Electronic Health Records

News

EHR analytics predict link between Type 2 diabetes, dementia

By Jennifer Bresnick

- Researchers following 30,000 Type 2 diabetic patients for a decade were able to use an analytics model to predict the risk of developing dementia while suffering from the endocrine disease, according to a new study published in The Lancet Diabetes & Endocrinology journal.  While researchers have known that diabetic patients are twice as likely to develop dementia as their healthier peers, a new analytical model has been able to identify specific risk factors and develop a predictive score to track higher risk patients.

The 20-point risk score is the first to be developed specifically for diabetic patients, and is based on each patient’s personal history.  Microvascular disease, diabetic foot problems, cardiovascular disease, age, depression, acute metabolic events, and other health factors each contribute a point to the score.  For patients with a score between 12 and 19, the risk of developing dementia was 73.3 percent.

“[The risk score] puts these patients on the radar of clinicians to know they are at high risk,” said Rachel Whitmer, a Kaiser Permanente researcher and the lead author on the study.  “The clinician could be especially attentive to looking out for memory or cognitive problems. If you are elderly and have Type 2 diabetes, that’s all the more reason to have appropriate glycemic control.”

An unrelated study recently found that using the same type of predictive analytics can identify Type 2 diabetes earlier than unaided physicians, potentially helping patients receive earlier treatment and prevent future complications like dementia, which are expensive and debilitating.  The algorithm was able to catch 25% of diabetes cases that may otherwise have been missed, and might be instrumental in reducing the delay between onset and diagnosis that can currently be as long as seven years.

Identifying risks before they develop using algorithms and risk scores based on EHR data is an emerging area of research intended to cut unnecessary healthcare costs and improve accountable care initiatives that keep patients healthier for longer.  Researchers are starting to focus heavily on preventing costly hospital readmissions, and risk scores based on analytics provide a concrete way to gauge the likelihood of negative complications.

“If we can predict future risk, we then know who to focus on, who to follow more, who to spend time with making sure they are taking care of themselves in terms of their type 2 diabetes,” Whitmer said.  “I think it’s important for this population to know that this disease can affect their brain and can greatly affect the quality of their life.”

Continue to site...