'Virtual' Management Improves In-Hospital Glucose Control

Miriam E Tucker

March 27, 2017

A "virtual" glucose management service may improve glycemic control among hospitalized adults with diabetes who develop hyperglycemia, new research suggests.

The findings were published online March 27 in Annals of Internal Medicine by Robert J Rushakoff, MD, of the division of endocrinology and metabolism, University of California, San Francisco (UCSF), and colleagues.

When UCSF introduced a new electronic medical record (EMR) in 2012, the authors built into it features that automatically detect and report uncontrolled blood glucose levels among nonobstetric medical and surgical inpatients.

A diabetes specialist reads the report remotely and, using information in the patients' charts, makes necessary recommendations for insulin dosing via an electronic note in the chart.

Use of this "virtual glucose management services (vGMS)" — comprising the automated reports, clinician review, and clinical notes — was associated with a 39% decrease in the daily number of inpatients with two or more glucose values of 12.5 mmol/L or greater (≥225 mg/dL), along with a 36% reduction in hypoglycemic readings in the year following implementation.

"An inpatient vGMS is a potentially scalable model that harnesses automated glucose screening and expedited clinical review to enhance the management of patients with diabetes," Dr Rushakoff and colleagues write, noting that between 30% and 40% of all hospitalized patients have diabetes or hyperglycemia.

Commendable Work but Opportunities Missed

In an accompanying editorial, Gerry Rayman, MD, of Ipswich Hospital National Health Service Trust Suffolk, United Kingdom, says: "Rushakoff and colleagues report impressive improvements in glycemic control with their use of a novel approach [that is] commendable."

However, Dr Rayman also cautioned, "The study…should be reviewed in the context of inpatient diabetes care practiced elsewhere. Unlike in the US, uptake of electronic health records and electronic prescribing is limited in many countries," including even in England and Wales, where just 17% of hospital trusts have both of those.

Moreover, Dr Rayman pointed out that important information such as consumption and timing of food isn't included in the EMR, and not all countries use a basal-bolus insulin approach to treat hyperglycemia in the hospital, the recommended practice in the United States.

Opportunities for patient and staff education are also missed using this approach, and another downside of the vGMS is the absence of patient and specialist interaction — as such, the system completely leaves the patient out of his or her own care, he notes.

Instead, "a combination of a vGMS with selected bedside care by diabetes specialists may lead to even better control. If such care resulted in shorter stays or fewer readmissions, it could even be cost saving."

Three Time Periods Compared; 30% of Patients Were Glucose Monitored

In his editorial, Dr Rayman notes that "more than 90% of patients with diabetes are admitted [to the hospital] for reasons unrelated to the disease and may be cared for by staff without specific diabetes expertise."

Indeed, in this 36-month cross-sectional study by Dr Rushakoff and colleagues, a high percentage of patients with elevated glucose levels were initially in the oncology and transplant services.

For their research, the authors compared the year before, the transition year, and the year following implementation of the vGMS.

Of the 68,505 total adult nonobstetric medical, surgical, and intensive-care hospitalizations at three UCSF campuses, just under a third, 19,294, were glucose-monitored, representing 12,535 individual patients (some had multiple hospitalizations).

Starting in June 2012, five- or six-times daily point-of-care glucose monitoring results (depending on whether the patient was eating or not) were uploaded into the system.

In October 2012, a report was added to document when patients had hyperglycemia — two or more glucose values of 2.5 mmol/L or greater (≥225 mg/dL); hypoglycemia — glucose level less than 3.9 mmol/L (<70 mg/dL); or used an insulin pump, all in the previous 24 hours.

The vGMS team consisted of a board-certified endocrinologist, a diabetes nurse educator, and a pharmacist diabetes educator, all with more than a decade of experience in diabetes management.

The percentage of hospitalizations with at least one vGMS note increased from 3.9% during the transition period to 4.8% during the vGMS period (P < .001).

Implementation of the vGMS was associated with significant decreases in hyperglycemia and hypoglycemia.

Patient-Day Mean Glucose Levels During the Three Study Periods and Proportion of Patient-Days With Hyperglycemia, at-Goal Glucose Levels, and Hypoglycemia

Variable Pre-vGMS Transition vGMS Risk ratio (vGMS vs pre- vGMS) Difference (vGMS vs pre- vGMS) P
Patient-day mean glucose, mmol/L (mg/dL) 9.48 (170.7) 9.08 (163.4) 9.24 (166.4)   -0.24 (-4.3) <0.001
Proportion per day per 100 hospitalized patients
Hyperglycemia 6.6 5.4 4.0 0.61 -2.5 <0.001
At goal 10.8 11.6 11.4 1.05 0.6 <0.001
Hypoglycemia 0.78 0.89 0.49 0.64 -0.28 <0.001
Severe hypoglycemia 0.032 0.028 0.010 0.31 -0.022 <0.001

Virtual Consultations May Result in Cost Savings

Since implementing the vGMS, Dr Rushakoff and colleagues note that the improvements in glycemic control with the vGMS have been sustained and the number of patients with hyperglycemia has decreased on each day following admission.

These changes haven't been due to an increase in formal on-site endocrinology consultations, which didn't vary across the study periods.

The authors echo the editorialist about the potential cost advantage, noting that "the vGMS and similar inpatient services may become economically important for cost savings as medicine moves toward bundled care. In a bundled-care payment model without additional payment for time-consuming in-person consultations, virtual consultations may result in significant cost savings."

The study was funded by the National Institutes of Health, the Wilsey Family Foundation, and the UCSF Clinical & Translational Science Institute. Dr Rushakoff has no relevant financial relationships. Disclosures for the coauthors are listed in the paper. Dr Rayman has no relevant financial relationships.

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

Ann Intern Med . Published online March 27, 2017. Abstract, Editorial


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.
Post as: