This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

Flourish Sound Bytes: The Surprising Challenges of Diabetes Care with Sherita Golden

[00:00:00]

Sarah Richardson: I'm Sarah Richardson, a principal here at this week Health where our mission is healthcare transformation, powered by community. This is Flourish Soundbites, unfiltered Conversations with healthcare leaders. Let's get real,

Welcome back to Flourish. I'm Sarah Richardson, and today's soundbite features a leader whose work sits at the critical intersection of clinical care, technology, and equity.

Dr. Sherita Golden Sida has spent much of her career focused on hospital-based diabetes management. One of the highest. Risk and most complex areas of inpatient care. She's also worked closely with hospitals pursuing the Joint Commission's, advanced inpatient diabetes certification, helping them understand how it systems, data, and workflows play a direct role in patient safety and outcomes. Today we're talking about what safe glucose management really requires in the inpatient setting. How technology can be a powerful clinical asset and why thoughtful system design is essential, especially for populations that have historically been [00:01:00] underserved. Sherita, welcome to the show.

Sherita Golden: Oh, it was great to be here. Thank you for. Having me.

Sarah Richardson: Of course, you have focused much of your work on hospital-based diabetes

Sherita Golden: Mm-hmm.

Sarah Richardson: Why is inpatient glucose management such a high risk environment compared to outpatient care?

Sherita Golden: Well I think there are some things that make the hospital-based care patients with diabetes unique. So one is that when patients come into the hospital, there are all of these, clinical conditions, some of which are related to whatever illness they've come in for, and others that we create in the

Hospital that they're not used to managing. for example, we give people medications, like steroids that make their glucose levels go up really high. We, make people what we call MPO, we stop their meals for a period of time so they can have a procedure. So we gotta figure out how do you adjust the medication around that.

And then we also, um, give people a real carbohydrate controlled diet in the. Hospital. So for some of our patients who are outpatients, they're used to actually eating more [00:02:00] carbs than are recommended. So when we give them the carbohydrate control diet in the hospital. Blood sugars go south. So there are, conditions.

And then we also give people, pain medications that make them drowsy. So all of these things impair their normal ability to manage their glucose. But then the other thing is that insulin is one of the top most, dangerous drugs. In the hospital, and that's because they can cause a really precipitous drop in blood sugar, if not used appropriately.

So, you know, they are right up there. Insulin is right up there with, opioids and also with anticoagulants as like the highest risk drugs that we use in the hospital.

Sarah Richardson: That is not something that you would normally believe to be true. When you hear that, you're like, how could insulin be as dangerous as these other drugs?

but when you start to think about. Breakdowns in care processes, data decision making that put patients at risk. Where is this happening most often?

Sherita Golden: Well, it typically is happening at the transitions of care, and I can think of three that [00:03:00] I've seen over the course of my career. One is from the emergency department to the inpatient unit. did they get the insulin in the ED before they came upstairs? Suddenly? Like, because documentation systems may not be clear.

It can be unclear to the receiving team whether the patient got their insulin. So that's one transition of care. The other is in moving from the intensive care unit to a step down unit or to the regular floor. So the patient may have been on, say, intravenous insulin in the intensive care unit when they were more ill.

Now we're moving them to the floor and we're transitioning them from IV to subcutaneous insulin. How do we do that in a way that's. Safe. And then the other is that last, transition, which is of course from the inpatient setting to home during that discharge, back to the ambulatory setting, the, you know, what, what they were taking in the hospital may actually no longer be appropriate for what they need as an outpatient or what they were taking as an outpatient was an adequate.

And we need to adjust based on what happened in the hospital. And sometimes the inpatient ambulatory communication doesn't [00:04:00] happen smoothly.

Sarah Richardson: How have you seen transition of care protocols or handoffs improve over time based on the type of knowledge you're sharing with us today?

Sherita Golden: Well, I think one of the common handoff errors that we used to see in our own system was, um, say a patient was coming from the intensive care unit down to the floor and they were on, their intravenous insulin. Now they're eating, they're no longer on a breathing machine, so, we don't wanna just stop the insulin drip 'cause the sugar will shoot up.

So we would, start their, subcutaneous insulin about two hours before we turn that insulin drip off. But somehow in our initial smart order set, which I joke was not very smart compared to what we do today, it wouldn't be clear whether or not the patient actually got that dose. In the ICU before they left.

So then they would get another dose when they got down on the floor and then guess what? Their sugar drops really low. So It's because, you know, it wasn't clear in our IT system whether or not they'd gotten that dose before they left the [00:05:00] ICU. So, you know, we really designed a system where, basically if you've got one dose of a long acting insulin that lasted for 24 hours, that you were unable to get another dose before 24 hours.

Like it would be like a hard stop in our system. So that's an example of how we addressed a transition like that.

Sarah Richardson: And then the joint commission's. Advanced inpatient diabetes certification includes a requirement around using IT systems to track glucose metrics. Why is that so important?

Sherita Golden: Well, I think it's important because datas can really, the data can help to identify the patients that are most at risk in the units that are most at risk for either having patients with high blood sugars, hyperglycemia, or hypoglycemia, both of which, can be dangerous in the hospital.

So, you know, much of that data now sits in our EHR. So kind of in the, the, a lot of work that, that, that you do. And so being able to pull the information from the EHR that's been documented by the clinicians can really, um, help to inform our clinical, QI processes and really say, okay, here are the unit where we're having the most [00:06:00] hypoglycemia.

We're going to implement an intervention on that unit. And then we're gonna re-look at that data to see whether Hypoglycemia has improved. So it really does allow us not only to identify, where the target areas are, but to figure out whether our interventions have actually improved the outcome that we're looking for.

Sarah Richardson: Well, and you've helped. Several hospitals on their certification journey. What have you seen separate organizations that struggle from those that are succeeding in this space?

Sherita Golden: I think the most important thing is leadership buy-in. at the highest level, executive leadership really, Championing, safe management of diabetes indicating that it's an institutional priority. I think it also, in addition to indicating it's a priority, it's showing that it's a priority by investing financially in.

The glucose management program. So that means the, you know, physician champion, these are some of the things that, the Joint Commission requires. It can be the clinical implementation team, the IT expert who is [00:07:00] contributing to those efforts, ensuring that the staff that write the policies and execute the policies are all funded.

So all of those things, are really important. it's not just the. I'm telling you that this is a priority for our institution, but it's showing that it's a priority by investing financially. Those are the systems that have been most, successful.

Sarah Richardson: the consistent investment. In the technology side of the aspect, the change the process, how

Sherita Golden: Mm-hmm.

Sarah Richardson: it systems, EHRs, dashboards, order sets. Ones that may not have been as smart as you wanted them to be, and now they're more informed.

Sherita Golden: Yep.

Sarah Richardson: contribute directly to safer glucose management?

Sherita Golden: Well, I think that when, they're built optimally, the IT systems communicate, they're able to pull data and summarize them in one application so that clinicians are not having to go to different places to piece things together. when I first started this work, we had the glucoses, like in one dashboard.

But then the fluid status and whether they were NPO in another dashboard and the amount of insulin they got was in another [00:08:00] dashboard. we just said, what if it could be like it was when we had paper charts at the bedside and everything was actually literally written down in a flow sheet.

Can we reproduce that in our IT system? So over a number of years, we were able to really juxtapose all of that information. So the blood glucose. The insulin they got in response to the blood glucose. you know, whether or not they were eating. When did the eating status change? And what are other critical lab values that we need to understand?

Like do they have, acute kidney, injury in the hospital, which can then contribute to a lower blood sugar causing them to need less insulin. But so having all those things lined up on one screen has been really critical to enabling our clinicians to make, the safest informed decisions.

And I also think that when the clinical IT systems follow the evidence-based guidelines and making recommendations, it can actually be a teaching tool. That then reinforces the evidence-based prescribing practices to the clinicians that use them on a regular [00:09:00] basis. So, those are things that I think are really critical.

It's like giving the clinicians the tools they need at the point of care to make the decisions.

Sarah Richardson: And how helpful it truly is when we think about data.

Sherita Golden: Mm-hmm.

Sarah Richardson: one place,

Sherita Golden: Yes.

Sarah Richardson: I appreciate the move from paper to electronic, and yet the ability for electronic to truly be an enabler in a way that's more efficient than what we had left behind in the past.

Sherita Golden: I joked for years and you date yourself when you say things. I was like, I used to go to the patient's bedside when I was in the mid nineties and pull that little black bedside chart out and it would all be right there and it was like, okay. It was very clear then what I had to do. when we went.

To our EHR. the nurses were documenting in one system and all of that had to be integrated into one system. we had kind of piecemealed our IT systems and they weren't all communicating. I've seen over the last 10 or 15 years that evolved so that now we're seeing everything in one [00:10:00] place.

But that was one of the biggest challenges. I was there for the paper EHR transition, and that was the biggest challenge.

Sarah Richardson: Oh yeah, we were there.

Sherita Golden: Yes, we were

Sarah Richardson: audience appreciates the perspective we're bringing to the table. Sharita, it took us a while. Took us a long time to have this perspective, didn't it?

Sherita Golden: Yes. And I think a lot of our younger clinicians, they don't even understand, what do you mean a paper bedside chart? And you know that this was kind of the error that we had to transition through. So it's been great to see that happen successfully.

Sarah Richardson: And yet hospitals that are well prepared for any kind of operational consistency during an outage for whatever reason it may be. If you have to go back to paper for a few days, it's pretty

Sherita Golden: Yes,

Sarah Richardson: how that learning curve is in reverse for populations that have never been outside of an EHR,

Sherita Golden: yes, exactly. Yes. Very interesting. When that happens.

Sarah Richardson: and you've said you see technology as an asset to clinical care delivery, what does that look like when it's done well?

Sherita Golden: You know, I think what it does is it supports the clinicians [00:11:00] in delivering, high quality care that's safe, but it still incorporates their best judgment. So it's not replacing their clinical expertise, but it is enabling them, giving them guidance. So that then they can focus on, well, how do I need to then adapt this for the patient in front of me?

So, you know, our diabetes order set based on whether the patient has type one or type two diabetes, whether they've eaten. Or are eating, how much insulin they've gotten in the last 24 hours will make a recommendation based on the current blood sugar readings, like what their insulin dosing should be, for the next day.

but you know, a clinician may look at that and say, okay, it's recommending this much. Insulin, but I know from the fact that every morning they've awakened with a low that I really should not raise their basal insulin. I'm gonna keep it the same, or in fact reduce it by 20%. To me, that's where the order set is giving you the foundation to start with, but then it's allowing the clinician to [00:12:00] still have the power to use their clinical judgment to inform the final decision.

Sarah Richardson: And the maturity of the order set is also something that you've shared with me. Why is it such. A critical element of inpatient diabetes workflows.

Sherita Golden: Well, I think it's important because, you know, hospitals, you know, the, the, it needs to be able to account for changing clinical conditions. like I mentioned, someone may come into the hospital for, one thing and then the next thing you know, they've got pneumonia and now they, um, you know, they, their kidney function isn't as great as it used to be.

So you need to be able to rapidly adjust. Insulin dosing, sometimes more than once a day in order to keep their blood sugars controlled in a safe manner and not induce hypoglycemia. So I think that's why we need to have mature order sets that are actually able. To allow the clinician to respond in real time.

You know, if you can only adjust insulin once a day, that's not going to really work in, a clinically dynamic, environment like most of us find ourselves in, in the [00:13:00] hospital. And I think the other thing, I joke about our first order set. we called it smart because we were so excited. We had spent.

Months. We met like every Wednesday for almost a year in the little conference room with our chief, information technology officer. We built, you know, this order set. And what it would do is, again, based on these clinical questions, what type of diabetes, whether or not the patient had eaten or not, um, how much insulin they got in the last 24 hours, and was it.

Intravenous was a subcutaneous. And, it would basically, you know, say, okay, here's the, here's the insulin doses they want, and it's a little gray box down at the bottom. You check the boxes that give you the dose, but then the clinicians still had to go in and put the orders in. So now what the order set does is it pre-populates.

It tells you how much basal insulin, how much insulin before each meal, and what type of correction scale that you need. and then it still allows the clinician to then say, okay, I can adjust this if needed, based on. This particular patient, but it pre-populates. And again, because it's [00:14:00] pre-populating based on American Diabetes Association recommended guidelines.

Then it also, again is teaching the clinician and reinforcing the guidelines as a part of the process. So to me, that's what a mature order set does.

Sarah Richardson: And a thoughtful system can identify patient demographic characteristics. I mean, talking about race, ethnicity, geography, insurance status, health related social needs. Why is this essential in managing populations that a health system serves?

Sherita Golden: We've been talking about the Joint Commission, advanced Inpatient Diabetes Certification, and the Joint Commission also recommends stratifying, patient clinical quality and safety data, by selected demographics. so to make sure that you're identifying who is, most at risk for poor outcomes in your system, and then how.

To, really, target your interventions to ensure that they reach their goal.

Another reason this is important is that the NCQA, which focuses on health plan quality, similarly requires data stratification by race and ethnicity, as well as, now [00:15:00] also including stratification recommendations by disability status and health related social needs.

So all of these types of, stratifications are necessary, to be documented by self-report in organization's, health IT system. And then there's also additional value to documenting patient's health related social needs because it enables a health system to say, okay, wow, we have. This percentage of our population that has food insecurity.

So we need to make sure that we have identified local resources to support them. And so that's the first level. And then the second level is, you know, we have, a community health needs assessment and to maintain our tax exempt status, you know, we need to invest. Resources in the community that support the needs of the community.

So again, your IT system can identify what are the needs of the patients in the community that you serve.

Sarah Richardson: Conditions like diabetes are disproportionately impactful to underserved populations. And as an African-American leader, how do you think about [00:16:00] bringing awareness? To these disparities while focusing on solutions.

Sherita Golden: So I think it's important, to point out that diabetes, um, doesn't just disproportionately impact racial and ethnic minoritized communities, but other communities. So, communities that live in poverty. regardless of race and then communities that live in rural areas. So, again, it's not only a concern for people of color and, colleagues of mine, and I highlighted this in a recent, national Academy of Medicine perspective in the fall.

And, you know, we use diabetes as an exemplar. So if you think about, areas of the country that have a higher than average, um, risk rate of diabetes, Appalachia. Is one area. Um, New Mexico is the one state, and then also, Mississippi. But interestingly, they all have very different racial.

Makeup and composition, but what they have in common is the poverty, which leads to lack of access to care, um, a, a difficulty in obtaining, healthy food [00:17:00] choices and difficulty with access to intentional physical activity. I think it's really important that we need to think more broadly about, innovative community-based interventions that reach those communities as well.

a couple of examples I think are really neat. There's a community paramedic. Program intervention that's really been shown to support diabetes, self-management and education, particularly in rural areas so they can get to patients that maybe can't come into a health system. And then it also, helps us to think more about how do we ensure that telemedicine and broadband access is really, um, reaching the rural communities that most needed.

and then like another. Population we don't often think about are the elders. So older adults where we're really thinking, you know, about how do we partner in their care with geriatricians, physical therapists, occupational therapists, social services, and long-term care facilities. So again, you know, by broadening the lens of, who are those who are disproportionately [00:18:00] impacted, then we can really think more broadly about the types of interventions that we need to design.

Sarah Richardson: I'm grateful for the work that you're doing because the. Aging population and the rural population are only going to be expanding over the next 20 plus years. And the ability to reach them where they are in a way that's meaningful to them

Sherita Golden: Mm-hmm.

Sarah Richardson: paramount, and that consumerism aspect of patient care and dealing with some of these more complex conditions, it just doesn't always get the attention that it really should to keep people healthy and safe.

Sherita Golden: No. Absolutely. And I think we also undervalue, you know, sort of, medical care extenders like community health workers and patient navigators who can actually meet. Those people and patients where they are to really ensure that the typical barriers to them getting care are addressed. and we know for all of those populations that I've mentioned, that community health worker types of interventions have been shown to improve clinical outcomes.

Sarah Richardson: I wanna jump to the fact that many communities have [00:19:00] had the lion's share of missed opportunities in healthcare. How do systems begin to rebuild trust and bring awareness in a way that's meaningful to the patient?

Sherita Golden: Yeah. So I think it's really, um, you know, important to, um, really engage the community in those conversations and ask them like kind of co-developing programs. So I think what we often will do in our system is we have an idea, this is what we think the community needs. But we didn't actually ask them.

Number one, is this what you need? And number two, if this is what you need, what is the best way to deliver that to you? So some sort of co-developed programs that have worked really well, particularly around diabetes are, there have been interventions where diabetes, self-management education has been incorporated into, a food bite.

And what that has resulted in is not only improvement in diabetes outcomes, but also it addresses the food insecurity, for example. So like those are examples where the partnership is very [00:20:00] beneficial and we did that really well during COVID, you know, we delivered, vaccines and, you know, church basements and on parking lots and in community centers.

So, you know, we really need to think about can we deliver chronic care? Management in the same way that we did. we have a model that works. It's a matter of figuring out how to extend and scale it.

Sarah Richardson: As we start to close out our. What role can clinicians, IT leaders and administrators each play in closing the gaps you've identified?

Sherita Golden: you know, we all have a role to play, so I think we have to really, find out what are the priorities of our community and make sure that we are addressing that. Don't assume what the needs are, and I think we need to actually share the data that we collect with our affected communities and then engage them in system design.

You know, like oftentimes they already know, but I think we need to look at our own data and then toengage them in system design. And I think that. Clinicians that interact directly with patients, we need to be the voice back to our executive leadership and our IT [00:21:00] leadership, to say these are the systems that we need in place in order to safely take care of our patients with diabetes or.

Any, chronic disease. So that's how I became very close friends with our, chief Medical Information Officer, Dr. Peter Green, who was a cardiac surgeon, turned, um, you know, it guru, but it was like, this is what, this is what the boots on the ground, this is what the physicians, the nurses, the advanced practice clinicians who care for patients with diabetes.

These are the challenges we are having, and that's how we started. The journey of developing IT solutions. and I think also these investments need to be sustained. And we talked about that earlier and again, you know, sometimes we have state grants. To implement great programs, we show that there's an improvement in outcomes, but often those grants are time limited and the grant goes away, and then our systems don't invest.

Um, you know, but, but we've shown it's actually gonna improve outcomes and it has a financial benefit and return on investment. So then our systems need to actually [00:22:00] invest in continuing those evidence-based practices in our own health system.

Sarah Richardson: Yeah, plan for the cost even after the grant funding may go away

Sherita Golden: Exactly. Particularly for those that are effective. Sometimes we find out, okay, we thought that would be effective, but it wasn't. But once we know it's effective, we should really plan to integrate it.

Sarah Richardson: which leads us to speed round. Are you ready?

Sherita Golden: I think so.

Sarah Richardson: what is a metric that every hospital leader should understand when it comes to inpatient diabetes care?

Sherita Golden: They should know how much risk litigation they've paid out in the last 10 years due to hypo or hyperglycemia. Mismanagement.

Sarah Richardson: Can you imagine the power of that being on the business case?

Sherita Golden: Yes. That's the business case that I used at our hospital to build our program.

Sarah Richardson: I've

Sherita Golden: Yes,

Sarah Richardson: that as a metric being brought forward before, and I love that you just said

Sherita Golden: yes, yes.

Sarah Richardson: What is one thing that technology gets blamed for but really isn't the problem?

Sherita Golden: Well, I guess blame for replacing a doctor's brain. Well, they should have known, and the reality is, no [00:23:00] matter how great our technology becomes, it's never going to replace the expertise of an astute clinician. we need to not expect it to do things that was not designed to do.

Sarah Richardson: Thank you for just illuminating the human in the loop component that we talk

Sherita Golden: Yes.

Sarah Richardson: often with all of the new technologies coming our way.

Sherita Golden: Yes,

Sarah Richardson: one, what is a word that describes what good impact? Patient diabetes care should feel like for patients?

Sherita Golden: it should feel safe.

Sarah Richardson: Yes, it should. Thank you for that. This conversation is such an important reminder that patient safety, technology, and equity are deeply connected, especially in high risk settings like inpatient diabetes care. When systems are designed thoughtfully, they don't just support clinicians, they protect patients and communities.

Thank you for sharing your expertise and vision with us today.

Sherita Golden: Thank you for having me.

Sarah Richardson: And to our listeners, whether you're a clinician, IT leader or an executive, this is a space where intentional design can save lives. Until next time, keep flourishing.

[00:24:00] that's flourish soundbites, find your community at this week, health.com/subscribe. Every healthcare leader needs a community to learn from and lean on. Share the wisdom.

That's all for now.