Model predicts hypoglycemia events in type 2 diabetics

Technology from Baltimore-based WellDoc has successfully demonstrated the ability to predict hypoglycemia (a dangerously low blood sugar level) in patients with type 2 diabetes, according to a company announcement.

WellDoc researchers tested a prediction model using tens of thousands of blood glucose data points from both the International Diabetes Center and WellDoc's proprietary databases. What they learned was that by using seven days of spot blood glucose monitoring data, with about one blood glucose test per day, WellDoc's technology can send an alert for hypoglycemic events that are likely to occur within the next 24 hours, claims the company.  

Initial results of the company's hypoglycemia prediction model were presented on Oct. 31 during the scientific poster session at the 2013 Diabetes Technology Society Meeting in San Francisco. 

"In its initial testing, WellDoc's hypoglycemia model correctly predicted hypoglycemic events using this sparse data 91 percent of the time," said Mansur Shomali, M.D., C.M., medical director at WellDoc and an endocrinologist at MedStar Union Memorial Hospital in Baltimore, in a written statement.

Bharath Sudharsan, the lead investigator and data scientist at WellDoc, added in a written statement that the WellDoc algorithms correctly predicted hypoglycemic incidents "42 percent more often than expert human endocrinologists, with low rates of false positives and false negatives."

After further validatation and once it receives clearance from the U.S. Food and Drug Administration, WellDoc plans to include the prediction algorithms into the company's BlueStar software for use on mobile phones or personal computers. BlueStar, the first FDA-cleared mobile prescription therapy for adults with type 2 diabetes, provides secure capture, storage, and transmission of blood glucose data as well as information to aid in diabetes self-management. 

BlueStar is the prescription version of WellDoc's DiabetesManager, which received 510(k) clearance from the FDA in August 2010.

In related news, a commentary published this month in the Journal of Diabetes Science and Technology argues that there has been little research on the use of diabetes management mobile apps, thereby "marginalizing the role of diabetes professionals and educators in a burgeoning market, hungry for information and an improved quality of life." Though the number of diabetes management mobile apps available on the market has grown exponentially since 2009, most patients lack the skills necessary for finding relevant healthcare information, assert the authors.

"Despite the importance of education in clinical guidelines, very few of these apps include education in their top functionalities," states the editorial. "Healthcare providers must not only educate patients about these resources, but take steps to ensure that mobile apps follow accepted best practices and guidelines."

To learn more:
- read the announcement
- read the commentary

Suggested Articles

The newly launched Center for Connected Health will be largest telehealth hub in the Philadelphia region, according to Penn Medicine.

The FDA commissioner wants to use additional funding under Trump's budget to advance digital health initiatives and integrate real-world data.

The FDA's approval of an app that uses AI to notify specialists of a potential stroke offers new possibilities for triage software that uses CDS.