Computational Statistics & Data Analysis 42 (2003) 243–262 www.elsevier.com/locate/csda Joint modelling of cause-specic hazard functions with cubic splines: an application to a large series of breast cancer patients Patrizia Boracchi a , Elia Biganzoli b; , Ettore Marubini a; b a Istituto di Statistica Medica e Biometria, Universit a degli Studi di Milano, Italy b Unit a di Statistica Medica e Biometria, Istituto Nazionale per lo Studio e la Cura dei Tumori, Via G.Venezian 1, 20133, Milano, Italy Received 1 November 2000; received in revised form 1 February 2002 Abstract The time of appearance of several kinds of relapses after a therapeutic intervention is of increasing interest in oncology. Typically, in breast cancer patients, events of clinical interest are intra-breast tumor recurrences and distant metastases, which act in a competitive way when considered as rst failure. The evaluation of dierential eects of clinical and biological vari- ables on each event can improve the knowledge on the course of the disease and the targeting of future therapy. A simple tool for the joint smoothed estimation of cause-specic hazards func- tions and continuous covariate eects has been developed. Within the framework of generalized linear models with Poisson error, an extension of the piecewise exponential model is proposed, based on grouping follow-up times and continuous covariates. Interpolation of cause-specic hazards is obtained by resorting to cubic splines, which are piecewise polynomials of simple implementation with standard statistical software; their exibility and smoothness are easily con- trolled by the number of knots and constraints on polynomial derivatives. The approach was applied to a data set of 2233 breast cancer patients treated with conservative surgery. It al- lowed modelling time-dependent and cause-specic eects of covariates on the hazard functions. c 2002 Published by Elsevier Science B.V. Keywords: Competing risks; Hazard regression; GLMs; Splines; Breast cancer * Corresponding author. Tel.: +39-0223903203; fax: +39-022362930. E-mail address: biganzoli@institutotumori.mi.it (E. Biganzoli). 0167-9473/02/$-see front matter c 2002 Published by Elsevier Science B.V. PII:S0167-9473(02)00122-6