This post was inspired by an interview with Shahram Ebadollahi, PhD,Vice President, Innovation & Chief Science Officer at IBM Watson Health.
In part I of the series, we learned how cognitive computing can play a role in everyday life for people with diabetes, but were left wondering what impact cognitive computing in diabetes management could have on the healthcare system. Cognitive computing is the simulation of human thought processes in a computerized model, which involves self-learning systems that use data mining, pattern recognition, and natural language processing to mimic the way the human brain works.
The growing data volume makes our health data like puzzle pieces. Doctors can see some of the pieces – prescriptions, medical images, test results – but they miss a lot of the data that is available, leaving them with an incomplete picture as to how we live our daily lives. In fact, about 80% of health data never makes it into a patient’s medical record.[i] And yet the amount of valuable health data continues to grow with experts estimating it to double every 73 days by 2020.[ii]
“People use and interact with data in different contexts and for different purposes,” Ebadollahi said. “You cannot code rules into computers for every possible need an individual may have.”
Physicians can only learn so much during a patient visit, which tends to happen once every 90 days and lasts only 10 minutes on average.[iii] During the few minutes the patient and doctor spend together, how much information can realistically be exchanged – considering that between doctor’s visits, the typical patient with diabetes has about 300 meals, 700 blood glucose (BG) readings, 1,000 alarms and 25,000 continuous glucose-monitoring (CGM) data points?[iv]
Imagine if physicians had the ability to – within three seconds – look through all of their patient’s medical information and could create a personalized care plan.
Real World Applications
Through our unique partnership with IBM Watson Health, we are working together to develop near real-time, dynamic personalized care plans for healthcare providers and their patients, with the goal of improving clinical outcomes, efficiency, and patient satisfaction. This could include identifying patients who have been admitted to the Emergency Room (ER) for a hypoglycemic event (low blood sugar), and are likely to return again in 30 days. We will look to establish targets for both outcomes and costs over a specified period of time, and partner with select healthcare providers to develop the protocols needed to intervene at the right time with the right therapies, so that we can reach the desired outcome.
These applications could lead to improved patient outcomes – including hypoglycemic episodes as well as increased savings for the healthcare system. Imagine the cost-savings that can be realized for patients and the health care system if we can help avoid a hospital visit. Preventing just five hypoglycemic episodes per person, per year, can reduce the cost of emergency services by $735 and $9,300 in in-patient services.[v],[vi],[vii] With 9% of the world’s adult population living with diabetes[viii], imagine how many people could benefit from such innovation.
The combination of this technology, data and insights, are leading to new solutions that integrate care throughout a person’s journey with diabetes and within the healthcare system.
“Medtronic brings in the domain expertise of diabetes management, and we bring the cognitive, the cloud and the scale with respect to analytics,” Ebadollahi said. “Put them together, and you’re co-innovating the future of diabetes care.”
Working together, we can bring diabetes management into the era of cognitive computing, and reduce the unpredictability of managing life with the disease. Cognitive computing will transform diabetes care.
[i] Healthcare Content Management White Paper. Data Mark. October 2013. https://www.datamark.net/uploads/files/unstructured_ehr_data_white_paper.pdf
[ii] IBM Watson Health. http://www.ibm.com/smarterplanet/us/en/ibmwatson/health/
[iii] Amount of time U.S. primary care physicians spent with each patient as of 2015. The Statics Portal. http://www.statista.com/statistics/250219/us-physicians-opinion-about-their-compensation/
[iv] Treatment burden & health related QOL in children with chronic disease; Tareq Zaian et al; West University Australia – J Paediatr Child Health. 2006 Oct;42 (10):596-600.
[v] Quilliam BJ, Simeone JC, Ozbay AB, Kogut SJ. The incidence and costs of hypoglycemia in type 2 diabetes. Am J Manag Care 2011;17(10):673-680)
[vi] Rhoads et al;. Contribution of hypoglycemia to medical care expenditures and short-term disability in employees with diabetes. J Occup Environ Med 2005;47(5):447-452
[vii] Frier, B. M. The economic costs of hypoglycaemia. The British Journal of Diabetes & Vascular Disease 2011; 11, 10-12
Tags: big data
, cognitive computing
, diabetes devices
, insulin pump