SAN FRANCISCO--(BUSINESS WIRE)--Researchers at WellDoc® today presented updated results of its hypoglycemia prediction technology at this week’s 2014 American Diabetes Association 74th Scientific Sessions. In previous work, using just one week of blood glucose data, WellDoc’s models were shown to predict correctly 90 percent of the time that hypoglycemia would occur the following day. Today, WellDoc announced an enhanced model that incorporates both glucose and medication data has demonstrated the ability to predict hypoglycemia within a specific hour.
“The challenge in predicting hypoglycemia is that most patients with type 2 diabetes measure their blood glucose once or twice per day. This so-called ‘sparse data’ makes mathematical forecasting difficult,” states WellDoc Chief Data Science Officer, Anand Iyer, Ph.D. “We used machine learning algorithms which allow the computer to detect patterns and make predictions after being trained on thousands of data points.”
In the current study, the medication model improves upon the specificity of the prediction by 32 percent (from 69.5 percent to 92.1 percent). Further, the time window of prediction was narrowed from any time within 24 hours to a specific hour. These improvements were seen after the WellDoc team trained the model with both medication information as well as blood glucose patterns. “High specificity means that the system will have fewer false positives,” explained Mansur Shomali, MD, a practicing diabetes specialist at MedStar Union Memorial Hospital in Baltimore, MD, and senior author of the study. “We don’t want to alert patients needlessly or else the intervention will not be very effective.”
Hypoglycemia is one of the most dangerous, costly, and difficult diabetes complications to manage. A recent study1 of 2.4 million type 2 diabetes patients found that hypoglycemia was especially common in patients treated with insulin or sulfonylureas with costs for a hypoglycemia-related hospital event averaging $10,362, regardless of a patient’s drug regimen. A 2005 study2 further reported that patients who had experienced hypoglycemia had 77 percent more disability days per year. Using the new methodology, the model also could accurately predict the hour of the hypoglycemia event over the next day when the patients’ medication information is known.
“Hypoglycemia is very dangerous to patients and a great burden on our health care system,” remarked Suzanne Sysko Clough, MD, Chief Medical Officer and co-founder of WellDoc. “We are certain that incorporating hypoglycemia prediction into a mobile health device will be a wonderful tool for patients.”
WellDoc intends to incorporate the hypoglycemia prediction algorithms into BlueStar™, the world’s first mobile prescription therapy for adults with type 2 diabetes. Once incorporated, BlueStar’s automated, real-time coaching will educate a patient about how to best manage and avoid hypoglycemia.
“People respond best to information that is personal, timely, and actionable,” explains Robin Anthony-Kouyate, Ph.D., Director of Behavioral Science at WellDoc. “Once a patient is armed with the knowledge that a hypoglycemia event is likely to occur in the next 24 hours they may become more receptive to learning about the consequence of the event and what they can do to avoid going hypo.”
About WellDoc: Engaging Patients, Enhancing Outcomes™
WellDoc, a healthcare behavioral science and technology company that develops solutions aimed at engaging patients and enhancing health outcomes, is focused on simplifying the way chronic diseases are managed through a collaborative, life-changing approach that breaks down the barriers of engagement and adherence by providing patients and their healthcare providers with real-time, actionable information. WellDoc has developed an industry-leading, technology-based solution designed to help people with chronic diseases better manage their conditions. Forbes selected WellDoc as one of “America's Most Promising Companies” based on its potential for future growth.
About The American Diabetes Association 74th Scientific
June 13 - 17, 2014, San Francisco, California
Scientific Sessions is the largest diabetes meeting in the world, bringing together nearly 18,000 participants—including more than 14,000 clinicians and researchers from all 50 states and 117 countries.
The 5-day annual meeting features timely and significant advances in basic science and the prevention, diagnosis, and treatment of diabetes. Attendees choose from over 3,000 original presentations which include symposia, oral abstract sessions, meet the expert sessions, interest group discussions, guided audio poster tours, and poster presentations. It also includes approximately 175 exhibitors showcasing the latest in diabetes-related products, services, and technology available to health care professionals.
1 Curkendall SM, Zhang B, Oh KS, Williams SA, Pollack MF, Graham J. “Incidence and cost of hypoglycemia among patients with Type 2 diabetes in the United States: analysis of a health insurance database.” J Clin Outcomes Manage (2011) Oct;18. 10:455-462.
2 Rhoads GG, Orsini LS, Crown W, Wang S, Getahun D, Zhang Q. “Contribution of hypoglycemia to medical care expenditures and short-term disability in employees with diabetes.” J Occup Environ Med (2005) 47:447–452