[Skip to Navigation]
Sign In
Views 373
Citations 0
Comment & Response
October 27, 2020

Evaluating Non–Statistically Significant Results From Trials in Practice—Reply

Author Affiliations
  • 1Medical Research Institute of New Zealand, Wellington, New Zealand
  • 2Alfred Hospital Intensive Care Unit, Alfred Hospital, Melbourne, Victoria, Australia
  • 3Department of Intensive Care, Rigshospitalet, Copenhagen, Denmark
JAMA. 2020;324(16):1680. doi:10.1001/jama.2020.15657

In Reply We submit that clinicians have a responsibility to evaluate major clinical trials and, at times, to implement their findings in advance of clinical guidelines being produced by scientific societies and government agencies. In our Viewpoint,1 we sought to provide a framework for them to do this in situations in which actionable findings might be hidden behind a non–statistically significant primary outcome. We agree with Dr Piovani and colleagues that when deciding whether to implement clinical trial findings, outcomes that matter to patients are the most important. However, because data pertaining to all potentially important outcomes are often not available, clinicians must make decisions about evidence under conditions of uncertainty. We consider that they should do so based on an assessment of the probability that a particular treatment will result in a net benefit for a patient, focusing first on outcomes that are likely to matter to patients.

Add or change institution