City Research Online

Link-based survival additive models under mixed censoring to assess risks of hospital-acquired infections

Marra, G., Farcomeni, A. & Radice, R. ORCID: 0000-0002-6316-3961 (2020). Link-based survival additive models under mixed censoring to assess risks of hospital-acquired infections. Computational Statistics and Data Analysis, 155, 107092. doi: 10.1016/j.csda.2020.107092


The majority of methods available to model survival data only deal with right censoring. However, there are many applications where left, right and/or interval censoring simultaneously occur. A methodology that is capable of handling all types of censoring as well as flexibly estimating several types of covariate effects is presented. The baseline hazard is modelled through monotonic P-splines. The model’s parameters are estimated using an efficient and stable penalised likelihood algorithm. The proposed framework is evaluated in simulation, and illustrated using an original data example on time to first hospital infection or in-hospital death in cirrhotic patients. A peak of risk in the first week since hospitalisation is identified, together with a non-linear effect of Model for End-Stage Liver Disease (MELD) score. The GJRM R package, with an implementation of our approach, is freely available on CRAN.

Publication Type: Article
Additional Information: © 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license
Publisher Keywords: additive predictor; link function; mixed censoring; penalised log-likelihood; regression splines; survival data
Subjects: H Social Sciences > HF Commerce > HF5601 Accounting
R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
Departments: Bayes Business School > Actuarial Science & Insurance
Text - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (330kB) | Preview



Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login