Inpatient Diabetes Management: Preface: Inpatient Diabetes Management: Should We Make It Simpler?

Inpatient diabetes management is challenging for many hospitals and practitioners. Some large academic centers around the United States have specialized diabetes teams, diabetes technology such as continuous glucose monitoring (CGM) and telehealth consultations for inpatient use, and even endocrine hospitalists available to tackle the challenge of inpatient hyperglycemia (1–8). However, most people with diabetes in the United States receive their inpatient care at small community hospitals, where robust (and complex) guidelines and inpatient glycemic protocols may not have significantly changed the culture of glycemic management (9). Some hospital-based practitioners (i.e., hospitalists) practicing within an antiquated culture may rely solely on “sliding-scale” rapid-acting insulin to “correct” high glucose levels, and nurses are desensitized to hyperglycemic events, with a fasting glucose level of, say, 112 mg/dL sometimes being labeled as “low” and prompting a phone call to the provider asking to “hold all insulins.”

The transition of care for people with diabetes from the hospital to home or a skilled nursing facility proves to be even more challenging than inpatient glycemic management and may include the task of medication reconciliation, the burden of obtaining prior authorizations, and patients’ differing insulin requirements in the hospital versus at home, among other issues. I frequently teach endocrinology fellows that at least some antidiabetic medications and supplies listed in the electronic health record are incorrect until proven otherwise.

Additionally, I have discovered that some hospitalists are reluctant to initiate newer medications such as sodium–glucose cotransporter 2 inhibitors and glucagon-like peptide 1 receptor agonists at the time of discharge. This reluctance is not because the practitioners do not know about the glycemic, renal, and cardiovascular benefits of agents in these drug classes; rather, it is because they fear that these medications will not be covered by the patient’s insurance plan or that prescribing one would generate a prior authorization request that would require extra effort on their part to secure approval.

Furthermore, hospital administrations are slow to adopt the latest diabetes-related technology for inpatient use because they may be concerned about additional costs, lack of insurance reimbursement, and unclear benefits.

So why is glycemic management still such a challenge? Are available practice guidelines too cumbersome and complex? Is the implementation of subcutaneous basal-bolus insulin therapy too difficult? How can we simplify inpatient protocols and management processes? Why is the discharge process for people with diabetes so painful for hospitalists? In this Diabetes Spectrum From Research to Practice section, we explore these and other questions and offer strategies for tackling these challenges together.

It is important to remember why glycemic management matters; it has been shown to improve patient outcomes, decrease mortality, decrease hospital lengths of stay and readmission rates, and reduce costs, which is especially important in the United States, which has one of the most expensive health care systems in the world (3). In this special article collection, leading experts in inpatient diabetes care share their “whys” as well as many valuable “hows.”

One problem, at present, is that many hospitals don’t know what they don’t know. However, in August 2021, the Centers for Medicare & Medicaid Services (CMS) sought to address this knowledge gap by implementing electronic clinical quality measures (eCQMs) as part of its Hospital Inpatient Quality Reporting Program. Certain measures of hypoglycemia and hyperglycemia will be now reported to the CMS. In the first article in our collection (p. 391), Sara Atiq Khan and I help to decipher these new metrics. We speculate that eCQMs will bring greater attention to the need for guideline-based glycemic management in the hospital. Although it is unclear exactly when, once these metrics become pay-for-performance measures, hospitals will likely be forced to develop dashboards to track them and make efforts to reduce their rates of inpatient dysglycemia. Dr. Khan and I also outline the basic steps hospitals must take to ensure that they follow the new CMS rules.

As expected, the use of diabetes technology has rapidly expanded in recent years (10). Many people with diabetes and their care providers are embracing the advanced features and improved glycemic outcomes possible through the use of diabetes devices such as CGM systems and advanced insulin pumps. Patients often now bring their devices to the hospital with them, which has increased the pressure on nursing, medical, compliance, and legal departments to develop and implement policies outlining safety measures and protocols for the safe use of these devices in the inpatient setting. In our second article (p. 398), Jillian Pattison et al. discuss best-practice guidelines for the continuation of personal diabetes technology use in the hospital and outline the specific roles of each diabetes care team member. They recommend screening patients on admission and involving a diabetes expert in the care of those who use diabetes technology. The authors provide additional recommendations about the use of automated insulin delivery systems, which connect an insulin pump and a CGM system with an algorithm to automatically adjust basal insulin delivery, and advise medical teams to prepare alternative diabetes management plans in case these systems become inappropriate for certain hospitalized patients.

In April 2020, in response to the coronavirus disease 2019 pandemic, the U.S. Food and Drug Administration expanded the availability and approved use of noninvasive patient monitoring devices. Several medical centers around the world have been studying CGM use in the hospital setting (11). Glucose telemetry systems are a promising alternative to periodic point-of-care blood glucose monitoring (BGM) with a traditional glucose meter, with many important advantages, including automatic measurements at 5-minute intervals, transmission of estimated glucose values to display devices, and even programmable alerts for impending dysglycemia. In their article (p. 405), Rebecca Rick Longo and Renu Joshi describe their successes and challenges with CGM use in the hospital and discuss the future of this modality for inpatient use. They showcase how robust CGM protocols may provide an opportunity for improvement in inpatient glycemic management. While reviewing current inpatient CGM guidelines, they point out that this technology should be used with select patients and must be maintained appropriately (including, in some cases, continuation of some BGM checks that may be required to document that the CGM system meets acceptable correlation criteria).

Our next article, by Samaneh Dowlatshahi et al. (p. 420), provides several clinical cases and some outside-the-box strategies for managing hyperglycemia in the noncritical care inpatient setting. The authors challenge current guidelines that all inpatients with hyperglycemia should be managed with basal-bolus insulin therapy (i.e., scheduled long-acting basal insulin, scheduled short-acting insulin with meals, and an insulin correctional dosing scale as needed). They describe their institutions’ experience with basal insulin only. Interestingly, practitioners have found improved glycemic outcomes when using this approach compared with basal-bolus insulin therapy (12). Additionally, the authors share an example of one hospitalized patient who was receiving daily steroids and whose glycemia was well managed with just NPH insulin and a dipeptidyl peptidase 4 inhibitor. They argue that some available noninsulin agents may effectively reduce mean postprandial glucose in patients with steroid-induced hyperglycemia without the risk of hypoglycemia and thus may decrease excessive glycemic excursions. Could it be possible that simpler is better?

Hyperglycemia has been noted to occur in nearly half of hospitalized patients who are receiving enteral or parenteral nutrition regardless of their diabetes status (13,14). Preethi Polavarapu et al., in the fifth article in this series (p. 427), describe the strategies they use at the University of Nebraska Medical Center to manage glycemia in hospitalized patients receiving nutrition support. They discuss targeted insulin therapy that matches glycemic profiles of the modes of enteral nutrition delivery. For example, they recommend using intravenous insulin infusion for hemodynamically unstable patients; for hemodynamically stable patients, they suggest adding 80% of the total daily insulin dose as regular insulin in the bag with total parenteral nutrition. Alternatively, practitioners could calculate the dose of regular insulin for in-bag use based on an insulin-to-dextrose ratio of 1:20 for patients without diabetes and 1:10 to 1:15 for those with diabetes. These authors also share their strategies for hypoglycemia prevention (e.g., nurse-driven initiation of D10% infusion when artificial nutrition is interrupted).

When is it time to discharge a patient with type 2 diabetes? In our final article (p. 440), Andrew P. Demidowich et al. argue that the lasting impact of inpatient diabetes management is achieved at the time of discharge and encourage hospitalists and other health care providers to take the time to reconcile antidiabetic medications and develop easy-to-understand, successful discharge plans. The authors share a guiding motto: KISD—Keep It Simple on Discharge. They challenge diabetes care providers to learn a lesson from behavioral economics: that the more difficult a task is to accomplish, the less likely a person is to do it consistently over time. Their article outlines a general algorithm for selecting discharge regimens for people with type 2 diabetes that maximize glycemic, renal, and cardiovascular benefits and minimize barriers to performing self-care. The authors speculate that diabetes care providers should reduce the number of “pricks and sticks” (i.e., insulin doses and fingersticks for BGM) and encourage case managers to ensure that each patient’s regimen is affordable.

I am hopeful that readers will enjoy this special section about effective inpatient diabetes management. It is possible that your institution may have different protocols and guidelines from those described herein. If so, I encourage you to not fear change and to consider trying these new strategies that have potential to improve patient care and outcomes.