|
Activity Number:
|
281
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #306289 |
|
Title:
|
Evaluation of Three Approaches To Correct for Ascertainment of Pedigrees for Random-Effects Cox Proportional Hazard Linkage Analysis
|
|
Author(s):
|
Susan Slager*+ and Shannon K. McDonnell and Vernon S. Pankratz and Antje Hoering and Terry M. Therneau and Mariza de Andrade
|
|
Companies:
|
Mayo Clinic College of Medicine and Mayo Clinic College of Medicine and Mayo Clinic College of Medicine and Cancer Research and Biostatistics and Mayo Clinic College of Medicine and Mayo Clinic College of Medicine
|
|
Address:
|
Division of Biostatistics, 200 1st Street, SW, Rochester, MN, 55905,
|
|
Keywords:
|
genetics ; age-at-onset ; linkage analysis
|
|
Abstract:
|
Many diseases have variable age of onset. By accounting for this variability in a linkage analysis, one may be able to obtain more statistical power to identify susceptibility loci than had one analyzed the data as a binary trait. Several analytical age-at-onset linkage approaches exist, one of which is the random effects Cox proportional hazard model. This approach is an extension of the Cox proportional hazard regression that includes both fixed and random effects. Recently, Pankratz et al. have found parameter estimates obtained from this random-effects hazard model are biased. This bias may be due to several factors, including pedigree ascertainment. We hypothesize that correcting for ascertainment may improve the parameters estimates. We investigate three approaches to correct for pedigree ascertainment and compare them to the approach that makes no correction.
|
