Sample sizes for estimation of exposure-specific disease rates in population-based case-control studies

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Accurate estimation of exposure-specific disease rates is vital for understanding disease etiology.
• Population-based case-control studies are a common design for investigating disease risk factors.
• Sample size determination is a critical step in study design to ensure adequate statistical power and precision.

Purpose of the Study:

• To derive formulas for calculating optimal sample sizes in population-based case-control studies.
• To determine sample sizes for precise estimation of exposure-specific disease rates.
• To minimize total study size while achieving desired precision in rate estimation.

Main Methods:

• Utilized Neutra and Drolette's confidence limits based on the log-ratio of binomial exposure rates.
• Derived expressions for optimal case and control sample sizes.
• Analyzed the impact of case exposure proportion on the precision of disease rate estimation.

Main Results:

• Precision of disease rate estimation depends on the proportion of exposed cases.
• Optimal control-to-case ratio equals the square root of the exposure odds ratio.
• For rare diseases, precise estimation of exposed disease rates requires smaller sample sizes than estimating exposure odds ratios.

Conclusions:

• The study provides a method for optimizing sample size allocation in case-control studies.
• Findings offer guidance for researchers aiming for precise estimation of exposure-specific disease rates.
• Optimized sample sizes can lead to more efficient and cost-effective epidemiological research.