Predicting postoperative opioid use with machine learning and insurance claims in opioid-naïve patients

• Published in: The American journal of surgery Vol. 222; no. 3; pp. 659 - 665
• Main Authors: Hur, Jaewon, Tang, Shengpu, Gunaseelan, Vidhya, Vu, Joceline, Brummett, Chad M., Englesbe, Michael, Waljee, Jennifer, Wiens, Jenna
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.09.2021; Elsevier Limited
• Subjects: Claims data; Codes; Electronic health records; Gastrointestinal surgery; Generic drugs; Insurance; Insurance claims; Laparoscopy; Learning algorithms; Machine learning; Narcotics; Opioids; Pain; Patients; Performance prediction; Postoperative opioid use; Postoperative period; Prescriptions; Support vector machines; Surgery; Postoperative opioid use; Claims data; Machine learning
• ISSN: 0002-9610; 1879-1883
• DOI: 10.1016/j.amjsurg.2021.03.058

The clinical impact of postoperative opioid use requires accurate prediction strategies to identify at-risk patients. We utilize preoperative claims data to predict postoperative opioid refill and new persistent use in opioid-naïve patients. A retrospective study was conducted on 112,898 opioid-naïve adult postoperative patients from Optum’s de-identified Clinformatics® Data Mart database. Potential predictors included sociodemographic data, comorbidities, and prescriptions within one year prior to surgery. Compared to linear models, non-linear models led to modest improvements in predicting refills – area under the receiver operating characteristics curve (AUROC) 0.68 vs. 0.67 (p < 0.05) – and performed identically in predicting new persistent use – AUROC = 0.66. Undergoing major surgery, opioid prescriptions within 30 days prior to surgery, and abdominal pain were useful in predicting refills; back/joint/head pain were the most important features in predicting new persistent use. Preoperative patient attributes from insurance claims could potentially be useful in guiding prescription practices for opioid-naïve patients. •A large retrospective study on opioid-naïve patient was conducted.•Machine learning models were trained using insurance claims data.•Non-linear models performed modestly better than linear models.•Opioid refills are associated with the nature of the surgery.•New persistent opioid use is associated with underlying chronic pain conditions.