Books+ Search Results

Pattern Mixture Model and Joint Modeling Approach to Assess Treatment Effects in a Major Depressive Disorder Clinical Trial with Non-ignorable Missing Data

Author
Title
Pattern Mixture Model and Joint Modeling Approach to Assess Treatment Effects in a Major Depressive Disorder Clinical Trial with Non-ignorable Missing Data [electronic resource].
ISBN
9780355777529
Published
Ann Arbor : ProQuest Dissertations & Theses, 2017.
Physical Description
1 online resource (33 p.)
Local Notes
Access is available to the Yale community.
Notes
Source: Masters Abstracts International, Volume: 57-05.
Adviser: Haiqun Lin.
Access and use
Access restricted by licensing agreement.
Summary
Major Depressive Disorder (MDD), also known as depression, is one of the most disabling and widespread mental disorders in the United States. It has high prevalence among adults in the U.S., especially in the veteran population. There are many medications to treat the disease, however, not all of them are effective. Therefore "next-step" treatment is needed when patients do not show satisfactory response. In this paper we looked at three "next-step" medications---Aripiprazole augmentation, Bupropion augmentation, and Switch to Bupropion monotherapy---and compared the effectiveness of the two augmentation treatments with the switching treatment. We used a pattern mixture model approach and a joint model approach to assess the treatment effects on lowering QIDS-C16 scores (a depressive symptom rating scale), while accounting for any impact of premature treatment termination on unobserved data. Both models captured a significant treatment effect of AD+ARI compared with Switch to BUP (PMM: --0.50, p = 0.0250; joint: --0.50, p = 0.0083). Premature dropout due to lack of treatment response also had an impact on QIDS-C16 scores (PMM: 1.78, p < 0.0001) compared to patients who did not dropout, and the impact got greater as the study progressed (PMM: 2.19, p < 0.0001). Dropout due to side effects also had greater influence on the QIDS-C16 scores as time went by (PMM: 1.74, p < 0.0001). However, we did not observe a differential dropout behavior between augmentation groups and switch treatment for both dropout due to lack of treatment response and due to side effects.
Format
Books / Online / Dissertations & Theses
Language
English
Added to Catalog
July 30, 2018
Thesis note
Thesis (M.P.H.)--Yale University, 2017.
Citation

Available from:

Online
Loading holdings.
Unable to load. Retry?
Loading holdings...
Unable to load. Retry?