BST 290 Seminar Series
PLEASE NOTE: This Seminar has been cancelled (5/26/15). We apologize for the inconvenience.
Tuesday, May 26th, 2015, 4:10pm, MSB 1147 (Colloquium Room)
Refreshments at 3:30pm in MSB 4110 (Statistics Lounge)
Speaker: Jonathan Schildcrout, Associate Professor, Department of Biostatistics, Vanderbilt University
Title: Targeted sampling designs to improve research efficiency with application to a longitudinal lung function study in children with asthma
Abstract: Substudies of the Childhood Asthma Management Program seek to identify patient characteristics associated with asthma symptoms and lung function. To determine if genetic measures are associated with trajectories of lung function as measured by forced vital capacity (FVC), children in the primary cohort study retrospectively had candidate loci evaluated. Given constraints on financial resources, it is often necessary in retrospective substudies to target a sub-sample for ascertainment of costly measures (e.g., IL10 cytokine polymorphisms from stored bloods samples). We will discuss considerations regarding whom to sample in longitudinal cohort sub-studies as well as analytical approaches once the sample has been obtained. We will show that targeted sampling combined with efficient analytical approaches are far more efficient than standard designs and analyses. Further, we will distinguish efficiency gains due to the design from those that are due to the efficient analytical procedures.