Related Experiment Videos
A more flexible regression-to-the-mean model with possible stratification
1Department of Preventive Medicine, Rush-Presbytrian-St. Luke's Medical Center, Chicago, Illinois 60612-3824, USA. schen@bstat.pvm.rpslmc.edu
Biometrics
|September 29, 1998
Summary
This study introduces a regression-to-the-mean model with additive and multiplicative treatment effects, applicable to screening trials with large initial samples. Efficient, simple estimators are proposed for stratified treatment effects based on initial measurements.
Area of Science:
- Statistics
- Biostatistics
- Clinical Trial Design
Background:
- Regression-to-the-mean phenomena are common in screening trials.
- Estimating treatment effects in such settings requires specialized models.
- Stratification by initial measurements can improve precision.
Purpose of the Study:
- To develop a regression-to-the-mean model incorporating additive and multiplicative treatment effects.
- To address scenarios with large initial samples and smaller second-stage subsamples.
- To propose efficient and computationally simple estimators for model parameters.
Main Methods:
- A regression-to-the-mean model framework was established.
- Treatment effects were allowed to be stratified by ranges of the first measurement.
- Asymptotically efficient estimators were derived for model parameters.
Main Results:
- The proposed estimators are computationally simple and efficient.
- The model accommodates both additive and multiplicative stratified treatment effects.
- The methodology is demonstrated using a large screening trial example.
Conclusions:
- The developed model and estimators are suitable for analyzing screening trial data.
- Stratification by initial measurements enhances the analysis of treatment effects.
- The approach provides a practical tool for statistical inference in large-scale screening studies.