- EHR algorithms scanning patient records for signs of diabetes can identify sufferers more than 90% of the time, and predict the exact date of a diagnosis for the disease in 78.4% of cases, according to research published in BioMedCentral. Using only data typically entered into an EHR, the algorithm can prevent a delayed diagnosis in 11% of patient cases, allowing physicians to prescribe treatment earlier than ever before.
Diabetes is seen as a prime example of how data analytics can improve care and reduce the costs associated with poorly controlled chronic diseases. With the disease affecting 25.8 million people, and costing $174 billion annually, diabetes is an effective test case for the principles of the patient-centered medical home (PCMH), accountable care organizations (ACOs), and the power of predictive EHR analytics. There is often a significant delay in the diagnosis and treatment of the condition, the researchers from the University of California San Francisco say, with a median delay between onset and treatment of 2.4 years, and 7% of cases going completely undiagnosed for a whopping seven years.
“Achieving early glycemic control in patients with newly diagnosed diabetes reduces the risk of microvascular complications, myocardial infarction, and all-cause mortality,” the study states. “The distinct advantage of our automated, real-time algorithm is the timely recognition of diabetes. Relying on only two ICD-9 encounter codes to establish the diagnosis date, a quarter of the cases in our cohort would have been missed.”
The researchers were able to look at how individual components of EHR data work together to build a picture of a diabetes patient, with the aim of helping health systems build diabetes prediction software in the future. Such software could help providers seeking financial incentives for quality accountable care to achieve their goals while getting patients the treatment they need as soon as possible.
“Healthcare systems may additionally apply this algorithm to provide feedback to providers on the quality of their care, generate letters to patients, identify underperforming clinics for quality improvement initiatives, link clinical decision support tools to inform decision making at the point-of-care, and risk stratify diabetic patients to direct limited resources to patients at greatest risk for developing complications.” Source