Related Experiment Videos
Nested case-control and case-cohort methods of sampling from a cohort: a critical comparison
1Department of Preventive Medicine, University of Southern California, School of Medicine, Los Angeles 90033-9987.
American Journal of Epidemiology
|January 1, 1990
Summary
The case-cohort design offers less efficiency than expected for intervention trials without random censoring. Simulations show it can perform worse than the nested case-control method with censoring or staggered entry.
Area of Science:
- Epidemiology
- Biostatistics
- Clinical Trials
Background:
- The case-cohort design is a recent sampling method for cohort studies.
- The nested case-control method is a traditional alternative for efficient case-control sampling within cohorts.
Purpose of the Study:
- To compare the relative efficiency of the case-cohort design versus the nested case-control design.
- To evaluate the performance of the case-cohort method under various censoring and entry conditions.
Main Methods:
- Asymptotic relative efficiency calculations were performed for both designs.
- Simulations were conducted to assess performance with random censoring and staggered entry.
Main Results:
- Corrected asymptotic relative efficiency indicates less improvement for the case-cohort design in intervention trials without random censoring.
- Simulations reveal that the case-cohort method can be substantially less efficient than the nested case-control method when moderate random censoring or staggered entry is present.
Conclusions:
- The case-cohort design may not offer the anticipated efficiency gains in all scenarios, particularly in intervention trials without random censoring.
- The nested case-control method may be preferable when dealing with moderate random censoring or staggered entry in cohort studies.