Unknown

Dataset Information

0

Bayesian methods for meta-analysis of causal relationships estimated using genetic instrumental variables.

ABSTRACT: Genetic markers can be used as instrumental variables, in an analogous way to randomization in a clinical trial, to estimate the causal relationship between a phenotype and an outcome variable. Our purpose is to extend the existing methods for such Mendelian randomization studies to the context of multiple genetic markers measured in multiple studies, based on the analysis of individual participant data. First, for a single genetic marker in one study, we show that the usual ratio of coefficients approach can be reformulated as a regression with heterogeneous error in the explanatory variable. This can be implemented using a Bayesian approach, which is next extended to include multiple genetic markers. We then propose a hierarchical model for undertaking a meta-analysis of multiple studies, in which it is not necessary that the same genetic markers are measured in each study. This provides an overall estimate of the causal relationship between the phenotype and the outcome, and an assessment of its heterogeneity across studies. As an example, we estimate the causal relationship of blood concentrations of C-reactive protein on fibrinogen levels using data from 11 studies. These methods provide a flexible framework for efficient estimation of causal relationships derived from multiple studies. Issues discussed include weak instrument bias, analysis of binary outcome data such as disease risk, missing genetic data, and the use of haplotypes.

SUBMITTER: Burgess S 

PROVIDER: S-EPMC3648673 | biostudies-literature | 2010 May

REPOSITORIES: biostudies-literature

Json Xml
altmetric image

Publications

Bayesian methods for meta-analysis of causal relationships estimated using genetic instrumental variables.

Burgess Stephen S   Thompson Simon G SG   Burgess S S   Thompson S G SG   Andrews G G   Samani N J NJ   Hall A A   Whincup P P   Morris R R   Lawlor D A DA   Davey Smith G G   Timpson N N   Ebrahim S S   Ben-Shlomo Y Y   Davey Smith G G   Timpson N N   Brown M M   Ricketts S S   Sandhu M M   Reiner A A   Psaty B B   Lange L L   Cushman M M   Hung J J   Thompson P P   Beilby J J   Warrington N N   Palmer L J LJ   Nordestgaard B G BG   Tybjaerg-Hansen A A   Zacho J J   Wu C C   Lowe G G   Tzoulaki I I   Kumari M M   Sandhu M M   Yamamoto J F JF   Chiodini B B   Franzosi M M   Hankey G J GJ   Jamrozik K K   Palmer L L   Rimm E E   Pai J J   Psaty B B   Heckbert S S   Bis J J   Anand S S   Engert J J   Collins R R   Clarke R R   Melander O O   Berglund G G   Ladenvall P P   Johansson L L   Jansson J-H JH   Hallmans G G   Hingorani A A   Humphries S S   Rimm E E   Manson J J   Pai J J   Watkins H H   Clarke R R   Hopewell J J   Saleheen D D   Frossard R R   Danesh J J   Sattar N N   Robertson M M   Shepherd J J   Schaefer E E   Hofman A A   Witteman J C M JC   Kardys I I   Ben-Shlomo Y Y   Davey Smith G G   Timpson N N   de Faire U U   Bennet A A   Sattar N N   Ford I I   Packard C C   Kumari M M   Manson J J   Lawlor Debbie A DA   Davey Smith George G   Anand S S   Collins R R   Casas J P JP   Danesh J J   Davey Smith G G   Franzosi M M   Hingorani A A   Lawlor D A DA   Manson J J   Nordestgaard B G BG   Samani N J NJ   Sandhu M M   Smeeth L L   Wensley F F   Anand S S   Bowden J J   Burgess S S   Casas J P JP   Di Angelantonio E E   Engert J J   Gao P P   Shah T T   Smeeth L L   Thompson S G SG   Verzilli C C   Walker M M   Whittaker J J   Hingorani A A   Danesh J J  

Statistics in medicine 20100501 12

Genetic markers can be used as instrumental variables, in an analogous way to randomization in a clinical trial, to estimate the causal relationship between a phenotype and an outcome variable. Our purpose is to extend the existing methods for such Mendelian randomization studies to the context of multiple genetic markers measured in multiple studies, based on the analysis of individual participant data. First, for a single genetic marker in one study, we show that the usual ratio of coefficient  ...[more]

Similar Datasets

Unknown Causal Genetic Inference Using Haplotypes as Instrumental Variables.
| S-EPMC5625343 | biostudies-literature
Unknown Network Mendelian randomization: using genetic variants as instrumental variables to investigate mediation in causal pathways.
| S-EPMC4469795 | biostudies-literature
Unknown Genetic markers as instrumental variables.
| S-EPMC4770870 | biostudies-literature
Unknown General Control Functions for Causal Effect Estimation from Instrumental Variables.
| S-EPMC8096518 | biostudies-literature
Unknown Using multiple genetic variants as instrumental variables for modifiable risk factors.
| S-EPMC3917707 | biostudies-literature
Unknown Falsification Testing of Instrumental Variables Methods for Comparative Effectiveness Research.
| S-EPMC4799892 | biostudies-other
Unknown Designs combining instrumental variables with case-control: estimating principal strata causal effects.
| S-EPMC3608098 | biostudies-literature
Unknown Bayesian hierarchical meta-analytic methods for modeling surrogate relationships that vary across treatment classes using aggregate data.
| S-EPMC7065251 | biostudies-literature
Unknown Handling Missing Data in Instrumental Variable Methods for Causal Inference.
| S-EPMC8025985 | biostudies-literature
Unknown CausalTrail: Testing hypothesis using causal Bayesian networks.
| S-EPMC4743151 | biostudies-other