Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups
| Autor: | Hale M Thompson, Matthew M. Churpek, Majid Afshar, Niranjan S. Karnik, Brihat Sharma, Connor McCluskey, Sameer Bhalla, Dmitriy Dligach, Randy A. Boley |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
AcademicSubjects/SCI01060
Health Informatics Research and Applications Machine Learning Electronic health record Statistics medicine Electronic Health Records Humans natural language processing AcademicSubjects/MED00580 Interpretability Uncategorized bias and fairness Framingham Risk Score structural racism Opioid use disorder opioid use disorder Hispanic or Latino medicine.disease Opioid-Related Disorders Confidence interval Substance abuse AcademicSubjects/SCI01530 Psychology interpretability Classifier (UML) Type I and type II errors |
| Zdroj: | Journal of the American Medical Informatics Association : JAMIA |
| ISSN: | 1527-974X 1067-5027 |
| Popis: | Objectives To assess fairness and bias of a previously validated machine learning opioid misuse classifier. Materials & Methods Two experiments were conducted with the classifier’s original (n = 1000) and external validation (n = 53 974) datasets from 2 health systems. Bias was assessed via testing for differences in type II error rates across racial/ethnic subgroups (Black, Hispanic/Latinx, White, Other) using bootstrapped 95% confidence intervals. A local surrogate model was estimated to interpret the classifier’s predictions by race and averaged globally from the datasets. Subgroup analyses and post-hoc recalibrations were conducted to attempt to mitigate biased metrics. Results We identified bias in the false negative rate (FNR = 0.32) of the Black subgroup compared to the FNR (0.17) of the White subgroup. Top features included “heroin” and “substance abuse” across subgroups. Post-hoc recalibrations eliminated bias in FNR with minimal changes in other subgroup error metrics. The Black FNR subgroup had higher risk scores for readmission and mortality than the White FNR subgroup, and a higher mortality risk score than the Black true positive subgroup (P Discussion The Black FNR subgroup had the greatest severity of disease and risk for poor outcomes. Similar features were present between subgroups for predicting opioid misuse, but inequities were present. Post-hoc mitigation techniques mitigated bias in type II error rate without creating substantial type I error rates. From model design through deployment, bias and data disadvantages should be systematically addressed. Conclusion Standardized, transparent bias assessments are needed to improve trustworthiness in clinical machine learning models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|