Activity Number:
|
153
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section*
|
Abstract - #300934 |
Title:
|
Modeling Grade IV Venous Gas Emboli using a Limited Failure Population Model with Random Effects
|
Author(s):
|
Laura Thompson*+ and Raj Chhikara and Johnny Conkin
|
Affiliation(s):
|
University of Houston, Clear Lake and University of Houston, Clear Lake and National Space Biomedical Research Institute
|
Address:
|
2700 Bay Area Blvd, Houston, Texas, 77058, USA
|
Keywords:
|
limited failure population ; decompression sickness ; cure rate ; interval censoring ; hypobaric environment
|
Abstract:
|
The presence of gas bubbles in venous blood (venous gas emboli, or VGE) is associated with an increased risk of decompression sickness (DCS) in hypobaric environments. A high grade of VGE (e.g., Grade IV) can be a precursor to serious DCS. We model time to Grade IV VGE considering a population that consists of a subset of individuals who are assumed never to experience the event. The data are from NASA's Hypobaric Decompression Sickness Databank, which contains results from volunteer subjects undergoing up to 13 denitrogenation test procedures prior to being exposed to a hypobaric environment. The onset time is recorded only as being contained within certain time intervals. We fit a lognormal mixture survival model to the interval- and right-censored data that accounts for the possibility of a subset of individuals who are immune to experiencing the event. Random subject effects are used to account for correlations between repeated measurements. Model assessments and cross-validation indicate that the data are better approximated by a limited failure population model than by a non-mixture model. We conclude with current work on a Bayesian implementation of our analysis.
|