Sample size for individually matched case-control studies

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Standard sample size calculations for matched case-control studies assume uniform exposure probability among controls.
• This assumption conflicts with the rationale for matching, where matching factors often create subgroups with differing exposure proportions.

Purpose of the Study:

• To develop and present novel formulas for calculating sample size in individually matched case-control studies.
• To address the inconsistency arising from the homogeneity assumption in existing formulas when applied to matched designs.

Main Methods:

• Derivation of new sample size formulas that incorporate heterogeneity of exposure probability across different matched sets.
• Comparison of proposed formulas with standard formulas, highlighting the impact of exposure variability.

Main Results:

• The proposed formulas provide a more consistent approach to sample size calculation in matched case-control studies.
• Illustrative examples demonstrate that heterogeneity in control exposure significantly impacts the required sample size.

Conclusions:

• The developed formulas offer a more accurate method for determining sample size in matched case-control studies with heterogeneous control populations.
• Accounting for exposure heterogeneity is crucial for robust study design and valid inferences in epidemiological research.