Closed-Loop Referral

The foundational technical capability that distinguishes a closed-loop from a traditional open referral. It requires a secure, interoperable channel—typically leveraging HL7 FHIR APIs—for community-based organizations (CBOs) to send outcome data back to the originating health system.

• Outbound: Structured referral with SDOH screening data and patient consent.
• Inbound: Status updates (accepted, scheduled, completed) and outcome codes.
• Standard: Gravity Project FHIR Implementation Guide defines the exchange payloads.

The system maintains a persistent, stateful record of the referral's lifecycle, moving beyond a simple 'sent' notification. Each transition updates a centralized case file visible to both the care team and the CBO.

• State Machine Logic: Tracks stages like Pending, Accepted, In Progress, Completed, or Unable to Reach.
• Temporal Auditing: Timestamps every state change to measure cycle time and identify bottlenecks.
• Proactive Alerting: Triggers notifications for care coordinators if a referral stalls in a non-terminal state beyond a defined SLA.

A legally compliant closed-loop system is architected around granular patient consent, not just a blanket HIPAA Business Associate Agreement (BAA). The system must manage purpose-based access control for sensitive SDOH data.

• Granular Consent: Patients can consent to share housing data but not substance use history.
• Revocation Handling: A technical mechanism to propagate consent revocation to the CBO and purge shared data.
• 42 CFR Part 2 Compliance: Specialized logic for referrals involving substance use disorder treatment information.

The 'closing' of the loop is not merely a CBO self-reporting completion. It involves deterministic verification logic to confirm that the patient's social need was actually resolved.

• Structured Outcome Codes: Uses LOINC-coded SDOH observations to report a resolved food insecurity score, not just a free-text note.
• Duplicate Resolution: Logic to prevent a single successful connection from being counted multiple times across different referral platforms.
• Re-Opening Triggers: Automatically re-activates a closed referral if a subsequent screening indicates the same social risk has recurred.

The system relies on a continuously curated, machine-readable directory of CBOs that includes real-time capacity and service eligibility criteria. This is not a static spreadsheet.

• Capacity Signaling: CBOs can broadcast their current availability (e.g., 'Waitlist Open', 'At Capacity') via API.
• Eligibility Matching: Algorithmic matching of patient needs and demographics (e.g., veteran status, geographic catchment) to specific program requirements.
• Network Health Analytics: Dashboards that identify service deserts where patient needs have no available, active CBO partner.

The final characteristic is the ability to aggregate de-identified loop-closure data for population health analytics and regulatory reporting, such as the NCQA HEDIS Social Need Screening and Intervention (SNS-E) measure.

• Funnel Analytics: Measures the conversion rate from Positive Screen → Referral Sent → Service Connected → Need Resolved.
• Equity Stratification: All metrics are cross-tabulated by race, ethnicity, and payer type to detect disparities in loop closure rates.
• ROI Calculation: Quantifies the reduction in avoidable ED utilization attributable to successfully closed social need referrals.

The digital process of matching a patient's identified social needs to specific, available community-based organizations (CBOs) and programs through an integrated platform. This involves maintaining a curated resource directory with real-time service availability, eligibility criteria, and referral protocols.

• Maps structured SDOH findings to specific services
• Often uses FHIR RESTful APIs to query resource directories
• Critical for bridging clinical and social care ecosystems

A Fast Healthcare Interoperability Resources (FHIR) resource used to represent a specific, screened social risk finding in a standardized, exchangeable format. It captures the screening instrument used, the value (e.g., a food insecurity score), and the interpretation of that finding.

• Based on the HL7 Gravity Project standards
• Enables structured data exchange between EHRs and referral platforms
• Supports the USCDI SDOH Data Elements mandate for certified health IT

A quality assurance workflow where a clinical reviewer audits and corrects AI-extracted SDOH data and referral recommendations, typically focusing on low-confidence predictions flagged by the model. This ensures clinical accuracy before a referral is initiated.

• Reviewers validate entity extraction and negation detection
• Provides active feedback to improve model performance
• Essential for maintaining trust in automated referral workflows

A HL7 standard for triggering real-time, context-aware clinical decision support within a clinician's EHR workflow. A CDS Hook can fire when a patient's SDOH screening is completed, prompting the provider to initiate a closed-loop referral directly from the encounter.

• Uses a RESTful service architecture
• Triggers on specific EHR events like patient-view or order-select
• Integrates seamlessly with SMART on FHIR authentication

A semantic network that connects social risk concepts, ICD-10 Z-codes, community resources, and patient data to enable complex reasoning and inference. It powers the intelligent matching logic in a closed-loop referral system by understanding the relationships between a patient's housing instability and the appropriate shelter referral.

• Uses ontologies aligned with the Gravity Project
• Enables contextual reasoning beyond simple keyword matching
• Supports longitudinal tracking of social need resolution

The ability of disparate health IT systems to seamlessly exchange standardized social determinant data using frameworks like the HL7 FHIR SDOH Implementation Guide. This is the technical prerequisite for a closed-loop system where the EHR, the referral platform, and the CBO's system all speak the same data language.

• Leverages FHIR QuestionnaireResponse and ServiceRequest resources
• Ensures bidirectional data flow for referral status updates
• Aligns with TEFCA and national interoperability mandates