Cargando…

Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis

Participants in epidemiologic and genetic studies are rarely true random samples of the populations they are intended to represent, and both known and unknown factors can influence participation in a study (known as selection into a study). The circumstances in which selection causes bias in an inst...

Descripción completa

Detalles Bibliográficos
Autores principales: Hughes, Rachael A., Davies, Neil M., Davey Smith, George, Tilling, Kate
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Lippincott Williams & Wilkins 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6525095/
https://www.ncbi.nlm.nih.gov/pubmed/30896457
http://dx.doi.org/10.1097/EDE.0000000000000972
_version_ 1783419654523322368
author Hughes, Rachael A.
Davies, Neil M.
Davey Smith, George
Tilling, Kate
author_facet Hughes, Rachael A.
Davies, Neil M.
Davey Smith, George
Tilling, Kate
author_sort Hughes, Rachael A.
collection PubMed
description Participants in epidemiologic and genetic studies are rarely true random samples of the populations they are intended to represent, and both known and unknown factors can influence participation in a study (known as selection into a study). The circumstances in which selection causes bias in an instrumental variable (IV) analysis are not widely understood by practitioners of IV analyses. We use directed acyclic graphs (DAGs) to depict assumptions about the selection mechanism (factors affecting selection) and show how DAGs can be used to determine when a two-stage least squares IV analysis is biased by different selection mechanisms. Through simulations, we show that selection can result in a biased IV estimate with substantial confidence interval (CI) undercoverage, and the level of bias can differ between instrument strengths, a linear and nonlinear exposure–instrument association, and a causal and noncausal exposure effect. We present an application from the UK Biobank study, which is known to be a selected sample of the general population. Of interest was the causal effect of staying in school at least 1 extra year on the decision to smoke. Based on 22,138 participants, the two-stage least squares exposure estimates were very different between the IV analysis ignoring selection and the IV analysis which adjusted for selection (e.g., risk differences, 1.8% [95% CI, −1.5%, 5.0%] and −4.5% [95% CI, −6.6%, −2.4%], respectively). We conclude that selection bias can have a major effect on an IV analysis, and further research is needed on how to conduct sensitivity analyses when selection depends on unmeasured data.
format Online
Article
Text
id pubmed-6525095
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher Lippincott Williams & Wilkins
record_format MEDLINE/PubMed
spelling pubmed-65250952020-05-01 Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis Hughes, Rachael A. Davies, Neil M. Davey Smith, George Tilling, Kate Epidemiology Methods Participants in epidemiologic and genetic studies are rarely true random samples of the populations they are intended to represent, and both known and unknown factors can influence participation in a study (known as selection into a study). The circumstances in which selection causes bias in an instrumental variable (IV) analysis are not widely understood by practitioners of IV analyses. We use directed acyclic graphs (DAGs) to depict assumptions about the selection mechanism (factors affecting selection) and show how DAGs can be used to determine when a two-stage least squares IV analysis is biased by different selection mechanisms. Through simulations, we show that selection can result in a biased IV estimate with substantial confidence interval (CI) undercoverage, and the level of bias can differ between instrument strengths, a linear and nonlinear exposure–instrument association, and a causal and noncausal exposure effect. We present an application from the UK Biobank study, which is known to be a selected sample of the general population. Of interest was the causal effect of staying in school at least 1 extra year on the decision to smoke. Based on 22,138 participants, the two-stage least squares exposure estimates were very different between the IV analysis ignoring selection and the IV analysis which adjusted for selection (e.g., risk differences, 1.8% [95% CI, −1.5%, 5.0%] and −4.5% [95% CI, −6.6%, −2.4%], respectively). We conclude that selection bias can have a major effect on an IV analysis, and further research is needed on how to conduct sensitivity analyses when selection depends on unmeasured data. Lippincott Williams & Wilkins 2019-05 2019-04-08 /pmc/articles/PMC6525095/ /pubmed/30896457 http://dx.doi.org/10.1097/EDE.0000000000000972 Text en Copyright © 2019 The Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the Creative Commons Attribution License 4.0 (CCBY) (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Methods
Hughes, Rachael A.
Davies, Neil M.
Davey Smith, George
Tilling, Kate
Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title_full Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title_fullStr Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title_full_unstemmed Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title_short Selection Bias When Estimating Average Treatment Effects Using One-sample Instrumental Variable Analysis
title_sort selection bias when estimating average treatment effects using one-sample instrumental variable analysis
topic Methods
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6525095/
https://www.ncbi.nlm.nih.gov/pubmed/30896457
http://dx.doi.org/10.1097/EDE.0000000000000972
work_keys_str_mv AT hughesrachaela selectionbiaswhenestimatingaveragetreatmenteffectsusingonesampleinstrumentalvariableanalysis
AT daviesneilm selectionbiaswhenestimatingaveragetreatmenteffectsusingonesampleinstrumentalvariableanalysis
AT daveysmithgeorge selectionbiaswhenestimatingaveragetreatmenteffectsusingonesampleinstrumentalvariableanalysis
AT tillingkate selectionbiaswhenestimatingaveragetreatmenteffectsusingonesampleinstrumentalvariableanalysis