Related Experiment Videos
Using missing data techniques to explore the lack of survival effect: illustration with the CABG patch trial
J N Rottman1, B Levin, M C Paik
1Departments of Medicine (Cardiology) and Pharmacology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA. jeff.rottman@mcmail.vanderbilt.edu
Statistics in Medicine
|August 10, 1999
Summary
The Multicenter Automatic Defibrillator Implantation Trial (MADIT) found implantable defibrillators reduced mortality by 54% in patients with inducible ventricular tachycardia (VT) and impaired heart function. The Coronary Artery Bypass Graft (CABG) Patch Trial showed no benefit, likely due to fewer VT-inducible patients.
Area of Science:
- Cardiology
- Clinical Trials
- Biostatistics
Background:
- The Multicenter Automatic Defibrillator Implantation Trial (MADIT) demonstrated a significant mortality reduction with implantable defibrillators in specific patient groups.
- The Coronary Artery Bypass Graft (CABG) Patch Trial, with a similar patient population, did not show a mortality benefit, highlighting a potential difference in patient selection criteria.
Purpose of the Study:
- To statistically estimate the proportion of patients in the CABG Patch Trial who were inducible for ventricular tachycardia (VT) at randomization.
- To reconcile the differing mortality outcomes between the MADIT and CABG Patch trials based on VT inducibility.
Main Methods:
- Utilized a statistical mixture model approach to estimate the prevalence of VT inducibility in the CABG Patch Trial control population.
- Employed missing data techniques to estimate the 'mixing percentage' of inducible patients.
- Developed a mathematical and physiological basis for the mixture model and its estimation.
Main Results:
- Presented a method to accurately estimate the fraction of CABG Patch patients who were VT inducible.
- The developed mixture model provides a basis for the CABG Patch Trial's electrophysiological substudy.
- This approach aims to explain the discrepancy in mortality benefits observed between the two trials.
Conclusions:
- The difference in the frequency of inducible VT between MADIT and CABG Patch trial populations may account for the observed differences in mortality benefit from implantable defibrillators.
- This statistical approach offers a framework for analyzing patient subgroups in clinical trials.