Bremer Kolloquium Epidemiologie - Public Health

Dr. Stefan Walter (School of  Medicine, University of California, San Francisco (UCSF)) hält am Donnerstag, den 14. Januar 2016, ab 14:15 Uhr am BIPS (Raum 1.550) einen Vortrag mit dem Titel "Mendelian Randomization Studies in Mental Health and Dementia".


Genetic Instrumental Variable Analysis, also referred to as Mendelian Randomization Studies, are often considered a solution to many of the causal inference problems in observational studies. Using the experiment of  "meiosis", the allocation of genes inherited from our parents is essentially random and the genetic code fixed even prior to birth thus circumventing many of the confounding problems usually encountered in observational research. Nonetheless - genetic IV studies rely on many assumptions - some of them cannot be refuted empirically and require sample sizes that are similar to those needed to identify genetic variants in the first place using genome-wide association studies. Fortunately, genetic IV studies can be done using published data alone and as such have a low threshold for investigating hypothesis of interest given that large exploratory genome-wide association studies commonly publish all association results between the single-nucleotide polymorphisms and the phenotype of interest. In this talk, we will discuss the assumptions of genetic IV analysis, how they can be verified, and how these type of studies have been applied to Mental Health and Dementia research.

Smith, George Davey, and Shah Ebrahim. "‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease?" International journal of epidemiology 32.1 (2003): 1-22.
Burgess, Stephen, Adam Butterworth, and Simon G. Thompson. "Mendelian randomization analysis with multiple genetic variants using summarized data." Genetic epidemiology 37.7 (2013): 658-665.
Østergaard, Søren D., et al. "Associations between potentially modifiable risk factors and Alzheimer Disease: A Mendelian Randomization Study." PLoS Med 12.6 (2015): e1001841.