The application of latent class analysis and latent transition analysis to large scale disaster data: Modeling PTSD in a population of disaster workers
Item
-
Title
-
The application of latent class analysis and latent transition analysis to large scale disaster data: Modeling PTSD in a population of disaster workers
-
Identifier
-
d_2009_2013:d4bf02c234fe:11706
-
identifier
-
12285
-
Creator
-
Wyka, Katarzyna,
-
Contributor
-
Jay Verkuilen
-
Date
-
2013
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Quantitative psychology | Clinical psychology
-
Abstract
-
Sophisticated statistical methodologies are needed in order to analyze large, population-based datasets, such as screening projects, following disasters. Currently, the most common methodology applied to disaster research uses marginal or population-averaged models. However, mixture models with latent variables have versatile applications that may provide additional insight into the psychiatric outcomes following disasters and capture population heterogeneity that is usually overlooked. Thus, this dissertation conducts a novel application of these methodologies to a longitudinal database following a disaster.;In the wake of the terrorist attacks on September 11th, the Weill Cornell Screening Project conducted annual psychological screenings with over 3,000 non-rescue, World Trade Center (WTC) disaster workers from 2002-2008. This dissertation applies two types of categorical mixture models to this dataset: latent class analysis (LCA) and its longitudinal extension, latent transition analysis (LTA). Both models are particularly well suited for the analysis of psychiatric screening data, because they allow individuals to be grouped into classes based on their symptomatology. Furthermore, these methods permit the course of symptoms to be examined over time by modeling individuals' developmental trajectories.;The goal of this dissertation was to assess the utility and feasibility of applying LCA and LTA in large scale disaster research, specifically within a study of the longitudinal course of posttraumatic stress symptoms in WTC disaster workers. The LCA model successfully captured the heterogeneity of posttraumatic stress symptoms in this population. Additionally, the LTA model yielded unique information regarding patterns of symptom changes over time. The multiple-group analysis provided information about racial and ethnic differences in PTSD presentation and longitudinal course.;The application of LCA and LTA methodologies in this dissertation yielded practical findings in the field of psychiatric and disaster research. These findings have the potential to inform criteria selection for diagnostic manuals and offer insight into the mechanisms involved in the maintenance and remission of posttraumatic stress symptoms. Challenges associated with the analysis of complex longitudinal data from large screening databases and future directions are discussed.
-
Type
-
dissertation
-
Source
-
2009_2013.csv
-
degree
-
Ph.D.
-
Program
-
Educational Psychology
