MODELING LONGITUDINAL MEASUREMENTS OF CLINICAL AND EPIDEMIOLOGICAL PARAMETERS USING JOINPOINT REGRESSION

Mihaela Moscalu, Ionela-Lăcrămioara Șerban, Norin Forna, Mihaela Pundiche, Loredana Mitran, Madalina-Elena Datcu, Dragomir Nicolae Șerban, Alin Constantin Pînzariu

DOI : 10.62610/RJOR.2025.2.17.16

ABSTRACT

Aim of the study The rapid advancement of medical technologies and the growing volume of data generated in clinical practice and research have created a continuous demand for sophisticated statistical tools for analysis. Currently, longitudinal data and information regarding time-to-event outcomes (e.g., death, recurrence, remission) are ubiquitous in medical studies. In this context, analytical methods capable of detecting and interpreting temporal changes in clinical parameters have become essential. This article aims to provide a comprehensive synthesis of the applications of Joinpoint regression in medicine, with a particular focus on the modelling of longitudinal measurements. Materials and methods The Joinpoint regression model was employed in integration with joint longitudinal–time-to-event models, as well as with Bayesian extensions. Results Joinpoint regression was applied to detect significant changes in temporal data trends. In the medical field, this method provides valuable opportunities for analysing the evolution of clinical indicators and modelling time-dependent risks of events. The conducted study required the modelling of temporal changes, as time-related trends in medicine are not always linear or stable. Sudden shifts may occur due to the introduction of new treatments, public health interventions, or unknown biological mechanisms. Joinpoint regression accurately identified inflection points, offering a clearer perspective on the underlying processes. Conclusions The utility of this method is highlighted in supporting evidence-based decision-making in both clinical practice and research, while also emphasizing the methodological challenges and future directions for development. Joinpoint regression is emerging as an indispensable tool in the arsenal of modern medical statistics, playing a critical role in enhancing the understanding of disease dynamics and optimizing informed clinical decision-making.