A general method for handling missing binary outcome data in randomized controlled trials. (December 2014)

• Record Type: Journal Article
• Authors: Jackson, Dan; White, Ian R.; Mason, Dan; Sutton, Stephen
• Is Part Of: Addiction. Volume 109:Number 12(2014:Dec.)
• Page Start: 1986
• Page End: 1993
• Publication Date: 2014-12
• Subjects: Alcoholism -- Periodicals; Drug addiction -- Periodicals; 616.86
• DOI: 10.1111/add.12721
• Languages: English
• ISSNs: 0965-2140

Abstract

<abstract abstract-type="main"> <title>Abstract</title> <sec id="add12721-sec-0001" sec-type="section"> <title>Aims</title> <p>The analysis of randomized controlled trials with incomplete binary outcome data is challenging. We develop a general method for exploring the impact of missing data in such trials, with a focus on abstinence outcomes.</p> </sec> <sec id="add12721-sec-0002" sec-type="section"> <title>Design</title> <p>We propose a sensitivity analysis where standard analyses, which could include 'missing = smoking' and 'last observation carried forward', are embedded in a wider class of models.</p> </sec> <sec id="add12721-sec-0003" sec-type="section"> <title>Setting</title> <p>We apply our general method to data from two smoking cessation trials.</p> </sec> <sec id="add12721-sec-0004" sec-type="section"> <title>Participants</title> <p>A total of 489 and 1758 participants from two smoking cessation trials.</p> </sec> <sec id="add12721-sec-0005" sec-type="section"> <title>Measurements</title> <p>The abstinence outcomes were obtained using telephone interviews.</p> </sec> <sec id="add12721-sec-0006" sec-type="section"> <title>Findings</title> <p>The estimated intervention effects from both trials depend on the sensitivity parameters used. The findings differ considerably in magnitude and statistical significance under quite extreme assumptions about the missing data, but are reasonably consistent under more moderate assumptions.</p> </sec> <sec id="add12721-sec-0007" sec-type="section"> <title>Conclusions</title> <p>A new method for undertaking sensitivity analyses when handling missing data in trials with binary outcomes allows a wide range of assumptions about the missing data to be assessed. In two smoking cessation trials the results were insensitive to all but extreme assumptions.</p> </sec> </abstract>