Lies, Half-Truths, and Statistics Part 2

Dr. Lyndon HernandezHow to Lie with Statistics: A Field Guide for GIE Reviewers

Written by Lyndon Hernandez, MD, MPH, FASGE, Chair of the GIE Editorial Review Board

There is a reason why the FDA and major medical journals insist that the primary endpoint of a study be specified, even before the study is initiated. This is because endpoints can be manipulated to confirm the bias of the researcher. Herein is the second of a series, intended for reviewers new to GIE, the next time they critique a study.

Tip 2: Choose the endpoint that will make your study shine

Commonly seen in: Large clinical trials, especially phase III pharmaceutical studies

Comments: Selective endpoint reporting is so hard to resist when there are several endpoints to choose from. Some endpoints are statistically significant and will highlight a study, others not. Indeed, some endpoints will even help advance your career.         

Example: A Study on Endoscopic Treatment X for GERD

This study focused on decrease in acid reflux symptom score and used it as main outcome and did not report some of the other non-significant outcomes.

The Solution: Specify the primary endpoint in advance, and stick with it. As a reviewer, politely ask if there were some other outcomes that were measured but were not reported, for a variety of reasons.  Also, the power analysis calculation should have been based on the main outcome, and so there should be no turning back once the study is done.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s