A MIXED GAMMA-MODEL FOR REGRESSION-ANALYSES OF QUANTITATIVE ASSAY DATA

Numerous biological factors can modify an individual's degree of immun e response to vaccine. Such factors may complicate an immunogenicity t rial by acting as counfounding variables; alternatively, their relatio nship to the measured antibody response may be the primary focus of an investigation. Standard regression analyses can adjust for many varia bles simultaneously and assess their relative importance, but require several conditions or assumptions. To reduce these requirements, we pr esent a flexible regression model that allows for: (I) a broad class o f shapes for the response distribution, (2) censoring of observations clue to detection limits,. and (3) the existence of a separate distrib ution of low-responders. We illustrate this modeling approach with neu tralizing antibody, data from a factorial study of measles vaccine. Th e effects of vaccine dose and strain, obscured by standard analyses, a re elucidated by the new model. Copyright (C) 1996 Elsevier Science Lt d.