Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/139067
Title: Estrategias para el abordaje de la problemática de los datos faltantes en ensayos clínicos longitudinales
Author: Pardo Montenegro, Beatriz
Tutor: Perez-Alvarez, Nuria  
Others: Ventura, Carles  
Abstract: Longitudinal clinical trials are done with repeated measures of some variables over a period of time. It is a common problem to find that some of these measures are missing. Developing appropriate strategies for dealing with missing data is a major challenge. One of the recommended alternatives is multiple imputation. Survival analyses look at the time it takes for an event of interest to occur, if it does not occur within the follow-up time of the trial, it is called censoring. There are numerous alternatives for performing survival analysis with censored data, the most commonly used conventional method is Kaplan Meier. In recent years, machine learning algorithms have been developed, one of which is the random survival forest. In this TFM, after performing data management tasks, imputation of missing data from the Lake trial database is performed with 2 imputation options available in R, in the mice and randomForestSRC libraries. The results of the Kaplan Meier survival function and the random survival forest algorithm applied to the result of both imputations are compared. The event studied is virologic failure in HIV-diagnosed patients treated with experimental treatment vs. control. The results of the C-index are very similar. By Kaplan Meier, it is concluded that experimental treatment has fewer virological failures than standard treatment, but the differences are significant only in the database imputed with the randomForestSRC library.
Keywords: missing data
imputation
survival analysis
longitudinal trials
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Dec-2021
Publication license: http://creativecommons.org/licenses/by-nc-sa/3.0/es  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
bpardomTFM1221memoria.pdfMemoria del TFM1,63 MBAdobe PDFThumbnail
View/Open