A1374
Title: Monitoring time to event in medical registry data using CUSUMs based on excess hazard models
Authors: Jimmy Huy Tran - University of Stavanger (Norway)
Jan Terje Kvaloy - University of Stavanger (Norway) [presenting]
Hartwig Korner - Stavanger University Hospital (Norway)
Abstract: In health registries, patients are routinely registered at diagnosis, and outcome data like survival times are added later. Based on such data, an aspect of interest could be to monitor whether the distribution of the time to an outcome of interest changes over time, for instance, if the survival time distribution of cancer patients changes over time, while adjusting for known risk factors. A challenge in monitoring survival times based on such registry data is that the cause of death is often not registered. To quantify the burden of disease in such cases, excess hazard methods, where the total hazard is modelled as the population hazard plus the excess hazard due to the disease, can be used. The aim is to propose a CUSUM procedure for monitoring for changes in the time to event distribution in such cases. The procedure is based on a survival loglikelihood ratio which extends previously suggested methods for monitoring of time to event to the excess hazard setting. The procedure considers changes in the population risk over time, as well as changes in the excess hazard explained by observed covariates. Properties, challenges, and an application to cancer registry data are presented.
