Exact equivalence test for risk ratio and its sample size determination under inverse sampling

Area of Science:

• Biostatistics
• Epidemiology
• Statistical Methods

Background:

• Equivalence testing is crucial for comparing risks in dichotomous data.
• Inverse sampling offers a valuable approach for specific study designs.
• Existing methods may not fully address risk ratio equivalence under inverse sampling.

Purpose of the Study:

• To develop an exact equivalence test for the risk ratio using inverse sampling.
• To explore the relationship between this test and conditional confidence limits.
• To provide procedures for determining the minimum required sample size for desired statistical power.

Main Methods:

• Development of an exact equivalence test tailored for risk ratios with inverse sampling.
• Investigation of the link between the exact equivalence test and exact conditional confidence limits.
• Formulation of exact and asymptotic procedures for sample size calculations.

Main Results:

• An exact equivalence test for risk ratios under inverse sampling has been developed.
• The relationship between the proposed test and conditional confidence limits is elucidated.
• Procedures and a summary table for minimum sample size determination are presented.

Conclusions:

• The proposed exact equivalence test is a valuable tool for risk ratio assessment in dichotomous data with inverse sampling.
• Accurate sample size calculations are provided to ensure adequate statistical power.
• This work offers practical guidance for researchers in biostatistics and epidemiology.