Case-control methods in the presence of multiple failure times and competing risks

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Cohort and case-control studies are common epidemiological designs.
• The Cox proportional hazards model facilitates analysis of follow-up data.
• Linking cohort data to case-control designs can streamline analysis.

Purpose of the Study:

• To extend the linkage between cohort and case-control studies.
• To address competing risks and multiple event times in disease analysis.
• To simplify the analysis of exposure effects in large cohorts.

Main Methods:

• Utilizing risk sets from incident cases for control selection.
• Applying the Cox proportional hazards model.
• Extending case-control sampling to competing risks and multiple event times.

Main Results:

• Demonstrated the feasibility of analyzing cohort data as case-control studies.
• Showcased the extension of this linkage to competing risks.
• Illustrated the application to diseases with multiple incidence or recurrence times.

Conclusions:

• The linkage between cohort and case-control studies is extendable to complex scenarios.
• Sampling controls from risk sets offers an efficient analytical approach.
• This methodology simplifies the analysis of exposure effects in epidemiological research.