An insightful three-part customer case study presentation that delves into the captivating stories of organizations on their glycemic management journeys. It showcases the distinct phases of these journeys – from inception to continual enhancement – offering a comprehensive view of the challenges, triumphs, and transformations that mark the path towards improved glycemic health.
Featured speakers include Peter Rosher, R.Ph. from SUNY Upstate Medical University Hospital, Aimee Fahey, MSN, RN, CCRN-K, NE-BC from Soin Medical Center (Kettering Health) and Angela Hodges, PharmD, LSSYB, BC-ADM from Texas Health Huguley Hospital.
Get the Slides!
TRANSCRIPT
Dave Cooper: Good afternoon, and thank you for joining me for this session, the glycemic management journey, navigating beginnings, progress, and ever evolving optimizations. My name is Dave Cooper, and I'm the Vice President of Customer Success, and it is my honor to welcome to the stage three of our incredible customers who will share with you their journeys in glycemic management.
First to the stage is Peter Rosher. Peter is a pharmacist at SUNY Upstate Medical, which, as you will hear, went live with Glucommander IV last fall and Glucommander SubQ this spring. He'll share what the early days of their journey were like and some of the successes they achieved along the way. Please join me in welcoming to the stage, Peter Rosher.
Peter Rosher: Good afternoon. My name is Peter Rosher, and I wanted to talk to you today about diabetes management at SUNY Upstate Healthcare System. A little bit about myself is I have been a pharmacist since 1999, I've worked in retail and hospital. I've concentrated in neurology and oncology until 2020, and then moved to endocrinology full time.
I precept numerous students and pharmacy residents, and I teach an elective to medical interns. I've done some posters and publications related to diabetes management, and in full disclosure, I joined the Glytec clinical support team per diem in September 2023. A little bit about our healthcare system at Upstate.
We have two teaching hospital locations. We have over 600 inpatient beds. We have a children's hospital, a cancer center, and numerous clinics. We have an endocrine safety multidisciplinary team that includes all members of the healthcare system taking care of patients with diabetes.
And our team is always striving to improve diabetes management at Upstate. And some accomplishments that we have done are a pharmacist reduction of home basal dose by 20 percent on admission, restriction of insulin and oral anti diabetic agents that have the highest risk for hypoglycemia. We have a best practice advisory for providers when long acting insulin is not prescribed for patients with type 1 diabetes.
We're working with nutrition to offer better snack options and sugar free beverages. We've had a long transition from the sliding scale to the carb counting basal bolus. We did a DKA pilot. We've automated metformin restriction. We've had a robust discharge program for patients with diabetes. We've instituted glycemic stewardship chart notes, and we've recently formed a glycemic team,
But we are still having issues with management of diabetes. And in 2018 we had 66 safety alert events that went up to 207 in 2021, and that's a 213% increase. And in 2017, Upstate's rate of hypoglycemia was double the National Action Plan for Adverse Drug Event Prevention's goal of 21.42 events or fewer per 1,000 patient discharges.
And that corresponds with our data that we have from 2021 for severe hypoglycemia less than 55 milligrams per deciliter. And this chart's a little hard to see, but it's a roughly about 140 severe hypoglycemic events between both campuses a month. Our Vizient baseline data, less than 50 milligrams per deciliter, pushed Upstate to pursue Glucommander technology to help lower our hypoglycemic events at our institutions.
So we decided to do a two stage Glucommander rollout. In November 22, we did the IV, and in June of 23, we did the subcutaneous. Stage one, our IV Glucommander rollout. Prior to Glucommander, our IV insulin infusion had the highest rate of hypoglycemia at approximately 50%. I've included the nurse driven insulin infusion protocol as a reference for everyone to see how confusing it was and not surprising that we had such a high rate of hypoglycemia.
We also concentrated this in our ICUs, so a limited number of units. By instituting IV Glucommander, we nearly eliminated all severe hypoglycemia at our downtown campus. We fully eliminated severe hypoglycemia with IV Glucommander at our community campus. So these positive metrics helped us with the momentum of our subcutaneous Glucommander rollout.
Because we knew it was going to involve more nurses, more pharmacists, more patients, and more providers. We knew there were going to be more decisions. And that we needed to make a solid plan so this could be adopted by all members of our healthcare system. So, we spent a lot of time aligning our current policies with the new Glucommander policies, so nursing would have one way to practice diabetes management in our institution.
We decided to go with 100 percent postprandial insulin administration, and we kept hypoglycemic rescue meds the same between Glucommander and non-Glucommander patients. And nursing started to buy-in. Our blood glucose measurement timeliness was over 10 percent higher than other Glucommander institutions.
So they... are doing an amazing job and the results are showing it. We also had a multi-tiered approach to support Glucommander implementation. Members of the glycemic team reviewed daily hypoglycemic events, hyperglycemic events, and issues related to Glucommander. We started a Glucommander support EPIC secure chat line and a Glucommander Vocera support group to help answer questions in real time.
We endorse safety alert events to foster a team approach to education. And doing all this, we can celebrate the positive metrics where we nearly eliminated severe hypoglycemia at our downtown campus with Glucommander SubQ, and the same at our community campus. SubQ is way more intensive than the IV and is still showing positive results.
In this slide, shows the percent patient days by glycemic control. And our downtown location, the green circle, is comparable to Glucommander average institutions, and our community location is slightly higher in terms of hyperglycemia. But both institutions are way better than other Glucommander institutions in terms of hypoglycemia.
So this is early data, but it is propelling us in the right direction. And our institution is totally committed to improving the safety and health of our patients with diabetes. And one way that they embraced and supported Glucommander and our team was awarding me the Outstanding Individual Clinical Contribution of the Year 2023.
But our team was huge and it wouldn't have been a success without the members of all of our team. Thank you for taking the time to listen to my presentation today.
Dave Cooper: Thank you so much, Peter, for sharing what the beginning of your Glucommander journey was like with all of us. Definitely a lot of hard work, but definitely a lot to be proud of.
Next up, we have Aimee Fahey, the Chief Nursing Officer at Soin Medical Center of Kettering Health. Her story covers the mid-stage of the journey and what it was like after they implemented Glucommander. I love this story because it emphasizes that Glucommander, same as all aspects of glycemic management, are not set it and forget it, and that excellent results are the fruit of ongoing labor. Aimee, take it away.
Aimee Fahey: Hi, my name is Aimee Fahey. It is my pleasure to serve as the Vice President of Patient Care and Chief Nursing Officer at Kettering Health's Soin Medical Center and Green Memorial. I have spent 30 years in nursing, continued those in leadership, and it's been an exciting journey with our Glucommander rollout for the past couple years, and I just want to share a little bit of that with you.
A little bit about our system. We are a faith-based center in Dayton, Ohio. We have 15 medical centers with 120 outpatient facilities and on the screen here you'll see Soine, which is where we have done a lot of our tests of change with Glycommander and it is a 240 bed level 3 acute care center in Beaver Creek, Ohio.
After years of system discussion and review, we were ready to launch our IV Glucommander module. And so, there was a lot of EPIC build, process mapping, socialization, and education. And then, in the middle of this COVID hit, so we had a couple of lulls for that. But eventually we were ready to go live August of 2021.
And one of our goals, change is hard. And so we wanted to reduce as many adaptive changes as possible by doing IV, Glucommander as the first choice in EPIC and the only choice for DKA treatment. This worked pretty well. We were able to launch to get everybody on board using IV Glucommander.
And we were tracking at that time, what was our timeliness between the time the blood sugars were due and the time the blood sugars and that went amazingly well. It was a limited number of providers and clinicians that needed to be engaged because this was in our ICUs. We knew that SubQ would be a bigger lift.
So we immediately started preparing in November of 22 for a launch in June of 22. And so we, again, had a plan for SubQ insulin to only be available as a Glucommander product. However, at the last minute, as we were finalizing order sets, we realized that Glucommander really needed 48 hours to optimize glucose control.
So we still needed other SubQ orders available for our observation patients. We decided to go forward with that. We worked with IT to create dashboards, and we knew we would find gaps and errors, but we had a longer time figuring out meaningful data to give to our team than we expected. And without that meaningful feedback, excitement and utilization over using the products started to wane.
In July and August, right after launch, we had great utilization. We were at 58%, you know, with that last minute hiccup of not having Glucommander be the only product available for SubQ insulin. We were happy to be hitting this and what, again, still looking at what data can we share with our clinicians to get them engaged in using this.
As we continued to not have meaningful data, we saw utilization continue to wane down to about 38 percent in February of 23.
In January of 23, however, GlucoMetrics dashboard was created from Glytec, and we had the ability to see hypoglycemia comparisons, hyperglycemia comparisons, and our average admission and discharge value for patients on Glucommander versus patients not on Glucommander.
So, we did have one site that was an early adopter, a success story. They had more than 80 percent utilization. It was a very small community hospital, but we looked at them to see what had they done to make this launch successful and to sustain their results. They had one hospitalist group, which was helpful because there was only one team to engage, and we found out that the nursing educator there had really championed Glucommander.
She worked with the providers. She put tip sheets in their workspaces, in their sleep rooms, on their units. And she worked with the nursing team. And every time there was a patient who had, SubQ insulin ordered, she was checking on that patient, checking on the providers, and following up.
And we said, well, how can we reproduce this success story at our other sites? To start our refresh at our first site, we had our physician champions come to discuss outcomes of our network and our site, looking at the GlucoMetrics data. So we looked at historical data from Kettering to look at the cost of care for patients who had elevated blood sugars on our inpatient units.
We saw they had a higher length of stay, a higher cost, more infections, more falls, and then we looked at the GlucoMetrics data which showed fewer hypoglycemic events and better glucose control on Glucommander. And we talked to the physicians about what barriers are there, why aren't they using Glucommander.
One thing that had been told to me by a system leader is physicians weren't using it because nurses don't like it. And I haven't had any providers not use something because another clinician doesn't like it. So I wanted to really peel back that information and figure out what was going on. What we found is nurses were calling providers with technical questions.
Or they were calling because they were worried about the dose. They were, they were used to reactive insulin versus proactive. So if a blood sugar was normal and the patient was about to eat lunch, Glucommander was recommending coverage. And that was different and cause for concern. So the docs wanted an easy button.
How did they quit getting all of these calls? And so we provided an easy button, which was our charge nurse leader. So, any question about the Glucommander, how to use the platform or a question about an order 1st goes to the charge nurse leader. We definitely want any medication concerns to go to the provider, but only once the charge nurse and the floor nurse are able to discuss the concern and see if there's still a concern once they have talked through the dose.
We also found out that there was a 30 second delay opening Glucommander from EPIC, and we recorded that and we're able to get assistance from Glytec in fixing that so that now the providers can quickly and seamlessly get into Glucommander. We also knew everyone had not completed education from a provider standpoint, and so we provided pocket cards, put up tip sheets like the educator at the successful site had done, kind of plastered them all over so the team would know how to order Glucommander, and the different tips and tricks that there were.
We also knew that providing data would be important, so daily emails go to the hospitalist leads to, to let them know who has blood sugars that are high with patients that aren't on Glucommander, and then what's our utilization?
We had two teams of hospitalists. Team A works one week, and is off the next, and then Team B comes on, and that's not a reflection of their care. But that's just a reflection of two different teams with what we found were two very different utilization rates. So this information helped our leader meet with the team that wasn't utilizing Glucommander as much as we wanted to talk to them and really get them engaged.
And it eventually became a little competition between the two groups on how they could ensure that they were having a better utilization rate than the other. And then the nursing team, we needed to tell them ‘the why’. We established a multidisciplinary committee that has pharmacy, a diabetic educator, staff nurse champion, educators, and nursing leaders to talk through what are the barriers for nursing using this?
What are the questions that they're asking their charge nurse leaders that we could be proactive in addressing? And our first order of business was to start bite sized weekly information that would go out. We shared outcomes. What are our patients glucose outcomes if they're on Glucommander versus if they are not?
What is reactive versus proactive dosing? What should you expect for basal insulin, like when does it peak, how long does it last? Encourage them to look at the patient's total daily insulin intake at home because what they're getting in the hospital should be similar, maybe slightly reduced to what they're getting at home.
I mentioned the charge nurse leaders. They are super users and they are the easy button for the bedside team and provide a little cushion before information goes to the providers. We also reminded them there's a 1-800 number they can use if they're having technical issues at night and encouraged them to use our safe event reporting if they had concerns so that we could easily collect concerns and get information back.
And our results, we started that at the end of April. You can quickly see on this slide, our blue line, which are the treated patients, went from, you know, around 30 percent back in February to 60 percent in May. And as we continue forward, we are hitting in the high, like 76 to 78 percent utilization right now at Soine, and we continue to see that our patients who are on Glucommander have fewer hypoglycemic days and their discharge blood sugar is more in align with what we want than our patients who are not treated on Glucommander.
After the success in May, we took this same, um, information and information and data and process to another site where we again saw quickly they got into the mid 70s in treatment.
But just in September, I noticed that their utilization had dropped off by 15%, so I contacted that team to see what might be going on and found out that they had a concern with the order set and how it was working.
So, we were able to engage our team who does the order sets, validate it was working correctly, and then get that information back to those providers. So, they would start using it again. And we are seeing in October that they are starting to utilize this again, but it just shows the importance of continued and frequent communication and monitoring and training.
You're never done. You're always, just, just moving forward, and providing updates to the team and answering questions. So, what's our next stop? We will continue this refresh at our other care sites. So, at each care site, pulling together the providers to show them data. Pulling together the other part of the clinical team to talk about the why, to show them data, and to provide an easy button for our providers so that nursing questions or concerns are first being vetted by charge nurse leaders versus going directly to the providers.
And then I would just say, communicate, communicate, communicate. Sending the data, showing how patients are doing, showing how the team is engaging is super important. We also are working with one of our process excellence team members to go back and look at the past year and a half to see what, what can we find with data science to show our team, um, outcomes of the patients who have good hypoglycemic control versus outcomes of our patients who don't have quite the same glucose control.
And then my final destination is hopefully at the end of all this, I have a trip planned to Ireland in March. So I hope to, um, be on maintenance of all these sites and go to Ireland. Thank you so much for your time. I wish you the best of luck on your, um, journey to adoption and utilization.
That's my contact number if I can be of any service.
Dave Cooper: Thank you, Aimee. And you know that I'll be tracking the ongoing utilization of Kettering Hospitals over the next few months, along with our Clinical Customer Success team, to see how Glytec can continue to help in the optimization efforts. Because at Glytec, we walk this journey with you. You are never alone.
My team of Clinical Customer Success Managers, not to mention the support team, our clinical practice team, our product team, and everyone else at Glytec is always just a call away for the lifetime of our partnership. And someone who knows that well is our next and final speaker, Angela Hodges. Angela is the AdventHealth Glycemic Collaborative Co-Chair and longtime friend of Glytec.
Perhaps you recognize her from previous Time to Targets. Join me in welcoming her to share a story of ongoing optimization and how the work is never finished. Angela Hodges.
Angela Hodges: I am here to tell you that the work is never finished with inpatient glycemic management. There is always something new, whether it be a new nurse, new pharmacist, a new hospitalist group, or a new electronic health record.
Here's a little bit about me. I am currently the manager of the Diabetes Center of Excellence at Texas Health Huguley Hospital, and I'm also the co-chair of the Glycemic Collaborative for AdvenHealth. And a proud graduate of Texas Southern University. So here is our beautiful Tower A at Texas Health Huguley.
And we are the only hospital in our county that has a Joint Commission Certification for Advanced Inpatient Diabetes Management. We also have a Certified Outpatient Diabetes Center through the Association of Diabetes Care and Education Specialists, and part of our maturity, we have a pharmacy consult model since 2011 for insulin dosing.
That was also published in the American Journal of Health System Pharmacy in 2017. So, on our way to making sure that we have progressed over the years, sometimes you are awarded and sometimes you are highlighted, as you can see with the diabetes certification and the publication.
So let's look at hardwiring, focusing on that. So, initially we had Cerner as our electronic health record, and we had really hardwired our process. So we had the automatic pharmacy consult for insulin dosing that included ICU and non ICU. And this consult model covered the ordering of insulin, the ordering of labs that were relevant to glycemic management monitoring.
And also making sure that all of the pharmacists have access to those patients so they can make adjustments as they got calls from nurses. On the nursing side, we start with orientation. So, during nursing orientation, I'm allowed to meet with them for the first time, and the nurses really own this process.
By having that initial introduction to them, they get, they understand the Huguley culture, and also the expectation. And so then they do produce those results. So in our system, the nurses are taught to quickly intervene when patients meet certain criteria for scheduled insulin. And they also document an education assessment.
So all patients with diabetes at Texas Health Huguley get a diabetes education assessment. And we had our insulin administration process hardwired, of course. On the physician's side, The physicians do recommend diabetes education on the outpatient spectrum for our patients either going through the outpatient diabetes center or connecting with one of our community partners that provides outpatient diabetes education.
So, what do you think happens when you go from a hardwired process with one electronic health record to a totally new electronic health record? So, when we went to EPIC, as you can imagine, it blew up all of our old processes and procedures and we pretty much had to start over again.
So, we did have good reason for going to EPIC. So, AdventHealth wanted to bring all systems together. All facilities in alignment, so to make sure that from coast to coast, wherever we had a facility with AdventHealth, we had similar orders, similar formulary, as much as possible, similar processes, so that was the decision to go to EPIC, and I know here locally, the physicians were very happy that we were switching to EPIC, and just as a side note, Glucommander is what we were using to dose our inpatient insulin infusion, and we had had that program in place for five years.
So, when we started out going to EPIC, we first had to map out all of our old processes. So, the old Glucommander procedure, basal bolus insulin dosing procedure, and then we crosswalk that to EPIC. So what was available in EPIC? If we ordered insulin this way in, in Cerner, how did we have to order it in EPIC?
If a nurse did a diabetes assessment this way in Cerner, how did they do that in EPIC? Once all of that was in place, we wanted to monitor the process to make sure that what we put in place was working. And when you find issues, of course, you have to make adjustments. You have to update the process. So, that's what we did.
So, for a highlight, I wanted to just look at one area of that big crosswalk. And I wanted to highlight the Glucommander IV, DKA, and HHS panel. So what happened is we mirrored the power plans in Cerner, so pretty much the same order. The insulin drip was there. The IV fluids aligned with clinical practice guidelines.
We had initial and serial labs still. We had nurse communication to strengthen safety. We added potassium options to all of the IV fluids so that if patients met that criteria, they would be able to have very easily evidence-based practice, potassium added to their IV fluids, and we also added the hypoglycemia medications to that panel just to make it very quick, easy for nurses to get to.
So that was our starting point when we crosswalked to EPIC and created this panel for Glucommander IV.
And as I stated earlier, we had to review this process, so sometimes you don't know what is missing. You don't know where the problems are going to be when you start. So we did our very best to make sure everything was even, but what we realized very quickly is once you order a panel within EPIC, those individual orders are not visible to you any longer.
So in Cerner, if some of you are still on Cerner, you can hit the little light bulb and all of the orders within the power plan are available for you and the practitioner can make choices and changes right within that power plan. We don't have that light bulb function in EPIC that we have currently.
The EPIC panels therefore disassociate from that order. Once it is, once the practitioner makes a decision. So if the doctor decided to give D5NS, then the D5NS with 20 of potassium, that order is no longer visible. You only get the orders on the MAR that the doctor, or in the system that the doctor actually orders.
And Glucommander, what we also realized is that the multiply and target range for our obstetric patients could use some adjusting, and so we went back and with feedback from the field, went back and made adjustments to our multiplier and our target range for our obstetric patients.
And with our partners Glytec, they also shared this important information with us. They looked at when patients are being treated for DKA and they're on Glucommander, how often are we having hypokalemia, so a low potassium level. So we had 17 percent of those patients. Had a moderate hypokalemia, so potassium between 2.5 and 3, less than 48 hours after initiating Glucommander.
But we had 3 percent with severe hypokalemia. So again, reviewing the process and making adjustments. So, it was determined the best route would be, because as I said earlier, once you order within your panel, you don't have access to those evidence based practice, um, IV fluids.
And if you know about DKA, they're taking labs every four hours, so your potassium level can change, your sodium level can change, and so you need to be able to make changes to your IV fluids. So our solution for that was to make all of the IV fluids within the DKA panel PRN and therefore visible on the MAR.
And this was to try to help prevent hypokalemia by utilizing the correct IV fluids based on the electrolytes and using the corrected sodium level. We sent education for the corrected sodium level to the nurses because we made this a nurse driven process. So the nurse would have all of the IV screenshot of the D5 containing IV fluids there, D5 half, D5NS, D5 half NS with 20 of K, D5NS with 20 of K.
So those are all the fluids that would be available for the nurse to have access to based on the labs that are present. And then the bottom part there is showing you that we got very specific with our description on the multipliers. So just for example, for Glucommander, you need a multiplier, you need a target range, and so as you can see 0.01, we have DKA patients, HHS patients, chronic kidney disease, low body weight. Type 1 diabetes, including obstetric patients and hypoglycemia risk patients in general would be recommended to start at that multiplier.
And that was from feedback and monitoring our system, even post EPIC. So for our obstetric patients, if they were type 1 diabetes and had DKA, patients were adjusted to get the most conservative multiplier, that 0.01.
So when patients are pregnant, patients usually use a much more aggressive dosing for insulin, but we didn't find this to be beneficial for our patients. We also adjusted our level when patients are on Glucommander and they're pregnant, their hypoglycemia level to less than 70 or less instead of 60 or less because what we found is when patients were below 70, Glucommander did not overcorrect or allow them to go, uh, on below 60.
So we can keep that patient 70 to 100 or 120 when they are pregnant. We did not need to go down to less than 60 for a hypoglycemia trigger. So we have our first quarter post implementation data from Glytec, looking at hypokalemia with DKA treatment. And in the moderate hypokalemia group, we went from 17 percent to 9%, and from 3 percent to 1 percent for severe hypokalemia.
What does it mean to hardwire a process? So in psychology, hardwiring means your innate behaviors or thoughts that are not learned. They are instincts, reflexes, and behaviors that you're born with. But when you hardwire a process, you're going to create daily workflows and ideas into your work culture, also to create those opportunities to get things right, but you don't know what you don't know.
So here on the Johari window, we look at our fourth pane. There are things that might be unknown to the system and things that are unknown to the patients or the surrounding community, and that's the area that we really want to find, and you have to use your data to find those opportunities. So here's a roadmap to hard wiring a process.
First, you have to review evidence-based practice. Start with that evidence-based practice. Then you're going to leverage data to understand your baseline and where your opportunities are. You have to create the process for your weakest link team member to make it clear. Remove the waste, make it easy.
And then you have to educate the team. Use available tech resources, whether you have lens boards or the teams app or GroupMe. Those are some technologies that we have access to, which you may have access to different technology. You have to meet your team where they are. So sending an email is great, but not everyone reads their email.
You may have to use huddles, those daily meetings where you're going over topics so that keep them fresh and fresh in mind. You need to layer that information with pocket cards and notebooks so they can easily go back and find the answers that they're looking for. And also utilize your education figures.
We also have the opportunity to use rounding. So I have some departments in the hospital that are using education rounding to keep their information fresh for team members. So these are all areas that we can work on and all things that we can do to try to hardwire a process. So thank you for joining us.
SOP #34