Use of Zero-Inflated Mixture Models to Compare Antibody Titers in Response to H1N1 Vaccination

• Activity Number: 412
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Consulting
• Author(s): Leela M. Aertker*+ and Daniel J. Zaccaro
• Companies: Rho and Rho
• Keywords: Vaccine ; Zero-Inflated Model ; Asthma ; Antibodies ; Influenza ; H1N1

Abstract

Pandemic H1N1 influenza vaccine was administered to participants with mild/moderate and severe asthma. Hemagglutination inhibition antibody titers were measured pre-vaccination (Day 1) and post-vaccination (Day 21). A preponderance of antibody titers below the lower limit of detection (36-75% of observations at Day 1, depending on subgroup) and antibody titers above the upper limit of detection (6-52% of observations at Day 21, depending on subgroup) were observed. Zero-inflated log-normal models that accounted for left- and right-censoring and a "point mass" were utilized to compare subgroups with respect to antibody titers. Results derived from the traditional analytical method of ordinary least squares (OLS) linear regression with imputation of censored values were compared to results from zero-inflated methods. By formal criteria, zero-inflated mixture models provided a better fit to the data and yielded results that were qualitatively and quantitatively different from the naïve models using OLS linear regression. Zero-inflated mixture models elucidated differences among subgroups that were otherwise obscured by traditional methods.