The impact of unmeasured within- and between-cluster confounding on the bias of effect estimatorsof a continuous exposure

Yun Li, Yoonseok Lee, Friedrich K. Port, Bruce M. Robinson

Research output: Contribution to journalArticle

Abstract

Unmeasured confounding almost always exists in observational studies and can bias estimates of exposure effects. Instrumental variable methods are popular choices in combating unmeasured confounding to obtain less biased effect estimates. However, we demonstrate that alternative methods may give less biased estimates depending on the nature of unmeasured confounding. Treatment preferences of clusters (e.g. physician practices) are the most frequently used instruments in instrumental variable analyses. These preference-based instrumental variable analyses are usually conducted on data clustered by region, hospital/facility, or physician, where unmeasured confounding often occurs within or between clusters. We aim to quantify the impact of unmeasured confounding on the bias of effect estimators in instrumental variable analysis, as well as several common alternative methods including ordinary least squares regression, linear mixed models, and fixed-effect models to study the effect of a continuous exposure (e.g. treatment dose) on a continuous outcome. We derive closed-form expressions of asymptotic bias of estimators from these four methods in the presence of unmeasured within- and/or between-cluster confounders. Simulations demonstrate that the asymptotic bias formulae well approximate bias in finite samples for all methods. The bias formulae show that instrumental variable analyses can provide consistent estimates when unmeasured within-cluster confounding exists, but not when between-cluster confounding exists. On the other hand, fixed-effect models and linear mixed models can provide consistent estimates when unmeasured between-cluster confounding exits, but not for within-cluster confounding. Whether instrumental variable analyses are advantageous in reducing bias over fixed-effect models and linear mixed models depends on the extent of unmeasured within-cluster confounding relative to between-cluster confounding. Furthermore, the impact of unmeasured between-cluster confounding on instrumental variable analysis estimates is larger than the impact of unmeasured within-cluster confounding on fixed-effect model and linear mixed model estimates. We illustrate the use of these methods in estimating the effect of erythropoiesis stimulating agents on hemoglobin levels. Our findings provide guidance for choosing appropriate methods to combat the dominant types of unmeasured confounders and help interpret statistical results in the context of unmeasured confounding.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Confounding
Instrumental Variables
Linear Models
Fixed Effects Model
Linear Mixed Model
Hematinics
Physicians
Asymptotic Bias
Consistent Estimates
Estimate
Least-Squares Analysis
Observational Studies
Biased
Hemoglobins
Estimator
Clustered Data
Observational Study
Hemoglobin
Least Squares Regression
Ordinary Least Squares

Keywords

  • Bias formula
  • causal inference
  • instrumental variables
  • linear mixed model
  • observational study
  • unmeasured confounders

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

The impact of unmeasured within- and between-cluster confounding on the bias of effect estimatorsof a continuous exposure. / Li, Yun; Lee, Yoonseok; Port, Friedrich K.; Robinson, Bruce M.

In: Statistical Methods in Medical Research, 01.01.2019.

Research output: Contribution to journalArticle

@article{128a9389f3724faf94041b503fd8172f,
title = "The impact of unmeasured within- and between-cluster confounding on the bias of effect estimatorsof a continuous exposure",
abstract = "Unmeasured confounding almost always exists in observational studies and can bias estimates of exposure effects. Instrumental variable methods are popular choices in combating unmeasured confounding to obtain less biased effect estimates. However, we demonstrate that alternative methods may give less biased estimates depending on the nature of unmeasured confounding. Treatment preferences of clusters (e.g. physician practices) are the most frequently used instruments in instrumental variable analyses. These preference-based instrumental variable analyses are usually conducted on data clustered by region, hospital/facility, or physician, where unmeasured confounding often occurs within or between clusters. We aim to quantify the impact of unmeasured confounding on the bias of effect estimators in instrumental variable analysis, as well as several common alternative methods including ordinary least squares regression, linear mixed models, and fixed-effect models to study the effect of a continuous exposure (e.g. treatment dose) on a continuous outcome. We derive closed-form expressions of asymptotic bias of estimators from these four methods in the presence of unmeasured within- and/or between-cluster confounders. Simulations demonstrate that the asymptotic bias formulae well approximate bias in finite samples for all methods. The bias formulae show that instrumental variable analyses can provide consistent estimates when unmeasured within-cluster confounding exists, but not when between-cluster confounding exists. On the other hand, fixed-effect models and linear mixed models can provide consistent estimates when unmeasured between-cluster confounding exits, but not for within-cluster confounding. Whether instrumental variable analyses are advantageous in reducing bias over fixed-effect models and linear mixed models depends on the extent of unmeasured within-cluster confounding relative to between-cluster confounding. Furthermore, the impact of unmeasured between-cluster confounding on instrumental variable analysis estimates is larger than the impact of unmeasured within-cluster confounding on fixed-effect model and linear mixed model estimates. We illustrate the use of these methods in estimating the effect of erythropoiesis stimulating agents on hemoglobin levels. Our findings provide guidance for choosing appropriate methods to combat the dominant types of unmeasured confounders and help interpret statistical results in the context of unmeasured confounding.",
keywords = "Bias formula, causal inference, instrumental variables, linear mixed model, observational study, unmeasured confounders",
author = "Yun Li and Yoonseok Lee and Port, {Friedrich K.} and Robinson, {Bruce M.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1177/0962280219883323",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - The impact of unmeasured within- and between-cluster confounding on the bias of effect estimatorsof a continuous exposure

AU - Li, Yun

AU - Lee, Yoonseok

AU - Port, Friedrich K.

AU - Robinson, Bruce M.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Unmeasured confounding almost always exists in observational studies and can bias estimates of exposure effects. Instrumental variable methods are popular choices in combating unmeasured confounding to obtain less biased effect estimates. However, we demonstrate that alternative methods may give less biased estimates depending on the nature of unmeasured confounding. Treatment preferences of clusters (e.g. physician practices) are the most frequently used instruments in instrumental variable analyses. These preference-based instrumental variable analyses are usually conducted on data clustered by region, hospital/facility, or physician, where unmeasured confounding often occurs within or between clusters. We aim to quantify the impact of unmeasured confounding on the bias of effect estimators in instrumental variable analysis, as well as several common alternative methods including ordinary least squares regression, linear mixed models, and fixed-effect models to study the effect of a continuous exposure (e.g. treatment dose) on a continuous outcome. We derive closed-form expressions of asymptotic bias of estimators from these four methods in the presence of unmeasured within- and/or between-cluster confounders. Simulations demonstrate that the asymptotic bias formulae well approximate bias in finite samples for all methods. The bias formulae show that instrumental variable analyses can provide consistent estimates when unmeasured within-cluster confounding exists, but not when between-cluster confounding exists. On the other hand, fixed-effect models and linear mixed models can provide consistent estimates when unmeasured between-cluster confounding exits, but not for within-cluster confounding. Whether instrumental variable analyses are advantageous in reducing bias over fixed-effect models and linear mixed models depends on the extent of unmeasured within-cluster confounding relative to between-cluster confounding. Furthermore, the impact of unmeasured between-cluster confounding on instrumental variable analysis estimates is larger than the impact of unmeasured within-cluster confounding on fixed-effect model and linear mixed model estimates. We illustrate the use of these methods in estimating the effect of erythropoiesis stimulating agents on hemoglobin levels. Our findings provide guidance for choosing appropriate methods to combat the dominant types of unmeasured confounders and help interpret statistical results in the context of unmeasured confounding.

AB - Unmeasured confounding almost always exists in observational studies and can bias estimates of exposure effects. Instrumental variable methods are popular choices in combating unmeasured confounding to obtain less biased effect estimates. However, we demonstrate that alternative methods may give less biased estimates depending on the nature of unmeasured confounding. Treatment preferences of clusters (e.g. physician practices) are the most frequently used instruments in instrumental variable analyses. These preference-based instrumental variable analyses are usually conducted on data clustered by region, hospital/facility, or physician, where unmeasured confounding often occurs within or between clusters. We aim to quantify the impact of unmeasured confounding on the bias of effect estimators in instrumental variable analysis, as well as several common alternative methods including ordinary least squares regression, linear mixed models, and fixed-effect models to study the effect of a continuous exposure (e.g. treatment dose) on a continuous outcome. We derive closed-form expressions of asymptotic bias of estimators from these four methods in the presence of unmeasured within- and/or between-cluster confounders. Simulations demonstrate that the asymptotic bias formulae well approximate bias in finite samples for all methods. The bias formulae show that instrumental variable analyses can provide consistent estimates when unmeasured within-cluster confounding exists, but not when between-cluster confounding exists. On the other hand, fixed-effect models and linear mixed models can provide consistent estimates when unmeasured between-cluster confounding exits, but not for within-cluster confounding. Whether instrumental variable analyses are advantageous in reducing bias over fixed-effect models and linear mixed models depends on the extent of unmeasured within-cluster confounding relative to between-cluster confounding. Furthermore, the impact of unmeasured between-cluster confounding on instrumental variable analysis estimates is larger than the impact of unmeasured within-cluster confounding on fixed-effect model and linear mixed model estimates. We illustrate the use of these methods in estimating the effect of erythropoiesis stimulating agents on hemoglobin levels. Our findings provide guidance for choosing appropriate methods to combat the dominant types of unmeasured confounders and help interpret statistical results in the context of unmeasured confounding.

KW - Bias formula

KW - causal inference

KW - instrumental variables

KW - linear mixed model

KW - observational study

KW - unmeasured confounders

UR - http://www.scopus.com/inward/record.url?scp=85074992718&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074992718&partnerID=8YFLogxK

U2 - 10.1177/0962280219883323

DO - 10.1177/0962280219883323

M3 - Article

C2 - 31694489

AN - SCOPUS:85074992718

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -