Year: 2016
Type: dissertation
Abstract: This thesis is concerned with statistical modeling and prediction of disease processes subject to intermittent \nobservation. \nTimes of disease progression are interval-censored when progression status is only known at a series of \nassessment times. \nThis situation arises routinely in clinical trials and cohort studies when events of interest are only \ndetectable upon imaging, based on blood tests, or upon careful clinical examination. \nThe work that follows is motivated by the study of demographic, genetic and clinical data available from the \nUniversity of Toronto Psoriasis \nRegistry and the University of Toronto Psoriatic Arthritis Registry, each involving cohorts of several hundred \npatients with the respective diseases. \n \nChapter 2 deals with the problem of selecting important prognostic biomarkers from a large set of candidates \nbiomarkers when the status with respect to an event of interest (e.g. disease progression) is only known at \nirregularly spaced and individual-specific assessment times. \nPenalized regression techniques (e.g. LASSO, adaptive LASSO and SCAD) are adapted to deal with the \ninterval-censored event times arising from this observation scheme. \nAn expectation-maximization algorithm is developed which is demonstrated to perform well in extensive simulation studies \ninvolving independent and correlated continuous and binary covariates. \nApplication to the motivating study of the development of arthritis mutilans in patients with psoriatic arthritis \nis given and several important human leukocyte antigen (HLA) variables are identified for further \ninvestigation. \nExtensions of this algorithm are developed for settings in which data from different sources \nwith distinct disease-related entry conditions are to be synthesized. \nThe extended Turnbull-type expectation-maximization algorithm is based on a complete data likelihood which \nincorporates missing information from individuals not meeting the entry criteria of the respective registries. \nSimulation studies demonstrate good empirical performance and an application to the motivating study identifies \nHLA markers associated with the onset of psoriatic arthritis among individuals with psoriasis. \nThis analysis is carried out using data from a psoriasis registry in which the times to psoriatic arthritis are left-truncated, and psoriatic arthritis registry in which the onset times are right-truncated. \n \nChapter 3 deals with the challenge of assessing the accuracy of a predictive model when response times are \ninterval-censored. \nInverse probability weighted (IPW) and augmented inverse probability weighted (AIPW) estimators of predictive accuracy are developed and evaluated based on the \nmean prediction error and the area under the receiver operating characteristic curve. \nThe weights are estimated from a multistate model which jointly considers the event process, the inspection process, \nand the right-censoring processes. \nWe investigate the performance of the proposed methods by simulation and illustrate their application in the context \nof a motivating rheumatology study in which HLA markers are used for predicting disease progression in \npsoriatic arthritis. \n \nA two-phase model is developed in Chapter 4 for chronic diseases which feature an indolent phase followed by a \nphase with more active disease resulting in progression and damage. \nThe time-scales for the intensity functions for the active phase are more naturally based on the time since the \nstart of the active phase, corresponding to a semi-Markov formulation. \nIn cohort studies for which the disease status is only known at a series of clinical assessment times, \ntransition times are interval-censored which means the time origin for phase II is interval-censored. \nWeakly parametric models with piecewise constant baseline hazard and rate functions are specified and an \nexpectation-maximization algorithm is described for model fitting. \nA computationally faster two-stage estimation procedure is also developed and the asymptotic variances of the \nresulting estimators are derived. \nSimulation studies examining the performance of the proposed model show good performance under both \nmaximum likelihood and two-stage estimation. \nAn application to data from the motivating study of disease progression in psoriatic arthritis illustrates \nthe procedure, and identifies new human leukocyte antigens associated with the duration of the indolent phase, and others associated with disease progression in the active phase. \n \nOpen problems and topics for ongoing and future research are discussed in Chapter 5.
Source: UWSpace (University of Waterloo)
Author: Ying Wu
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2515440222