Dissertation
USING PARALLEL LATENT GROWTH CURVE MODELS TO BETTER UNDERSTAND CO-MORBIDITY IN TWO RANDOMIZED CLINICAL TRIALS: SMOKING WITH CO-MORBID ADHD AND CO-MORBID SUD
Doctor of Philosophy (PhD), Washington State University
01/2015
Handle:
https://hdl.handle.net/2376/6216
Abstract
This study investigated relationships between cigarette smoking and ADHD symptoms (Study A) and cigarette smoking and stimulant use (Study B). While cigarette smoking is highly comorbid with these two disorders, the relationships are rarely evaluated simultaneously in order to fully understand the association between the two disorders and how they influence one another over time while receiving treatment for one or both of them. To more precisely evaluate these relationships, a parallel latent growth curve model (LGCM) was applied to two randomized clinical trials datasets. The impact of treatment for the targeted and non-targeted disorders to further understand the link between the two comorbid disorders, and how one treatment may indirectly affect the non-targeted disorder were evaluated.
Study A involved adults diagnosed with ADHD with positive carbon monoxide (CO) levels. Participants were randomly assigned to either the experimental (osmotic-release methylphenidate, OROS-MPH) or placebo group to treat ADHD (targeted disorder), and smoking (non-targeted disorder). Study B included adults diagnosed with stimulant use disorder (SUD) and had positive CO levels. Participants were randomly assigned to the experimental group (smoking cessation treatment with treatment-as-usual) or placebo group (treatment-as-usual).
LGCMs were first used to determine best fitting models for each disorder within each study. Due to the complexity of the model within Study A, a parallel LGCM was then applied only to Study B to help determine whether initial levels of one disorder predicted growth trajectories of the other, whether initial levels are related, and whether growth trajectories are related over time. There were significant relationships across disorders for Study B. Treatment was added to both studies (separate LGCMs for Study A, and a parallel LGCM for Study B) to predict change scores and to examine whether treatment had an impact on either disorder. Both studies found a significant treatment effect on the targeted disorders and no significant effect on the non-targeted disorders, in line with the original findings. Through the use of a parallel LGCM, researchers can test general hypotheses of how treatment may have an effect on the off-target disorders, especially when the effect is routed through change in the targeted disorder.
Metrics
6 File views/ downloads
19 Record Views
Details
- Title
- USING PARALLEL LATENT GROWTH CURVE MODELS TO BETTER UNDERSTAND CO-MORBIDITY IN TWO RANDOMIZED CLINICAL TRIALS
- Creators
- Mary Rose Mamey
- Contributors
- G. Leonard Burns (Advisor)Sterling McPherson (Advisor)Celestina Barbosa-Leiker (Committee Member)Craig Parks (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Psychology, Department of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 102
- Identifiers
- 99900581727101842
- Language
- English
- Resource Type
- Dissertation