Researchgradient boostingrisk predictionelectronic health recordsopioid use disorder
Machine Learning Predicts 3-Month Incident OUD
9.1
Researchers used 2017–2022 OneFlorida+ EHR data to develop and validate a machine learning model predicting 3-month incident opioid use disorder among 182,083 adults initiating opioid therapy. A gradient boosting machine achieved a C-statistic of 0.879 in internal validation and 0.756 on external UPMC data; top decile risk stratification captured ~68% of cases (PPV 3.26%, NNE 31).
Key Points
- 1Develops GBM model predicting 3-month incident OUD with 0.879 C-statistic in validation
- 2Demonstrates top decile captures ~68% of OUD cases, PPV 3.26%, NNE 31
- 3Enables clinicians to stratify patients for early intervention and targeted monitoring
Scoring Rationale
Strong external validation and high discrimination drive score; limited PPV and deployment challenges constrain immediate impact.
Sources
Public references used for this report.
Practice with real Ad Tech data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
Active Search Campaigns by BudgetEasyHigh CPC Clicks & Poor Landing PagesMediumCampaign ROAS by Attribution ModelHard
250 free problems · No credit card
See all Ad Tech problems
