Date of Award

2009

Document Type

Thesis

Degree Name

Master of Science in Clinical Research

Department

Biostatistics, Bioinformatics, and Epidemiology

College

College of Graduate Studies

First Advisor

Thomas Hulsey

Second Advisor

Carol Wagner

Third Advisor

Paul Nietert

Fourth Advisor

Bruce Hollis

Fifth Advisor

Barbara Tilley

Abstract

Background: Results of a trial of prenatal Vitamin D supplementation are analyzed for associations of pill count data with serum-based measures of adherence. Obiective: It is hypothesized that adherence as measured by pill count will have a significant association with serum-based measures of adherence. Design: The study from which the data for this analysis were derived is a large randomized, controlled, and double-blinded trial of vitamin D supplements in pregnant women. The women have been stratified by race and randomized into three groups, which receive a 400IU (control), 20001U, or 40001U dose of vitamin D[subscript 3] once daily. Women enter the study at or before 12 weeks gestation and continue their supplementation throughout pregnancy and the first year postpartum. Results: A series of 5 multivariate logistic regression models was created to examine whether mean percent adherence by pill count was significantly associated with the thresholds of adherence by serum 25(OH)D when controlling for race, dose, age, season at enrollment, and initial BMI. The analysis revealed that mean percentage of adherence by pill count was not a significant predictor of adherence by serum 25-OHD at any time point. Odds ratios ranged from 0.9 to 1.1, and a significance level of 0.05 was not reached at any time point Conclusions: In a multivariate logistic regression, there was not a significant association between adherence as defined by serum and mean percent pill count adherence for any time point.

Rights

All rights reserved. Copyright is held by the author.

Download

Share

COinS