Researchers at Boston University May Be Able to Predict Who Will Need Life Support

They hope it will allow families of at-risk patients to begin planning for end of life issues earlier.

Researchers at Boston University School of Medicine (BUSM) found a way to predict the future—sort of.

BUSM’s team developed an algorithm to help determine which patients are at highest risk of one day needing life support with mechanical ventilation. The goal of the project is to give families of at-risk patients a way to plan for potential end of life issues earlier, before any health issues occur.

Using data from the Framingham Heart Study, which examined data from patients 65 and older who were enrolled in Medicare, the researchers identified and assigned point values to traits that often go along with eventually needing life support, including older age, alcohol use, diabetes, high blood pressure, prior hospitalizations, and more. After assigning point values to each trait, the researchers estimated the probability of each patient needing life support within five years based on his or her total; they then verified the accuracy of their predictions against clinical data from a healthcare system in Utah.

Of course, no research tool can foresee serious accidents or unexpected health conditions. But corresponding author and BUSM professor Allan Walkey said in a statement that a better understanding of the likelihood of needing life support could help families with end of life planning:

“A tool that improves the identification of people who are at risk for needing life support will allow for better communication between patients, family and physicians regarding patient wishes should these patients become incapacitated by critical illness. Improved early communication may lead to later care more in-line with patient wishes, increasing patient autonomy and improving our ability to care for patients.”