Cargando…

The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model

OBJECTIVE: To demonstrate the adverse impact of ignoring statistical interactions in regression models used in epidemiologic studies. STUDY DESIGN AND SETTING: Based on different scenarios that involved known values for coefficient of the interaction term in Cox regression models we generated 1000 s...

Descripción completa

Detalles Bibliográficos
Autores principales: Vatcheva, KP, Lee, M, McCormick, JB, Rahbar, MH
Formato: Online Artículo Texto
Lenguaje:English
Publicado: 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4918637/
https://www.ncbi.nlm.nih.gov/pubmed/27347436
http://dx.doi.org/10.4172/2161-1165.1000216
_version_ 1782439142702972928
author Vatcheva, KP
Lee, M
McCormick, JB
Rahbar, MH
author_facet Vatcheva, KP
Lee, M
McCormick, JB
Rahbar, MH
author_sort Vatcheva, KP
collection PubMed
description OBJECTIVE: To demonstrate the adverse impact of ignoring statistical interactions in regression models used in epidemiologic studies. STUDY DESIGN AND SETTING: Based on different scenarios that involved known values for coefficient of the interaction term in Cox regression models we generated 1000 samples of size 600 each. The simulated samples and a real life data set from the Cameron County Hispanic Cohort were used to evaluate the effect of ignoring statistical interactions in these models. RESULTS: Compared to correctly specified Cox regression models with interaction terms, misspecified models without interaction terms resulted in up to 8.95 fold bias in estimated regression coefficients. Whereas when data were generated from a perfect additive Cox proportional hazards regression model the inclusion of the interaction between the two covariates resulted in only 2% estimated bias in main effect regression coefficients estimates, but did not alter the main findings of no significant interactions. CONCLUSIONS: When the effects are synergic, the failure to account for an interaction effect could lead to bias and misinterpretation of the results, and in some instances to incorrect policy decisions. Best practices in regression analysis must include identification of interactions, including for analysis of data from epidemiologic studies.
format Online
Article
Text
id pubmed-4918637
institution National Center for Biotechnology Information
language English
publishDate 2015
record_format MEDLINE/PubMed
spelling pubmed-49186372016-06-23 The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model Vatcheva, KP Lee, M McCormick, JB Rahbar, MH Epidemiology (Sunnyvale) Article OBJECTIVE: To demonstrate the adverse impact of ignoring statistical interactions in regression models used in epidemiologic studies. STUDY DESIGN AND SETTING: Based on different scenarios that involved known values for coefficient of the interaction term in Cox regression models we generated 1000 samples of size 600 each. The simulated samples and a real life data set from the Cameron County Hispanic Cohort were used to evaluate the effect of ignoring statistical interactions in these models. RESULTS: Compared to correctly specified Cox regression models with interaction terms, misspecified models without interaction terms resulted in up to 8.95 fold bias in estimated regression coefficients. Whereas when data were generated from a perfect additive Cox proportional hazards regression model the inclusion of the interaction between the two covariates resulted in only 2% estimated bias in main effect regression coefficients estimates, but did not alter the main findings of no significant interactions. CONCLUSIONS: When the effects are synergic, the failure to account for an interaction effect could lead to bias and misinterpretation of the results, and in some instances to incorrect policy decisions. Best practices in regression analysis must include identification of interactions, including for analysis of data from epidemiologic studies. 2015-01-15 2015-02 /pmc/articles/PMC4918637/ /pubmed/27347436 http://dx.doi.org/10.4172/2161-1165.1000216 Text en http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Article
Vatcheva, KP
Lee, M
McCormick, JB
Rahbar, MH
The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title_full The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title_fullStr The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title_full_unstemmed The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title_short The Effect of Ignoring Statistical Interactions in Regression Analyses Conducted in Epidemiologic Studies: An Example with Survival Analysis Using Cox Proportional Hazards Regression Model
title_sort effect of ignoring statistical interactions in regression analyses conducted in epidemiologic studies: an example with survival analysis using cox proportional hazards regression model
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4918637/
https://www.ncbi.nlm.nih.gov/pubmed/27347436
http://dx.doi.org/10.4172/2161-1165.1000216
work_keys_str_mv AT vatchevakp theeffectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT leem theeffectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT mccormickjb theeffectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT rahbarmh theeffectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT vatchevakp effectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT leem effectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT mccormickjb effectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel
AT rahbarmh effectofignoringstatisticalinteractionsinregressionanalysesconductedinepidemiologicstudiesanexamplewithsurvivalanalysisusingcoxproportionalhazardsregressionmodel