โก Quick Summary
This study developed and validated machine learning models to predict hospital admissions and 90-day readmissions among patients with cardiovascular risk factors using community health survey data. The Extra Trees model achieved impressive performance metrics, including an AUROC of 0.99 for readmissions, highlighting the potential of integrating patient-reported data into healthcare systems.
๐ Key Details
- ๐ Dataset: 1,318 participants with cardiovascular risk factors
- ๐งฉ Features used: Patient-reported survey data
- โ๏ธ Technology: Eight supervised machine learning models, primarily Extra Trees
- ๐ Performance: ET model: AUROC 0.93 for hospitalisation, AUROC 0.99 for readmission
๐ Key Takeaways
- ๐ 35.0% of participants reported at least one hospitalisation.
- ๐ 10.4% experienced a 90-day readmission.
- ๐ Extra Trees model outperformed other models in predicting both outcomes.
- ๐ก Key predictors included heart disease, medication burden, race/ethnicity, employment, and insurance status.
- ๐ High accuracy of patient-reported data complements traditional EHR models.
- ๐ Study conducted from July 2021 to December 2022 in the US.
- ๐ SHAP analysis provided insights into influential predictors.
- ๐ฉบ Potential for early identification of high-risk individuals through integrated data.
๐ Background
Hospitalisation and readmission rates are critical metrics in healthcare, particularly for patients with cardiovascular risk factors. Traditional methods of predicting these events often rely heavily on electronic health records (EHRs), which may not capture the full spectrum of patient experiences. This study explores the integration of patient-reported survey data to enhance predictive accuracy and improve healthcare outcomes.
๐๏ธ Study
Conducted between July 2021 and December 2022, this cross-sectional survey involved US adults aged 18 and older with at least one cardiovascular risk factor. Participants were recruited through various channels, including social media and outpatient clinics, resulting in a final sample of 1,318 individuals. The study aimed to develop machine learning models to predict all-cause hospital admissions and 90-day readmissions based on structured survey data.
๐ Results
The Extra Trees model emerged as the most effective, achieving an AUROC of 0.93 for predicting hospitalisations and an outstanding AUROC of 0.99 for readmissions. The model also demonstrated a precision of 0.83 and recall of 0.87 for hospitalisations, while for readmissions, it achieved a precision of 0.95 and recall of 0.96. These results underscore the model’s robustness in accurately predicting critical healthcare events.
๐ Impact and Implications
The findings of this study have significant implications for healthcare delivery. By leveraging patient-reported data, healthcare providers can identify high-risk individuals earlier and implement targeted preventive interventions. This approach not only enhances patient care but also has the potential to reduce healthcare costs associated with avoidable hospitalisations and readmissions.
๐ฎ Conclusion
This study highlights the transformative potential of machine learning in predicting hospitalisation and readmission risks among patients with cardiovascular risk factors. By integrating patient-reported data into EHRs, healthcare systems can improve their predictive capabilities and ultimately enhance patient outcomes. The future of healthcare may very well depend on such innovative approaches to data integration and analysis.
๐ฌ Your comments
What are your thoughts on the integration of machine learning and patient-reported data in healthcare? We would love to hear your insights! ๐ฌ Join the conversation in the comments below or connect with us on social media:
Development of machine learning models to predict risk of hospitalisation and 90-day readmission among patients with cardiovascular risk factors using community health survey data.
Abstract
OBJECTIVES: This study aimed to develop and validate machine learning (ML) models to predict all-cause hospital admissions and 90-day readmissions using structured, patient-reported survey data.
METHODS: A cross-sectional survey was conducted between 3 July 2021 and 18 December 2022, among US adults aged โฅ18 years with at least one cardiovascular risk factor. Participants were recruited through social media, community pharmacies and outpatient clinics. The final sample included 1318 participants. Primary outcomes were any all-cause hospitalisation and readmission within 90 days. Eight supervised ML models were trained using an 80:20 train-test split and 10-fold cross-validation. Model performance was evaluated using area under the receiver operating characteristic curve (AUROC), precision, recall, F1 score and calibration metrics. SHapley Additive exPlanations (SHAP) values identified key predictors.
RESULTS: Among 1318 participants, 35.0% reported at least one hospitalisation and 10.4% reported a 90-day readmission. The Extra Trees (ET) model demonstrated the best performance across both outcomes. For hospitalisation, ET achieved an AUROC of 0.93, precision of 0.83 and recall of 0.87. For readmission, AUROC was 0.99 with precision of 0.95 and recall of 0.96. SHAP analysis identified heart disease, medication burden, race/ethnicity, employment and insurance status as the most influential predictors.
DISCUSSION: Patient-reported data reflecting behavioural, social and clinical factors can predict hospitalisations with high accuracy, complementing traditional EHR-based models.
CONCLUSIONS: Integrating such patient-reported and behavioural data into electronic health records could enable earlier identification of high-risk individuals and support targeted, preventive interventions to improve healthcare outcomes.
Author: [‘Nkemdirim Okere A’, ‘Li T’, ‘Islam MM’, ‘Ali AA’, ‘Buxbaum SG’, ‘Diaby V’]
Journal: BMJ Health Care Inform
Citation: Nkemdirim Okere A, et al. Development of machine learning models to predict risk of hospitalisation and 90-day readmission among patients with cardiovascular risk factors using community health survey data. Development of machine learning models to predict risk of hospitalisation and 90-day readmission among patients with cardiovascular risk factors using community health survey data. 2025; 32:(unknown pages). doi: 10.1136/bmjhci-2025-101742