Heart disease remains the nation’s number one health issue and leading cause of death. Yet when aging people with heart conditions turn to the health care system for treatment, their physicians are guided by clinical practice guidelines that are largely based on “lower levels of evidence or expert opinion,” a new survey in the Journal of the American Medical Association showed.

The survey of 16 clinical practice guidelines (CPGs) considered current by the American College of Cardiology and the American Heart Association found that just 11 percent of 2,711 separate recommendations for diagnosing and treating various forms of heart disease were based on multiple randomized clinical trials, which is considered the highest level of evidence. Nearly half were based on the lowest level of acceptable evidence – expert opinions from the clinicians writing the guidelines.

In some common areas of cardiology, the paucity of evidence behind the recommendations was shocking. Take stable angina, for instance, a fairly common condition. Fully one-third or 78 of 235 recommendations for managing this condition received a strong endorsement from the most recent guidelines, which were issued in 2002. Yet only 15 of those recommendations (6.4 percent of the total) had the highest level evidence – multiple randomized clinical trials – backing the conclusions.

Secondary prevention of heart disease – preventive medications for blood pressure control or lowering cholesterol – wasn’t much better despite a massive investment in clinical trials over many years by industry. The 2006 ACC/AHA guidelines made 48 recommendations, of which 38 or nearly 80 percent received a strong endorsement. Yet just 11 of those recommendations were based on multiple, confirmative clinical trials while 33 or 69 percent were based on a single trial – presumably the industry-funded trial that brought the drug to market.

An accompanying editorial by Terrence M. Shaneyfelt and Robert M. Centor of the University of Alabama School of Medicine offered the obvious reason for CPGs offering strong opinions based on weak evidence. “Guidelines often have become marketing tools for device and pharmaceutical manufacturers. Financial ties between guideline panel members and industry are common,” they said.

But other biases may be affecting the quality of guidelines, they noted. Specialty societies that sponsor guidelines are often looking to expand the reach of their own practices in a crowded marketplace; and government sponsors of guideline-writing committees might be influenced by the need to limit costs.

Meanwhile, there’s too many guidelines, and many are out of date. The result? Most physicians ignore them.

The editorial offered a radical solution to the problem. They called for the total elimination of conflicts of interest on guideline-writing panels, and creating a centralized authority in the government at the Agency for Healthcare Research and Quality to coordinate guideline writing. Their model: the U.S. Preventive Services Task Force.

Centor and Shaneyfelt, an occasional blogger at Medical Rants, at several points in their editorial expressed deep skepticism about the entire process. Their views seemed to reflect the traditional clinicians' perspective that every patient is different, and CPGs can be abused by payers seeking to limit physician autonomy. “If all that can be produced are biased, minimally applicable consensus statements, perhaps guidelines should be avoided completely,” they wrote.

I found that conclusion dispiriting. Evidence-based guidelines written by unconflicted panels, reinforced by payment incentives, can be a powerful tool for reducing variation in medical practice and improving outcomes. Today, physicians who ignore guidelines practice what can best be called absence-of-evidence-based medicine. To suggest, as the editorialists do, that physicians should simply rely on their own reading of the literature to inform their decision-making is a prescription for maintaining the status quo. That’s unacceptable.