Paying For Quality

Friday, January 14, 2011 // Uncategorized

Rewarding doctors for quality of care is something that makes for a good sound bite, but is far more difficult to do.  The problem is, it is hard to define and measure quality.  In addition, adherence to certain guidelines which are often used as markers of quality of care  frequently turn out to be based on poor  data as summarized in the article from Journal Watch.

Study Finds Half of Guideline Recommendations Are Based on Low-Quality Evidence
More than half of recommendations included in guidelines from the Infectious Diseases Society of America rely on low-quality evidence, according to a study in the Archives of Internal Medicine.
Researchers examined 41 guidelines published by IDSA since 1994. Of the 4200 individual recommendations in those guidelines, 55% were supported by level III quality of evidence (e.g., expert opinions), while only 14% were guided by level I evidence (e.g., randomized controlled trials).
Five guidelines were updated during the study interval. In these updates, the number of recommendations increased between 20% and 400%, but only two updates saw an increase in the number of recommendations based on high-quality evidence.
An editorialist said that one of the main take-home messages of this study “is to be wary of falling into the trap of ‘cookbook medicine.’ The existence of guidelines is probably better than no guidelines, but guidelines will never replace critical thinking in patient care.”
Archives of Internal Medicine article (Free abstract)
Archives of Internal Medicine editorial (Subscription required)

4200 individual recommendations! Now we have guidelines for guidelines.  There are so many that no physician could ever adhere to all the recommendations.  Many things that sounded like a good idea at the time, turn out not to be when carfeully scrutinized.  This doesn’t mean that we shouldn’t keep trying, but spotlights how difficult it is.


Leave a Reply

Your email address will not be published. Required fields are marked *