Newswise — August 29, 2024 —Machine learning algorithms are good at predicting risk of persistent opioid use after surgeryIn the August survey, Problems Plastic and Reconstructive Surgery®, The official medical journal of American Academy of Plastic Surgeons (ASPS). This journal is published by Lippincott Portfolio. Wolters Kluwer.
“We found that machine learning models are good at identifying patients undergoing surgical procedures who are likely to become addicted to opioids,” said ASPS-member surgeon Kevin C. Chung, MD, MSc, of the University of Michigan, Ann Arbor. “This may lead to more efficient strategies to identify high-risk patients and implement measures to prevent opioid addiction. Similarly, artificial intelligence could enable a more personalized approach to prescribing the right pain medication in the optimal dose for a specific patient undergoing a specific procedure.”
Two machine learning models tested to predict persistent opioid use
The study evaluated two previously published machine learning models. Michigan Genomics Initiative There are two models based on the Medical Grade Index (MGI) and insurance claims data. The models were first evaluated in a large sample of general surgery patients and then SurgerySuch as carpal tunnel or wrist fracture surgery.
The study focused on whether a machine learning model could predict which patients would develop persistent opioid use, based on prescriptions filled up to six months after surgery. The MGI model included 889 patients, about half of whom had previously used opioids. The claims model was limited to 439 “opioid-naive” patients who had not recently used opioids.
In the MGI model that included past opioid users, 21% of patients developed persistent opioid use. In the claims model that excluded past opioid users, 10% of patients developed persistent opioid use.
In an “area under the curve” analysis, the MGI model performed very well in identifying patients with persistent opioid use: 84% for models trained on surgical data and 85% for general surgery patients. In contrast, the claims model had a predictive ability of 69% based on surgical data but only 52% with the complete data set.
Machine learning could streamline assessment of postoperative opioid risk
In the MGI model, having an opioid prescription preoperatively was the strongest predictor of postoperative opioid use. Other predictors included generalized pain and prescriptions for hydrocodone, a relatively strong opioid commonly prescribed for postoperative pain.
As with any type of surgery, persistent opioid use poses a risk for patients undergoing surgery. Although several risk factors have been identified, assessing postoperative opioid risk is a challenging and time-consuming process, given the diversity of patient populations and variability in surgical complexity. New research suggests that machine learning could provide a more integrated, straightforward approach to identifying high-risk patients.
Models that include patient-reported data on factors such as pain and mental health, such as those collected in the MGI, appear to provide the highest predictive value. “With access to comprehensive datasets, machine learning may be able to streamline the identification and analysis of detailed factors that influence patients' postoperative pain experience,” the researchers wrote.
The authors note that their study has several limitations and may not reflect changes in prescribing patterns in response to the opioid epidemic. Dr. Chung and his co-authors conclude that “in practice, these models could be implemented as decision support tools to help clinicians efficiently identify patients who are most susceptible to addiction and who require customized pain management and counseling.”
Read the article: Predicting persistent opioid use after surgery: A machine learning approach
Wolters Kluwer delivers trusted clinical technology and evidence-based solutions to empower clinicians, patients, researchers and students to make effective decisions and achieve outcomes across the entire spectrum of healthcare. We support clinical effectiveness, learning and research, clinical oversight and compliance, and data solutions. To learn more about our solutions, please visit: https://www.wolterskluwer.com/en/health.
###
About Wolters Kluwer
Wolters Kluwer (EURONEXT: WKL) is a global leader in information, software solutions and services for healthcare, tax & accounting, financial & corporate compliance, legal & regulatory, company performance and ESG professionals. We help our clients make critical decisions every day: Expert Solutions It combines deep expertise with technology and services.
Wolters Kluwer reported annual revenues of €5.6 billion in 2023. The Group serves clients in over 180 countries, has operations in over 40 countries and employs approximately 21,400 people worldwide. The company is headquartered in Alphen aan den Rijn, the Netherlands.
For more information, Learn morefollow me LinkedIn, Facebook, YouTube and Instagram.