Algorithm could sort out the most pressing individual health recommendations

Primary care physicians soon may be able to use a mathematical algorithm to help them quickly prioritize their recommendations for individual patients.

The U.S. Preventive Services Task Force has issued recommendations for 60 distinct clinical services, but physicians tend to focus on the ones that take the least time. Those aren't necessarily the most important ones for improving a patient's health, according to an article at amednews.com.

In a study published at the Annals of Internal Medicine, researchers used the algorithm, connected to EHRs, to sort out the recommendations most closely tied to life expectancy based on a particular patient's condition.

For instance, a 62-year-old obese man who smoked and had high cholesterol, hypertension and a family history of colorectal cancer would benefit most from task force recommendations on tobacco cessation, weight loss and blood pressure control, the study found.

Overall, the top three recommendations in the study were diabetes control, tobacco cessation and weight loss.

So far, the algorithm is in the "proof of concept" stage, and is being piloted in a primary care clinic at Bellevue Hospital in New York.

Meanwhile, researchers at the University of Notre Dame have developed a computerized assessment tool to help doctors offer patients a personalized disease management and wellness plan. It sorts through reams of data on patients with similar conditions to make lifestyle recommendations based on the patient's individual disease risk.

All the data collected on routine doctor visits could be put to work to improve care for all, a recent Institute of Medicine discussion paper proposed.

Meanwhile, Indiana University researchers have created an artificial intelligence framework that shows how simulation modeling that "understands and predicts" the outcomes of health treatments could reduce healthcare costs and improve patient outcomes by about 50 percent. They previously showed how machine learning could decide a patient's best treatment.

To learn more:
- read the article
- find the research