New Diabetes App Could Predict Dangerous Health Events Hours Earlier

Diabetes is a complex autoimmune condition that requires many treatment and education strategies to achieve health goals beyond glycemic control. Food, sleep, physical activity, insulin level, and near-constant monitoring are all lifestyle factors that individuals with diabetes must manage. Even when they successfully do so, there can be uncertainty about whether their blood sugar levels will drop suddenly.

To help manage that uncertainty, technology giant IBM and medical device company Medtronic have teamed up to provide a new predictive smartphone app designed to give patients with diabetes early warnings of potential emergencies. Watson, the supercomputer that infamously defeated 2 Jeopardy champions in 2011, powers the app. The app, which was presented at the International Consumer Electronic Show in January 2016, analyzes Medtronic’s data from insulin pumps and glucose monitoring devices to establish possible patterns that could suggest something is amiss in patients.

The algorithms are derived from de-identified data from 150 million patient days recorded via Medtronic wearable devices. An early research project involving 600 patient cases revealed that the app could predict near-term hypoglycemic events, including seizure, coma, and death, up to 3 hours prior to symptom onset.

“The solutions we’re codeveloping with IBM may one day enable Medtronic to provide real-time insights and coaching to help people understand the impact of daily activities on their diabetes and make adjustments as needed,” wrote Annette Brüls, president, Global Diabetes Service & Solutions, Medtronic, in a blog post on the company’s website.

As more patients turn to their smartphones for daily health management, pharmacists should become more familiar with mobile health apps and prepare to integrate them into their practice. The IBM/ Medtronic app, in particular, could be a game changer for patients with diabetes looking to better monitor their health.

That apps could prove integral in diabetes management is not a novel concept. A study, published in Diabetes Care, showed that behavioral mobile coaching using blood glucose data, lifestyle behaviors, and patient self-management data substantially reduced glycated hemoglobin levels over 1 year.

What remains to be seen is how the FDA will regulate the app. In a final guidance, released in September 2013, the FDA said it does not intend to enforce requirements under the Federal Drug & Cosmetic Act for the majority of mobile apps, as they “pose minimal risk to consumers.”

The Medtronic/IBM diabetes app, however, could fall under the umbrella of apps that the FDA will oversee because use of the app could be interpreted as “an accessory to a regulated medical device” and could, therefore, “present a greater risk to patients if [it does] not work as intended.”

Better management of diabetes could have widespread implications for the US health care market. At present, 29 million individuals have diabetes in the United States, and the costs to treat these patients is approximately $240 billion annually.

“It’s essential for Medtronic to make this transition because the world’s health care systems are shifting from payment for services to payment for outcomes,” wrote Brüls. “Because of that shift, we believe we have an imperative to make use of data that could help people with diabetes use our devices in the most effective way to manage their disease.”

The diabetes partnership with Medtronic is the latest among IBM’s initiatives aimed at leveraging technology in chronic disease prevention and management. In October 2015, CVS Health and IBM forged an agreement to use the pharmacy’s health records to create predictive analytics, and allow health providers to maximize patient encounters.