Medicine Overdose Prediction Using Machine Learning
Keywords:
Medicine Overdose, Machine Learn- ing, Predictive Analytics, Healthcare AI, Risk AssessmentAbstract
The opioid addiction crisis in the United States has gained national attention due to the alarming rise in opioid overdose (OD)-related deaths. Addressing this epidemic has become a top priority for governments and healthcare providers, requiring critical insights into the risk factors associated with opioid overdose. In this paper, we present our work on developing machine learning-based prediction models to assess the likelihood of opioid overdose using patients’ electronic health records (EHR).
We conducted two studies utilizing New York State claims data (SPARCS) with 440,000 patients and Cerner’s Health Facts database with 110,000 patients. Our experiments demonstrated that EHR-based prediction models achieved the highest recall using the random forest method (precision: 95.3%, recall: 85.7%, F1 score: 90.3%), while deep learning yielded the highest pre- cision (precision: 99.2%, recall: 77.8%, F1 score: 87.2%). Addi- tionally, we identified clinical events as key features contributing to the accuracy of these predictions.
Downloads
References
Han B, Compton WM, Blanco C, Crane E, Lee J, Jones CM. Prescription opioid use, misuse, and use disorders in US adults: 2015 National Survey on Drug Use and Health. Annals of Internal Medicine. 2017 Sep 5;167(5):293–301.
Rudd RA. Increases in drug and opioid-involved overdose deaths—United States, 2010–2015. MMWR Morbidity and Mortality Weekly Report. 2016:65.
Scholl L, Seth P, Kariisa M, Wilson N, Baldwin G. Drug and opioid- involved overdose deaths—United States, 2013–2017. Morbidity and Mortality Weekly Report. 2019 Jan 4;67(5152):1419.
Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain—United States, 2016. 2016;315(15):1624–45.
Bruneau J, Ahamad K, Goyer ME` , Poulin G, Selby P, Fischer B, Wild TC, Wood E. Management of opioid use disorders: a national clinical practice guideline. Canadian Medical Association Journal. 2018 Mar 5;190(9):E247–57.
Henry J, Pylypchuk Y, Searcy T, Patel V. Adoption of electronic health record systems among US non-federal acute care hospitals: 2008-2015. ONC Data Brief. 2016 May;35:1–9.
Cheng Y, Wang F, Zhang P, Hu J, editors. Risk prediction with electronic health records: A deep learning approach. Proceedings of the 2016 SIAM International Conference on Data Mining; 2016: SIAM.
Miotto R, Wang F, Wang S, Jiang X, Dudley JT. Deep learning for healthcare: review, opportunities, and challenges. 2017;19(6):1236–46.
Shickel B, Tighe PJ, Bihorac A, Rashidi P. Deep EHR: A survey of recent advances in deep learning techniques for electronic health record (EHR) analysis. 2018;22(5):1589–604.
Wang F, Casalino LP, Khullar D. Deep Learning in Medicine—Promise, Progress, and Challenges. 2018.
Statewide Planning Research Cooperative System. [Internet]. Health.ny.gov. [cited 7 March 2019]. Available from: https://www.health.ny.gov/statistics/sparcs/
Rajkomar A, Oren E, Chen K, Dai AM, Hajaj N, Hardt M, et al. Scalable and accurate deep learning with electronic health records. 2018;1(1):18.
Avati A, Jung K, Harman S, Downing L, Ng A, Shah NH. Improving palliative care with deep learning. BMC Medical Informatics and Decision Making. 2018 Dec;18(4):122.
Lipton ZC, Kale DC, Elkan C, Wetzel R. Learning to diagnose with LSTM recurrent neural networks. 2015 Nov 11; arXiv preprint arXiv:1511.03677.
Esteban C, Staeck O, Baier S, Yang Y, Tresp V, editors. Predicting clinical events by combining static and dynamic information using recurrent neural networks. 2016 IEEE International Conference on Healthcare Informatics (ICHI); 2016: IEEE.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
