A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes

Longitudinal growth patterns are routinely seen in medical studies where developments of individuals on one or more outcome variables are followed over a period of time. Many current methods for modeling growth presuppose a parametric relationship between the outcome and time (e.g., linear, quadrati...

Full description

Bibliographic Details
Main Author: Kliethermes, Stephanie Ann
Other Authors: Oleson, Jacob J.
Format: Others
Language:English
Published: University of Iowa 2013
Subjects:
Online Access:https://ir.uiowa.edu/etd/2546
https://ir.uiowa.edu/cgi/viewcontent.cgi?article=4675&context=etd
id ndltd-uiowa.edu-oai-ir.uiowa.edu-etd-4675
record_format oai_dc
spelling ndltd-uiowa.edu-oai-ir.uiowa.edu-etd-46752019-11-13T09:14:12Z A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes Kliethermes, Stephanie Ann Longitudinal growth patterns are routinely seen in medical studies where developments of individuals on one or more outcome variables are followed over a period of time. Many current methods for modeling growth presuppose a parametric relationship between the outcome and time (e.g., linear, quadratic); however, these relationships may not accurately capture growth over time. Functional mixed effects (FME) models provide flexibility in handling longitudinal data with nonparametric temporal trends because they allow the data to determine the shape of the curve. Although FME methods are well-developed for continuous, normally distributed outcome measures, nonparametric methods for handling categorical outcomes are limited. In this thesis, we propose a Bayesian hierarchical FME model to account for growth curves with non-Gaussian outcomes. In particular, we extend traditional FME models which assume normally distributed outcomes by modeling the probabilities associated with the binomially distributed outcomes and adding an additional level to the hierarchical model to correctly specify the outcomes as binomially distributed. We then extend the proposed binomial FME model to the multinomial setting where the outcomes consist of more than two nominal categories. Current modeling approaches include modeling each category of a multinomial outcome separately via linear and nonlinear mixed effects models; yet, these approaches ignore the inherent correlation among the categories of the outcome. Our model captures this correlation through a sequence of conditional binomial FME models which results in one model simultaneously estimating probabilities in all categories. Lastly, we extend our binomial FME model to address a common medical situation where multiple outcomes are measured on subjects over time and investigators are interested in simultaneously assessing the impact of all outcomes. We account for the relationship between outcomes by altering the correlation structure in the hierarchical model and simultaneously estimating the outcome curves. Our methods are assessed via simulation studies and real data analyses where we investigate the ability of the models to accurately predict the underlying growth trajectory of individuals and populations. Our applications include analyses of speech development data in adults and children with cochlear implants and analyses on eye-tracking data used to assess word processing in cochlear implant patients. 2013-01-01T08:00:00Z dissertation application/pdf https://ir.uiowa.edu/etd/2546 https://ir.uiowa.edu/cgi/viewcontent.cgi?article=4675&context=etd Copyright 2013 Stephanie Kliethermes Theses and Dissertations eng University of IowaOleson, Jacob J. Bayesian Cochlear Implants Functional Mixed Effects Growth Curves Longitudinal Visual World Paradigm Biostatistics
collection NDLTD
language English
format Others
sources NDLTD
topic Bayesian
Cochlear Implants
Functional Mixed Effects
Growth Curves
Longitudinal
Visual World Paradigm
Biostatistics
spellingShingle Bayesian
Cochlear Implants
Functional Mixed Effects
Growth Curves
Longitudinal
Visual World Paradigm
Biostatistics
Kliethermes, Stephanie Ann
A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
description Longitudinal growth patterns are routinely seen in medical studies where developments of individuals on one or more outcome variables are followed over a period of time. Many current methods for modeling growth presuppose a parametric relationship between the outcome and time (e.g., linear, quadratic); however, these relationships may not accurately capture growth over time. Functional mixed effects (FME) models provide flexibility in handling longitudinal data with nonparametric temporal trends because they allow the data to determine the shape of the curve. Although FME methods are well-developed for continuous, normally distributed outcome measures, nonparametric methods for handling categorical outcomes are limited. In this thesis, we propose a Bayesian hierarchical FME model to account for growth curves with non-Gaussian outcomes. In particular, we extend traditional FME models which assume normally distributed outcomes by modeling the probabilities associated with the binomially distributed outcomes and adding an additional level to the hierarchical model to correctly specify the outcomes as binomially distributed. We then extend the proposed binomial FME model to the multinomial setting where the outcomes consist of more than two nominal categories. Current modeling approaches include modeling each category of a multinomial outcome separately via linear and nonlinear mixed effects models; yet, these approaches ignore the inherent correlation among the categories of the outcome. Our model captures this correlation through a sequence of conditional binomial FME models which results in one model simultaneously estimating probabilities in all categories. Lastly, we extend our binomial FME model to address a common medical situation where multiple outcomes are measured on subjects over time and investigators are interested in simultaneously assessing the impact of all outcomes. We account for the relationship between outcomes by altering the correlation structure in the hierarchical model and simultaneously estimating the outcome curves. Our methods are assessed via simulation studies and real data analyses where we investigate the ability of the models to accurately predict the underlying growth trajectory of individuals and populations. Our applications include analyses of speech development data in adults and children with cochlear implants and analyses on eye-tracking data used to assess word processing in cochlear implant patients.
author2 Oleson, Jacob J.
author_facet Oleson, Jacob J.
Kliethermes, Stephanie Ann
author Kliethermes, Stephanie Ann
author_sort Kliethermes, Stephanie Ann
title A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
title_short A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
title_full A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
title_fullStr A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
title_full_unstemmed A Bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
title_sort bayesian nonparametric approach to modeling longitudinal growth curves with non-normal outcomes
publisher University of Iowa
publishDate 2013
url https://ir.uiowa.edu/etd/2546
https://ir.uiowa.edu/cgi/viewcontent.cgi?article=4675&context=etd
work_keys_str_mv AT kliethermesstephanieann abayesiannonparametricapproachtomodelinglongitudinalgrowthcurveswithnonnormaloutcomes
AT kliethermesstephanieann bayesiannonparametricapproachtomodelinglongitudinalgrowthcurveswithnonnormaloutcomes
_version_ 1719290803902218240