Robustness of estimation methods in a survival cure model with mismeasured covariates

In medical applications, one frequently encounters time-to-event data. While classical survival methods are well known and broadly used to analyze such data, they do not allow one to take into account two phenomena which appear quite often in practice: individuals who will never experience the event of inter- est (they are cured from this event) and measurement error in the continuous covariates. This paper deals with a model designed to take both features into account. Two approaches exist in the literature to estimate such a model. However, while they work well in many settings, they require information about the distribution of the measurement error which is rarely fully known in practice. In this paper, we _rst justify the need to take the measurement error into account, via a theoretical study of the bias. We then present the results of an extensive simulation study investigating the robustness of both correction approaches with respect to their assumptions. The conclusions allow us to give some practical recommendations for similar situations. We conclude by analyzing the time until recurrence after surgery for rectal cancer patients, taking into account the advice from the simulation results. Both correction methods have been implemented in the R package miCoPTCM.

• metadata
• relations

Document type	Document de travail (Working Paper)
Access type	Accès libre
Publication date	2016
Language	Anglais
Number of pages	31 pages
Collection	ISBA Discussion Paper - 2016/06
Affiliation	UCL - SSH/LIDAM/ISBA - Institut de Statistique, Biostatistique et Sciences Actuarielles
Keywords	Semiparametric method ; Bias correction ; Cure fraction ; Measurement error ; Promotion time cure model
Links	http://hdl.handle.net/2078.1/171508[Handle]

See also
	[boreal:183241]	Robustness of estimation methods in a survival cure model with mismeasured covariates