Infrastructure for Medication Reconciliation Automation

Medication reconciliation has documented Joint Commission requirements and facility med-rec policies, but the clinical judgment criteria for resolving discrepancies—whether a dose difference represents an intentional change or a transcription error, whether a missing medication represents omission or discontinuation—are not formally codified. Documentation practice exists for the reconciliation process itself (completion attestation, discrepancy logging) but the decision logic for automated discrepancy classification remains undocumented pharmacist judgment, making this an L2 environment where structure exists but clinical reasoning isn't formalized.

Medication reconciliation automation requires systematic capture of all medication sources: EHR active medication list, Surescripts external fill data, patient-reported medications from structured intake workflows, and prior discharge summaries. Template-driven capture ensures each reconciliation event records sources consulted, discrepancies identified, and resolution actions taken with consistent fields. This systematic capture enables NLP extraction pipelines to receive structured inputs and produces the audit documentation required for Joint Commission compliance.

Medication reconciliation automation requires consistent schema across all medication source records: drug name normalized to RxNorm, dose, route, frequency, source (EHR, pharmacy fill, patient-reported), and last fill date. Without consistent schema across sources, the NLP system can't computationally match 'metoprolol succinate 50mg daily' from the EHR against 'metoprolol XL 50' from a Surescripts fill record as the same medication. RxNorm normalization and standardized dose fields enable automated matching and discrepancy detection.

Medication reconciliation automation requires API access to the EHR medication list, Surescripts external pharmacy fill data, prior discharge summary NLP extraction, and the clinical documentation system to record reconciliation completion. These connections enable the system to automatically assemble medication information from all sources at the point of care transition without manual data retrieval. The existing pharmacy-to-EHR integration and Surescripts connectivity in most health systems provides this foundation.

Medication reconciliation logic must update when formulary changes affect drug name mappings (generic substitutions create new RxNorm matching requirements), when new medication sources are added (new Surescripts data partners), or when Joint Commission standards update reconciliation documentation requirements. Event-triggered maintenance ensures the NLP extraction and matching logic stays aligned with current drug databases and regulatory requirements without requiring continuous monitoring.

Medication reconciliation automation requires API-based connections between the EHR, Surescripts (external fill data), pharmacy dispensing system, patient intake workflows, and clinical documentation. Care transitions are the trigger—admission, transfer, and discharge events must automatically initiate the reconciliation workflow, pulling data from all connected sources. The existing medication use process integration provides the internal EHR-pharmacy connection; Surescripts API extends it to external pharmacies critical for identifying home medication discrepancies.