Online Program Home
  My Program

Abstract Details

Activity Number: 634 - Dynamic Modeling for Timely Health Care Decisions
Type: Invited
Date/Time: Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
Sponsor: WNAR
Abstract #321979
Title: Dynamic Landmark Prediction and Model Selection for Genetic Mixture Models
Author(s): Tanya Garcia* and Layla Parast
Companies: Texas A&M University and RAND Corporation
Keywords: Dynamic modeling ; Landmark Prediction ; Huntington's disease ; Kin-cohort study
Abstract:

In kin-cohort studies, clinicians and genetic counselors are interested in providing their patients the most current cumulative risk of a disease arising from a rare deleterious mutation. Estimating the cumulative risk is difficult, however, when the genetic mutation status in patients is unknown and, instead, only estimated probabilities of a patient having the mutation are available. We propose to estimate the cumulative risk for this scenario using a novel, consistent nonparametric estimator that incorporates landmark intermediate events. The landmark events provide additional information about the disease process and ultimately allows us to update predicted probabilities conditional on survival up to the landmark time. Our contributions are three-fold. First, we better inform patients of their disease risk by using the landmark events to dynamically adjust the estimated probability a patient is at risk. Second, we provide personalized dynamic predictions over time that can be incorporated into treatment or diagnostic decision-making. Third, we achieve more accurate predictions by incorporating patient covariate information even when genetic mutation status is unknown.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association