Parkinson’s? Data mining may diagnose it

At the Washington University School of Medicine in St. Louis, researchers were able to predict with up to 83 percent accuracy whether a person would develop Parkinson’s Disease just by looking at an individual’s medical history.

Before Parkinson’s is diagnosed, a person usually comes to the doctor with complaints about tremors, posture problems and digestive difficulties, among other ills. Researchers reviewed the health insurance claims of about 208,000 people, of which about 90,000 developed Parkinson’s.

The group discovered that a particular pattern of health insurance claim codes was a strong predictor that Parkinson’s was on the way.