Related Experiment Videos
Flexible covariate effects in the proportional hazards model
T Hastie1, L Sleeper, R Tibshirani
1Statistics and Data Analysis Research Department, AT&T Bell Laboratories, Murray Hill, New Jersey.
Breast Cancer Research and Treatment
|January 1, 1992
Summary
This study introduces flexible semiparametric models for analyzing right-censored clinical trial data. These models allow prognostic factors to influence outcomes in nonlinear ways, improving treatment effect analysis.
Area of Science:
- Biostatistics
- Clinical Trials
- Survival Analysis
Background:
- The proportional hazards model is standard for analyzing right-censored clinical trial data.
- This model measures treatment effects and adjusts for prognostic factors.
- Existing models may not fully capture complex relationships between prognostic factors and outcomes.
Purpose of the Study:
- To present a class of semiparametric models for clinical trial analysis.
- To allow modeling of prognostic factors with nonlinear effects.
- To enable data to inform the functional form of prognostic factor effects.
Main Methods:
- Development of semiparametric models.
- Incorporation of nonlinear modeling for prognostic factors.
- Application to right-censored survival data.
Main Results:
- The proposed models effectively handle right-censored data.
- Nonlinear effects of prognostic factors can be identified and modeled.
- The methods were successfully illustrated using breast cancer clinical trial data.
Conclusions:
- Semiparametric models offer a flexible approach to survival data analysis in clinical trials.
- These models enhance the ability to understand and adjust for prognostic factors.
- The methodology provides a valuable tool for clinical trial research, particularly in oncology.