Abstract

In veterinary biologics, clinical studies conducted to support the licensure of a vaccine generally include a demonstration of efficacy in the species of interest. Typically, these studies are designed to assess a vaccine’s ability to prevent or mitigate clinical disease. Study designs utilize two or more treatment groups, and often incorporate blocking structure restrictions to accommodate animal housing or litter-related effects. When assessing a vaccine’s ability to prevent clinical disease, the prevented fraction (PF), a function of the group proportions of affected animals, is often utilized. Typically the sample size per treatment group is limited, and each block is represented by only a few experimental units per treatment group. Thus, it is a common occurrence for group proportion estimates to be 0 or 1 at the block level. Typical methods utilized in analyzing study data include generalized linear mixed model with delta method for confidence interval (GLMM), Cochran-Mantel-Haenszel (CMH) and Gart & Nam (GN). Through simulation, we compare the performance characteristics (power, bias, coverage) of these methods for a range of study designs, sample sizes and PF values, including an assessment of type 1 error rates. Simulation results suggest CMH generally performs well whereas the GN can perform poorly with regards to type 1 error. GLMM performance varies, depending on the simulated situation. Further, upon closer investigation of GN simulated results, it was determined that the method does not always result in a unique solution.

Keywords

Prevented Fraction, binomial response

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
May 1st, 1:00 PM

SIMULATION COMPARISON OF STATISTICAL METHODS USED IN ASSESSING VACCINE EFFICACY IN VETERINARY BIOLOGICS

In veterinary biologics, clinical studies conducted to support the licensure of a vaccine generally include a demonstration of efficacy in the species of interest. Typically, these studies are designed to assess a vaccine’s ability to prevent or mitigate clinical disease. Study designs utilize two or more treatment groups, and often incorporate blocking structure restrictions to accommodate animal housing or litter-related effects. When assessing a vaccine’s ability to prevent clinical disease, the prevented fraction (PF), a function of the group proportions of affected animals, is often utilized. Typically the sample size per treatment group is limited, and each block is represented by only a few experimental units per treatment group. Thus, it is a common occurrence for group proportion estimates to be 0 or 1 at the block level. Typical methods utilized in analyzing study data include generalized linear mixed model with delta method for confidence interval (GLMM), Cochran-Mantel-Haenszel (CMH) and Gart & Nam (GN). Through simulation, we compare the performance characteristics (power, bias, coverage) of these methods for a range of study designs, sample sizes and PF values, including an assessment of type 1 error rates. Simulation results suggest CMH generally performs well whereas the GN can perform poorly with regards to type 1 error. GLMM performance varies, depending on the simulated situation. Further, upon closer investigation of GN simulated results, it was determined that the method does not always result in a unique solution.