How much of the past matters? Using dynamic survival models for the monitoring of potassium in heart failure patients using electronic health records

Statistical methods to study the association between a longitudinal biomarker and the risk of death are very relevant for the long-term care of subjects affected by chronic illnesses, such as potassium in heart failure patients. Particularly in presence of comorbidities or pharmacological treatments, sudden crises can cause potassium to undergo very abrupt yet transient changes. In the context of the monitoring of potassium, there is the need for a dynamic model that can be used in clinical practice to assess the risk of death related to an observed patient's potassium trajectory. We considered different dynamic survival approaches, starting from the simple approach considering the most recent measurement, to the joint model. We then propose a novel method based on wavelet filtering and landmarking to retrieve the prognostic role of short-term potassium oscillations. The data used comes from over 2000 subjects, with a total of over 80 000 repeated potassium measurements collected through Administrative Health Records and Outpatient and Inpatient Clinic E-chart. We claim that while taking into account for past information is important, not all past information is equally informative. Recent, even if short-term changes in potassium contain relevant prognostic information. However, we show that these oscillations are not captured by state-of-the-art dynamic survival models. The latter are prone to give more importance to the mean long-term value of potassium. As a consequence, a novel dynamic survival approach is proposed in this work for the monitoring of potassium in heart failure. The proposed wavelet landmark method shows promising results revealing the prognostic role of short-term oscillations, according to their different duration, and achieving higher performances in predicting survival probability based on potassium over time.