AI model predicted older adults' opioid risk classification with macro-F1 of 0.72.
Responsible AI for safer opioid risk management in older adults.
Sumedha Bakshi … Niam Yaraghi
npj health systems · Jun 2026
Background
Older adults face elevated opioid-related risks driven by multimorbidity, altered pharmacokinetics, and polypharmacy.
Results
correctly classified risk level in most cases, though 0.72 leaves real error rate
0.72macro-
Opioid risk classifica
0.77macro-
Beers medication safet
0.84macro-
PIM status
More results
We present a hybrid digital health framework that integrates a Long Short-Term Memory (Long Short-Term Memory) network for temporal risk prediction with a Retrieval-Augmented Generation (Retrieval-Augmented Generation) module for evidence-grounded clinical explanation tailored to adults aged ≥65 years.
More results
Retrieval-grounded reasoning, uncertainty signaling, and clinician-review cautions are embedded to mitigate unsupported inferences and automation bias.
“
Conclusion
By coupling temporal prediction with verifiable, guideline-grounded explanations, this work illustrates an operational, responsible-AI design approach for transparent and clinician-centered decision support in opioid management for ageing populations.