SDOH Integration

SDOH Identification & Action Agent

ai

Identifies social determinants from clinical notes and screenings, maps findings to Z-codes, matches needs to community resources, initiates referrals, and tracks outcomes.

Care ManagerPopulation HealthPatient Services
4x
SDOH capture rate vs. screening-only
60%
Referral completion rate
Z-code ready
CMS VBC submission compliant
Use Case 1

Population SDOH Sweep

Trigger: Monthly population health refresh

1.Run NLP on clinical notes, screening responses, structured data
2.Identify unaddressed social needs
3.Map findings to ICD-10 Z-codes
4.Cross-match to community resource database by location and language
5.Prepare Z-code update recommendations
6.Initiate referral packages

SDOH needs captured systematically, coded for CMS submission, and matched to real community resources.

Use Case 2

Closed-Loop Referral Tracking

Trigger: Referral initiated

1.Send patient their resource options (preferred channel and language)
2.Notify community partner of referral
3.Track connection, engagement, and service delivery
4.Check in with patient to verify need resolution
5.Update care plan and Z-codes
6.Feed outcomes back into resource quality scoring

Referrals don't disappear into voids—outcomes tracked and fed back into resource selection.

Integrations

EHR Clinical Notes & Screenings (PRAPARE, AHC)
Community Resource Directory (Unite Us, findhelp)
ICD-10 Z-Code Reference
Patient Messaging (multilingual)
Care Management Platform
CMS VBC Reporting Module

Industries

healthcare

See It In Action

A sample conversation with this agent

Scan our attributed population for unaddressed SDOH needs.
Running NLP against clinical notes, screening responses, and structured data across 4,218 members. Surfacing: 312 with food insecurity signals, 187 with housing instability, 156 with transportation barriers, 94 with social isolation, 73 with utility insecurity. Of these, only 340 have matching Z-codes documented—the remaining 482 are captured informally in notes but not coded. I'll prepare Z-code update recommendations and cross-match everyone to community resources in their ZIP code.
Maria Gonzalez's intake screening flagged food insecurity. What happens next?
Pulling her profile. She's in ZIP 08103 and speaks Spanish primarily. Matching community resources: Food Bank of South Jersey delivers to her area Tuesdays/Fridays, Camden SNAP outreach office (0.8 mi away, Spanish-speaking staff), and two local pantries. Generating a referral package: I'll send her an SMS in Spanish with her top 3 options, notify the Food Bank of her referral, add Z59.41 (food insecurity) to her problem list for physician review, and queue a care manager check-in in 2 weeks to confirm she accessed services.
How are our referrals performing?
90-day rolling view: 487 referrals initiated, 291 confirmed connections (60%), 142 verified service delivery (29%), 54 patients reported needs resolved. Strongest channel: transportation (LogistiCare, 78% completion). Weakest: housing (18% completion—waitlist bottleneck). Top gaps by resource scarcity: affordable housing in ZIP 08103, mental health counseling for Spanish speakers. Flagging the housing gap for population health leadership.