May 2nd, 2011

Standard Guidelines Compared with Individualized Guidelines

Should patients be treated by standard guidelines, or should guidelines be individualized for patients? In a study published in the Annals of Internal Medicine, David Eddy and colleagues used data from the Atherosclerosis Risk in Communities (ARIC) Study to calculate and compare the expected benefit from hypertension treatment based on JNC 7 guidelines with the expected benefit from treatment based on individualized guidelines. (Individualized guidelines were derived from a calculator developed as part of the Archimedes computer model designed to simulate clinical trials.)

The authors (all of whom are affiliated with the Archimedes project) found that “individualized guidelines could prevent more than an additional 43% of MIs and strokes than JNC 7 guidelines for the same cost.” To achieve the same benefit as the JNC 7 guidelines, 52% of people indicated for treatment by JNC 7 would not be indicated for treatment by the individualized guidelines, while 36% of people indicated for treatment by the individualized guidelines would not be indicated for treatment by JNC 7.

The authors conclude: “Our results show that individualized guidelines can prevent the same number of MIs and strokes at a lower cost (and by exposing fewer people to the risks and adverse effects of treatments) or can prevent more MIs and strokes at the same cost as the JNC 7 guideline.”

In an accompanying editorial, Douglas Owens notes that “tailoring recommendations is easier said than done.” He writes that “tailored recommendations must meet 3 challenges” before being generally accepted. First, for a specific clinical problem, individualized guidelines must be shown to have the potential to produce better results. Second, clinical trials must demonstrate these potential benefits. Third, the individualized guidelines must be “easy enough to use that clinicians will adopt them.”

One Response to “Standard Guidelines Compared with Individualized Guidelines”

  1. Saurav Chatterjee, MD says:

    sounds fascinating the way a computer predicted one’s individual health risk-can someone elucidate bit more about the exact way in which this was done?