Development of a Predictive Model for Potentially Inappropriate Medications in Older Patients with Cardiovascular Disease.

Abstract

Background: Older patients with cardiovascular disease (CVD) are highly susceptible to adverse drug reactions due to age-related physiological changes and the presence of multiple comorbidities, polypharmacy, and potentially inappropriate medications (PIMs).

Objective: This study aimed to develop a predictive model to identify the use of PIMs in older patients with CVD.

Methods: Data from 2012 to 2021 from the Changhua Christian Hospital Clinical Research Database (CCHRD) and the Kaohsiung Medical University Hospital Research Database (KMUHRD) were analyzed. Participants over the age of 65 years with CVD diagnoses were included. The CCHRD data were randomly divided into a training set (80% of the database) and an internal validation set (20% of the database), while the KMUHRD data served as an external validation set. The training set was used to construct the prediction models, and both validation sets were used to validate the proposed models.

Results: A total of 48,569 patients were included. Comprehensive data analysis revealed significant associations between the use of PIMs and clinical factors such as total cholesterol, glycated hemoglobin (HbA1c), creatinine, and uric acid levels, as well as the presence of diabetes, hypertension, and cerebrovascular accidents. The predictive models demonstrated moderate power, indicating the importance of these factors in assessing the risk of PIMs.

Conclusions: This study developed predictive models that improve understanding of the use of PIMs in older patients with CVD. These models may assist clinicians in making informed decisions regarding medication safety.