American Academy of Ophthalmology Web Site:
Original URL:

April 2008

Clinical Practice Guidelines: Too Easily Purloined?
By Richard P. Mills, MD, MPH
Chief Medical Editor, EyeNet

While putatively synonymous with “stolen,” the word “purloined” carries Edgar Allan Poe’s nuance of a violation of trust: The thief has the trust of the victim. Why is this connotation important in the context of guidelines? Let’s go together to the caboose of my train of thought and move forward from there.

Clinical practice guidelines have proliferated; a search of PubMed found that in the past decade, 8,275 of them have been published in the peer-reviewed literature. As we start the new millennium, guidelines are assuming new importance since they are the basis for rating quality of care of HMOs, for testing physicians for certification, and will eventually be used for pay for performance criteria. Physicians who follow guidelines will be rewarded; those who do not will be penalized. When properly prepared, guidelines are evidence-based, or at least consensus-based, and they are as free from bias as possible. Our Academy was a leader when it produced its first set of Preferred Practice Patterns in 1988. The process was methodologically sound, an expensive proposition, but the result was critical acclaim.

Of course, by their very nature, guidelines produce winners and losers, and not just among physicians. Favored treatments stand to gain, while others lose. It should be obvious why medical industry has a great interest in care guidelines. So much of an interest that many companies have assembled panels of experts, convened them in resort destinations, often under the umbrella of an allegedly independent third party educational foundation, and produced guidelines that (surprise!) favor the companies’ products. The companies then have paid for dissemination of the report via a supplement to a widely read tabloid, or even, in some cases, a peer-reviewed journal.

More insidious is the relationship between industry and authors of clinical practice guidelines endorsed by medical societies. A 2002 analysis across medicine of the behavior of authors of such guidelines revealed that 87 percent of authors had some form of interaction with the pharmaceutical industry. Fifty-eight percent had received financial support to perform research, and 38 percent had served as employees or consultants for a pharmaceutical company. Notably, in the published versions of the guidelines, specific declarations of the authors’ financial interactions were made in only two of 44 guidelines reviewed.1 Last year, in the New England Journal of Medicine, Robert Steinbrook proposed a publicly funded process for guideline development as the only way to eliminate bias favoring industry. Absent that, he urged scrutiny of guidelines for conflicts of interest in authors and sponsoring organizations.2

So whose trust is being violated when guidelines are purloined? The patient’s, of course—the ultimate beneficiary of quality care. And who is the thief? Our colleagues who have personal financial relationships with those organizations that stand to profit if guidelines are favorable to their cause. But there is hope! If physicians and organizations are willing to carefully eliminate bias in the creation of clinical practice guidelines, and reject those with bias injected by authors or funders, then will clinical practice guidelines be easily purloined? As Poe’s raven quoth,

1 JAMA 2002;287:612–617.
2 New Engl J Med 2007;356:331–333.

For more on this topic, see (“Ask the Ethicist”)