|
Activity Number:
|
185
- Novel Methods for Clinical Trial Design and Characterizing Heterogeneity
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Mental Health Statistics Section
|
|
Abstract #312282
|
|
|
Title:
|
Practice Effects: What, Why and How
|
|
Author(s):
|
Rajesh Talluri*
|
|
Companies:
|
University of Mississippi Medical Center
|
|
Keywords:
|
Practice Effects;
Dementia;
Longitudinal data;
Bayesian;
cognitive outcome
|
|
Abstract:
|
In dementia research, cognitive function and cognitive decline measures are necessary to evaluate novel treatments to improve cognitive function. It is therefore of paramount importance that cognitive measures reflect an individual’s true cognitive ability. As participants are exposed to the same cognitive test repeatedly, an individual may learn and adjust to the test(practice effect). Ignoring practice effects may lead to underestimation of cognitive decline and overestimation of baseline cognition function of an individual which may bias the results of the study. We propose a statistical framework to model the effect of practice effects on the cognitive measurement in a longitudinal cohort. The dependence of practice effects on time dependent covariates and the time elapsed between the tests was modeled. Simulations show that the proposed model is identifiable and the model parameter estimates are unbiased. Finally data from the Atherosclerosis Risk in Communities (ARIC) Neurocognitive Study cohort was analyzed using the proposed model to estimate the practice effects. The effect of practice effects on downstream analyses was evaluated using further simulations.
|
Authors who are presenting talks have a * after their name.
