No abstract

In this article, we propose a new approach to the problem of dynamic prediction of survival data in the presence of competing risks as an extension of the landmark model for ordinary survival data. The key feature of our method is the introduction of dynamic pseudo-observations constructed from the prediction probabilities at different landmark prediction times. They specifically address the issue of estimating covariate effects directly on the cumulative incidence scale in competing risks. A flexible generalized linear model based on these dynamic pseudo-observations and a generalized estimation equations approach to estimate the baseline and covariate effects will result in the desired dynamic predictions and robust standard errors. Our approach has a number of attractive features. It focuses directly on the prediction probabilities of interest, avoiding in this way complex modeling of cause-specific hazards or subdistribution hazards. As a result, it is robust against departures from these omnibus models. From a computational point of view an advantage of our approach is that it can be fitted with existing statistical software and that a variety of link functions and regression models can be considered, once the dynamic pseudo-observations have been estimated. We illustrate our approach on a real data set of chronic myeloid leukemia patients after bone marrow transplantation.

We propose an extension of the landmark model for ordinary survival data as a new approach to the problem of dynamic prediction in competing risks with time-dependent covariates. We fix a set of landmark time points tLM within the follow-up interval. For each of these landmark time points tLM , we create a landmark data set by selecting individuals at risk at tLM ; we fix the value of the time-dependent covariate in each landmark data set at tLM . We assume Cox proportional hazard models for the cause-specific hazards and consider smoothing the (possibly) time-dependent effect of the covariate for the different landmark data sets. Fitting this model is possible within the standard statistical software. We illustrate the features of the landmark modelling on a real data set on bone marrow transplantation.

Background In older adults pneumococcal disease is strongly associated with respiratory viral infections, but the impact of viruses on Streptococcus pneumoniae carriage prevalence and load remains poorly understood. Here, we investigated the effects of influenza-like illness (ILI) on pneumococcal carriage in community-dwelling older adults. Methods We investigated the presence of pneumococcal DNA in saliva samples collected in the 2014/2015 influenza season from 232 individuals aged ≥60 years at ILI-onset, followed by sampling 2-3 weeks and 7-9 weeks after the first sample. We also sampled 194 age-matched controls twice 2-3 weeks apart. Pneumococcal DNA was detected with quantitative-PCRs targeting piaB and lytA genes in raw and in culture-enriched saliva. Bacterial and pneumococcal abundances were determined in raw saliva with 16S and piaB quantification. Results The prevalence of pneumococcus-positive samples was highest at onset of ILI (18% or 42/232) and lowest among controls (13% or 26/194, and 11% or 22/194, at the first and second sampling moment, respectively), though these differences were not significant. Pneumococcal carriage was associated with exposure to young children (OR:2.71, 95%CI 1.51-5.02, p<0.001), and among asymptomatic controls with presence of rhinovirus infection (OR:4.23; 95%CI 1.16-14.22, p<0.05). When compared with carriers among controls, pneumococcal absolute abundances were significantly higher at onset of ILI (p<0.01), and remained elevated beyond recovery from ILI (p<0.05). Finally, pneumococcal abundances were highest in carriage events newly-detected after ILI-onset (estimated geometric mean 1.21E -5, 95%CI 2.48E -7-2.41E -5, compared with pre-existing carriage). Conclusions ILI exacerbates pneumococcal colonization of the airways in older adults, and this effect persists beyond recovery from ILI.

We study an alternative approach for estimation in the competing risks framework, called vertical modeling. It is motivated by a decomposition of the joint distribution of time and cause of failure. The two elements of this decomposition are (1) the time of failure and (2) the cause of failure condition on time of failure. Both elements of the model are based on observable quantities, namely the total hazard and the relative cause-specific hazards. The model can be implemented using the standard software. The relative cause-specific hazards are flexibly estimated using multinomial logistic regression and smoothing splines. We show estimates of cumulative incidences from vertical modeling to be more efficient statistically than those obtained from the standard nonparametric model. We illustrate our methods using data of 8966 leukemia patients from the European Group for Blood and Marrow Transplantation.