Using real-world electronic health record data to predict the development of 12 cancer-related symptoms in the context of multimorbidity
Open Access
- 1 July 2024
- journal article
- research article
- Published by Oxford University Press (OUP) in JAMIA Open
- Vol. 7 (3) , ooae082
Abstract
Objective: This study uses electronic health record (EHR) data to predict 12 common cancer symptoms, assessing the efficacy of machine learning (ML) models in identifying symptom influencers.Materials and Methods: We analyzed EHR data of 8156 adults diagnosed with cancer who underwent cancer treatment from 2017 to 2020. Structured and unstructured EHR data were sourced from the Enterprise Data Warehouse for Research at the University of Iowa Hospital and Clinics. Several predictive models, including logistic regression, random forest (RF), and XGBoost, were employed to forecast symptom development. The performances of the models were evaluated by F1-score and area under the curve (AUC) on the testing set. The SHapley Additive exPlanations framework was used to interpret these models and identify the predictive risk factors associated with fatigue as an exemplar.Results: The RF model exhibited superior performance with a macro average AUC of 0.755 and an F1-score of 0.729 in predicting a range of cancer-related symptoms. For instance, the RF model achieved an AUC of 0.954 and an F1-score of 0.914 for pain prediction. Key predictive factors identified included clinical history, cancer characteristics, treatment modalities, and patient demographics depending on the symptom. For example, the odds ratio (OR) for fatigue was significantly influenced by allergy (OR = 2.3, 95% CI: 1.8-2.9) and colitis (OR = 1.9, 95% CI: 1.5-2.4).Discussion: Our research emphasizes the critical integration of multimorbidity and patient characteristics in modeling cancer symptoms, revealing the considerable influence of chronic conditions beyond cancer itself.Conclusion: We highlight the potential of ML for predicting cancer symptoms, suggesting a pathway for integrating such models into clinical systems to enhance personalized care and symptom management.Keywords
Funding Information
- Betty Irene Moore Fellowship for Nurse Leaders and Innovators
- College of Nursing, University of Iowa
- Center for Advancing Multimorbidity Science
- NINR
- National Institute for Nursing Research (P20 1P20NR018081)
- Holden Comprehensive Cancer Center
- University of Iowa
- National Cancer Institute (P30 P30CA086862)
- Iowa Health Data Resource
- University of Iowa
- Institute for Clinical and Translational Science
- CTSA University of Iowa (UL1TR002537)
This publication has 55 references indexed in Scilit:
- Multiple Chronic Conditions Among US Adults: A 2012 UpdatePreventing Chronic Disease, 2014
- Future of Electronic Health Records: Implications for Decision SupportMount Sinai Journal of Medicine: A Journal of Translational and Personalized Medicine, 2012
- Association of symptoms of colon cancer patients with tumor location and TNM tumor stageScandinavian Journal of Gastroenterology, 2012
- MissForest—non-parametric missing value imputation for mixed-type dataBioinformatics, 2011
- Prediction models in cancer careCA: A Cancer Journal for Clinicians, 2011
- Prediction Modeling Using EHR DataMedical Care, 2010
- Assessing the Symptoms of Cancer Using Patient-Reported Outcomes (ASCPRO): Searching for StandardsJournal of Pain and Symptom Management, 2010
- Symptom Burden: Multiple Symptoms and Their Impact as Patient-Reported OutcomesJournal of the National Cancer Institute Monographs, 2007
- Cancer Symptom Assessment Instruments: A Systematic ReviewJournal of Clinical Oncology, 2006
- Natural Language ProcessingScience, 1991