Virtual Inpatient Diabetes Management Tools Minimize COVID-19 Risk for Staff and Improve Patient Outcomes

Minimizing interactions with patients infected with COVID-19 is challenging when glycemic management is required. Automated virtual tools for inpatient diabetes management in use at UC San Francisco have helped the medical staff deliver safe and effective patient care while reducing their exposure to SARS-CoV-2.

Robert J. Rushakoff, MD, MS, is a professor of medicine and medical director for inpatient diabetes care at UCSF. He and his team developed and implemented a virtual glucose management service (vGMS) and an automated, self-adjusting subcutaneous insulin algorithm (SQIA). UCSF was already using these innovative tools to manage diabetes in hospitalized patients before the COVID-19 pandemic began. 

“The outcomes for patients with COVID-19 and diabetes have been very positive at UCSF,” Rushakoff said. “The mortality rate for these patients is not significantly different than for those without diabetes.”

The vGMS increases efficiency and improves patient outcomes 

Each morning, the vGMS generates an automated report of all inpatients with uncontrolled blood glucose, which is integrated into the electronic medical record (EMR). A diabetes specialist reviews this report remotely, along with the insulin-glucose chart, and enters insulin dosing recommendations into each patient’s EMR. These recommendations are available for clinician review by 6:30 a.m. daily. 

Since implementing the vGMS in 2013, UCSF has seen a 45% decrease in the number of inpatients on the daily high-glucose report and rates of hypoglycemia have been consistently low. 

“In the past five years, the patient census at UCSF has significantly increased,” Rushakoff said. “But with the vGMS in place, the time to monitor glucose levels has not increased. This system is fast.”

Because patients’ glucose levels are monitored constantly, any problems that arise can be observed and addressed quickly. For example, when increased numbers of surgical patients with diabetes appeared on the daily high-glucose report, the specialists monitoring the vGMS reviewed patient records to determine likely causes. Some patients had received preoperative carbohydrate beverages and others had been prescribed high-dose dexamethasone for nausea or pain management. Rushakoff and the team recommended that these practices be limited for patients with diabetes. Protocols were changed, and high numbers of surgical patients with diabetes no longer appeared on the daily report.

“Through the vGMS, we saw these issues within days and were able to manage them quickly,” Rushakoff said.

Self-adjusting algorithm automates inpatient glucose maintenance

The automated, self-adjusting SQIA is designed for adult patients who are nil per os (NPO), on continuous enteral tube feedings or on total parenteral nutrition (TPN). The SQIA is integrated into the EMR.

A calculator that requires input of the patient’s current glucose level is embedded into the electronic medical administration record (eMAR). Once the glucose level is entered, the SQIA titrates insulin doses based on the patient’s metabolic requirements.

Previously, new orders had to be written to change a patient’s insulin dose. “It could take days of having to rewrite orders over and over again to get the right amount of insulin,” Rushakoff said. “The SQIA automates this process.” To date, this algorithm has been used in more than 2,000 inpatients at UCSF to efficiently maintain target glucose levels.

Maximizing patient safety with insulin-pen barcodes

To further ensure patient safety, the SQIA has been integrated into UCSF’s patient-specific barcode labeling system for insulin pens. Developed by Rushakoff and his team in 2017, this system is designed to prevent wrong-patient insulin pen errors and is built into the EMR.

Patients and their insulin pens are given unique barcode labels, which are scanned before insulin administration to ensure a correct match. With the SQIA programmed into this system, when the barcode is scanned, a notification appears requiring the nurse to confirm whether feeding has been stopped for that patient. The algorithm then calculates the appropriate insulin dose based on input from the nurse.

Innovative tools for safe and efficient inpatient glycemic management 

Approximately one-third of hospitalized patients require glycemic management. Virtual and automated tools like the vGMS and SQIA contribute to the safe and efficient care of inpatients with diabetes and have been especially valuable during the COVID-19 pandemic. These tools enable constant virtual glucose monitoring, identify issues quickly and minimize the need to rewrite orders. In addition, remote glucose management teams using such tools can help fill gaps in care created by staff shortages.

“The vGMS, SQIA and similar inpatient services that leverage technology may also become economically important for cost savings, as medicine moves toward bundled care,” Rushakoff said.

UCSF Medical Center is ranked No. 6 in the nation for diabetes and endocrinology by U.S. News & World Report’s 2021-2022 Best Hospitals survey.

To learn more

UCSF Diabetes Clinic at Mount Zion

Phone: (415) 885-3868 | Fax: (415) 885-7724

Refer a patient

Clinical trials