The main aim of mediation analysis is to study the direct and indirect effects of an exposure on an outcome. To date, the literature on mediation analysis with multiple mediators has mainly focused on continuous and dichotomous outcomes. However, the development of methods for multiple mediation analysis of survival outcome is still limited. Here we extend to survival outcomes a method for multiple mediation analysis based on the computation of appropriate weights. The approach considered has the advantage of not requiring specific models for mediators, allowing non-independent mediators of any nature and not relying on the assumption of rare outcomes. Simulation studies show a good performance of the proposed estimator, in terms of bias and coverage probability. The method is further applied to an example from a published study on mortality for prostate cancer aimed at understanding to what extent the effect of DNA methyltransferase genotype on mortality was explained by DNA methylation and tumor aggressiveness. The approach can be used to quantify the marginal time-dependent direct and indirect effects carried by multiple indirect pathways and a code is provided to facilitate its application.

Marginal Time-Dependent Causal Effects in Mediation Analysis With Survival Data

Fasanelli, Francesca;Giraudo, Maria Teresa;Ricceri, Fulvio;Zugna, Daniela
Last
2019

Abstract

The main aim of mediation analysis is to study the direct and indirect effects of an exposure on an outcome. To date, the literature on mediation analysis with multiple mediators has mainly focused on continuous and dichotomous outcomes. However, the development of methods for multiple mediation analysis of survival outcome is still limited. Here we extend to survival outcomes a method for multiple mediation analysis based on the computation of appropriate weights. The approach considered has the advantage of not requiring specific models for mediators, allowing non-independent mediators of any nature and not relying on the assumption of rare outcomes. Simulation studies show a good performance of the proposed estimator, in terms of bias and coverage probability. The method is further applied to an example from a published study on mortality for prostate cancer aimed at understanding to what extent the effect of DNA methyltransferase genotype on mortality was explained by DNA methylation and tumor aggressiveness. The approach can be used to quantify the marginal time-dependent direct and indirect effects carried by multiple indirect pathways and a code is provided to facilitate its application.
188
5
967
974
http://aje.oxfordjournals.org/
multiple mediation analysis; proportional hazards model; pure direct effect; total indirect effect; weighting approach;
Fasanelli, Francesca; Giraudo, Maria Teresa; Ricceri, Fulvio; Valeri, Linda; Zugna, Daniela
File in questo prodotto:
File Dimensione Formato  
30689682_zugna.doc

Accesso riservato

Tipo di file: POSTPRINT (VERSIONE FINALE DELL’AUTORE)
Dimensione 8.62 MB
Formato Adobe PDF
8.62 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
30689682_ZUGNA.pdf

Accesso riservato

Tipo di file: PDF EDITORIALE
Dimensione 470.08 kB
Formato Adobe PDF
470.08 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
30689682_zugna.pdf

Accesso aperto

Tipo di file: POSTPRINT (VERSIONE FINALE DELL’AUTORE)
Dimensione 392.76 kB
Formato Adobe PDF
392.76 kB Adobe PDF Visualizza/Apri
kwz016.pdf

Accesso riservato

Tipo di file: PDF EDITORIALE
Dimensione 470.5 kB
Formato Adobe PDF
470.5 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
kwz016_giraudo_web_material_final.pdf

Accesso riservato

Tipo di file: PDF EDITORIALE
Dimensione 115.36 kB
Formato Adobe PDF
115.36 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/1695465
Citazioni
  • ???jsp.display-item.citation.pmc??? 2
  • Scopus 11
  • ???jsp.display-item.citation.isi??? 11
social impact