-
The Brier Score under Administrative Censoring: Problems and Solutions
Authors:
Håvard Kvamme,
Ørnulf Borgan
Abstract:
The Brier score is commonly used for evaluating probability predictions. In survival analysis, with right-censored observations of the event times, this score can be weighted by the inverse probability of censoring (IPCW) to retain its original interpretation. It is common practice to estimate the censoring distribution with the Kaplan-Meier estimator, even though it assumes that the censoring dis…
▽ More
The Brier score is commonly used for evaluating probability predictions. In survival analysis, with right-censored observations of the event times, this score can be weighted by the inverse probability of censoring (IPCW) to retain its original interpretation. It is common practice to estimate the censoring distribution with the Kaplan-Meier estimator, even though it assumes that the censoring distribution is independent of the covariates. This paper discusses the general impact of the censoring estimates on the Brier score and shows that the estimation of the censoring distribution can be problematic. In particular, when the censoring times can be identified from the covariates, the IPCW score is no longer valid. For administratively censored data, where the potential censoring times are known for all individuals, we propose an alternative version of the Brier score. This administrative Brier score does not require estimation of the censoring distribution and is valid even if the censoring times can be identified from the covariates.
△ Less
Submitted 18 December, 2019;
originally announced December 2019.
-
Continuous and Discrete-Time Survival Prediction with Neural Networks
Authors:
Håvard Kvamme,
Ørnulf Borgan
Abstract:
Application of discrete-time survival methods for continuous-time survival prediction is considered. For this purpose, a scheme for discretization of continuous-time data is proposed by considering the quantiles of the estimated event-time distribution, and, for smaller data sets, it is found to be preferable over the commonly used equidistant scheme. Furthermore, two interpolation schemes for con…
▽ More
Application of discrete-time survival methods for continuous-time survival prediction is considered. For this purpose, a scheme for discretization of continuous-time data is proposed by considering the quantiles of the estimated event-time distribution, and, for smaller data sets, it is found to be preferable over the commonly used equidistant scheme. Furthermore, two interpolation schemes for continuous-time survival estimates are explored, both of which are shown to yield improved performance compared to the discrete-time estimates. The survival methods considered are based on the likelihood for right-censored survival data, and parameterize either the probability mass function (PMF) or the discrete-time hazard rate, both with neural networks. Through simulations and study of real-world data, the hazard rate parametrization is found to perform slightly better than the parametrization of the PMF. Inspired by these investigations, a continuous-time method is proposed by assuming that the continuous-time hazard rate is piecewise constant. The method, named PC-Hazard, is found to be highly competitive with the aforementioned methods in addition to other methods for survival prediction found in the literature.
△ Less
Submitted 15 October, 2019;
originally announced October 2019.
-
Time-to-Event Prediction with Neural Networks and Cox Regression
Authors:
Håvard Kvamme,
Ørnulf Borgan,
Ida Scheel
Abstract:
New methods for time-to-event prediction are proposed by extending the Cox proportional hazards model with neural networks. Building on methodology from nested case-control studies, we propose a loss function that scales well to large data sets, and enables fitting of both proportional and non-proportional extensions of the Cox model. Through simulation studies, the proposed loss function is verif…
▽ More
New methods for time-to-event prediction are proposed by extending the Cox proportional hazards model with neural networks. Building on methodology from nested case-control studies, we propose a loss function that scales well to large data sets, and enables fitting of both proportional and non-proportional extensions of the Cox model. Through simulation studies, the proposed loss function is verified to be a good approximation for the Cox partial log-likelihood. The proposed methodology is compared to existing methodologies on real-world data sets, and is found to be highly competitive, typically yielding the best performance in terms of Brier score and binomial log-likelihood. A python package for the proposed methods is available at https://github.com/havakv/pycox.
△ Less
Submitted 13 September, 2019; v1 submitted 1 July, 2019;
originally announced July 2019.
-
Do Japanese and Italian women live longer than women in Scandinavia?
Authors:
Ørnulf Borgan
Abstract:
Life expectancies at birth are routinely computed from period life tables. Such period life expectancies may be distorted by selection when comparing countries where the living conditions improved earlier (like Norway and Sweden) with countries where they improved later (like Italy and Japan). One way to get a fair comparison between the countries, is to use cohort data and consider the expected n…
▽ More
Life expectancies at birth are routinely computed from period life tables. Such period life expectancies may be distorted by selection when comparing countries where the living conditions improved earlier (like Norway and Sweden) with countries where they improved later (like Italy and Japan). One way to get a fair comparison between the countries, is to use cohort data and consider the expected number of years lost before a given age a. Contrary to the results based on period data, one then finds that Italian women may expect to lose more years than women in Norway and Sweden, while there are no indications that Japanese women will lose fewer years than Scandinavian women.
△ Less
Submitted 21 September, 2016;
originally announced September 2016.
-
Dynamic path analysis - A useful tool to investigate mediation processes in clinical survival trials
Authors:
Susanne Strohmaier,
Kjetil Røysland,
Rune Hoff,
Ørnulf Borgan,
Terje Pedersen,
Odd O. Aalen
Abstract:
When it comes to clinical survival trials, regulatory restrictions usually require the application of methods that solely utilize baseline covariates and the intention-to-treat principle. Thereby a lot of potentially useful information is lost, as collection of time-to-event data often goes hand in hand with collection of information on biomarkers and other internal time-dependent covariates. Howe…
▽ More
When it comes to clinical survival trials, regulatory restrictions usually require the application of methods that solely utilize baseline covariates and the intention-to-treat principle. Thereby a lot of potentially useful information is lost, as collection of time-to-event data often goes hand in hand with collection of information on biomarkers and other internal time-dependent covariates. However, there are tools to incorporate information from repeated measurements in a useful manner that can help to shed more light on the underlying treatment mechanisms. We consider dynamic path analysis, a model for mediation analysis in the presence of a time-to-event outcome and time-dependent covariates to investigate direct and indirect effects in a study of different lipid lowering treatments in patients with previous myocardial infarctions. Further, we address the question whether survival in itself may produce associations between the treatment and the mediator in dynamic path analysis and give an argument that, due to linearity of the assumed additive hazard model, this is not the case. We further elaborate on our view that, when studying mediation, we are actually dealing with underlying processes rather than single variables measured only once during the study period. This becomes apparent in results from various models applied to the study of lipid lowering treatments as well as our additionally conducted simulation study, where we clearly observe, that discarding information on repeated measurements can lead to potentially erroneous conclusions.
△ Less
Submitted 24 April, 2015;
originally announced April 2015.
-
History of applications of martingales in survival analysis
Authors:
Odd O. Aalen,
Per Kragh Andersen,
Ørnulf Borgan,
Richard D. Gill,
Niels Keiding
Abstract:
The paper traces the development of the use of martingale methods in survival analysis from the mid 1970's to the early 1990's. This development was initiated by Aalen's Berkeley PhD-thesis in 1975, progressed through the work on estimation of Markov transition probabilities, non-parametric tests and Cox's regression model in the late 1970's and early 1980's, and it was consolidated in the early 1…
▽ More
The paper traces the development of the use of martingale methods in survival analysis from the mid 1970's to the early 1990's. This development was initiated by Aalen's Berkeley PhD-thesis in 1975, progressed through the work on estimation of Markov transition probabilities, non-parametric tests and Cox's regression model in the late 1970's and early 1980's, and it was consolidated in the early 1990's with the publication of the monographs by Fleming and Harrington (1991) and Andersen, Borgan, Gill and Keiding (1993). The development was made possible by an unusually fast technology transfer of pure mathematical concepts, primarily from French probability, into practical biostatistical methodology, and we attempt to outline some of the personal relationships that helped this happen. We also point out that survival analysis was ready for this development since the martingale ideas inherent in the deep understanding of temporal development so intrinsic to the French theory of processes were already quite close to the surface in survival analysis.
△ Less
Submitted 8 December, 2022; v1 submitted 28 February, 2010;
originally announced March 2010.
