Cost-Effective Extreme Case-Control Design Using a Resampling Method.

Area of Science:

• Epidemiology
• Biostatistics
• Medical Research

Background:

• Nested case-control designs are crucial for rare events in cohort studies.
• Conventional methods involve random control selection, potentially with matching, to manage confounding.
• Challenges include specifying risk sets and determining optimal control numbers.

Purpose of the Study:

• To propose a new nested case-control sampling design.
• To integrate extreme case-control design principles and resampling techniques.
• To overcome limitations of traditional nested case-control methods, particularly for rare outcomes.

Main Methods:

• Developed a novel algorithm combining extreme case-control design with resampling.
• Eliminated the need to specify risk sets at each event time.
• Enabled flexible control selection based on resource constraints, including under-sampling (fewer controls than cases).

Main Results:

• Simulation studies confirmed the algorithm's effectiveness.
• The method performs well even with a control-to-case ratio less than one.
• The algorithm was successfully applied to the "Thorotrast Study" dataset.

Conclusions:

• The proposed sampling design offers advantages over conventional nested case-control methods.
• It enhances efficiency and flexibility in selecting controls for rare events.
• This approach provides a robust tool for epidemiological research with rare outcomes.