Services on Demand
Journal
Article
English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mailIndicators
-
Cited by SciELO -
Access statistics
Related links
-
Cited by Google -
Similars in
SciELO -
Similars in Google
Share
Permalink
Revista Colombiana de Estadística
Print version ISSN 0120-1751
Rev.Colomb.Estad. vol.45 no.1 Bogotá Jan./June 2022 Epub Jan 17, 2023
https://doi.org/10.15446/rce.v45n1.94553
Artículos originales de investigación
Causal Mediation for Survival Data: A Unifying Approach via GLM
Mediación causal para datos de supervivencia: un enfoque unificador a través de GLM
1 DEPARTMENTO DE ESTATÍSTICA, INSTITUTO DE MATEMÁTICA E ESTATÍSTICA, UNIVERSIDADE FEDERAL DA BAHIA, SALVADOR, BRAZIL
Mediation analysis has been receiving much attention from the scientific community in the last years, mainly due to its ability to disentangle causal pathways from exposures to outcomes. Particularly, causal mediation analysis for time-to-event outcomes has been widely discussed using accelerated failures times, Cox and Aalen models, with continuous or binary mediator. We derive general expressions for the Natural Direct Effect and Natural Indirect Effect for the time-to-event outcome when the mediator is modeled using generalized linear models, which includes existing procedures as particular cases. We also define a responsiveness measure to assess the variations in continuous exposures in the presence of mediation. We consider a community-based prospective cohort study that investigates the mediation of hepatitis B in the relationship between hepatitis C and liver cancer. We fit different models as well as distinct distributions and link functions associated to the mediator. We also notice that estimation of NDE and NIE using different models leads to non-contradictory conclusions despite their effect scales. The survival models provide a compelling framework that is appropriate to answer many research questions involving causal mediation analysis. The extensions through GLMs for the mediator may encompass a broad field of medical research, allowing the often necessary control for confounding.
Key words: causal inference; generalized linear models; mediation; survival analysis
El análisis de mediación ha recibido mucha atención en los últimos años, principalmente debido a su capacidad para desenredar las vías causales. Particularmente, mediación causal para el tiempo hasta el evento se ha discutido ampliamente utilizando tiempos de falla acelerados, modelos de Cox y Aalen, con mediador continuo o binario. Derivamos expresiones generales para el efecto directo natural y el efecto indirecto natural para el el tiempo hasta el evento cuando el mediador se modela utilizando modelos lineales generalizados, que incluyen procedimientos existentes como casos particulares. Definimos una medida para evaluar variaciones en exposiciones continuas en presencia de mediación. Consideramos un estudio de cohorte prospectivo que investiga la mediación de la hepatitis B en la relación entre la hepatitis C y el cáncer de hígado. Encajamos diferentes modelos, así como distintas distribuciones y funciones de enlace. Todos los enfoques dan como resultado evaluaciones consistentes de los effectos considerando sus correspondientes escalas. Los modelos de supervivencia proporcionan un marco convincente apropiado para responder a muchas preguntas de investigación que involucran mediación causal. Las extensiones a través de GLM para el mediador pueden abarcar un amplio campo de investigación médica, lo que permite el control necesario para los factores de confusión.
Palabras clave: análisis de supervivencia; inferencia causal; mediación; modelos lineales generalizados
References
Aalen, O. O., Borgan, O. & Gjessing, H. K. (2008), Survival and Event History Analysis: A Process Point of View, Springer, New York. [ Links ]
Aalen, O., R0ysland, K. & Gran, J. (2012), 'Causality, mediation and time: a dynamic viewpoint', Journal of the Royal Statistical Society - Series A 175(4), 831-861. [ Links ]
Aalen, O., Stensrud, M., Didelez, V., Daniel, R., Røysland, K. & Strohmaier, S. (2020), 'Time-dependent mediators in survival analysis: Modeling direct and indirect effects with the additive hazards model', Biometrical Journal 62, 532-549. [ Links ]
Albert, J. & Nelson, S. (2011), 'Generalized Causal Mediation Analysis', Biometrics 67, 1028-1038. [ Links ]
Baron, R. M. & Kenny, D. A. (1986), 'The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations', Journal of Personality and Social Psychology 51, 1173-1182. [ Links ]
Bender, R., Augustin, T. & Blettner, M. (2005), 'Generating survival times to simulate Cox proportional hazards models', Statistics in Medicine 24, 1713-1723. [ Links ]
Daniel, R., Stavola, B., Cousens, S. & Vansteelandt, S. (2015), 'Causal mediation analysis with multiple mediators', Biometrics 71, 13-14. [ Links ]
Didelez, V. (2019), 'Defining causal mediation with a longitudinal mediator and a survival outcome', Lifetime Data Analysis 25, 593-610. [ Links ]
Fasanelli, F., Giraudo, M., Ricceri, F., Valeri, L. & Zugna, D. (2019), 'Marginal Time-Dependent Causal Effects in Mediation Analysis With Survival Data', American Journal of Epidemiology 188(5), 967-974. [ Links ]
Fulcher, I., Tchetgen Tchetgen, E. & Williams, P. (2017), 'Mediation Analysis for Censored Survival Data Under an Accelerated Failure Time Model', Epidemiology 28(5), 660-666. [ Links ]
Huang, Y.-T., Jen, C.-L., Yang, H.-I., Lee, M.-H., Lu, S.-N., Iloeje, U. & Chen, C.-J. (2011), 'Lifetime risk and sex difference of hepatocellular carcinoma among patients with chronic hepatitis B and C', Journal of Clinical Oncology 29(27), 3643-3650. [ Links ]
Huang, Y.-T. & Yang, H.-I. (2017), 'Causal mediation analysis of survival outcomes with multiple mediators', Epidemiology 28(3), 370-378. [ Links ]
Iacobucci, D. (2012), 'Mediation analysis and categorical variables: the final frontier', Journal of Consumer Psychology 22, 582-594. [ Links ]
Imai, K., Keele, L. & Tingley, D. (2010), 'A general approach to causal mediation analysis', Psychological Methods 15, 309-334. [ Links ]
Kalbfleisch, J. D. & Prentice, R. L. (2002), The Statistical Analysis of Failure Time Data, 2 edn, Wiley, Hoboken, N.J. [ Links ]
Lange, T. & Hansen, J. (2011), 'Direct and indirect effects in a survival contex', Epidemiology 22, 575-581. [ Links ]
Lange, T., Vansteelandt, S. & Bekaert, M. (2012), 'A simple unified approach for estimating natural direct and indirect effects', American Journal of Epidemiology 176, 190-195. [ Links ]
Lin, S.-H., Young, J., Logan, R. & VanderWeele, T. (2017), 'Mediation analysis with a survival outcome with time-varying exposures, mediators and confounders', Statistics in Medicine 36, 4153-4166. [ Links ]
Loyes, T., Moerkerke, B., De Smet, O., Byusse, A.S,. & J. Vansteelandt, S. (2013), 'Flexible mediation analysis in the presence of nonlinear relations: beyond the mediation formula', Multivariate Behavioral Research 48(6), 871-894. [ Links ]
MacKinnon, D. (2008), Introduction to statistical mediation analysis, Taylor and Francis, New York. [ Links ]
MacKinnon, D. P. & Dwyer, J. H. (1993), 'Estimating mediated effects in prevention studies', Evaluation Review 17, 144-158. [ Links ]
MacKinnon, D., Warsi, G. & Dwyer, J. (1995), 'A simulation study of mediated effect measures', Multivariate Behavioral Research 30, 41-62. [ Links ]
Miles, C., Shpitser, I., Kanki, P., Meloni, S. & Tchetgen Tchetgen, E. (2020), 'On semiparametric estimation of apath-specific effect in the presence of mediator-outcome confounding', Biometrika 107(1), 159-172. [ Links ]
Pearl, J. (2001), 'Direct and indirect effects', Proceedings of the 17th Annual Conference on Uncertainty in Artificial Intelligence (UAI-01) pp. 411-442. [ Links ]
Pratschke, J., Haase, T., Comber, H., Sharp, L., Cancela, M. & Johnson, H. (2016), 'Mechanisms and mediation in survival analysis: towards an integrated analytical framework', BMC Medical Research Methodology pp. 16-27. [ Links ]
Robins, J., Hernán, M. & Brumback, B. (2000), 'Marginal structural models and causal inference in epidemiology', Epidemiology 11, 550-560. [ Links ]
Smyth, G. (1989), 'Generalized linear models with varying dispersion', Journal of the Royal Statistical Society: Series B 51(1), 47-60. [ Links ]
Tchetgen Tchetgen, E. (2013), 'Inverse odds ratio-weighted estimation for causal mediation analysis', Statistics in Medicine 32, 4567-4580. [ Links ]
Tchetgen Tchetgen, E. & Shpitser, I. (2012), 'Semiparametric theory for causal mediation analysis: efficiency bounds, multiple robustness and sensitivity analysis', Annals of statistics 40(3), 1816-1845. [ Links ]
VanderWeele, T. (2011), 'Causal mediation analysis with survival data', Epidemiology 22(4), 582-585. [ Links ]
VanderWeele, T. (2015), Explanation in Causal Inference: Methods for Mediation and Interaction, Oxford University Press, New York. [ Links ]
VanderWeele, T. (2016), 'Mediation Analysis: a Practitioner's Guide', Annual Review of Public Health 37, 17-32. [ Links ]
VanderWeele, T. & Vansteelandt, S. (2009), 'Conceptual issues concerning mediation, interventions and composition', Statistics and its Interface 2, 457-468. [ Links ]
VanderWeele, T. & Vansteelandt, S. (2010), 'Odds ratios for mediation analysis with a dichotomous outcome', American Journal of Epidemiology 172, 1339-1348. [ Links ]
Vansteelandt, S., Linder, M., Vandenberghe, S., Steen, J. & Madsen, J. (2019), 'Mediation analysis of time-to-event endpoints accounting for repeatedly measured mediators subject to time-varying confounding', Statistics in Medicine 38, 4828-4840. [ Links ]
Received: March 2021; Accepted: December 2021
This is an open-access article distributed under the terms of the Creative Commons Attribution License
