Activity Number:
|
126
- Diagnostics, Classification, and Agreement
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 31, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Medical Devices and Diagnostics
|
Abstract #322878
|
|
Title:
|
Sample Size and Confidence Interval Planning for Estimating the Optimal Thresholds in a Sequence of Diagnostic Tests
|
Author(s):
|
Christine M Schubert* and Donna McClish and Amber R Wilk
|
Companies:
|
Air Force Institute of Technology/ENC and VCU and Virginia Commonwealth University
|
Keywords:
|
diagnostic test ;
sequence ;
optimal thresholds ;
BP ;
BN ;
sample size
|
Abstract:
|
In this work, we present a method to estimate the width of confidence intervals (CIs) for the thresholds associated with the optimal operating point of a planned study of a sequence of diagnostic tests. We assume an anticipated total sample size of 40-200 subjects and known biomarkers. Two diagnostic test sequences are considered: Believe the Positive (BP) and Believe the Negative (BN). We present quick computational equations that provide reasonable estimates of the resulting CI widths for the thresholds of each biomarker in both the BP and the BN diagnostic sequence. These formulas are based upon characteristics of the diseased and non-diseased population as well as the ratio of patients with/without disease and the correlation between tests. These widths may be used in the planning process to determine a priori feasibility of a study designed to estimate the optimal thresholds in a sequence of diagnostic tests. The computational equations provided are demonstrated using data examples of potential BP and BN sequences.
|
Authors who are presenting talks have a * after their name.