Browsing by Subject "Huntington’s disease"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Bayesian quantile regression joint models: Inference and dynamic predictions.
    (Statistical methods in medical research, 2018-01) Yang, Ming; Luo, Sheng; DeSantis, Stacia
    In the traditional joint models of a longitudinal and time-to-event outcome, a linear mixed model assuming normal random errors is used to model the longitudinal process. However, in many circumstances, the normality assumption is violated and the linear mixed model is not an appropriate sub-model in the joint models. In addition, as the linear mixed model models the conditional mean of the longitudinal outcome, it is not appropriate if clinical interest lies in making inference or prediction on median, lower, or upper ends of the longitudinal process. To this end, quantile regression provides a flexible, distribution-free way to study covariate effects at different quantiles of the longitudinal outcome and it is robust not only to deviation from normality, but also to outlying observations. In this article, we present and advocate the linear quantile mixed model for the longitudinal process in the joint models framework. Our development is motivated by a large prospective study of Huntington's disease where primary clinical interest is in utilizing longitudinal motor scores and other early covariates to predict the risk of developing Huntington's disease. We develop a Bayesian method based on the location-scale representation of the asymmetric Laplace distribution, assess its performance through an extensive simulation study, and demonstrate how this linear quantile mixed model-based joint models approach can be used for making subject-specific dynamic predictions of survival probability.
  • Thumbnail Image
    ItemOpen Access
    Predicting the Risk of Huntington's Disease with Multiple Longitudinal Biomarkers.
    (Journal of Huntington's disease, 2019-06-22) Li, Fan; Li, Kan; Li, Cai; Luo, Sheng; PREDICT-HD and ENROLL-HD Investigators of the Huntington Study Group
    BACKGROUND:Huntington's disease (HD) has gradually become a public health threat, and there is a growing interest in developing prognostic models to predict the time for HD diagnosis. OBJECTIVE:This study aims to develop a novel prognostic model that leverages multiple longitudinal biomarkers to inform the risk of HD. METHODS:The multivariate functional principal component analysis was used to summarize the essential information from multiple longitudinal markers and to obtain a set of prognostic scores. The prognostic scores were used as predictors in a Cox model to predict the right-censored time to diagnosis. We used cross-validation to determine the best model in PREDICT-HD (n = 1,039) and ENROLL-HD (n = 1,776); external validation was carried out in ENROLL-HD. RESULTS:We considered six commonly measured longitudinal biomarkers in PREDICT-HD and ENROLL-HD (Total Motor Score, Symbol Digit Modalities Test, Stroop Word Test, Stroop Color Test, Stroop Interference Test, and Total Functional Capacity). The prognostic model utilizing these longitudinal biomarkers significantly improved the predictive performance over the model with baseline biomarker information. A new prognostic index was computed using the proposed model, and can be dynamically updated over time as new biomarker measurements become available. CONCLUSION:Longitudinal measurements of commonly measured clinical biomarkers substantially improve the risk prediction of Huntington's disease diagnosis. Calculation of the prognostic index informs the patient's risk category and facilitates patient selection in future clinical trials.