Vance W. Berger, PhD
Potential Financial Conflicts of Interest: None disclosed.
Berger VW. Do Methodological Flaws Invalidate a Randomized Trial of Lifestyle Modification Programs in Obese Patients?. Ann Intern Med. 2009;151:70. doi: 10.7326/0003-4819-151-1-200907070-00014
Download citation file:
Published: Ann Intern Med. 2009;151(1):70.
TO THE EDITOR:
The article by Digenio and colleagues (1) has several methodological errors. Can we trust the results of a trial in which patients were randomly assigned inappropriately, patient withdrawals substantially shrank the cohort, the “modified” intention-to-treat analysis did not account for the many patients who withdrew, and the analysis was based on distributional and other assumptions that could not possibly be true? This is a bit hard to swallow.
In an unmasked trial, such as this one, the worst form of randomization is permuted blocks—especially with a small, fixed block size (2, 3). With 5 treatments, in this trial, each block was allocated to each treatment only once; therefore, 5 is the smallest possible block size, and future allocations are easy to predict. Allocation concealment is thus impossible. In addition, race is poorly distributed among treatment groups. Digenio and colleagues should have used a more appropriate method of randomization, such as the maximal procedure (2, 3). The numerous patient withdrawals may have been unavoidable, but neglecting to include them in the analysis is inexcusable. We all know the merits of the intention-to-treat approach and the rationale for using it. But if using a method (such as intention-to-treat) is good, then surely not using it must be bad. That is true in this case, because the exclusion of randomly assigned patients opens the analysis to many biases, especially when the number of patients who withdrew is large and uneven across treatment groups, as it is here. Moreover, the primary analysis was a repeated-measures linear model, which piles assumption on assumption, not the least of which is normality. When the analysis drowns out rather than reflects the data, it cannot be taken seriously.
Learn more about subscription options.
Register Now for a free account.
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only