AWavelet-mixed Effect Landmark Model for the Effect of Potassium and Biomarkers Profiles on Survival in Heart Failure Patients

Statistical methods to study the association between a longitudinal biomarker and the risk of death are a very relevant problem for the long-term monitoring of biomarkers. In this context, sudden crises can cause the biomarker to undergo very abrupt changes. Although these oscillations are typically short-term, they often contain relevant prognostic information. We propose a method that couples a linear mixed-model with a wavelet smoothing to extract both the long-term component and the short-term oscillations of the individual longitudinal biomarker profiles, and describe them as functional data. We then use them as predictors in a landmark model to study their association with the risk of death. To illustrate the method, we use clinical application which motivated our work, i.e. the monitoring of potassium and related biomarkers in Heart Failure patients. The dataset consists of real-world data coming from the integration of Administrative Health Records and Outpatient and Inpatient Clinic E-chart from Trieste (Italy). Our method not only allows us to identify the short-term oscillations, but also reveals their prognostic role, according to their duration, demonstrating the importance of including them in the modeling. Compared to other state of the art methods (e.g., landmark analyses and joint models), our proposal archives higher predictive performances. Our analysis has also an important clinical implications, since it allows us to derive a dynamic score that can be used in clinical practice to assess the risk related to an observed patient’s potassium trajectory and then tune the actual drug therapy she/he has to undergo.