Real life outcomes vs. clinical trial results

• Published in: Journal of ophthalmic & vision research Vol. 14; no. 1; pp. 88 - 92
• Main Author: Zarbin, Marco
• Format: Journal Article
• Language: English
• Published: Iran Wolters Kluwer India Pvt. Ltd 01.01.2019; Medknow Publications and Media Pvt. Ltd; Medknow Publications & Media Pvt. Ltd; Medknow Publications & Media Pvt Ltd; Knowledge E
• Subjects: Clinical medicine; Clinical Practice; Clinical Trial; Confidence Interval; P value; Rule of Three; Statistical Significance; Clinical trials; Confidence intervals; Ethical aspects; Evaluation; Evidence-based practice; Experiments; Medical errors; Patients; Stroke; Studies; Treatment outcome; P value; Statistical Significance; Clinical Trial; Confidence Interval; Rule of Three; Clinical Practice
• ISSN: 2008-322X; 2008-2010; 2008-322X
• DOI: 10.4103/jovr.jovr_279_18

Two questions that clinicians should answer as they attempt to apply the results of clinical trials to clinical practice are: (1) will statistically significant results be reproduced in their clinical practice; and (2) if nothing goes wrong in a clinical trial, is everything alright? Regarding the first question, when considering the results of a randomized, multicenter, prospective, controlled clinical trial, two questions that cannot be addressed by simply by reading the trial results and that only the practicing clinician can answer are: (1) is the study population representative of the patient about to be treated; and (2) is the totality of evidence outside the trial (including the clinician's own extensive experience) consistent with the trial result? Regarding the second question, clinicians are advised to recognize that most studies, even Phase 3 trials, are underpowered to accurately assess the risk of low frequency events.