user avatar
Gordon H. Guyatt
@GuyattGH
Promoting use of the best evidence and patient values and preferences to inform optimal clinical and health policy decisions. threads.net/@guyattgh
Hamilton, On. Canada
linkedin.com/in/guyattgh/
Joined June 2014
501
Following
15.2K
Followers
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Mar 29, 2023
    Current American guidelines for managing gender dysphoria in adolescents untrustworthy. Don’t acknowledge the very low certainty evidence regarding alternatives and do not make the very guarded weak/conditional recommendations appropriate for such evidence
    bmj.com
    Gender dysphoria in young people is rising—and so is professional disagreement
    More children and adolescents are identifying as transgender and are being offered medical treatment, especially in the US—but some providers and European authorities are urging caution because of a...
    251K
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    May 12, 2022
    Great article: frequent mistakes people make in interpreting p-values, confidence intervals and statistical power and how to correctly interpret. Enhanced my understanding of comparing two results and relation between CI overlap and stat sig differences. bit.ly/3kWcKaF
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Aug 7, 2022
    An advanced EBM evidence pyramid acknowledges that randomized trials sometimes yield only very low certainty evidence and that observational studies sometimes yield high certainty evidence.
    New evidence pyramid
    From ebm.bmj.com
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Feb 18, 2022
    This decade old #JAMA users guide for #non-inferiority #trials still drives home key message: investigators’ and #FDA non-inferiority margins completely untrustworthy. Need to think of how #patients value benefits and harms/burdens. bit.ly/2L6gCW4
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Jan 12, 2024
    Just published: ground-breaking study of over 23,000 RCTs in Cochrane reviews. Trials hugely underpowered for true effects. Consequently, “statistically significant” estimates seriously overestimate actual treatment effects, can’t trust them.
    evidence.nejm.org
    A New Look at P Values for Randomized Clinical Trials
    We have examined the primary efficacy results of 23,551 randomized clinical trials from the Cochrane Database of Systematic Reviews. We estimate that the great majority of trials have much lower st...
    50K
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Mar 27, 2023
    Our group has developed an inventory of all estimates of minimal important differences for all patient-reported outcomes known to humankind and rated the credibility of each MID. We are keen to work with anyone interested in using the fruits of our work. doi.org/10.1016/j.jcli…
    72K
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Jul 2, 2022
    Nice paper on what one can deduce about p-values from looking at overlap in confidence intervals between two independent means. Many do not realize there can be considerable overlap in confidence intervals and p-value still appreciably below 0.05 bit.ly/3y5HkoG
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Jan 4, 2023
    Best compliment ever received as mentor: student telling me he was sure I was treating him as special - until he discussed with my other students who all felt the same way. Worthwhile goal to set oneself: making all students feel as if they are getting special mentorship.
    28K
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    May 28, 2022
    With my daughter Claire on her graduation from McGill university's Arts and Science program. She'll soon be off with two friends for a two month tourist adventure in Southeast Asia.
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Dec 19, 2022
    I was asked to prepare a very brief introduction to GRADE for new members of a WHO guideline panel. The 11.5 minute result could be useful for any guideline group wanting to give their panel members a brief bird’s eye view of the GRADE process.
    33K
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Aug 5, 2022
    Making a distinction between evidence-based and consensus-based guidelines is both misguided and misleading: all trustworthy guidelines demand both rigorous evaluation of the evidence and a process to reach a panel consensus.
    Evidence vs Consensus in Clinical Practice Guidelines
    From pmc.ncbi.nlm.nih.gov
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Jan 26, 2022
    Recently published #JAMA #Users’ #Guides for #Guidelines best around for navigating the way to identifying and using a trustworthy guideline. bit.ly/3rz9Mff @rominabrigpet
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Nov 28, 2021
    The patient revolution 1 A paper from the most brilliantly original thinker writing about #healthcare today. Victor Montori tells us about careful and kind care and how caring for each other is a social determinant of health. bit.ly/310ROcA @vmontori @patientrev
  • user avatar
    Gordon H. Guyatt
    @GuyattGH
    Jan 20, 2022
    #SampleSize calculations are a hoax. Sample sizes seldom approximate requirements to detect small important differences. True determinants: feasible recruitment, available funding. #Clinicians, #guideline developers: ignore authors’ superiority thresholds. bit.ly/34TZb76