UCSF Researchers See 39 Percent Reduction in High Glucose Levels
An innovative virtual glucose management service (vGMS) for hospitalized patients with diabetes is highly effective at maintaining appropriate glucose levels, according to a study led by researchers at UC San Francisco.
The system, which was developed by UCSF diabetes specialists, combines an automated daily report of all patients experiencing high or low glucose levels, with individual reviews by a trained diabetes specialist via the electronic medical record (EMR). These are integrated into each patient’s record, along with a program that customizes ideal insulin levels based on whether the patient has eaten, or other factors. Together, these enable providers who are not diabetes specialists to immediately assess the correct amount of insulin to provide.
Their study appears online March 28, 2017, in Annals of Internal Medicine.
“With this new system, there has been a 39 percent decrease in patients on the daily high-glucose hospital report,” said lead author Robert Rushakoff, MD, MS, professor of medicine at UCSF and medical director for inpatient diabetes at UCSF Medical Center. “The vGMS and similar inpatient-services leveraging technology may also become economically important for cost savings, as medicine moves toward bundled care.”
Diabetes (hyperglycemia) causes blood glucose (sugar) levels to rise higher than normal. More than 29 million Americans, or about 9.3 percent of the population, are living with diabetes, and an estimated 86 million are living with prediabetes, a serious health condition that increases the risk of diabetes, according to the U.S. Centers for Disease Control and Prevention.
About a third of hospitalized patients have diabetes, which is associated with increased risks for complications, length of stay, costs of care and mortality. Blood glucose levels in inpatients often exceed recommended targets due to low priority given to glycemic control, suboptimal regimens chosen for simplicity over effectiveness and difficulty titrating insulin in acutely ill patients. Diabetes treatment also can cause hypoglycemia, or low blood sugar levels, which is linked to mortality, increased likelihood of readmission and longer hospital stays.
Diabetes specialists and inpatient diabetes teams performing formal consultations can reduce length of stay and improve glycemic control and clinical outcomes, but these are time and resource intensive.
vGMS Simplifies Diabetes Management Process
In the Annals of Internal Medicine study, Rushakoff and his colleagues analyzed 12,535 non-obstetric adult patients at three UCSF Medical Center hospitals who had two or more elevated or lowered glucose values in the previous 24 hours, as identified from a daily glucose report. The analysis occurred during three 12-month periods from June 2012 to May 2015: pre-vGMS (June 2012 to May 2013), transition (June 2013 to May 2014) and vGMS (June 2014 to May 2015).
Rather than rely on manual contact of clinical teams individually, the research team of diabetes specialists developed an electronic method to convey recommendations via a new glucose management note. The automated glucose reports, clinician review and clinical notes combined to create the vGMS.
The UCSF vGMS team consisted of three providers, all with several years of inpatient diabetes experience: a board-certified endocrinologist (MD), a nurse educator (DNP and certified diabetes educator) and a pharmacist diabetes educator (PharmD and certified diabetes educator).
The vGMS system generates an automated daily report by 5:30 each morning of all inpatients with uncontrolled blood glucose. The report is then reviewed remotely by a diabetes specialist, along with an insulin/glucose chart, via the EMR. Together, these elements contain sufficient information about the patient’s insulin regimen to make appropriate recommendations without interviewing the patient.
The vGMS template includes drop-down boxes with specific recommendations based on whether the patient was eating or receiving glucocorticoids or enteral feedings, followed by suggested insulin doses. The note, which is available for clinician review by 6:30 a.m. daily, also includes a disclaimer that it is solely a recommendation, and advises the treating team to take into account the patient’s current clinical condition and obtain a formal endocrinology consultation if necessary.
As a result, the proportion of hyperglycemic patients decreased 39 percent, from 6.6 per 100 patient-days in the pre-vGMS period to 4.0 in the vGMS period. The hypoglycemic proportion in the vGMS period was 36 percent lower than in the pre-vGMS period. Overall, 40 severe hypoglycemic events occurred during the pre-vGMS period and only 15 during the vGMS period.
The UCSF Medical Center vGMS continues to operate now without interruption.
“The whole process takes about 45 minutes,” Rushakoff said. “If you had a formal consult service to intervene on these patients, it would take all day to see everyone and then talk to the teams. The note also reinforces the training of resident physicians just at the time it is needed, and improvements then can be made for that patient and future patients.”
Inpatient Diabetes Monitoring Past and Future
The vGMS is the latest effort to effectively monitor inpatient diabetes by Rushakoff and his colleagues, which began in the 1990s with online, interactive training for nurses and physicians, and continued into the early 2000s as they standardized the approach and reduced errors. The launch of UCSF’s electronic health record system in 2012 enabled complex paper orders to be translated into online orders, simplifying the process and eventually leading to the vGMS.
Besides vGMS, Rushakoff is working with UCSF Medical Center colleagues to launch a prediction model to decrease patient hypoglycemia risk via automated text message to the clinical team, a self-titrating insulin algorithm that will accurately deliver the proper insulin dose to patients on enteral feedings and a general inpatient diabetes brochure for incoming patients.
The researchers noted that UCSF is an academic medical center with residents and attending physicians rotating regularly, so with more stable medical staff, the effect may be greatest initially and decrease over time. Information also was not collected on patients’ concurrent illnesses and treatment or physicians’ responses to the vGMS notes, and the vGMS is possible only with a well-established infrastructure and staff.
Other UCSF contributors to the Annals of Internal Medicine study were Michael Kohn, MD, MPP, associate professor of medicine; Mary Sullivan, DNP, professor of nursing; Heidemarie Windham MacMaster, PharmD, pharmacist; Alvin Rajkomar, MD, assistant professor of medicine; and David Glidden, PhD, professor of medicine. Arti Shah, MD, of UCLA and formerly UCSF, also contributed. Funding was provided by National Institutes of Health training grant 5T32DK007418-34, the Wilsey Family Foundation and UCSF Clinical and Translational Science Institute (CTSI) grant UL1 TR000004.
UC San Francisco (UCSF) is a leading university dedicated to promoting health worldwide through advanced biomedical research, graduate-level education in the life sciences and health professions, and excellence in patient care. It includes top-ranked graduate schools of dentistry, medicine, nursing and pharmacy; a graduate division with nationally renowned programs in basic, biomedical, translational and population sciences; and a preeminent biomedical research enterprise. It also includes UCSF Health, which comprises three top-ranked hospitals, UCSF Medical Center and UCSF Benioff Children’s Hospitals in San Francisco and Oakland, and other partner and affiliated hospitals and healthcare providers throughout the Bay Area.