Abstract
Diagnostic tests are used in a wide range of behavioral, medical, psychosocial, and healthcare-related research. Test sensitivity and specificity are the most popular measures of accuracy for diagnostic tests. Available methods for analyzing longitudinal study designs assume fixed gold or reference standards and as such do not apply to studies with dynamically changing reference standards, which are especially popular in psychosocial research. In this article, we develop a novel approach to address missing data and other related issues for modeling sensitivity and specificity within such a time-varying reference standard setting. The approach is illustrated with real as well as simulated data.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1213-1230 |
| Number of pages | 18 |
| Journal | Journal of Applied Statistics |
| Volume | 37 |
| Issue number | 7 |
| DOIs | |
| State | Published - Jul 2010 |
Funding
This research was supported in part by NIH grants R01-DA012249, U54 RR023480, and UL1 RR024160. We sincerely thank Ms. Cheryl Bliss-Clark at the University of Rochester, two anonymous reviewers, and Editor Prof. Robert Aykroyd for their constructive comments to help improve the presentation of the article.
Keywords
- Augmented inverse probability weighted (AIPW) estimate
- Bivariate monotone missing data pattern (BMMDP)
- Diagnostic test
- Double robust estimate
- Inverse probability weighted (IPW) estimate
- Missing data
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty