A moratorium on strong recommendations is neededBMJ 2021; 375 doi: https://doi.org/10.1136/bmj.n2887 (Published 25 November 2021) Cite this as: BMJ 2021;375:n2887
- Victor M Montori, professor of medicine1,
- Dominique Allwood, honorary senior clinical lecturer2,
- Anja Fog Heen, consultant3
- 1Mayo Clinic, Rochester, MN, USA
- 2Department of Primary Care and Public Health, Imperial College London, London, UK
- 3Department of Medicine, Lovisenberg Diaconal Hospital, Oslo, Norway
- Correspondence to: V M Montori
Medicine is addicted to so-called hopium, an unwarranted confidence in the value of its tests and treatments. Clinicians overestimate their value,1 while experts on guideline panels make strong recommendations about care backed by untrustworthy evidence.2 Rigorous guideline methods can highlight this problem, but they are far from a perfect antidote—it is time for a moratorium on strong recommendations.
In a linked paper (doi:10.1136/BMJ-2021-066045),3 Yao and colleagues report that almost half (1246 of 2528) of the recommendations issued by the leading American cardiology and oncology professional societies were strong, “just do it” recommendations.4 About a quarter (354 of these were based on low certainty evidence. Compared to a consensus process, an evidence based guideline process reduced the risk of issuing such inappropriately strong recommendations, but not completely: the evidence based approach produced about a third (105 of 354) of the inappropriately strong recommendations in this study. Most of them simply conveyed the panels’ overconfidence in the benefit of following their recommendation.
The evolution …