Predicting hospitalization of COVID-19 positive patients using clinician-guided machine learning methods
Wenyu Song1,2, Linying Zhang3, Luwei Liu1
1Department of Medicine, Brigham & Women's Hospital, Boston, Massachusetts, USA.
Insights
Machine learning models accurately predict hospitalization risk in older adults with COVID-19. Albumin levels were a key predictor, aiding in timely healthcare decisions for high-risk patients.
Area of Science:
- Medical Informatics
- Epidemiology
- Geriatric Medicine
Background:
- The COVID-19 pandemic presents a significant global health challenge requiring efficient resource allocation.
- Identifying high-risk COVID-19 patients, particularly older adults, is crucial for timely medical intervention.
- Predictive models can support healthcare systems in managing pandemic-related patient loads.
Purpose of the Study:
- To develop and evaluate machine learning models for predicting hospitalization in older adults (age 65+) who tested positive for COVID-19.
- To identify key clinical features associated with COVID-19 hospitalization in this demographic.
- To assess the potential of these models for clinical decision support in patient triage.
Main Methods:
- A cohort of 1495 older adult outpatients with COVID-19 was established from electronic health records across 11 hospitals.
- A 3-stage feature selection process involving literature review, expert opinion, and EHR data exploration identified 44 predictive features.
- Four machine learning models (logistic regression, SVM, random forest, neural network) were trained and compared.
Main Results:
- All four models demonstrated strong predictive performance, with AUCs exceeding 0.80.
- The Random Forest model achieved the highest predictive accuracy with an AUC of 0.83.
- Serum albumin, an indicator of nutritional status, emerged as the most significant predictor of hospitalization.
Conclusions:
- Four machine learning models were successfully developed to predict hospitalization risk in COVID-19 positive older adults.
- The study identified crucial clinical factors and temporal patterns influencing hospitalization.
- The developed models and methodology offer a promising tool for enhancing decision support and efficient triage of COVID-19 patients.
Objectives:
The coronavirus disease 2019 (COVID-19) is a resource-intensive global pandemic. It is important for healthcare systems to identify high-risk COVID-19-positive patients who need timely health care. This study was conducted to predict the hospitalization of older adults who have tested positive for COVID-19.
Methods:
We screened all patients with COVID test records from 11 Mass General Brigham hospitals to identify the study population. A total of 1495 patients with age 65 and above from the outpatient setting were included in the final cohort, among which 459 patients were hospitalized. We conducted a clinician-guided, 3-stage feature selection, and phenotyping process using iterative combinations of literature review, clinician expert opinion, and electronic healthcare record data exploration. A list of 44 features, including temporal features, was generated from this process and used for model training. Four machine learning prediction models were developed, including regularized logistic regression, support vector machine, random forest, and neural network.
Results:
All 4 models achieved area under the receiver operating characteristic curve (AUC) greater than 0.80. Random forest achieved the best predictive performance (AUC = 0.83). Albumin, an index for nutritional status, was found to have the strongest association with hospitalization among COVID positive older adults.
Conclusions:
In this study, we developed 4 machine learning models for predicting general hospitalization among COVID positive older adults. We identified important clinical factors associated with hospitalization and observed temporal patterns in our study cohort. Our modeling pipeline and algorithm could potentially be used to facilitate more accurate and efficient decision support for triaging COVID positive patients.
More Related Videos
Related Concept Videos
01:18Steps in Outbreak Investigation
01:29Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...