RESUMO
Cardiovascular disease (CVD) in cancer patients can affect the risk of unplanned readmissions, which have been reported to be costly and associated with worse mortality and prognosis. We aimed to demonstrate the feasibility of using machine learning techniques in predicting the risk of unplanned 180-day readmission attributable to CVD among hospitalized cancer patients using the 2017-2018 Nationwide Readmissions Database. We included hospitalized cancer patients, and the outcome was unplanned hospital readmission due to any CVD within 180 days after discharge. CVD included atrial fibrillation, coronary artery disease, heart failure, stroke, peripheral artery disease, cardiomegaly, and cardiomyopathy. Decision tree (DT), random forest, extreme gradient boost (XGBoost), and AdaBoost were implemented. Accuracy, precision, recall, F2 score, and receiver operating characteristic curve (AUC) were used to assess the model's performance. Among 358,629 hospitalized patients with cancer, 5.86% (n = 21,021) experienced unplanned readmission due to any CVD. The three ensemble algorithms outperformed the DT, with the XGBoost displaying the best performance. We found length of stay, age, and cancer surgery were important predictors of CVD-related unplanned hospitalization in cancer patients. Machine learning models can predict the risk of unplanned readmission due to CVD among hospitalized cancer patients.
Assuntos
Doenças Cardiovasculares , Insuficiência Cardíaca , Neoplasias , Humanos , Doenças Cardiovasculares/complicações , Doenças Cardiovasculares/epidemiologia , Readmissão do Paciente , Neoplasias/complicações , Neoplasias/epidemiologia , Neoplasias/terapia , Aprendizado de MáquinaRESUMO
Background: Cardiovascular disease (CVD) can lead to unplanned care in patients with cancer, which may affect their prognosis and survival. We aimed to compare the rates, timing, and length of stay of unplanned CVD readmission in hospitalized patients with and without cancer. Methods: This study used the 2017-2018 Nationwide Readmissions Database to identify adult hospitalized patients with and without cancer. The primary outcome was 180-day unplanned CVD readmission rates. CVD was defined based on a composite variable that included atrial fibrillation, coronary artery disease, cardiomegaly, cardiomyopathy, heart failure, peripheral artery disease, and stroke. For patients readmitted due to CVD, the timing between admissions (based on the mean number of days between index hospitalization and readmission) and length of stay were further identified. Results: After matching, 300,398 patients were included in the two groups. The composite CVD readmission rates were significantly higher in patients with cancer (5.92% vs 4.10%; odds ratio (OR) 1.47, 95% CI 1.44-1.51, p < 0.001). Patients with cancer were also associated with shorter mean number of days to composite CVD readmission (60.48 days vs 68.32 days, p < 0.001) and longer length of stay of composite CVD readmission (8.21 days vs 7.13 days, p < 0.001). These trends were maintained in analyses of the individual CVD. Conclusions: Hospitalized patients with cancer experienced higher rates of unplanned readmission due to CVD, and their CVD readmissions occurred sooner and required longer lengths of stay compared to patients without cancer. Efforts to reduce unplanned CVD readmissions, such as providing optimized chronic post-discharge care, may improve the health outcomes of patients with cancer.