The Analysis of 2x2 Crossover Design with Repeated Baseline Measurement within Period Open Access

Abstract

Background: Neuroimaging study suggests precuneus is related to visuospatial imagery. Evidence suggests, the precuneus of Neanderthal is less developed than the ancestor of modern man and Neanderthal is not able to hunt prey by throwing from long distance away; while the ancestor of modern man has a more developed precuneus, and is known for throwing long-distance weapon in hunting. Thus, it is hypothesized that, the developed precuneus, provided the ancestor of modern man with the improved visuospatial skills needed in hunting prey with projectiles from a distance.

Objective: Identify that precuneus is statistical significant related to visuospatial ability, which is reflected by throw accuracy. Then the result of this study serves as evidence for the proposed hypothesis.

Methods: 2x2 crossover design was applied. Precuneus’s function was oppressed by TMS treatment, while Sham treatment serves as a placebo. Throw score was chose as outcome to reflect visuospatial skill. Baseline measurement was implemented at each period before delivery of treatment. In analysis, four methods of handling baseline, modeling mean score or individual score, and various covariance structures were applied to the data. Analysis was done by applying linear mixed model.

Result: The model which handling baseline with “change from baseline” method, modeling individual change score, and assuming equal variance, cross-visit correlation is zero and within visit correlation is , was chosen. The estimated difference among the effect of Sham and the effect of TMS on throw score is 0.1715, p=0.3548. After investigation into subject characteristics and experiment characteristics, temperature was adjusted in model, the estimated difference is -0.0258, p=0.8899. There is no statistical significant difference among the effects of TMS and Sham.

Conclusion: This study did not prove that precuneus function is related to visuospatial ability of human. Future study design may consider the following: 1. Indoor experiment controlling for temperature; 2. Apply randomized parallel design instead of crossover design; 3. May consider uniform MRI intensity in treatments, rather than subject specific MRI intensity.

Table of Contents

Introduction. 1

Material and Method. 3

1.   Study Design. 3

2.   Data Collection. 5

3.   Statistical Aim.. 6

4. Statistical Method. 6

Result 32

1.   Descriptive Statistics 32

2.   Model Result 38

Discussion. 50

1.   The Recommended Model 50

2.   Covariate Adjustment 59

Conclusion. 68

Future Study. 69

Reference. 71

Appendix. 72

 

About this Master's Thesis

Rights statement

• 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.

School	Rollins School of Public Health
Department	Biostatistics
Subfield / Discipline	Biostatistics - MPH & MSPH
Degree	M.S.P.H.
Submission	Master's Thesis
Language	English
Research Field	Health Sciences, Public Health Biology, Biostatistics
Keyword	Repeated measurement within period Crossover design Baseline
Committee Chair / Thesis Advisor	Renee Moore, Emory University
Committee Members	C. Christina Mehta, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	The Analysis of 2x2 Crossover Design with Repeated Baseline Measurement within Period ()	2019-04-09 14:12:14 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions