Copyright 2007 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
The ARCHIVES has published 2 randomized trials of duct tape therapy for warts.1 - 2 A commentary3 regarding the most recent of these trials2 pointed out that confidence intervals for the size of the treatment effect were not given in the results. The commentary3 provided confidence intervals in a table, but these were intervals for the observed outcome proportions in each trial arm, not intervals for the estimated effect of treatment.
Any study that compares 2 or more groups should calculate statistics that compare the group outcomes, along with estimates of precision for those comparisons, such as confidence intervals. This advice is given by the International Committee of Medical Journal Editors4 and by the Consolidated Standards of Reporting Trials guidelines for the reporting of randomized controlled trials.5 The ARCHIVES recommends the use of point estimates and confidence intervals in its instructions to authors6 and has endorsed this practice in editorials.7 - 9 Despite these recommendations, neither trial of duct tape therapy reported estimates or confidence intervals for the effect of treatment on wart resolution.1 - 2
The main outcomes for the duct tape trials were binary (Table 1), so appropriate statistics for treatment effects include risk ratios, risk differences, and odds ratios (Table 2).10 Odds ratios have a desirable symmetry; if we compare the treatment arm with the control arm, the odds ratio for the undesirable outcome (wart remaining) will be the inverse of the odds ratio for the desirable outcome (wart resolution). But when more than 10% of participants have the outcome, odds ratios are difficult to interpret because they will not approximate risk ratios well.10 - 14 The risk difference has a clear meaning, but a difference (additive) model assumes that treatment changes the outcome by adding a fixed proportion to the outcome proportion in the control group. Often a ratio (multiplicative) model is more plausible or fits the data more closely; this model assumes that treatment changes the outcome by multiplying the control group outcome proportion by a fixed ratio.15 - 17 Risk ratios are often a useful summary statistic for treatment effects.17 However, they lack the symmetry of odds ratios; the risk ratio for the harmful outcome is usually not the inverse of the risk ratio for the beneficial outcome. This asymmetry is apparent for the duct tape trials; the trial by de Haen et al2 found a greater treatment effect using the risk ratio for wart resolution, whereas the trial by Focht et al1 found a greater treatment effect using the risk ratio for the wart remaining at the end of follow-up. A reasonable approach would be to report both risk ratios with their confidence intervals.
In their commentary, Van Cleve et al3 used power calculations to estimate that the study by de Haen et al2 had only 26% power to find a statistically significant difference between the observed treatment risk difference of 10% and a hypothetical risk difference of 0%. When the observed trial difference is not statistically significant, such a post hoc calculation will always estimate that power was less than 50%.18 Power calculations are an inefficient method for interpreting results, as they provide an estimate for only one assumed effect size compared with the null hypothesis. After a study is completed, the size of the effect can be estimated from the data, and confidence intervals allow us to see how compatible all possible effects are with the observed results.18 - 20
Editor's Note: Matthew M. Davis, MD, MAPP, has read this letter but declined to reply.
Correspondence: Dr Cummings, 250 Grandview Dr, Bishop, CA 93514 (peterc@u.washington.edu).
Financial Disclosure: None reported.
Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature
Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal
Instructions
Comments are moderated and will appear on the site at the discretion of the Archives of Pediatrics and Adolescent Medicine editors. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest* Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Register Now
Enter your username and email address. We'll send you a reminder to the email address on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.