Issue: July 2015
July 16, 2015
14 min read

Hospitals work to improve inpatient diabetes management

Issue: July 2015
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Each year, millions of people with diabetes are hospitalized in the United States, according to the CDC. Although poorly controlled diabetes in the hospital has a direct effect on outcomes — patients develop more complications, endure longer hospital stays and incur higher costs — the disease is still routinely considered secondary to the primary reason for admission.

Perhaps that is because most people with diabetes are not being hospitalized due to a hyperglycemic crisis. “They’re being admitted for other acute medical illnesses like pneumonia, cellulitis; diabetes is sort of a bystander,” said Nestoras Mathioudakis, MD, an assistant professor of medicine and the associate director of the Inpatient Diabetes Management Program at Johns Hopkins Hospital.

Increasingly, U.S. hospitals are recognizing that diabetes is not merely a secondary diagnosis and that managing inpatient diabetes well is critical to delivering top-notch patient care.

“Good care in the hospital, beginning with education in the hospital and a better planned discharge can prevent the next readmission,” said Kenneth Cusi, MD, professor of medicine and chief of the division of endocrinology, diabetes and metabolism at the University of Florida in Gainesville.

Many hospitals are identifying shortcomings in their inpatient diabetes management programs and designing appropriate solutions to improve diabetes care. In this issue, Endocrine Today talks to experts about the challenges of managing diabetes in the hospital environment, as well as ongoing quality initiatives that may help improve care.

The timing for this shift in care priorities is good, as obesity rates and diabetes rates continue to skyrocket. According to the CDC, 29 million people in the United States have the disease. Another 86 million people have prediabetes, which may put them at increased risk for developing diabetes.

Challenge to health care

Diabetes management challenges health care on several levels. The first is identifying which hospital team is responsible for managing these patients. “Who owns diabetes control in the hospital?” Mathioudakis asked. “Is it the responsibility of the primary team? Is it the responsibility of the diabetes team or the endocrine service?”

Good inpatient diabetes care, beginning with education and ending with a well-planned discharge, can prevent readmissions, according to Endocrine Today Editorial Board member Kenneth Cusi, MD.

Source: John Jernigan

The care team varies from hospital to hospital, according to David W. Lam, MD, assistant professor of medicine, endocrinology, diabetes and bone disease at Mount Sinai Hospital in New York City. Most patients with diabetes are managed by the primary team, typically a hospitalist, Lam said.

Typically, endocrinologists are not responsible for the majority of inpatient diabetes care. “There are not enough endocrinologists out there to handle diabetes [in the hospital],” he said.

There also are numerous clinical challenges, the most significant of which is achieving stable glucose control. Patients are stressed, which increases their risk for hyperglycemia. They are not eating normally, which affects glucose levels.

“If you’re giving somebody insulin and they have a poor appetite, there’s the potential for hypoglycemia,” Mathioudakis said. “They may have altered kidney function, which affects how insulin is cleared from the system, so there’s increased risk of hypoglycemia there.”

New medications can cause problems. “A number of medications that we give to treat acute illnesses worsen glucose tolerance and make diabetes, more difficult to control” Cusi said.

Hyperglycemia common

Hyperglycemia is a common and costly problem in hospitalized patients. Bersoux and colleagues found that the prevalence of hyperglycemia was 32.3% in non-ICU patients and 28.2% in ICU patients.

“About one-third of patients in the hospital have sugars that are at least about 250 mg/dL,” Cusi said. The stress of being ill and hospitalized can cause stress hyperglycemia even in patients who do not have diabetes.

Elevated blood glucose has serious implications for patients. “Hyperglycemia leads to increased complications: decreased wound healing, increased risk of infections, etc,” said Priyathama Vellanki, MD, assistant professor in the division of endocrinology, metabolism and lipids at Emory University School of Medicine.


In the hospital, typically, hyperglycemia is treated with sliding scale insulin. “With that, you chase the blood sugar, essentially,” Vellanki said. “You wait for the blood sugar to be high and then you lower it.”

Despite its popularity in the hospital setting, sliding-scale insulin therapy “is a very poor way of managing diabetes,” Cusi said. Instead, the American Diabetes Association and Endocrine Society recommend the use of a basal-bolus treatment.

“But that has not broken the traditional culture of the sliding scale,” Cusi said.

But with time and the weight of good results, perhaps the basal-bolus approach will gain a footing. In their research, Vellanki’s group has compared sliding scale insulin to basal-bolus insulin therapy.

Two trials, RABBIT 2 and RABBIT 2 Surgery, showed that basal-bolus treatment with long-acting Lantus (insulin glargine, Sanofi Aventis) and short-acting Apidra (insulin glulisine, Sanofi Aventis) improved glycemic control.

“We showed that clearly the long-acting insulin and the short-acting insulin for meals is clearly superior at controlling blood sugars compared to a sliding-scale insulin regimen,” Vellanki said.

Now, the team is working to simplify the regimen, which currently involves at least four injections and four glucose level checks each day.

In recent years, tight control of hyperglycemia has led to hypoglycemia in some cases. Hypoglycemia, which affects 10% to 20% of patients treated with glucose-lowering medications in the hospital, has negative effects as well, according to Mathioudakis. It can lead to increased length of stay, increased complications and increased mortality.

Optimal blood glucose target

Although it is clear that glucose control is important, “the target, I think, has been more controversial,” Lam said. “In terms of what is the perfect blood glucose, that depends on the patient.”

Nestoras Mathioudakis

The NICE-SUGAR study results demonstrated that tight glycemic control in critically ill patients increased risk for mortality.

According to the American Association of Clinical Endocrinologists’ most recent diabetes guidelines, the glucose targets for ICU patients with hyperglycemia should be a range of 140 mg/dL to 180 mg/dL. For non-ICU patients, AACE recommends a pre-meal target of less than 140 mg/dL and a random blood glucose level of less than 180 mg/dL.

The ADA recommends the same target for ICU patients; however, its target for certain non-ICU patients is between 110 mg/dL and 140 mg/dL, according to Kellie Antinori-Lent, MSN, RN, ACNS-BC, BC-ADM, CDE, programmatic nurse specialistat UPMC Shadyside Hospital in Pittsburgh and spokeswoman for the American Association of Diabetes Educators.

In all, there are seven U.S. inpatient glucose management guidelines available. Mathioudakis and colleagues assessed these guidelines, which cover both the critical care and non-critical care settings, in a recent study, and they found that discrepancies in glucose targets and the frequency of blood glucose monitoring could “introduce confusion for health care providers,” Mathioudakis said. “It would be helpful to have an inpatient guideline that trumps or unifies all of these guidelines.”


Getting patients the proper nutrition at the proper time is another high hurdle for diabetes management. “As far as diet goes, one of the more challenging aspects of diabetes management in the hospital is people who are not eating regular meals,” Mathioudakis said.

“As soon as somebody comes in to the hospital and they have diabetes, they should be ordered a diabetic-friendly diet, or carb-controlled diet,” Mathioudakis said. “Nutrition education should be provided by a certified diabetes educator or dietician. There’s going to be nutrition education provided either by the nurse practitioner, physician assistant or dietitian. The hospital is a great time to intervene in the care of someone. You’ve got a captive audience stuck in a hospital bed.”

At Mount Sinai Hospital, besides the carb-controlled diet, registered dietitians follow hospitalized patients with diabetes to assess the quantity and quality of nutrition, according to Lam.


Quality improvement initiatives

There are several key strategies that are important to improving the quality of inpatient diabetes management. Foremost is having a glucose management system, according to Mathioudakis. Another critical component for quality improvement is securing a physician champion and getting buy-in from hospital leadership, he said.

Making clinical decision support tools available is important, Mathioudakis said. Residents at Johns Hopkins carry a small pocket guide that contains detailed algorithms on initiating insulin, adjusting insulin doses, transitioning patients from an insulin drip to injections, and more.

Ongoing training is critical. Johns Hopkins employs a “train the trainer” approach. “Our two dedicated diabetes nurse practitioners have monthly nursing rounds … where we train nurses on the units to be ‘diabetes super users,’” Mathioudakis said. These super users serve as diabetes point persons for their nursing units. “That’s been an effective strategy of reaching out [and] getting our message out through the hospital system.”

As part of its quality improvement efforts, in January 2014 The University of Florida and Shands Hospital, at Gainesville, Florida, established a Diabetes Care Task Force, which offers an integrated inpatient and outpatient diabetes care program and meets monthly to discuss issues and progress. The task force is focused on involving all hospital stakeholders in the effort to improve diabetes care, according to Cusi.

One of the task force’s first jobs was to review all of the inpatient insulin infusion protocols. After finding that that their old protocol required intense nursing staff involvement, they chose a simpler process.

“We chose a protocol that used a lower insulin infusion rate and that led to patient, nursing staff and house staff satisfaction that was extremely valuable,” Cusi said. “We have a much greater knowledge and awareness among nursing staff about diabetes and a willingness to identify patients who have knowledge deficits.”

Since its inception, the task force has been focused on educating the nurses, Cusi said. The primary tool for this effort is a diabetes resource nurse, who is responsible for educating the other nurses. There is one diabetes resource nurse for each floor of the hospital. House staff diabetes education is also important.

The task force has increased the number of inpatient diabetes consults from about 30 to 50 consults per month to 150 to 200 per month, which Cusi sees as a sign that diabetes is “slowly coming to the forefront.”

The hospital pharmacy has developed a tool that will help identify patients at greatest risk for hypoglycemia, according to Cusi. “We plan to notify the primary team that they have a patient who is at risk for hypoglycemia and [should] lead to a safer management of these patients,” he said.

Ultimately, the task force would like to become a flagship of concierge diabetes care, at least regionally, Cusi said.

Kellie Antinori-Lent

“This is the kind of effort that over time, not overnight, leads to success,” he said. “It’s creating a collective consciousness that this is a big problem.”

Discharge planning: early and often

Another critical point for improving inpatient diabetes care is discharge planning, which ideally, should begin upon admission, as soon as learning need is identified, Antinori-Lent said.

“Too often, diabetes education is left until the last minute — literally,” Antinori-Lent said. “I am willing to bet every inpatient diabetes educator you ask would tell you they have received a call to come and see a patient for diabetes education, only to find out that the patient is waiting to be discharged. The education may get done, but the benefit is questionable. Quality diabetes education cannot be performed on the fly.”

Even though the hospital is not the ideal setting in which to educate patients about their disease, patients must be prepared to safely care for themselves when they return home, according to Antinori-Lent. The focus should be on learning what is called “survival skills” which includes medication administration, blood glucose monitoring, meal planning and hypoglycemia identification and treatment. Further, this type of education has been successful. Magee and colleagues showed that patient knowledge and medication adherence was improved following a hospital-based survival skills diabetes education program. There was also an association with a reduction in ED and hospital admissions.


“Two key points about diabetes education in the hospital: first, begin early, on admission preferably and second, repeat it — repetition breeds confidence and helps master self-management skills, especially with insulin administration and blood glucose testing,” Antinori-Lent said. The more patients practice in the hospital, the better they will understand what they need to do, and the more likely they will be to it correctly at home. “Until the culture of diabetes education changes from a last-minute consideration to a ‘get it started upon admission’ standard of care, we will struggle with adequately preparing our patients for discharge, which negatively impacts patient lives and causes readmission,” Antinori-Lent said.

The literature supports the positive effect of inpatient diabetes education and of diabetes educators on outcomes. Healy and colleagues demonstrated that in patients with poor glycemic control, formal inpatient diabetes education resulted in a lower rate of hospital readmission within 30 days.

Mount Sinai has a rapid discharge clinic that helps patients get follow-up care soon after leaving the hospital. Patients are typically seen within a week of discharge, often by the same physicians or nurses who treated them in the hospital. “The nicest thing is the transition of care; our nurse practitioners who see the patients in our hospital, as well as our physicians, are the same people who are in the rapid discharge clinic,” Lam said. This continuity of care can identify and address any problems that may arise after the patient is discharged.

To further ease discharge, physicians at Mount Sinai have created a diabetes-specific discharge order set with short education components, Lam said.

Education about diabetes management should not be limited to patients. In their study, Vaidya and colleagues showed that a computer-based training program to teach house staff at Brigham and Women’s Hospital about inpatient diabetes care increased provider knowledge, potentially improving insulin administration. Desimone and colleagues also demonstrated that an education program aimed at health care providers could increase their knowledge of diabetes, in particular how to manage hyperglycemia.

Patients, too, can play an important role in improving quality. They need to know the signs and symptoms of hypoglycemia, and they must learn to be their own advocate. If they are not feeling well, they must alert their caregivers, Lam said.

David W. Lam

If they are well enough, patients must be actively involved in their care. They should ask questions, stay informed and take notes about information and personnel, Antinori-Lent said.

Quality improvement will continue to evolve. In the future, hospitals may turn to automated diabetes algorithms for glucose control, according to Lam.

Other technology, which is used often in outpatient settings, such as continuous glucose monitors and insulin pumps, may play a bigger role in the inpatient management of diabetes.

“In this day and age, I don’t think there’s a hospital in America that’s not doing quality improvement specific for diabetes, just because it affects so many patients, we know it affects long-term outcomes and, of course, it’s good for our patients,” Lam said. – by Colleen Owens

  • Anwar H, et al. Diabet Med. 2011;doi:10.1111/j.1464-5491.2011.03432.x.
  • Bersoux S, et al. Endocr Pract. 2014;doi:10.4158/EP13516.OR.
  • Desimone ME, et al. Endocr Pract. 2012;doi:10.4158/EP11277.OR.
  • Gibbons DC, et al. Diabet Med. 2014;doi:10.1111/dme.12444.
  • Handelsman Y, et al. Endocr Pract. 2015;doi:0.4158/EP15672.GL.
  • Healy SJ, et al. Diabetes Care. 2013;doi:10.2337/dc13-0108.
  • Magee MF, et al. Diabetes Educ. 2014;40:344-350.
  • Mathioudakis N, Golden SH. Curr Diab Rep. 2015;doi:10.1007/s11892-015-0583-8.
  • NICE-SUGAR Study Investigators. N Engl J Med. 2012;doi:10.1056/NEJMoa1204942.
  • Umpierrez GE, et al. Diabetes Care. 2007;30:2181-2186.
  • Umpierrez GE, et al. Diabetes Care. 2011;doi:10.2337/dc10-1407.
  • Umpierrez GE, et al. Diabetes Care. 2013;doi:10.2337/dc12-1988.
  • Vaidya A, et al. Diabetes Technol Ther. 2012;doi:10.1089/dia.2011.0258.
For more information:
  • Kellie Antinori-Lent, MSN, RN, ACNS-BC, BC-ADM, CDE, can be reached at 201 Preservation Hall, 5230 Centre Ave., Pittsburgh, PA 15232; email:
  • Kenneth Cusi, MD, can be reached at 1600 SW Archer Road, Room H-2, PO Box 100226, Gainesville, FL 32610; email:
  • David W. Lam, MD, can be reached at 5 E. 98th St., 3rd floor, New York, NY 10029; email:
  • Nestoras Mathioudakis, MD, can be reached at 601 N. Caroline St., Baltimore, MD 21287; email:
  • Priyathama Vellanki, MD, can be reached at 49 Jesse Hill Drive, SE, Room 498, FOB, Atlanta, GA 30303; email:

  • Antinori-Lent, Cusi, Lam, Mathioudakis and Vellanki report no relevant financial disclosures.


Who should provide diabetes education to patients newly diagnosed in the hospital: The hospital staff or the primary provider?


The hospital may provide a unique educational environment.

Alarmingly, less than 7% of patients in the United States with newly diagnosed diabetes receive any diabetes self-management education within the first year of diagnosis.

When diabetes is newly diagnosed in the hospital setting during an admission — whether related to diabetes or not — one should seize the opportunity presented by the hospital stay to initiate diabetes self-management education (DSME). The American Diabetes Association and The Joint Commission delineate educational content that should be communicated to all hospital patients with diabetes mellitus before discharge. Traditionally considered a suboptimal environment in which to provide education, the hospital may actually provide a unique opportunity to educate patients with diabetes mellitus. Recent studies suggest that providing inpatient DSME, improving communication of discharge instructions and involving patients more in prescription reconciliation may reduce risk for early readmissions and that inpatient DSME improves outcomes.

Michelle Magee

Hospitals face challenges in delivering DSME to all patients with diabetes as recommended in these national guidelines for best practices. Survival skills education, defined as teaching the topics essential for safe patient discharge, is recommended as an approach to inpatient DSME. Hospital-based endocrinologists/diabetes educators cannot reach all patients who need DSME and not all hospitals have access to inpatient endocrine consultants and/or diabetes educators. In those hospitals where a diabetes team exists, patients with a new diagnosis of diabetes would certainly most appropriately be served by these diabetes specialists for initiation of DSME, and most teams could accommodate provision of this service for this subgroup of diabetes patients.

In hospitals where such expertise is not available, the challenge is even greater, and creative solutions for delivery of DSME are required. Nursing already faces challenges in meeting competing demands for provision of services to patients at the bedside. Nonetheless, nurses in this circumstance are the best qualified to provide skills training when a patient will be discharged home with a new insulin start and/or will be using a blood glucose monitor after discharge, and nurses are qualified to teach the patient survival skills related to hypoglycemia if antihyperglycemic agents are prescribed prior to discharge. Allied health providers including nurse practitioners, physician assistants and pharmacists are also qualified to provide inpatient DSME at the bedside and may be engaged in a hospital-based strategy to care for patients with a new diabetes diagnosis. Rapid follow-up postdischarge with a diabetes center or with a diabetes educator will allow delivery of further survival skills DSME content.

Few, if any, clinical trials have utilized robust pre-implementation assessment methods and designs coupled with established implementation effectiveness evaluation frameworks to help inform and assess implementation practices in this realm. Such studies are needed.

Michelle Magee, MD, is director of the MedStar Diabetes Institute and is associate professor of medicine of Georgetown University School of Medicine in Washington, D.C. Disclosure: Magee has received prior research support from Sanofi for the conduct of an inpatient diabetes self-management education study.


Hospital and primary provider should work together.

As a human factors engineer, I design systems that support the work of people. The health care system is complex and includes components such as tools, technology, processes, environments and organizations that must be coordinated and designed to work optimally for providers to achieve the best patient outcomes. From my systems perspective, I believe that the best outcome for diabetes patients will result with systematically designed and purposefully coordinated education provided by both the hospital and the primary provider. Ideally, each would serve a different and complementary purpose. For less ideal circumstances, each provides a possible safety net to meet varying patient needs.


Vicki R. Lewis

Patients who are newly diagnosed in the hospital have unique learning needs, and hospitals have the opportunity to provide critical knowledge and survival skills prior to discharging patients. [Evidence has demonstrated] that it is possible to deliver diabetes self-management education in the emergency department with successful results. At the first follow-up visit, patients demonstrated a significant increase in diabetes knowledge test scores. In addition, significantly fewer patients required blood glucose meter instruction than had during the initial visit, and none required further insulin injection instruction. Education in the hospital provides the foundation for self-care and the opportunity to assist a patient in establishing primary care support. For those without a primary provider, these types of survival skills may reduce future hospital visits.

Of course, choosing to deliver education in the hospital is only the first step, and creating the system to support successful education delivery for improved outcomes is a challenge. Management support is needed for the development of established and clear learning objectives, acquisition of training supplies (for example, blood glucose meters), and training for those who will be providing the education. Financial resources are necessary to support an education program that will achieve the long-term objectives of improved patient outcomes and, ultimately, overall reduced cost of care.

Vicki R. Lewis, PhD, is a human factors engineer with Healthcare Safety Strategies LLC in Washington, D.C. Disclosure: Lewis reports no relevant financial disclosures. Reference: Lewis VR, et al. Diabetes Educator. 2015;doi:10.1177/0145721715577484.