A Comparative Study of Approaches for Predicting Prostate Cancer from Longitudinal Data

Abstract

Disease prediction based on longitudinal data can be done using various modeling approaches. Alternative approaches are compared using data from a longitudinal study to predict the onset of disease. The data are modeled using linear mixed-effects models. Posterior probabilities of group membership are computed starting with the first observation and sequentially adding observations until the subject is classified as developing the disease or until the last measurement is used. Individuals are classified by computing posterior probabilities using the marginal distributions of the mixed-effects models, the conditional distributions (conditional on the group-specific random effects), and the distributions of the random effects.

Keywords:

• Classification
• Cross Validation
• Linear mixed-effects models
• Sensitivity
• Specificity

Mathematics Subject Classification:

• 62J

Acknowledgment

We thank Dr. Dan L. Longo for encouraging us to pursue the comparison of different prediction approaches using the mixed model. This research was supported in part by the Intramural Research Program of the NIH, National Institute on Aging. We also thank the anonymous referees whose comments led to a great improvement of this article.

This article not subject to U.S. copyright law.

Notes

*Values in parentheses are from the 37 men predicted as cancers by all three approaches.

*Values in parentheses are from the 56 men predicted as cancers by all three approaches.

Note. The data priors are 0.6635, 0.2780, 0.0585.