Constructing a prognostic tool for predicting the risk of non-adherence to antiplatelet therapy in discharged patients with coronary heart disease: a retrospective cohort study

Abstract

Objective To investigate the incidence and influencing factors affecting the non-adherence behavior of patients with coronary heart disease (CHD) to antiplatelet therapy after discharge and to construct a personalized predictive tool. Methods In this retrospective cohort study, 289 patients with CHD who were admitted to the Department of Cardiology of The First Affiliated Hospital of the University of Science and Technology of China between June 2021 and September 2021 were enrolled. The clinical data of all patients were retrospectively collected from the hospital information system, and patients were followed up for 1 year after discharge to evaluate their adherence level to antiplatelet therapy, analyze their present situation and influencing factors for post-discharge adherence to antiplatelet therapy, and construct a nomogram model to predict the risk of non-adherence. Results Based on the adherence level to antiplatelet therapy within 1 year after discharge, the patients were divided into the adherence (n = 216) and non-adherence (n = 73) groups. Univariate analysis revealed statistically significant differences between the two groups in terms of variable distribution, including age, education level, medical payment method, number of combined risk factors, percutaneous coronary intervention, duration of antiplatelet medication, types of drugs taken at discharge, and CHD type (P < 0.05). Furthermore, multivariate logistic regression analysis revealed that, except for the medical payment method, all the seven abovementioned variables were independent risk factors for non-adherence to antiplatelet therapy (P < 0.05). The areas under the receiver operating characteristic curve before and after the internal validation of the predictive tool based on the seven independent risk factors and the nomogram were 0.899 (95% confidence interval [CI]: 0.858–0.941) and 0.89 (95% CI: 0.847–0.933), respectively; this indicates that the tool has good discrimination ability. The calibration curve and Hosmer–Lemeshow goodness of fit test revealed that the tool exhibited good calibration and prediction consistency (χ2 = 5.17, P = 0.739). Conclusion In this retrospective cohort study, we investigated the incidence and influencing factors affecting the non-adherence behavior of patients with CHD after discharge to antiplatelet therapy. For this, we constructed a personalized predictive tool based on seven independent risk factors affecting non-adherence behavior. The predictive tool exhibited good discrimination ability, calibration, and clinical applicability. Overall, our constructed tool is useful for predicting the risk of non-adherence behavior to antiplatelet therapy in discharged patients with CHD and can be used in personalized intervention strategies to improve patient outcomes.