Abstract
AimsThe 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.DesignWe 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.SettingWe apply our general method to data from two smoking cessation trials.ParticipantsA total of 489 and 1758 participants from two smoking cessation trials.MeasurementsThe abstinence outcomes were obtained using telephone interviews.FindingsThe 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.ConclusionsA 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.
Highlights
Missing outcome data are a common problem in randomized controlled trials
Because smoking cessation trials have this standard approach for handling missing outcome data, we use smoking as our example and incorporate the Russell Standard into our methods
Based on an informal review, we estimate that around 80% of reports of smoking cessation trials make the ‘missing = smoking’ assumption
Summary
A general method for handling missing binary outcome data in randomized controlled trials.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have