2020
DOI: 10.3390/toxics8040125
|View full text |Cite
|
Sign up to set email alerts
|

The Confounder-Mediator Dilemma: Should We Control for Obesity to Estimate the Effect of Perfluoroalkyl Substances on Health Outcomes?

Abstract: Confounding adjustment is important for observational studies to derive valid effect estimates for inference. Despite the theoretical advancement of confounding selection procedure, it is often challenging to distinguish between confounders and mediators due to the lack of information about the time-ordering and latency of each variable in the data. This is also the case for the studies of perfluoroalkyl substances (PFAS), a group of synthetic chemicals used in industry and consumer products that are persisten… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
15
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
9

Relationship

1
8

Authors

Journals

citations
Cited by 25 publications
(19 citation statements)
references
References 47 publications
(68 reference statements)
0
15
0
Order By: Relevance
“…2017 ; Jain and Ducatman 2019 ). However, cautious interpretation is recommended because stratification by baseline BMI could introduce (collider stratification) bias ( Inoue et al. 2020 ).…”
Section: Discussionmentioning
confidence: 99%
“…2017 ; Jain and Ducatman 2019 ). However, cautious interpretation is recommended because stratification by baseline BMI could introduce (collider stratification) bias ( Inoue et al. 2020 ).…”
Section: Discussionmentioning
confidence: 99%
“…Fifth, we added body mass index (BMI) as an additional covariate. BMI was not included in the primary analyses, because conditioning on BMI may introduce collider stratification bias [ 35 , 36 ] and because the sample size was reduced through missing data on height and weight. The sixth sensitivity analysis repeated the analyses with alcohol intake modelled across the full range of consumptions (eight points from ‘almost daily’ to ‘not at all in the past 12 months’) instead of categorization as a binary variable.…”
Section: Methodsmentioning
confidence: 99%
“…In PA studies, given the negative feedback system of RAAS and the HPA axis, it is often challenging to assume which variables (e.g., biomarker concentrations and medical conditions) occurred first at cross-sectional data. When the variables can be both confounders and mediators, researchers need to carefully interpret the results of stratified analysis by such variables because the seemingly heterogeneous findings could be biased (74). Temporal ordering can be established in longitudinal cohorts where the extraction of biosample and other information precedes the outcome assessment.…”
Section: Overadjustment and Collider Biasmentioning
confidence: 99%