In-Hospital Glucose Management Program Gives Dramatic Savings

Miriam E. Tucker

May 05, 2021

Initiatives targeting hypoglycemia and insulin pen wastage could lead to dramatic cost savings in small community hospitals, new data suggest.

The two projects are part of a dedicated inpatient glucose management service led by Mihail ("Misha") Zilbermint, MD, one of the few full-time endocrine hospitalists in the United States and one of even fewer who work at a small community hospital.

In 2019, Zilbermint and colleagues reported that their inpatient glucose management program resulted in a 27% reduction in length of stay and a 10.7% lower 30-day readmission rate. The projected cost savings for the period January 2016 to May 2017 was $953,578.

Zilbermint's team has written two new articles that document cost savings for specific elements of the program, namely, a set of hospital-wide hypoglycemia prevention measures, and an initiative that reduced duplicate inpatient insulin pen dispensing.

About 1 in 4 people in US hospitals have diabetes or hyperglycemia. Large academic hospitals have endocrine divisions and training programs, but 85% of people receive care at small community hospitals.

"There are management guidelines, but they're not always followed.... That's why I've been advocating for endocrine hospitalists to be deployed nationally," Zilbermint told Medscape Medical News. He is chief and director of endocrinology, diabetes, and metabolism at Johns Hopkins Community Physicians at Suburban Hospital, Bethesda, Maryland.

Asked to comment on behalf of the Society of Hospital Medicine (SHM), Greg Maynard, MD, program lead for SHM's Electronic Quality Improvement Programs, told Medscape Medical News that Suburban's overall program goals align with those of the SHM.

"Dedicated inpatient glycemic control teams are very important and desirable to improve the quality and safety of care for inpatients with hyperglycemia and diabetes," he said.

Regarding specific initiatives, such as those aimed at reducing hypoglycemia and insulin pen wastage, Maynard said, "All of these are feasible in a wide variety of institutions. The main barrier is getting the institutional support for people to work on these interventions. This series of studies can help spread the word about the positive return on investment."

Another barrier ― the current lack of publicly reported measures or pay-for-performance programs for hypoglycemia prevention and glycemic control ― may soon change, added Maynard, who is also chief quality officer at the University of California, Davis, Medical Center.

"The National Quality Forum has endorsed new measures, and the CDC's [Centers for Disease Control and Prevention's] National Healthcare Safety Network is working on ways to augment those measures and embed them into their infrastructure," he said.

Although SHM doesn't specifically endorse full-time glycemic control hospitalists over endocrinology-trained glycemic control experts, "certainly hospitalists who accrue added training are very well positioned to be an important part of these interdisciplinary teams," Maynard said.

"The Nurses Were So Afraid of Hypoglycemia"

Tackling hypoglycemia was Zilbermint's first priority when he started the glycemic management program at Suburban in late 2015.

"One of the most common complaints from the nurses was that a lot of their patients had hypoglycemia, especially in the ICU [intensive care unit] when patients were placed on insulin infusion protocols.... Every time, the nurse would have to call the attending and ask what to do," he explains.

In addition, Zilbermint says, there was no standard for treating hypoglycemia. A nurse in one unit would give two cups of juice, another [a 50% dextrose] infusion, or another, milk. Even more concerning, "the nurses were so afraid of hypoglycemia they would reflexively discontinue all insulin, including basal."

So one of the new initiatives, led by Carter Shelton, MSHCM, an administrative fellow at the Medical University of South Carolina, Charleston, South Carolina, was to implement a set of hospital-wide hypoglycemia prevention measures, as described in an article published online April 21 in the Journal of Diabetes Science and Technology.

Inpatient Hypoglycemia Rate Was Cut Nearly in Half

This began in 2016, when the multidisciplinary Suburban Hospital Glucose Steering Committee identified four main causes of insulin-induced hypoglycemia (defined as a blood glucose level of ≤70 mg/dL in a patient who had received at least one dose of insulin in the past 24 hours) and devised solutions for each:

1. Lack of a unified hypoglycemia protocol. A formal, evidence-based, nurse-driven treatment protocol with clinical decision support in the electronic medical record (EMR) was developed. The Suburban team adapted much of the protocol from one that had been recently implemented at the flagship Johns Hopkins Hospital, in Baltimore, Maryland.

Arrording to that protocol, if patients are able to swallow, they are given 15 g or 30 g of carbohydrates in order to achieve a blood glucose level of 50 to 70 mg/dL and <50 mg/dL, respectively. Levels are checked 15 minutes later. Intravenous D50 or glucagon is reserved for patients who can't swallow.

2. For patients in critical care, the insulin infusion protocol that had been in use set blood glucose targets of 80 to 110 mg/dL, which resulted in hypoglycemia in nearly every patient who received an insulin infusion. This protocol was changed to the currently recommended 140 to 180 mg/dL.

3. Most patients were managed with sliding-scale insulin, an outdated yet still widely used regimen whereby insulin is given based only on current blood glucose without accounting for carbohydrates consumed with meals and not corrected until the subsequent meal. This was changed so that nurses give insulin after the patient has consumed at least 50% of their meal carbohydrates.

4. Lack of hypoglycemia reporting. A glucometrics dashboard ― now used throughout the Johns Hopkins system ― was adopted to produce daily hypoglycemia reports in the EMR system that could be reviewed by the inpatient glucose management service to track quality metrics and plan further interventions.

Between January 1, 2016, and September 30, 2019, out of a total 49,315 patient-days, there were 2682 days on which any hypoglycemia occurred and 874 days on which moderate hypoglycemia occurred (≤54 mg/dL). Type 2 diabetes accounted for 84.4% of the total patient-days; type 1 accounted for 4.4%.

The overall frequency of any hypoglycemia patient-days per month decreased from 7.5% to 3.9% during the study period (P = .001). This was significant for the patients with type 2 diabetes (7.4% to 3.8%; P < .0001) but not for those with type 1 diabetes (18.5% to 18.0%; P = .08).

Rates of moderate hypoglycemia also decreased significantly among the patients with type 2 diabetes (1.9% to 1.0%; P = .03) but not for those with type 1 diabetes (7.4% to 6.0%; P = .14).

On the basis of these rates in reducing hypoglycemia, in which the inpatient hypoglycemia rate was cut nearly in half, the estimated savings in cost of care to the hospital was $98,635 during the period January 2016 to September 2019.

Reducing Insulin Pen Waste by Minimizing Duplicate Prescriptions

Suburban Hospital had been using insulin vials and syringes when Zilbermint first arrived there. He lobbied the administration to allow use of pens, because they're easier to use and they reduce the risk for needlestick injuries. Nurses were educated and retrained monthly in their use.

The switch to pens ― aspart (Novolog Flexpen) for bolus insulin and glargine (Lantus SoloSTAR) ― took place in 2018. The cost of the aspart pen was $16.19, and the cost of glargine was $25.08. Each holds 300 units of insulin.

After the first month, the team noticed a large increase in expenses. A quality improvement project was devised to address the issue.

"We were dispensing sometimes three or four pens per person. That's a lot. Each pen holds 300 units, so one pen should last the entire hospital stay of an average 4- or 5-day stay," Zilbermint explained. "We had to figure out where we were bleeding the money and where the pens were going."

When pens disappeared, the pharmacy would have to dispense new ones. One problem was that when patients were transferred from one unit to another, the pen would be left behind and the room would be cleaned. Sometimes the pens weren't stored properly or were misplaced. Often, they'd end up in a nurse's pocket.

The second intervention was led by Urooj Najmi, MD, of the American International School of Medicine, Atlanta, Georgia. A program was instituted to reduce duplicate inpatient insulin pen dispensing, as detailed in an article published in the same issue of the Journal of Diabetes Science and Technology.

Solutions to reduce duplicate pen dispensing included having pharmacy track daily insulin pen reports and monitor duplicate orders, with "do not dispense" instructions conveyed via the EMR system. All multidose medications, including insulin pens, were to be placed in patients' bins at the nursing station, and nurses were instructed to look for patients' insulin pens prior to their being transferred to another unit, rather than ask for a replacement pen.

From July 2018 to July 2019, 3121 patients received insulin, of whom 95% received aspart and 47% received glargine. Of the 9516 pens dispensed, 68% were for aspart and 32% were for glargine. During the study period, the number of pens dispensed per patient dropped from 2.2 to 1.2 for aspart and from 2.1 to 1.3 for glargine; differences were highly significant (P = .0002 and P = .0005, respectively).

The total amount of unnecessary dispensing during the first 4 months after initiating the pen implementation program was 58%. The average monthly cost was $11,820.68; the projected cost per year was $141,848.

Six months after the waste reduction strategies were implemented, monthly waste had dropped to 42%, translating to an estimated potential cost savings of $66,261 over 12 months.

Because Suburban Hospital doesn't have an outpatient dispensing license, there is still wastage when patients are discharged, because they can't take their pens home with them. That remains a challenge, Zilbermint noted.

The team is working on implementing automatic A1c testing for patients admitted with hyperglycemia who either have a history of diabetes or whose blood glucose level is >140 mg/dL. Zilbermint said, "it's in the guidelines, but it's not always done."

Zilbermint is a consultant for Guidepoint. Maynard, Shelton, and Najmi have disclosed no relevant financial relationships.

J Diabetes Sci Technol. Published online April 21, 2021. Shelton et al, Abstract; Najmi et al, Abstract

Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape, with other work appearing in the Washington Post, NPR's Shots blog, and Diabetes Forecast magazine. She is on Twitter @MiriamETucker.

For more diabetes and endocrinology news, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.