it'll be public reporting. And then, uh, financial penalties began for those who don't report these eCQMs in 2025.
So. Like I said, uh, CMS can do what they will with these measures. I think that a lot depends on, uh, how they're received when they come out, and what the performance is when we start doing, uh, more analysis to compare hospitals to each other.
What can you do to prepare and take action to get ready for this. Well, the first thing I'd have to say is these measures will really help you garner support, but they are not very good measures to help you improve. They're basically telling you when you've had a car wreck and when you run off the road, but they give you not much other guidance, so you really need better metrics in this.
Um, and realize you will not get any analysis by units or services. So when you come back and show this number, your doctors or nurses will say well about, what about my unit? What about my service? You won't be able to tell them anything from those eCQMs, um, necessarily. They lump critical care and acute care units together, there is no risk adjustment. So hospitals with more complex patients, a longer length of stay, will probably be penalized. And they're only looking at that severe, less than 40 milligrams per deciliter cutoff and greater than 300 with nothing really in between.
So, uh, if you're trying to improve, you want to improve so that you never have those less than 40 events. Um, and you take action when it's earlier than that. And you have to, you know, start billing protocols for hypoglycemia prevention at much higher numbers, than less than 40.
Uh, and then there is no measures for timeliness of treatment or recurrent hypoglycemic events, which is a marker of how well you're paying attention to the hypoglycemic event the first time.
Uh, there's no measures for in range and there's no measures for insulin use patterns. So clearly not enough to help you run a good quality improvement program, but useful for garnering support.
You really need some robust local metrics to drive improvement, whether they're homegrown or purchased. Um, certainly, uh, I understand Glytec has quite a few measures that you can take advantage of, SHM has some, you can build your own, there's others that are built into some services with, uh, uh, with, with the glucometers. Um, but you have to use something and more than what you're going to have in those eCQMs.
I think in addition to the monthly and quarterly reporting and benchmarking, you really also need real-time reporting um, to tell you who's almost in trouble now so that you can intervene before they get lower, or who's out of control on the high side so that your team can take action in some way to, uh, intervene in real time.
So you're not just looking back at people who are already out of the hospital, you're looking at people in real time and intervening at the same time, which we call active surveillance for sometimes a measure-vention.
Uh, and I do think separate reporting for critical care units and acute care units is basically essential since they're such different environments with different rates of monitoring and different kinds of insulin being used most of the time.
Glycemic control teams are essential, you've heard something about that here. You need, you know, to have standardization and then build that standard protocol into your EMR order sets and workflows, tools that nudge clinical decision support.
Uh, and you know, there are so many areas to improve besides just acute care units and critical care, there's peri-op and all the transitions and so on. So you really need all these different sorts of things that have been highlighted somewhat already in the, uh, conference.
So, there is, I think, a really exciting thing coming. I've been part of a subject matter expert team working with the CDC and, uh, The National Hospital Surveillance Network. Um, NHSN. That are, that are really interested in using their framework to enhance this reporting on, uh, anti-diabetic drug, adverse drug event reporting.
Um, so CMS and CDC have been actively partnering on this and CDC totally recognizes the shortcomings of the current eCQMs. So what they have in mind is using their platform to replicate the eCQMs but also add measures by patient-day, um, in addition to the patient-stay measures, they'll add the measures with the more cutoffs, um, like less than 70 and less than 54, uh, they'll have more measures on glycemic control and just like you can do for hospital acquired infections.
It will all allow you to, um, look at units and groups of like units, um, and how they're performing and it will do risk adjustment and we'll have benchmarking. So you'll have a way to gauge yourself against other hospitals, I guess, in some sense, it is my hope that this makes the SHM platform obsolete. If enough people use it and this actually comes to fruition.
They've been talking about this since 2014 or 2015 as well, but I really feel like this effort has legs and that it will happen. They put out an HL 7 implementation guide for hospitals and vendors. I think that will turbocharge interest in inpatient glycemic efforts and really broadened public reporting. Um, the estimated time of arrival, I, realistically, it's probably 2024, uh, but I hope for 2023.
So those are the basics of glucometrics and the eCQMs version out of them and I, what I hope will be a greatly enhanced version of them coming from the CDC. Uh, I think all of them will be helpful and, um, that none of them will be enough in and of themselves to drive improvement, but they will certainly help. I think, uh, with that, um, we can stop for questions or comments.
Uh, Jordan if you've been looking at the chat, You can let me know.
[00:26:44] Jordan Messler: Great. Thanks so much Dr. Maynard, that was a really fascinating insight of the world of metrics and the world to come around metrics. Uh, don't forget to answer, uh, to put your questions into the Q and A, so Dr. Maynard could answer your questions.
Uh, we do have a couple of questions coming through, um, one, um, relating to eCQMs and what you just mentioned around the CDC.
It seems initially these will be, uh, volunteer elective metrics. Uh, what do, w what,with elective metrics, um, um, who sees these metrics and what do you think you'll still get that level of interest and awareness from, uh, the fact that they're elective and not mandatory, uh, early on?
[00:27:27] Gregory Maynard: Well, only one, there's only one mandatory eCQM that I know of that's the opioid one. So, um, it's mandatory to pick 4, including that mandatory one. So I'm sure a good portion of hospitals will, will pick those hypo and hyperglycemic eCQMs, and it will, it will stir an interest.
Um, for the CDC, they still haven't worked out how that's gonna work exactly, exactly how will, um, be reported. They have the same HL 7 connections as uh, as they do for the eCQMs and you know, it could become mandatory reporting, just like hospital acquired infections are for CAUTI and CLABSI and C. Diff and so on, but I don't think those things are really worked out yet. I think anything that gets public reported, uh, or has even the option of getting benchmarked, will gain attention and will help quality improvement teams.
Um, uh, certainly it would be more impactful if they were all mandatory. Uh, and uh personally, I think sooner or later, the NHSN ones probably will become that way, but, um, I don't really have insider knowledge about that, those discussions, and can't comment further other than that I think, you know, it, that is their MO.
Their MO's have pay for reporting, uh, and then pay for performance and move them into mandatory, uh, slots later on as they mature.
[00:29:11] Jordan Messler: Great. Thanks. Um, we have a question about critical access hospitals. How does this, uh, apply to critical access hospitals in rural areas?
[00:29:21] Gregory Maynard: Oh, geez. I wish I had the expertise to tell you about that. I'm sorry. I, I, we can, I can ask about those things and try to get back to, uh, offline.
[00:29:33] Jordan Messler: I think, um, um, one thing I could add to critical access hospitals is, uh, I believe they don't apply to if that beds, uh, less than 25 beds, but if you're, um, but, uh, I believe greater than 25 beds, uh, would apply, but we can get back a more concrete answer to that one.
Um, I, you, another question that came through, I know you answered some of this throughout the talk, uh what would you suggest, uh, that my hospital start doing now to get ready for these measures.
[00:30:02] Gregory Maynard: Well, yeah, you have to start with their self-assessment. Right. I think the, uh, the, uh, the guide of, the implementation guide I showed up front from Society of Hospital Medicine, sort of walks you through those steps, uh, making sure you have an interdisciplinary team that's empowered to standardize, uh, and that you start working on measures.
And before you go, trying to build your own, you should do a good amount of research about products that are out there that can help you do that, um, in some standardized way and, uh, and augment those measures with some local active surveillance measures, uh, that you can use in real time. So there's lots you can do.
But most of it's basically following a quality improvement format that we're, that was discussed, uh, in earlier sessions today.
[00:31:02] Jordan Messler: Terrific. Um, I think, uh, I think that's all the time we have, uh, I don't see any more questions coming through the chat, uh anyway. Um, so we'll be sure to follow up with you if you didn't get your question, or if you have a question that comes to you afterwards, um, let us know.
I'd like to thank again, Dr. Maynard for joining us tonight and sharing his expertise and insight into this very important topic.
On behalf of the Glytec team, I'd also like to thank all of our attendees for logging in today. We appreciate you being here, uh, and uh hope you log in again tomorrow and enjoy the rest of the time at our Time to Target conference. Thanks again, Dr. Maynard.
[00:31:39] Gregory Maynard: Thank you.
END OF FILE