Loading…

Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates

•We model potential biases that may arise in Mendelian randomization analysis.•Genetic variants should robustly associate with exposures in independent samples.•If not, Mendelian randomization can suggest causality despite no true associations. Mendelian randomization methods, which use genetic vari...

Full description

Saved in:
Bibliographic Details
Published in:Economics and human biology 2014-03, Vol.13 (100), p.99-106
Main Authors: Taylor, Amy E., Davies, Neil M., Ware, Jennifer J., VanderWeele, Tyler, Smith, George Davey, Munafò, Marcus R.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03
cites cdi_FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03
container_end_page 106
container_issue 100
container_start_page 99
container_title Economics and human biology
container_volume 13
creator Taylor, Amy E.
Davies, Neil M.
Ware, Jennifer J.
VanderWeele, Tyler
Smith, George Davey
Munafò, Marcus R.
description •We model potential biases that may arise in Mendelian randomization analysis.•Genetic variants should robustly associate with exposures in independent samples.•If not, Mendelian randomization can suggest causality despite no true associations. Mendelian randomization methods, which use genetic variants as instrumental variables for exposures of interest to overcome problems of confounding and reverse causality, are becoming widespread for assessing causal relationships in epidemiological studies. The main purpose of this paper is to demonstrate how results can be biased if researchers select genetic variants on the basis of their association with the exposure in their own dataset, as often happens in candidate gene analyses. This can lead to estimates that indicate apparent “causal” relationships, despite there being no true effect of the exposure. In addition, we discuss the potential bias in estimates of magnitudes of effect from Mendelian randomization analyses when the measured exposure is a poor proxy for the true underlying exposure. We illustrate these points with specific reference to tobacco research.
doi_str_mv 10.1016/j.ehb.2013.12.002
format article
fullrecord <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_3989031</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S1570677X1300124X</els_id><sourcerecordid>1502336309</sourcerecordid><originalsourceid>FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03</originalsourceid><addsrcrecordid>eNp9kUGLFDEQhYMo7u7oD_AiOXrpNunMdNIKgiy6CiteXPAW0klluobuZEwyA_rrzTDrohchUAn11cujHiEvOGs54_3rXQvT2HaMi5Z3LWPdI3LJlRRNzwV7XO8byZpeyu8X5CrnXQVEHXtKLrq1UIp38pLMXyA4mNEEmkxwccFfpmAMFAOdwMxlogkymGSnN_QuY9hSs9-nuE9oCtAtBCho6dHUdyiZVg1qjhHdiRzRZHAUcsGl0vkZeeLNnOH5fV2Ru48fvl1_am6_3ny-fn_b2I3ipXG9X3sDarNee6mcV27wflTK1iPsMHgANYyWcy6tdVJ4L3o5esmkk511TKzIu7Pu_jAu4CyEksysq-fFpJ86GtT_dgJOehuPWgxqYIJXgVf3Ain-OFT_esFsYZ5NgHjImm_qKkUv2FBRfkZtijkn8A_fcKZPKemdrinpU0qad_oUwoq8_Nvfw8SfWCrw9gxA3dIRIelsEYIFhwls0S7if-R_Ax8dpxc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1502336309</pqid></control><display><type>article</type><title>Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates</title><source>Elsevier</source><creator>Taylor, Amy E. ; Davies, Neil M. ; Ware, Jennifer J. ; VanderWeele, Tyler ; Smith, George Davey ; Munafò, Marcus R.</creator><creatorcontrib>Taylor, Amy E. ; Davies, Neil M. ; Ware, Jennifer J. ; VanderWeele, Tyler ; Smith, George Davey ; Munafò, Marcus R.</creatorcontrib><description>•We model potential biases that may arise in Mendelian randomization analysis.•Genetic variants should robustly associate with exposures in independent samples.•If not, Mendelian randomization can suggest causality despite no true associations. Mendelian randomization methods, which use genetic variants as instrumental variables for exposures of interest to overcome problems of confounding and reverse causality, are becoming widespread for assessing causal relationships in epidemiological studies. The main purpose of this paper is to demonstrate how results can be biased if researchers select genetic variants on the basis of their association with the exposure in their own dataset, as often happens in candidate gene analyses. This can lead to estimates that indicate apparent “causal” relationships, despite there being no true effect of the exposure. In addition, we discuss the potential bias in estimates of magnitudes of effect from Mendelian randomization analyses when the measured exposure is a poor proxy for the true underlying exposure. We illustrate these points with specific reference to tobacco research.</description><identifier>ISSN: 1570-677X</identifier><identifier>EISSN: 1873-6130</identifier><identifier>DOI: 10.1016/j.ehb.2013.12.002</identifier><identifier>PMID: 24388127</identifier><language>eng</language><publisher>Netherlands: Elsevier B.V</publisher><subject>Bias ; Causal inference ; Causality ; Confounding Factors, Epidemiologic ; Genome-Wide Association Study - methods ; Humans ; Instrumental variable ; Mendelian randomization ; Mendelian Randomization Analysis - methods ; Research Design ; Smoking ; Smoking - genetics ; Tobacco</subject><ispartof>Economics and human biology, 2014-03, Vol.13 (100), p.99-106</ispartof><rights>2013 The Authors</rights><rights>Copyright © 2013 The Authors. Published by Elsevier B.V. All rights reserved.</rights><rights>2013 The Authors 2013</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03</citedby><cites>FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03</cites><orcidid>0000-0002-4049-993X</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/24388127$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Taylor, Amy E.</creatorcontrib><creatorcontrib>Davies, Neil M.</creatorcontrib><creatorcontrib>Ware, Jennifer J.</creatorcontrib><creatorcontrib>VanderWeele, Tyler</creatorcontrib><creatorcontrib>Smith, George Davey</creatorcontrib><creatorcontrib>Munafò, Marcus R.</creatorcontrib><title>Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates</title><title>Economics and human biology</title><addtitle>Econ Hum Biol</addtitle><description>•We model potential biases that may arise in Mendelian randomization analysis.•Genetic variants should robustly associate with exposures in independent samples.•If not, Mendelian randomization can suggest causality despite no true associations. Mendelian randomization methods, which use genetic variants as instrumental variables for exposures of interest to overcome problems of confounding and reverse causality, are becoming widespread for assessing causal relationships in epidemiological studies. The main purpose of this paper is to demonstrate how results can be biased if researchers select genetic variants on the basis of their association with the exposure in their own dataset, as often happens in candidate gene analyses. This can lead to estimates that indicate apparent “causal” relationships, despite there being no true effect of the exposure. In addition, we discuss the potential bias in estimates of magnitudes of effect from Mendelian randomization analyses when the measured exposure is a poor proxy for the true underlying exposure. We illustrate these points with specific reference to tobacco research.</description><subject>Bias</subject><subject>Causal inference</subject><subject>Causality</subject><subject>Confounding Factors, Epidemiologic</subject><subject>Genome-Wide Association Study - methods</subject><subject>Humans</subject><subject>Instrumental variable</subject><subject>Mendelian randomization</subject><subject>Mendelian Randomization Analysis - methods</subject><subject>Research Design</subject><subject>Smoking</subject><subject>Smoking - genetics</subject><subject>Tobacco</subject><issn>1570-677X</issn><issn>1873-6130</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2014</creationdate><recordtype>article</recordtype><recordid>eNp9kUGLFDEQhYMo7u7oD_AiOXrpNunMdNIKgiy6CiteXPAW0klluobuZEwyA_rrzTDrohchUAn11cujHiEvOGs54_3rXQvT2HaMi5Z3LWPdI3LJlRRNzwV7XO8byZpeyu8X5CrnXQVEHXtKLrq1UIp38pLMXyA4mNEEmkxwccFfpmAMFAOdwMxlogkymGSnN_QuY9hSs9-nuE9oCtAtBCho6dHUdyiZVg1qjhHdiRzRZHAUcsGl0vkZeeLNnOH5fV2Ru48fvl1_am6_3ny-fn_b2I3ipXG9X3sDarNee6mcV27wflTK1iPsMHgANYyWcy6tdVJ4L3o5esmkk511TKzIu7Pu_jAu4CyEksysq-fFpJ86GtT_dgJOehuPWgxqYIJXgVf3Ain-OFT_esFsYZ5NgHjImm_qKkUv2FBRfkZtijkn8A_fcKZPKemdrinpU0qad_oUwoq8_Nvfw8SfWCrw9gxA3dIRIelsEYIFhwls0S7if-R_Ax8dpxc</recordid><startdate>20140301</startdate><enddate>20140301</enddate><creator>Taylor, Amy E.</creator><creator>Davies, Neil M.</creator><creator>Ware, Jennifer J.</creator><creator>VanderWeele, Tyler</creator><creator>Smith, George Davey</creator><creator>Munafò, Marcus R.</creator><general>Elsevier B.V</general><general>Elsevier Science</general><scope>6I.</scope><scope>AAFTH</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-4049-993X</orcidid></search><sort><creationdate>20140301</creationdate><title>Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates</title><author>Taylor, Amy E. ; Davies, Neil M. ; Ware, Jennifer J. ; VanderWeele, Tyler ; Smith, George Davey ; Munafò, Marcus R.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2014</creationdate><topic>Bias</topic><topic>Causal inference</topic><topic>Causality</topic><topic>Confounding Factors, Epidemiologic</topic><topic>Genome-Wide Association Study - methods</topic><topic>Humans</topic><topic>Instrumental variable</topic><topic>Mendelian randomization</topic><topic>Mendelian Randomization Analysis - methods</topic><topic>Research Design</topic><topic>Smoking</topic><topic>Smoking - genetics</topic><topic>Tobacco</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Taylor, Amy E.</creatorcontrib><creatorcontrib>Davies, Neil M.</creatorcontrib><creatorcontrib>Ware, Jennifer J.</creatorcontrib><creatorcontrib>VanderWeele, Tyler</creatorcontrib><creatorcontrib>Smith, George Davey</creatorcontrib><creatorcontrib>Munafò, Marcus R.</creatorcontrib><collection>ScienceDirect Open Access Titles</collection><collection>Elsevier:ScienceDirect:Open Access</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Economics and human biology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Taylor, Amy E.</au><au>Davies, Neil M.</au><au>Ware, Jennifer J.</au><au>VanderWeele, Tyler</au><au>Smith, George Davey</au><au>Munafò, Marcus R.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates</atitle><jtitle>Economics and human biology</jtitle><addtitle>Econ Hum Biol</addtitle><date>2014-03-01</date><risdate>2014</risdate><volume>13</volume><issue>100</issue><spage>99</spage><epage>106</epage><pages>99-106</pages><issn>1570-677X</issn><eissn>1873-6130</eissn><abstract>•We model potential biases that may arise in Mendelian randomization analysis.•Genetic variants should robustly associate with exposures in independent samples.•If not, Mendelian randomization can suggest causality despite no true associations. Mendelian randomization methods, which use genetic variants as instrumental variables for exposures of interest to overcome problems of confounding and reverse causality, are becoming widespread for assessing causal relationships in epidemiological studies. The main purpose of this paper is to demonstrate how results can be biased if researchers select genetic variants on the basis of their association with the exposure in their own dataset, as often happens in candidate gene analyses. This can lead to estimates that indicate apparent “causal” relationships, despite there being no true effect of the exposure. In addition, we discuss the potential bias in estimates of magnitudes of effect from Mendelian randomization analyses when the measured exposure is a poor proxy for the true underlying exposure. We illustrate these points with specific reference to tobacco research.</abstract><cop>Netherlands</cop><pub>Elsevier B.V</pub><pmid>24388127</pmid><doi>10.1016/j.ehb.2013.12.002</doi><tpages>8</tpages><orcidid>https://orcid.org/0000-0002-4049-993X</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1570-677X
ispartof Economics and human biology, 2014-03, Vol.13 (100), p.99-106
issn 1570-677X
1873-6130
language eng
recordid cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_3989031
source Elsevier
subjects Bias
Causal inference
Causality
Confounding Factors, Epidemiologic
Genome-Wide Association Study - methods
Humans
Instrumental variable
Mendelian randomization
Mendelian Randomization Analysis - methods
Research Design
Smoking
Smoking - genetics
Tobacco
title Mendelian randomization in health research: Using appropriate genetic variants and avoiding biased estimates
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T19%3A49%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Mendelian%20randomization%20in%20health%20research:%20Using%20appropriate%20genetic%20variants%20and%20avoiding%20biased%20estimates&rft.jtitle=Economics%20and%20human%20biology&rft.au=Taylor,%20Amy%20E.&rft.date=2014-03-01&rft.volume=13&rft.issue=100&rft.spage=99&rft.epage=106&rft.pages=99-106&rft.issn=1570-677X&rft.eissn=1873-6130&rft_id=info:doi/10.1016/j.ehb.2013.12.002&rft_dat=%3Cproquest_pubme%3E1502336309%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c581t-d6f4fae8544f78df8d9ffb88c88c3c99fee89bc1117ccd73ff367bf707d72cd03%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1502336309&rft_id=info:pmid/24388127&rfr_iscdi=true