A simple method for analyzing matched designs with double controls: McNemar's test can be extended.

Area of Science:

• Biostatistics
• Epidemiology
• Statistical Methods

Background:

• Traditional matched study analyses often use one control per case.
• Extending these methods to accommodate multiple controls per case is crucial for robust statistical inference.
• Existing methods may not be optimal for one-to-two matched designs.

Purpose of the Study:

• To introduce a new analytic approach for matched studies with double controls (one case to two controls).
• To extend McNemar's test for one-to-two matching for binary predictors and outcomes.
• To offer a simple, efficient, and visually verifiable analytical tool.

Main Methods:

• Review of McNemar's test for matched data.
• Demonstration of the Mantel-Haenszel approach for integrating estimates.
• Introduction of a novel method with visual display and computational simplicity, compared against conditional logistic regression.

Main Results:

• The new approach was illustrated using real-world data on traffic crashes and weather.
• Results closely align with conditional logistic regression.
• The method is simple enough for handheld calculator computation and validated through simulations.

Conclusions:

• The developed approach offers a feasible, simple, and efficient method for analyzing matched designs with double controls.
• This extends existing statistical capabilities for epidemiological and biostatistical research.
• The method provides a practical alternative for analyzing complex matched data.