Bayesian Meta-Analysis to Account for Heterogeneity in Studies Relating Life Events to Disease
Authors: Elizabeth Stojanovski
Associations between life events and various forms of cancers have been identified. The purpose of a recent random-effects meta-analysis was to identify studies that examined the association between adverse events associated with changes to financial status including decreased income and breast cancer risk. The same association was studied in four separate studies which displayed traits that were not consistent between studies such as the study design, location, and time frame. It was of interest to pool information from various studies to help identify characteristics that differentiated study results. Two random-effects Bayesian meta-analysis models are proposed to combine the reported estimates of the described studies. The proposed models allow major sources of variation to be taken into account, including study level characteristics, between study variance and within study variance, and illustrate the ease with which uncertainty can be incorporated using a hierarchical Bayesian modelling approach.
Digital Object Identifier (DOI): doi.org/10.5281/zenodo.2021865Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 279
 C. L. Cooper and R. Payne, Personality and stress: individual differences in the stress process. Chichester, UK: Wiley, (1991).
 A. Rozanski, J. A. Blumenthal and J. Kaplan, Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation; 99, 2192-217 (1999).
 C. C. Chen, A. S. David, H. Nunnerley, M. Michell, J. L. Dawson, H. Berry, J. Dobbs and T. Fahy, Adverse life events and breast cancer: case-control study. British Medical Journal 311, 1527-30 (1995).
 S. F. Duijts, M. P. Zeegers and B. V. Borne, The association between stressful life events and breast cancer risk: A meta-analysis. International Journal of Cancer 107, 1023–1029 (2003).
 F. D. Roberts, P. A. Newcomb, A. Trentham-Dietz, B. E. Storer, Self-Reported Stress and Risk of Breast Cancer. Cancer 77, 1089-1093 (1996).
 W. H. DuMouchel, Bayesian meta analysis in statistical methodology. Pharmaceutical Sciences, D Berry Ed Marcel Dekker (1990).
 D. G. Fryback, N. K. Stout and M. A. Rosenberg, An Elementary Introduction to Bayesian Computing Using WinBUGS. International Journal of Technology Assessment in Health Care, 17, 98-113 (2001).
 D. J. Spiegelhalter, A. Thomas and N. Best, “WinBugs Version 1.4 User Manual”, MRC Biostatistics Unit, software available at http://www.mrcbsu.cam.ac.uk/bugs/ winbugs/contents.shtml (2000).