Contact Us


Frequently Asked Questions

ETD Help

Policies and Procedures

Copyright and Patents

Access Restrictions

Search ETDs:
Advanced Search
Browse by:
Browse ProQuest
Search ProQuest

Laney Graduate School

Rollins School of Public Health

Candler School of Theology

Emory College

Emory Libraries

New ETD website is now LIVE and located here:

The Structure of the Prodrome of Psychosis

Shi, Meng (2013)
Master's Thesis (37 pages)
Committee Chair / Thesis Adviser: Hanfelt, John
Committee Members: Huang, Eugene
Research Fields: Biology, Biostatistics; Health Sciences, Public Health
Partnering Agencies: Does not apply (no collaborating organization)
Keywords: Psychosis; Latent Class Analysis
Program: Rollins School of Public Health, Biostatistics and Bioinformatics (Biostatistics)
Permanent url:


Psychotic disorder is a group of serious illness that affects the mind. The symptoms are severe and it affects over 5% of the population. Among illness that affect people aged in 15 to 44, schizophrenia is the 8th leading cause of the disability worldwide. The first aim of this analysis is to conduct a latent class analysis on the clinical characteristics of adolescents at high risk of psychosis using the software Latent Gold. The second aim is to model the time to onset of psychosis in high-risk youth based on latent classes of comprehensive clinical information and socio-demographic variables. In this analysis, we used the Latent Class Analysis (LCA) approach to analyze variables collected as the North American Prodrome Longitudinal Study (NAPLS). The results showed that the four-class model was preferred according to the model selection criteria, such as AIC, BIC, and ICL-BIC. Based on these four subgroups, a proportional hazards model was used to characterize the relationship. In comparing the proportional hazards regression models with and without covariate measurement error, we found that the standard errors of the coefficients in the model with measurement error are smaller than the ones without measurement error.

Table of Contents

1. Introduction 1
2. Method 6
2.1 Coding Psychosis 6
2.2 Latent Class Analysis 7
2.2.1 Basic components of a Latent Class Cluster model 7
2.2.2 Probability Structure 8
2.2.3 Conditional distributions 9
2.2.4 Latent variable 10
2.2.5 Local independence 11
2.3 Proportional hazards model 12
2.4 Measurement with error 14
3. Results 17
4. Discussion 21
4.1 Discussion of results 21
4.2 Future work 21
5. Reference 24
6. Appendices 28
Appendix 1 28
Appendix 2 29


application/pdf Master Thesis 37 pages (11.1 MB) [Access copy of Dissertation/Thesis]
Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.