Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project

Evangelia Samoli; Sophia Rodopoulou; Ulla A. Hvidtfeldt; Kathrin Wolf; Massimo Stafoggia; Bert Brunekreef; Maciej Strak; Jie Chen; Zorana J. Andersen; Richard Atkinson; Mariska Bauwelinck; Tom Bellander; Jørgen Brandt; Giulia Cesaroni; Francesco Forastiere; Daniela Fecht; John Gulliver; Ole Hertel; Barbara Hoffmann; Kees de Hoogh; Nicole A.H. Janssen; Matthias Ketzel; Jochem O. Klompmaker; Shuo Liu; Petter Ljungman; Gabriele Nagel; Bente Oftedal; Göran Pershagen; Annette Peters; Ole Raaschou-Nielsen; Matteo Renzi; Doris T. Kristoffersen; Gianluca Severi; Torben Sigsgaard; Danielle Vienneau; Gudrun Weinmayr; Gerard Hoek; Klea Katsouyanni

doi:10.1016/j.envint.2020.106371

Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project

Evangelia Samoli, Sophia Rodopoulou, Ulla A. Hvidtfeldt, Kathrin Wolf, Massimo Stafoggia, Bert Brunekreef, Maciej Strak, Jie Chen, Zorana J. Andersen, Richard Atkinson, Mariska Bauwelinck, Tom Bellander, Jørgen Brandt, Giulia Cesaroni, Francesco Forastiere, Daniela Fecht, John Gulliver, Ole Hertel, Barbara Hoffmann, Kees de HooghNicole A.H. Janssen, Matthias Ketzel, Jochem O. Klompmaker, Shuo Liu, Petter Ljungman, Gabriele Nagel, Bente Oftedal, Göran Pershagen, Annette Peters, Ole Raaschou-Nielsen, Matteo Renzi, Doris T. Kristoffersen, Gianluca Severi, Torben Sigsgaard, Danielle Vienneau, Gudrun Weinmayr, Gerard Hoek, Klea Katsouyanni

Onderzoeksoutput: Article › peer review

23 Citaten (Scopus)

Samenvatting

Background: We evaluated methods for the analysis of multi-level survival data using a pooled dataset of 14 cohorts participating in the ELAPSE project investigating associations between residential exposure to low levels of air pollution (PM_2.5 and NO₂) and health (natural-cause mortality and cerebrovascular, coronary and lung cancer incidence).

Methods: We applied five approaches in a multivariable Cox model to account for the first level of clustering corresponding to cohort specification: (1) not accounting for the cohort or using (2) indicator variables, (3) strata, (4) a frailty term in frailty Cox models, (5) a random intercept under a mixed Cox, for cohort identification. We accounted for the second level of clustering due to common characteristics in the residential area by (1) a random intercept per small area or (2) applying variance correction. We assessed the stratified, frailty and mixed Cox approach through simulations under different scenarios for heterogeneity in the underlying hazards and the air pollution effects.

Results: Effect estimates were stable under approaches used to adjust for cohort but substantially differed when no adjustment was applied. Further adjustment for the small area grouping increased the effect estimates’ standard errors. Simulations confirmed identical results between the stratified and frailty models. In ELAPSE we selected a stratified multivariable Cox model to account for between-cohort heterogeneity without adjustment for small area level, due to the small number of subjects and events in the latter.

Conclusions: Our study supports the need to account for between-cohort heterogeneity in multi-center collaborations using pooled individual level data.

Originele taal-2	English
Artikelnummer	106371
Aantal pagina's	8
Tijdschrift	Environment International
Volume	147
DOI's	https://doi.org/10.1016/j.envint.2020.106371
Status	Published - feb. 2021

Bibliografische nota

Funding Information:
This work was supported by Health Effects Institute (HEI) Research Agreement [grant number #4954-RFA14-3/16-5-3].

Funding Information:
“Research described in this article was conducted under contract to the Health Effects Institute (HEI), an organization jointly funded by the United States Environmental Protection Agency (EPA) (Assistance Award No. R-82811201) and certain motor vehicle and engine manufacturers. The contents of this article do not necessarily reflect the views of HEI, or its sponsors, nor do they necessarily reflect the views and policies of the EPA or motor vehicle and engine manufacturers.”

Publisher Copyright:
© 2021 The Authors

Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.

Toegang tot document

10.1016/j.envint.2020.106371

Handle.net

Permanente koppeling

Andere bestanden en links

Link to publication in Scopus

Citeer dit

Samoli, E., Rodopoulou, S., Hvidtfeldt, U. A., Wolf, K., Stafoggia, M., Brunekreef, B., Strak, M., Chen, J., Andersen, Z. J., Atkinson, R., Bauwelinck, M., Bellander, T., Brandt, J., Cesaroni, G., Forastiere, F., Fecht, D., Gulliver, J., Hertel, O., Hoffmann, B., ... Katsouyanni, K. (2021). Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project. Environment International, 147, Artikel 106371. https://doi.org/10.1016/j.envint.2020.106371

@article{ab4859ca75a04fc18e654695395aec38,

title = "Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project",

abstract = "Background: We evaluated methods for the analysis of multi-level survival data using a pooled dataset of 14 cohorts participating in the ELAPSE project investigating associations between residential exposure to low levels of air pollution (PM2.5 and NO2) and health (natural-cause mortality and cerebrovascular, coronary and lung cancer incidence).Methods: We applied five approaches in a multivariable Cox model to account for the first level of clustering corresponding to cohort specification: (1) not accounting for the cohort or using (2) indicator variables, (3) strata, (4) a frailty term in frailty Cox models, (5) a random intercept under a mixed Cox, for cohort identification. We accounted for the second level of clustering due to common characteristics in the residential area by (1) a random intercept per small area or (2) applying variance correction. We assessed the stratified, frailty and mixed Cox approach through simulations under different scenarios for heterogeneity in the underlying hazards and the air pollution effects.Results: Effect estimates were stable under approaches used to adjust for cohort but substantially differed when no adjustment was applied. Further adjustment for the small area grouping increased the effect estimates{\textquoteright} standard errors. Simulations confirmed identical results between the stratified and frailty models. In ELAPSE we selected a stratified multivariable Cox model to account for between-cohort heterogeneity without adjustment for small area level, due to the small number of subjects and events in the latter.Conclusions: Our study supports the need to account for between-cohort heterogeneity in multi-center collaborations using pooled individual level data.",

keywords = "Air pollution, Cox model, Frailty models, Health effects, Mixed models, Multi-level analysis",

author = "Evangelia Samoli and Sophia Rodopoulou and Hvidtfeldt, {Ulla A.} and Kathrin Wolf and Massimo Stafoggia and Bert Brunekreef and Maciej Strak and Jie Chen and Andersen, {Zorana J.} and Richard Atkinson and Mariska Bauwelinck and Tom Bellander and J{\o}rgen Brandt and Giulia Cesaroni and Francesco Forastiere and Daniela Fecht and John Gulliver and Ole Hertel and Barbara Hoffmann and {de Hoogh}, Kees and Janssen, {Nicole A.H.} and Matthias Ketzel and Klompmaker, {Jochem O.} and Shuo Liu and Petter Ljungman and Gabriele Nagel and Bente Oftedal and G{\"o}ran Pershagen and Annette Peters and Ole Raaschou-Nielsen and Matteo Renzi and Kristoffersen, {Doris T.} and Gianluca Severi and Torben Sigsgaard and Danielle Vienneau and Gudrun Weinmayr and Gerard Hoek and Klea Katsouyanni",

note = "Funding Information: This work was supported by Health Effects Institute (HEI) Research Agreement [grant number #4954-RFA14-3/16-5-3]. Funding Information: “Research described in this article was conducted under contract to the Health Effects Institute (HEI), an organization jointly funded by the United States Environmental Protection Agency (EPA) (Assistance Award No. R-82811201) and certain motor vehicle and engine manufacturers. The contents of this article do not necessarily reflect the views of HEI, or its sponsors, nor do they necessarily reflect the views and policies of the EPA or motor vehicle and engine manufacturers.” Publisher Copyright: {\textcopyright} 2021 The Authors Copyright: Copyright 2021 Elsevier B.V., All rights reserved.",

year = "2021",

month = feb,

doi = "10.1016/j.envint.2020.106371",

language = "English",

volume = "147",

journal = "Environment International",

issn = "0160-4120",

publisher = "Elsevier Limited",

}

Samoli, E, Rodopoulou, S, Hvidtfeldt, UA, Wolf, K, Stafoggia, M, Brunekreef, B, Strak, M, Chen, J, Andersen, ZJ, Atkinson, R, Bauwelinck, M, Bellander, T, Brandt, J, Cesaroni, G, Forastiere, F, Fecht, D, Gulliver, J, Hertel, O, Hoffmann, B, de Hoogh, K, Janssen, NAH, Ketzel, M, Klompmaker, JO, Liu, S, Ljungman, P, Nagel, G, Oftedal, B, Pershagen, G, Peters, A, Raaschou-Nielsen, O, Renzi, M, Kristoffersen, DT, Severi, G, Sigsgaard, T, Vienneau, D, Weinmayr, G, Hoek, G & Katsouyanni, K 2021, 'Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project', Environment International, vol. 147, 106371. https://doi.org/10.1016/j.envint.2020.106371

TY - JOUR

T1 - Modeling multi-level survival data in multi-center epidemiological cohort studies

T2 - Applications from the ELAPSE project

AU - Samoli, Evangelia

AU - Rodopoulou, Sophia

AU - Hvidtfeldt, Ulla A.

AU - Wolf, Kathrin

AU - Stafoggia, Massimo

AU - Brunekreef, Bert

AU - Strak, Maciej

AU - Chen, Jie

AU - Andersen, Zorana J.

AU - Atkinson, Richard

AU - Bauwelinck, Mariska

AU - Bellander, Tom

AU - Brandt, Jørgen

AU - Cesaroni, Giulia

AU - Forastiere, Francesco

AU - Fecht, Daniela

AU - Gulliver, John

AU - Hertel, Ole

AU - Hoffmann, Barbara

AU - de Hoogh, Kees

AU - Janssen, Nicole A.H.

AU - Ketzel, Matthias

AU - Klompmaker, Jochem O.

AU - Liu, Shuo

AU - Ljungman, Petter

AU - Nagel, Gabriele

AU - Oftedal, Bente

AU - Pershagen, Göran

AU - Peters, Annette

AU - Raaschou-Nielsen, Ole

AU - Renzi, Matteo

AU - Kristoffersen, Doris T.

AU - Severi, Gianluca

AU - Sigsgaard, Torben

AU - Vienneau, Danielle

AU - Weinmayr, Gudrun

AU - Hoek, Gerard

AU - Katsouyanni, Klea

N1 - Funding Information: This work was supported by Health Effects Institute (HEI) Research Agreement [grant number #4954-RFA14-3/16-5-3]. Funding Information: “Research described in this article was conducted under contract to the Health Effects Institute (HEI), an organization jointly funded by the United States Environmental Protection Agency (EPA) (Assistance Award No. R-82811201) and certain motor vehicle and engine manufacturers. The contents of this article do not necessarily reflect the views of HEI, or its sponsors, nor do they necessarily reflect the views and policies of the EPA or motor vehicle and engine manufacturers.” Publisher Copyright: © 2021 The Authors Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021/2

Y1 - 2021/2

N2 - Background: We evaluated methods for the analysis of multi-level survival data using a pooled dataset of 14 cohorts participating in the ELAPSE project investigating associations between residential exposure to low levels of air pollution (PM2.5 and NO2) and health (natural-cause mortality and cerebrovascular, coronary and lung cancer incidence).Methods: We applied five approaches in a multivariable Cox model to account for the first level of clustering corresponding to cohort specification: (1) not accounting for the cohort or using (2) indicator variables, (3) strata, (4) a frailty term in frailty Cox models, (5) a random intercept under a mixed Cox, for cohort identification. We accounted for the second level of clustering due to common characteristics in the residential area by (1) a random intercept per small area or (2) applying variance correction. We assessed the stratified, frailty and mixed Cox approach through simulations under different scenarios for heterogeneity in the underlying hazards and the air pollution effects.Results: Effect estimates were stable under approaches used to adjust for cohort but substantially differed when no adjustment was applied. Further adjustment for the small area grouping increased the effect estimates’ standard errors. Simulations confirmed identical results between the stratified and frailty models. In ELAPSE we selected a stratified multivariable Cox model to account for between-cohort heterogeneity without adjustment for small area level, due to the small number of subjects and events in the latter.Conclusions: Our study supports the need to account for between-cohort heterogeneity in multi-center collaborations using pooled individual level data.

AB - Background: We evaluated methods for the analysis of multi-level survival data using a pooled dataset of 14 cohorts participating in the ELAPSE project investigating associations between residential exposure to low levels of air pollution (PM2.5 and NO2) and health (natural-cause mortality and cerebrovascular, coronary and lung cancer incidence).Methods: We applied five approaches in a multivariable Cox model to account for the first level of clustering corresponding to cohort specification: (1) not accounting for the cohort or using (2) indicator variables, (3) strata, (4) a frailty term in frailty Cox models, (5) a random intercept under a mixed Cox, for cohort identification. We accounted for the second level of clustering due to common characteristics in the residential area by (1) a random intercept per small area or (2) applying variance correction. We assessed the stratified, frailty and mixed Cox approach through simulations under different scenarios for heterogeneity in the underlying hazards and the air pollution effects.Results: Effect estimates were stable under approaches used to adjust for cohort but substantially differed when no adjustment was applied. Further adjustment for the small area grouping increased the effect estimates’ standard errors. Simulations confirmed identical results between the stratified and frailty models. In ELAPSE we selected a stratified multivariable Cox model to account for between-cohort heterogeneity without adjustment for small area level, due to the small number of subjects and events in the latter.Conclusions: Our study supports the need to account for between-cohort heterogeneity in multi-center collaborations using pooled individual level data.

KW - Air pollution

KW - Cox model

KW - Frailty models

KW - Health effects

KW - Mixed models

KW - Multi-level analysis

UR - http://www.scopus.com/inward/record.url?scp=85098956014&partnerID=8YFLogxK

U2 - 10.1016/j.envint.2020.106371

DO - 10.1016/j.envint.2020.106371

M3 - Article

AN - SCOPUS:85098956014

SN - 0160-4120

VL - 147

JO - Environment International

JF - Environment International

M1 - 106371

ER -

Modeling multi-level survival data in multi-center epidemiological cohort studies: Applications from the ELAPSE project

Samenvatting

Bibliografische nota

Toegang tot document

Handle.net

Andere bestanden en links

Vingerafdruk

Citeer dit