Our good friend at Diabetes Investor, David Kliff, has made a point of focusing on the difficulty of a patient’s success just because we have “whizbang” apps and devices to make glucose control almost automatic. The data shows he is probably right, as keeping patients on medications and causing them …

Woman, 55 years of age visited with me 3 weeks ago. Her glucose levels were in the 200-250 mg/dL range. She was actually a new patient of ours; her A1C was 9.2%. We recommended a CGM. She purchased one and brought it to the office for me to teach her how to use it. I did so. This patient is well-educated and seemed tech-savvy. When she left the visit, she "seemed" to understand CGM, what it is, what to do, and how to insert the sensor, get readings, interpret readings, etc.

Lisa M. Latts MD, MSPH, MBA, FACP is Deputy Chief Health Officer for IBM Watson Health. She is a national leader in health care quality and has nearly 20 years’ experience in health care delivery working to improve lives for patients.  She has worked as a health insurance executive, a consultant and for one of the nation’s leading academic medical centers helping to deliver high quality care to its faculty and employees. She is currently helping IBM Watson Health bring cognitive computing and machine learning to advance the idea of Value-Based Care and help solve the pressing problems that physicians, employers and payers are facing.

In this week's Homerun Slides, we determine what makes a diabetes app useful.

The relationship between pathogenic factors and beta-cell destruction remains poorly understood. Longitudinal studies of newborns and children at genetic risk and follow-up of at-risk first-degree relatives in natural history studies link the triggering of autoimmunity with the appearance of autoantibodies to one or more islet autoantigens; individuals with multiple autoantibodies have higher risk of diabetes progression.

Estimated difference in test scores 0.45 after adjustment for variables such as socioeconomic status