Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding
Summary Unmeasured confounding is a threat to causal inference in observational studies. In recent years, the use of negative controls to mitigate unmeasured confounding has gained increasing recognition and popularity. Negative controls have a long‐standing tradition in laboratory sciences and epid...
Gespeichert in:
Veröffentlicht in: | Journal of the Royal Statistical Society. Series B, Statistical methodology Statistical methodology, 2020-04, Vol.82 (2), p.521-540 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 540 |
---|---|
container_issue | 2 |
container_start_page | 521 |
container_title | Journal of the Royal Statistical Society. Series B, Statistical methodology |
container_volume | 82 |
creator | Shi, Xu Miao, Wang Nelson, Jennifer C. Tchetgen Tchetgen, Eric J. |
description | Summary
Unmeasured confounding is a threat to causal inference in observational studies. In recent years, the use of negative controls to mitigate unmeasured confounding has gained increasing recognition and popularity. Negative controls have a long‐standing tradition in laboratory sciences and epidemiology to rule out non‐causal explanations, although they have been used primarily for bias detection. Recently, Miao and colleagues have described sufficient conditions under which a pair of negative control exposure and outcome variables can be used to identify non‐parametrically the average treatment effect (ATE) from observational data subject to uncontrolled confounding. We establish non‐parametric identification of the ATE under weaker conditions in the case of categorical unmeasured confounding and negative control variables. We also provide a general semiparametric framework for obtaining inferences about the ATE while leveraging information about a possibly large number of measured covariates. In particular, we derive the semiparametric efficiency bound in the non‐parametric model, and we propose multiply robust and locally efficient estimators when non‐parametric estimation may not be feasible. We assess the finite sample performance of our methods in extensive simulation studies. Finally, we illustrate our methods with an application to the post‐licensure surveillance of vaccine safety among children. |
doi_str_mv | 10.1111/rssb.12361 |
format | Article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7768794</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2377826748</sourcerecordid><originalsourceid>FETCH-LOGICAL-c4811-94ee8b088ae44c3cacd0eed62aebee5b85b9fb1b19b2dfdb4180292bd656cc123</originalsourceid><addsrcrecordid>eNp9kc1qFTEUxwdR7IdufAAZcCPC1MkkNx8bQYtWoUWwug75OHObSya5TSYtd-cj9Bl9kuZ626IuPJsE8ju_nOTfNC9Qf4RqvU056yM0YIoeNfuIUNYJTvnjusdUdIygYa85yHnV16IMP232MMaMEiL2m8uz4me39ps2RV3y3BpVsvKtCyMkCAbaazdftDYW7eHXz5sASzW7K2hNDHOKvlV2VdsmCHM7xlTbZ1jG5Ex1lDCByiWB3dJjLMG6sHzWPBmVz_D8bj1sfnz6-P34c3f69eTL8fvTzhCOUCcIANc95woIMdgoY3sASwcFGmCh-UKLUSONhB7saDVBvB_EoC1dUGPqbxw273beddETWFMnTMrLdXKTShsZlZN_nwR3IZfxSjJGOROkCl7fCVK8LJBnOblswHsVIJYsB8KwqHf2fUVf_YOuYkmhPk8OmDE-UEZ4pd7sKJNizgnGh2FQL7dJym2S8neSFX755_gP6H10FUA74Np52PxHJb-dn3_YSW8BXVavPQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2377826748</pqid></control><display><type>article</type><title>Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding</title><source>EBSCOhost Business Source Complete</source><source>Access via Wiley Online Library</source><source>Oxford University Press Journals All Titles (1996-Current)</source><creator>Shi, Xu ; Miao, Wang ; Nelson, Jennifer C. ; Tchetgen Tchetgen, Eric J.</creator><creatorcontrib>Shi, Xu ; Miao, Wang ; Nelson, Jennifer C. ; Tchetgen Tchetgen, Eric J.</creatorcontrib><description>Summary
Unmeasured confounding is a threat to causal inference in observational studies. In recent years, the use of negative controls to mitigate unmeasured confounding has gained increasing recognition and popularity. Negative controls have a long‐standing tradition in laboratory sciences and epidemiology to rule out non‐causal explanations, although they have been used primarily for bias detection. Recently, Miao and colleagues have described sufficient conditions under which a pair of negative control exposure and outcome variables can be used to identify non‐parametrically the average treatment effect (ATE) from observational data subject to uncontrolled confounding. We establish non‐parametric identification of the ATE under weaker conditions in the case of categorical unmeasured confounding and negative control variables. We also provide a general semiparametric framework for obtaining inferences about the ATE while leveraging information about a possibly large number of measured covariates. In particular, we derive the semiparametric efficiency bound in the non‐parametric model, and we propose multiply robust and locally efficient estimators when non‐parametric estimation may not be feasible. We assess the finite sample performance of our methods in extensive simulation studies. Finally, we illustrate our methods with an application to the post‐licensure surveillance of vaccine safety among children.</description><identifier>ISSN: 1369-7412</identifier><identifier>EISSN: 1467-9868</identifier><identifier>DOI: 10.1111/rssb.12361</identifier><identifier>PMID: 33376449</identifier><language>eng</language><publisher>England: Oxford University Press</publisher><subject>Bias ; Causal inference ; Computer simulation ; Epidemiology ; Inference ; Licensing ; Negative control ; Observational studies ; Parameter estimation ; Popularity ; Regression analysis ; Robustness ; Semiparametric inference ; Simulation ; Statistical methods ; Statistics ; Surveillance ; Unmeasured confounding</subject><ispartof>Journal of the Royal Statistical Society. Series B, Statistical methodology, 2020-04, Vol.82 (2), p.521-540</ispartof><rights>2020 Royal Statistical Society</rights><rights>Copyright © 2020 The Royal Statistical Society and Blackwell Publishing Ltd</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c4811-94ee8b088ae44c3cacd0eed62aebee5b85b9fb1b19b2dfdb4180292bd656cc123</citedby><cites>FETCH-LOGICAL-c4811-94ee8b088ae44c3cacd0eed62aebee5b85b9fb1b19b2dfdb4180292bd656cc123</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1111%2Frssb.12361$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1111%2Frssb.12361$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>230,314,780,784,885,1417,27924,27925,45574,45575</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33376449$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Shi, Xu</creatorcontrib><creatorcontrib>Miao, Wang</creatorcontrib><creatorcontrib>Nelson, Jennifer C.</creatorcontrib><creatorcontrib>Tchetgen Tchetgen, Eric J.</creatorcontrib><title>Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding</title><title>Journal of the Royal Statistical Society. Series B, Statistical methodology</title><addtitle>J R Stat Soc Series B Stat Methodol</addtitle><description>Summary
Unmeasured confounding is a threat to causal inference in observational studies. In recent years, the use of negative controls to mitigate unmeasured confounding has gained increasing recognition and popularity. Negative controls have a long‐standing tradition in laboratory sciences and epidemiology to rule out non‐causal explanations, although they have been used primarily for bias detection. Recently, Miao and colleagues have described sufficient conditions under which a pair of negative control exposure and outcome variables can be used to identify non‐parametrically the average treatment effect (ATE) from observational data subject to uncontrolled confounding. We establish non‐parametric identification of the ATE under weaker conditions in the case of categorical unmeasured confounding and negative control variables. We also provide a general semiparametric framework for obtaining inferences about the ATE while leveraging information about a possibly large number of measured covariates. In particular, we derive the semiparametric efficiency bound in the non‐parametric model, and we propose multiply robust and locally efficient estimators when non‐parametric estimation may not be feasible. We assess the finite sample performance of our methods in extensive simulation studies. Finally, we illustrate our methods with an application to the post‐licensure surveillance of vaccine safety among children.</description><subject>Bias</subject><subject>Causal inference</subject><subject>Computer simulation</subject><subject>Epidemiology</subject><subject>Inference</subject><subject>Licensing</subject><subject>Negative control</subject><subject>Observational studies</subject><subject>Parameter estimation</subject><subject>Popularity</subject><subject>Regression analysis</subject><subject>Robustness</subject><subject>Semiparametric inference</subject><subject>Simulation</subject><subject>Statistical methods</subject><subject>Statistics</subject><subject>Surveillance</subject><subject>Unmeasured confounding</subject><issn>1369-7412</issn><issn>1467-9868</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNp9kc1qFTEUxwdR7IdufAAZcCPC1MkkNx8bQYtWoUWwug75OHObSya5TSYtd-cj9Bl9kuZ626IuPJsE8ju_nOTfNC9Qf4RqvU056yM0YIoeNfuIUNYJTvnjusdUdIygYa85yHnV16IMP232MMaMEiL2m8uz4me39ps2RV3y3BpVsvKtCyMkCAbaazdftDYW7eHXz5sASzW7K2hNDHOKvlV2VdsmCHM7xlTbZ1jG5Ex1lDCByiWB3dJjLMG6sHzWPBmVz_D8bj1sfnz6-P34c3f69eTL8fvTzhCOUCcIANc95woIMdgoY3sASwcFGmCh-UKLUSONhB7saDVBvB_EoC1dUGPqbxw273beddETWFMnTMrLdXKTShsZlZN_nwR3IZfxSjJGOROkCl7fCVK8LJBnOblswHsVIJYsB8KwqHf2fUVf_YOuYkmhPk8OmDE-UEZ4pd7sKJNizgnGh2FQL7dJym2S8neSFX755_gP6H10FUA74Np52PxHJb-dn3_YSW8BXVavPQ</recordid><startdate>202004</startdate><enddate>202004</enddate><creator>Shi, Xu</creator><creator>Miao, Wang</creator><creator>Nelson, Jennifer C.</creator><creator>Tchetgen Tchetgen, Eric J.</creator><general>Oxford University Press</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8BJ</scope><scope>8FD</scope><scope>FQK</scope><scope>JBE</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7X8</scope><scope>5PM</scope></search><sort><creationdate>202004</creationdate><title>Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding</title><author>Shi, Xu ; Miao, Wang ; Nelson, Jennifer C. ; Tchetgen Tchetgen, Eric J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c4811-94ee8b088ae44c3cacd0eed62aebee5b85b9fb1b19b2dfdb4180292bd656cc123</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Bias</topic><topic>Causal inference</topic><topic>Computer simulation</topic><topic>Epidemiology</topic><topic>Inference</topic><topic>Licensing</topic><topic>Negative control</topic><topic>Observational studies</topic><topic>Parameter estimation</topic><topic>Popularity</topic><topic>Regression analysis</topic><topic>Robustness</topic><topic>Semiparametric inference</topic><topic>Simulation</topic><topic>Statistical methods</topic><topic>Statistics</topic><topic>Surveillance</topic><topic>Unmeasured confounding</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shi, Xu</creatorcontrib><creatorcontrib>Miao, Wang</creatorcontrib><creatorcontrib>Nelson, Jennifer C.</creatorcontrib><creatorcontrib>Tchetgen Tchetgen, Eric J.</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>International Bibliography of the Social Sciences (IBSS)</collection><collection>Technology Research Database</collection><collection>International Bibliography of the Social Sciences</collection><collection>International Bibliography of the Social Sciences</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of the Royal Statistical Society. Series B, Statistical methodology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shi, Xu</au><au>Miao, Wang</au><au>Nelson, Jennifer C.</au><au>Tchetgen Tchetgen, Eric J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding</atitle><jtitle>Journal of the Royal Statistical Society. Series B, Statistical methodology</jtitle><addtitle>J R Stat Soc Series B Stat Methodol</addtitle><date>2020-04</date><risdate>2020</risdate><volume>82</volume><issue>2</issue><spage>521</spage><epage>540</epage><pages>521-540</pages><issn>1369-7412</issn><eissn>1467-9868</eissn><abstract>Summary
Unmeasured confounding is a threat to causal inference in observational studies. In recent years, the use of negative controls to mitigate unmeasured confounding has gained increasing recognition and popularity. Negative controls have a long‐standing tradition in laboratory sciences and epidemiology to rule out non‐causal explanations, although they have been used primarily for bias detection. Recently, Miao and colleagues have described sufficient conditions under which a pair of negative control exposure and outcome variables can be used to identify non‐parametrically the average treatment effect (ATE) from observational data subject to uncontrolled confounding. We establish non‐parametric identification of the ATE under weaker conditions in the case of categorical unmeasured confounding and negative control variables. We also provide a general semiparametric framework for obtaining inferences about the ATE while leveraging information about a possibly large number of measured covariates. In particular, we derive the semiparametric efficiency bound in the non‐parametric model, and we propose multiply robust and locally efficient estimators when non‐parametric estimation may not be feasible. We assess the finite sample performance of our methods in extensive simulation studies. Finally, we illustrate our methods with an application to the post‐licensure surveillance of vaccine safety among children.</abstract><cop>England</cop><pub>Oxford University Press</pub><pmid>33376449</pmid><doi>10.1111/rssb.12361</doi><tpages>20</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1369-7412 |
ispartof | Journal of the Royal Statistical Society. Series B, Statistical methodology, 2020-04, Vol.82 (2), p.521-540 |
issn | 1369-7412 1467-9868 |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7768794 |
source | EBSCOhost Business Source Complete; Access via Wiley Online Library; Oxford University Press Journals All Titles (1996-Current) |
subjects | Bias Causal inference Computer simulation Epidemiology Inference Licensing Negative control Observational studies Parameter estimation Popularity Regression analysis Robustness Semiparametric inference Simulation Statistical methods Statistics Surveillance Unmeasured confounding |
title | Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T17%3A52%3A20IST&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=Multiply%20robust%20causal%20inference%20with%20double%E2%80%90negative%20control%20adjustment%20for%20categorical%20unmeasured%20confounding&rft.jtitle=Journal%20of%20the%20Royal%20Statistical%20Society.%20Series%20B,%20Statistical%20methodology&rft.au=Shi,%20Xu&rft.date=2020-04&rft.volume=82&rft.issue=2&rft.spage=521&rft.epage=540&rft.pages=521-540&rft.issn=1369-7412&rft.eissn=1467-9868&rft_id=info:doi/10.1111/rssb.12361&rft_dat=%3Cproquest_pubme%3E2377826748%3C/proquest_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2377826748&rft_id=info:pmid/33376449&rfr_iscdi=true |