On the choice of parameterisation and priors for the Bayesian analyses of Mendelian randomisation studies
- PMID: 22415699
- DOI: 10.1002/sim.4499
On the choice of parameterisation and priors for the Bayesian analyses of Mendelian randomisation studies
Abstract
Mendelian randomisation is a form of instrumental variable analysis that estimates the causal effect of an intermediate phenotype or exposure on an outcome or disease in the presence of unobserved confounding, using a genetic variant as the instrument. A Bayesian approach allows current knowledge to be incorporated into the analysis in the form of informative prior distributions, and the unobserved confounder can be modelled explicitly. We consider Bayesian methods for Mendelian randomisation in the case where all relationships are linear and there are no interactions. A 'full' model in which the unobserved confounder is included explicitly is not completely identifiable, although the causal parameter can be estimated. We compare inferences from this general but non-identified model with a reduced parameter model that is identifiable. We show that, theoretically, additional information about the causal parameter can be obtained by using the non-identifiable full model, rather than the identifiable reduced model, but that this is advantageous only when realistically informative priors are used and when the instrument is weak or the sample size is small. Furthermore, we consider the impact of using 'vague' versus 'informative' priors.
Copyright © 2012 John Wiley & Sons, Ltd.
Similar articles
-
Effects of globally obtained informative priors on bayesian safety performance functions developed for Australian crash data.Accid Anal Prev. 2019 Aug;129:55-65. doi: 10.1016/j.aap.2019.04.023. Epub 2019 May 17. Accid Anal Prev. 2019. PMID: 31108237
-
A review of instrumental variable estimators for Mendelian randomization.Stat Methods Med Res. 2017 Oct;26(5):2333-2355. doi: 10.1177/0962280215597579. Epub 2015 Aug 17. Stat Methods Med Res. 2017. PMID: 26282889 Free PMC article. Review.
-
Adaptive allocation for binary outcomes using decreasingly informative priors.J Biopharm Stat. 2014;24(3):569-78. doi: 10.1080/10543406.2014.888441. J Biopharm Stat. 2014. PMID: 24697793
-
Use of Mendelian randomisation to assess potential benefit of clinical intervention.BMJ. 2012 Nov 6;345:e7325. doi: 10.1136/bmj.e7325. BMJ. 2012. PMID: 23131671 Review. No abstract available.
-
Bayesian methods for instrumental variable analysis with genetic instruments ('Mendelian randomization'): example with urate transporter SLC2A9 as an instrumental variable for effect of urate levels on metabolic syndrome.Int J Epidemiol. 2010 Jun;39(3):907-18. doi: 10.1093/ije/dyp397. Epub 2010 Mar 25. Int J Epidemiol. 2010. PMID: 20348110 Free PMC article.
Cited by
-
Bayesian mendelian randomization with study heterogeneity and data partitioning for large studies.BMC Med Res Methodol. 2022 Jun 3;22(1):162. doi: 10.1186/s12874-022-01619-4. BMC Med Res Methodol. 2022. PMID: 35658839 Free PMC article.
-
Statistical methods for Mendelian randomization in genome-wide association studies: A review.Comput Struct Biotechnol J. 2022 May 14;20:2338-2351. doi: 10.1016/j.csbj.2022.05.015. eCollection 2022. Comput Struct Biotechnol J. 2022. PMID: 35615025 Free PMC article. Review.
-
Overlapping-sample Mendelian randomisation with multiple exposures: a Bayesian approach.BMC Med Res Methodol. 2020 Dec 7;20(1):295. doi: 10.1186/s12874-020-01170-0. BMC Med Res Methodol. 2020. PMID: 33287714 Free PMC article.
-
Inferring the direction of a causal link and estimating its effect via a Bayesian Mendelian randomization approach.Stat Methods Med Res. 2020 Apr;29(4):1081-1111. doi: 10.1177/0962280219851817. Epub 2019 May 30. Stat Methods Med Res. 2020. PMID: 31146640 Free PMC article.
-
Epidemiology, genetic epidemiology and Mendelian randomisation: more need than ever to attend to detail.Hum Genet. 2020 Jan;139(1):121-136. doi: 10.1007/s00439-019-02027-3. Epub 2019 May 27. Hum Genet. 2020. PMID: 31134333 Free PMC article.