Why Guidelines Should Be Waged Like War


Here’s a modest proposal: we need fewer and shorter guidelines. In fact, I’d like to propose that guidelines, like war, should be waged only when there is absolute consensus and overwhelming evidence.

Anyone interested in the subject is aware that guidelines are in a complete mess. Just in the past two weeks I’ve written about major controversies regarding the hypertension and salt guidelines. (I think there are now more than five separate and quite divergent hypertension guidelines.) One of the biggest stories of the past year has been the ongoing battle over statins. And there are similar, and perhaps even more intense, battles in the world of cancer screening, including passionate disagreements over the proper role of mammography and PSA tests. The guideline minefield is so treacherous that a few years ago even the NHLBI decided to abandon its central role in developing cardiovascular guidelines.

The fact is that most guidelines are not based on solid evidence and probably don’t represent a genuine expert consensus. Instead, a lot of experts get together in a room and agree to agree, even though they often really disagree. This is a recipe for disaster. Sripal Bangalore is a hypertension expert who recently came out in favor of more aggressive blood pressure goals. He objected to an editorial that argued that guidelines should be based only on randomized controlled trials (RCTs). Here’s what he told me:

However, if one were to base guidelines only on RCT evidence all our guidelines would have been only a couple of pages in length. For example, 13.5% of STEMI guideline recommendations are based on RCT data. I am all for RCTs to drive our recommendations, but in the absence of that, we have to take the totality of evidence to make a rational decision.

In response to the promise of short guidelines, only a few pages long, one reader wrote Bangalore:

That’s the best argument you’ve ever made for RCT-only guidelines!

So let’s think about this for a minute. Do we really need so many guidelines, and do they really need to be so long?

In the cardiovascular area alone there are scores and scores of guidelines, many of them quite long. No one reads them cover to cover. Many of the authors– and there can be quite a few– haven’t even read every word, and they certainly haven’t tracked down and confirmed each of the hundreds of references and details. Quite simply, the situation is unmanageable, not only for clinicians but even for the academics who produce these guidelines.

Think a minute about Bangalore’s assertion that most guidelines lack an evidence base. It seems to me that the medical organizations that produce guidelines should freely admit this lack of evidence for most recommendations. Then, instead of getting their panties all in a bunch trying to defend the indefensible– as we saw recently with the salt guideline– they could advocate for better evidence. After all, it’s hard to convince anyone that more trials and data are necessary when there’s already a hundred pages of densely-referenced guidelines on the topic, and the assemblage of learned experts all express unanimous agreement about the best course of treatment. Who needs data in that case?

So if they want to make the case for more data they will have to first acknowledge their ignorance. And that will first require that rather than browbeat their committees into unanimity, they agree to disagree and express a variety of opinions.

It seems to me that we would be better off if the guideline-producing organizations–  most notably the American Heart Association, the American College of Cardiology, and the European Society of Cardiology– would spend  a whole lot more effort persuading everyone how little they know rather than how much they know. Perhaps then we would never again have a situation in which it could be creditably asserted that one single guideline may have been responsible for 800,000 deaths in five years.


I don’t want to leave the impression that the above is an original idea. Others have had the same idea, of course. Sanjay Kaul, an old friend of this blog, stated it much more soberly a while back on CardioSource:

“Guidelines that are driven by scientifically documented and rigorously appraised high-quality evidence rather than ‘filtered’ opinions are more likely to be accepted by the stakeholders, thereby reducing the variability in care and improving the quality and reducing the cost of care. It is our collective responsibility to ensure that the ACCF/AHA guidelines subscribe to the standards set forth by the IOM for developing rigorous and trustworthy CPGs to optimize patient care.”

In addition to the salt and the non-cardiac surgery guidelines, I wanted to mention one other example of the potentially disastrous effects of guidelines. To this day we still don’t know the precise effects of saturated fats, but it is very clear that guidelines from the American Heart Association and others which demonized fats starting in the 1980s may have had the catastrophic consequence of pushing people to consume more carbohydrates, including sugar. We will probably never know the full extent of the damage, but many have speculated that this may have contributed to the obesity and diabetes epidemics.


United Nations Security Council Resolution 707

United Nations Security Council Resolution 707 (Photo credit: Wikipedia)



  1. Guidelines should be like labels: address ONLY the population studied, the intervention studied, and the outcome studied. “Delphic” guidelines are not useful in practice, but quite useful for med mal lawyers.

    We need to stop pretending to know things that are actually opinions. Share opinions, but don’t make them laws.

    This will, of course, threaten the egos of the leaders of the guideline industry, but will serve patients and doctors much better.

    Then we need to COLLECT the data we currently think we know!

Speak Your Mind