Portal de Periódicos da CAPES

Factor Analysis and Modelling of 30-Day Hospital Readmission after Discharge for Copd Patients: A Retrospective Study of Macao

2021; RELX Group (Netherlands); Linguagem: Inglês

1556-5068

Meng Li, Kun Cheng, Keisun Ku, Junlei Li, Hao Hu, Carolina Oi Lam Ung,

Chronic Disease Management Strategies

Background: Chronic Obstructive Pulmonary Disease (COPD) is the third most common chronic disease in China, resulting in increased hospitalization and readmission rate. 30-day readmission after discharge is an important indicator of care transitions and disease management. This study aimed to develop a 30-day readmission prediction model using a decision tree model based on the data extracted from the electronic health record of COPD patients in Macao. Methods: Health records data from 2018 to 2019 of COPD inpatients was analyzed. Data discretization and balancing were conducted during data preprocessing, risk factors were identified by feature selection. A machine learning model was developed by a decision tree algorithm in MATLAB environment. Findings: 782 COPD hospitalizations were enrolled, 30-day readmission rate was 26·5%. Age, length of stay, smoking status, hemoglobin, systemic steroids use, antibiotics use and number of hospital admission due to COPD in last 12 months were found to be significant risk factors for 30-day readmission (P< 0·01). A data-driven decision tree-based modelling approach was developed. The mean precision, recall and AUC value for the decision tree classifier were 73·85%, 73·7% and 0·7506. Interpretation: Discharged COPD patients with high risks of being readmitted within 30 days could be identified by the model developed in this study. Number of hospital admission due to COPD in last 12 months, smoke status and patients’ age were the top factors for 30-day readmission in Macao. Funding: The study was supported by the University of Macau (SRG2021-00007-ICMS). Declaration of Interests: The author reports no conflicts of interest in this work. Ethics Approval Statement: Ethical approval for this study was obtained from the ethical review committee at the Kiang Wu Hospital, Macao [Ref. No. 2019-015] with a waiver regarding informed consent, because all data were retrospectively collected and anonymized in the standardized case report form of the hospital. It was also confirmed that all data collection was performed in accordance with relevant guidelines and regulations.