|
Activity Number:
|
28
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #320673
|
|
|
Title:
|
Joint Dichotomization of Continuous Variables to Discriminate Disease Status
|
|
Author(s):
|
Bethany Wolf* and Sybil Prince-Nelson and Paul J. Nietert and Viswanathan Ramakrishnan and Paula S. Ramos and Diane Kamen
|
|
Companies:
|
Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina
|
|
Keywords:
|
dichotomization ;
optimal thresholds ;
decision tools
|
|
Abstract:
|
Continuous variables are often dichotomized to develop decision tools in clinical practice. Development of complex diseases may be influenced by interactions between genetic and environmental factors. Appropriate clinical management of a patient requires consideration of these factors when assessing patient risk for disease. If interactions between two continuous factors lead to increased disease risk, then dichotomizing them independently may lead to an incorrect choice of thresholds. Many methods exist for optimizing a threshold to dichotomize a continuous variable to discriminate a binary outcome; however, there is limited methodology for simultaneously optimizing thresholds for multiple variables. We propose a method for jointly dichotomizing two variables and conduct simulations to compare joint to independent dichotomization. Our results show that thresholds selected jointly exhibit smaller mean square error and similar bias relative to those selected independently. This method can help in optimizing patient management strategies.
|
Authors who are presenting talks have a * after their name.
