Missing Data in Longitudinal-controlled Clinical Trials: A Power Comparison for Intent-to-Treat Analysis
Chakraborty, H. (2005, August). Missing Data in Longitudinal-controlled Clinical Trials: A Power Comparison for Intent-to-Treat Analysis. Presented at Joint Statistical Meetings, Minneapolis, MN.
Missing values and drop-outs are common issues in the longitudinal studies in all areas of medicine and public health and intent-to-treat (ITT) analysis has become widely accepted method for the analysis of controlled clinical trials in the pharmaceutical industry. In most controlled clinical trials, some patients do not complete their intended follow-up according to the protocol for a variety of reasons, generate missing values. Missing values leads to concern and confusion on identifying the ITT population, which makes the data analysis more complex and challenging. There is no adequate strategy for ITT analyses of longitudinal controlled clinical trial data with missing values. Several ad hoc strategies for dealing with missing values for an ITT analysis are common in the practice of controlled clinical trials. A detailed investigation based on simulation studies was performed to develop recommendations for ITT analysis of longitudinal controlled clinical trial data with missing values. We compared sizes, and powers between some popular ad hoc approaches and linear mixed model approach under different missing scenarios. Our results suggest that for the high percentage of missing values mixed model approach without any ad hoc imputation provide more powerful comparison.