Subject-Specific Analysis of Longitudinal Binary Data from Outcome-Dependent Sampling Designs

• Activity Number: 318
• Type: Invited
• Date/Time: Tuesday, August 2, 2011 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): John Neuhaus*+ and Alastair Scott and Chris J. Wild and Charles McCulloch
• Companies: University of California at San Francisco and University of Auckland and University of Auckland and University of California at San Francisco
• Keywords: Mixed-effects models ; Profile likelihood ; Retrospective sampling

Abstract

Investigators often use case-control or, more generally, outcome dependent sampling designs to study rare binary outcomes. In a series of papers, Neuhaus, Scott and Wild developed a profile/pseudo-likelihood approach that addresses the clustering and differential sampling rates that such designs feature. Longitudinal follow up of subjects gathered in an initial outcome dependent sample can be used to study the trajectories of responses over time and to assess the association of changes in predictors within subjects with change in response. This talk shows that by augmenting the response of the original profile/pseudo-likelihood approach we can extend it to accommodate longitudinal data from a wide variety of complex and novel outcome dependent sampling designs. Data from a study of Attention Deficit Hyperactivity Disorder in children motivates the work and illustrates the findings.