DOI

10.17077/etd.uksuke92

Document Type

Dissertation

Date of Degree

Spring 2017

Access Restrictions

Access restricted until 07/13/2019

Degree Name

PhD (Doctor of Philosophy)

Degree In

Biostatistics

First Advisor

Smith, Brian J.

First Committee Member

Cavanaugh, Joseph E.

Second Committee Member

Olsen, Jacob J.

Third Committee Member

Cowles, Mary K.

Fourth Committee Member

Field, R.W.

Abstract

Measurement error is a type of non-sampling error that could attenuate the effect of a risk factor on an outcome variable if no correction is made. Therefore, an effect might not be detectable, even if there is one. If a classical error type is present, then the power of the analysis will be lowered or a bigger sample size will be needed in order to maintain the desirable power. Thus, a correction should be made before drawing any conclusions from the analysis. The regression calibration and simulation extrapolation methods are some of the available methods developed to deal with this kind of problem.

This dissertation proposes a Bayesian method that uses a hierarchical approach to jointly model true radon exposure (measurement error model) and its effect on lung cancer (excess odds model). This method takes subject-specific characteristics into account when making the correction, and uses random effects when missing data are present. We carried out a simulation study in order to compare this method to the regression calibration and simulation extrapolation (SIMEX). Different scenarios were simulated and the simulated data were analyzed with the three methods. This is the first time that these three methods have been compared in the context of radon risk assessment.

The simulation results showed that the proposed Bayesian method had a consistent coverage through out the scenarios. However, the SIMEX method had the lowest bias and mean squared error and, most of the time, its coverage was the closest to the nominal coverage of 95%. The regression calibration was the fastest method to be implemented, but it was outperformed by the other methods.

The dissertation finalizes by performing individual and pooled analyses using data from five case-control North America radon studies (Iowa, Missouri, Winnipeg, Connecticut, and Utah/South Idaho). The data from each study were analyzed individually, first without making any correction, and then using the three correction methods. Finally, the data were combined and the methods were applied to this bigger sample. To the best of our knowledge, regression calibration and SIMEX have not been implemented using this combined dataset.

Public Abstract

Knowledge obtained from epidemiological studies can impact policies regarding public health and, as a result, improve the quality of life for individuals in society. When we understand the relationship between potential risk factors and disease outcomes, we can take steps to minimize the exposure to such risks. The assessment of this exposure may be susceptible to measurement error, which can bias risk estimates, often toward the null hypothesis of no effect. If no effect is found due to measurement error bias, then no action will be taken to prevent the exposure.

This dissertation uses Bayesian methods to assess the relationship between risk factors and disease status, while correcting for measurement error. Specifically, the focus is to estimate the lung cancer risk associated with radon exposure, which is measured with error. We conducted simulation studies to compare the proposed method to two existing methods that have been used to correct for measurement error: regression calibration and simulation extrapolation (SIMEX). This is the first time the performance among these three methods has been compared in the context of radon risk assessment.

Our simulation results on a simple case scenario suggest that correction for measurement error is necessary to remove bias that causes attenuation of lung cancer risk estimates and even suggests a protective effect of radon exposure. Therefore, it is important to correct for measurement error before analyzing radon study data and drawing conclusions from it.

A combined dataset containing information from five North American radon studies (Iowa, Missouri, Winnipeg, Connecticut, and Utah/South Idaho) will be used to apply the proposed Bayesian methods to correct for measurement error in a pooled analysis. For comparison, the same data will be analyzed using regression calibration and SIMEX. To date, these data had not been analyze using either of the three methods mentioned above.

Pages

xii, 100 pages

Bibliography

Includes bibliographical references (pages 97-100).

Copyright

Copyright © 2017 Keyla Pagán-Rivera

Included in

Biostatistics Commons

Share

COinS