Deterministic and stochastic interventions in reducing drug-drug interactions in inappropriate prescribing: A systematic review.
Muhammad Fahmi Ahmad Zuber, Nur Aishah Che Roos, Ruzanna Mat Jusoh et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review evaluated interventions aimed at reducing drug-drug interactions (DDIs) in inappropriate prescribing, focusing on deterministic and stochastic models. Ten studies were included, revealing that while stochastic models showed strong predictive metrics, they lacked external validation. Deterministic approaches improved prescribing processes but had inconsistent patient outcomes. Overall, the evidence does not confirm improved clinical safety.
Key findings
- Deterministic interventions showed modest improvements in prescribing processes but inconsistent patient-level outcomes.
- Stochastic models demonstrated strong internal performance metrics but had high risk of bias due to limited external validation.
- Current evidence does not show a proportional improvement in clinically reliable decision support from advanced models.
Why it matters
Understanding the effectiveness of different intervention models for reducing DDIs is crucial for clinicians managing polypharmacy, as it informs the selection of tools that enhance patient safety and prescribing practices.
What to keep in mind
The review identified a high risk of bias in AI-driven studies, primarily due to limited external validation and unclear handling of overfitting.