Legacy EHR connected. OASIS acceptance from 71% to 94%. Notes in the home.
How a 22-branch home health agency connected a 12-year-old on-premise EHR to modern payer portals, mobile field documentation, and automated OASIS submission — without replacing a single system.
“We were terrified of touching a system with 12 years of patient data. Metricoid ran both systems in parallel for 6 months and gave us complete confidence before we cut over. Our field clinicians now finish notes in the home, not in their cars.”
A home health agency trapped by a system it was afraid to touch
A Medicare-certified home health agency operating 22 branches across the Mid-Atlantic region. The organization had grown organically over 18 years and was running a 12-year-old on-premise EHR that had become a liability rather than an asset.
The system worked — in the sense that it stored patient records and supported clinical documentation — but it had become a barrier to everything the organization wanted to do next. No mobile access for field clinicians. No FHIR API for AI integration. No connectivity to modern payer portals. Manual OASIS submissions by fax. And a vendor that had reduced support to email-only with week-long response times.
Leadership wanted to move to a modern platform and had been discussing it for three years. The fear of losing 12 years of patient records in a migration had paralyzed every conversation. This engagement proved that fear was manageable — with the right parallel run approach.
This engagement was a full EHR integration project — connecting a legacy system to a modern ecosystem without replacing it. The primary deliverables were mobile field documentation, automated OASIS submission, payer portal connectivity, and PDGM billing automation. CS-05 covers the full legacy migration that followed 6 months later.
Field clinicians finishing notes in parking lots — every day
The most visible problem was field documentation. Home health nurses saw 6–8 patients per day in their homes across a wide geographic area. The legacy EHR had no mobile app — it required a desktop computer and office connectivity. So every nurse either completed paper notes during visits and re-entered them at the office later, or carried a laptop to patient homes and struggled with connectivity. Either way, documentation was routinely finished hours after visits — in parking lots, in patient driveways, or back at the office after a full day of patient care.
The OASIS submission problem was less visible but more expensive. OASIS forms were completed in the EHR, printed, and faxed to CMS. Manual completion and transcription errors were generating a 71% first-pass acceptance rate — meaning 29% of OASIS submissions required correction and resubmission. Each failed submission delayed Medicare payment by an average of 3.2 weeks per episode.
The root cause of both problems was the same: the legacy EHR was an island. It had no connection to mobile devices, no connection to CMS systems, no connection to payer portals, and no ability to support the AI documentation tools the organization wanted to implement. Everything had to flow through manual human processes because the system couldn't connect to anything.
The critical constraint: The legacy EHR had no REST API and no modern integration interfaces — only HL7 v2 messaging over MLLP. Any integration had to be built on what the system actually exposed, not what modern integration standards would prefer. This required custom connector engineering that most integration vendors don't do.
Picks up paper intake forms and visit schedules from office. Reviews paper notes from previous visits to prepare.
Three patient visits. Handwrites visit notes on paper during each visit. Takes paper vitals, wound measurements, medication reconciliation.
Parked outside patient 3 — typing visit notes into EHR on laptop using phone hotspot. Poor signal. System slow. Takes 35 minutes per patient.
Three more patient visits. Paper notes again. Laptop battery low.
Back at office or at home — entering the afternoon visit notes into the EHR from memory and paper. Two hours of documentation for six patient visits.
- OASIS first-pass acceptance rate: 71% — 29% requiring correction and resubmission
- Medicare payment delay from OASIS rejections: avg 3.2 weeks per affected episode
- Field documentation completed after visit: avg 90 min/day per nurse in after-hours
- Manual eligibility checks: 45 min per new patient — calling payers by phone
- PDGM grouping calculated manually: billing team step separate from clinical documentation
Three layers. One connected system. Built on what the legacy EHR actually exposed.
Because the legacy EHR had only HL7 v2 interfaces, the integration required a custom translation layer — converting HL7 v2 messages to FHIR R4 resources that modern systems could consume. This is the engineering work that most integration vendors decline to do.
Legacy on-premise EHR
12-year-old system with patient records, clinical documentation, and billing history. No REST API. No FHIR support. Only HL7 v2 ADT, ORM, and ORU messages over MLLP TCP connection. The database schema was undocumented — reverse-engineered during discovery.
HL7 → FHIR translation & routing layer
The custom-built translation layer that sits between the legacy EHR and the modern ecosystem. Receives HL7 v2 messages from the legacy system, validates and transforms them to FHIR R4 resources, routes them to appropriate downstream systems, and handles bidirectional sync so data written in modern systems flows back to the legacy EHR.
Mobile app · CMS iQIES · Availity · Waystar · PDGM
All modern systems now connect through the integration hub — receiving structured FHIR data from the legacy EHR and sending data back. Field clinicians use the iOS/Android app with offline mode. OASIS submissions route to CMS iQIES automatically. Eligibility and claims route to Availity and Waystar. PDGM grouping calculated from clinical data without manual billing steps.
Six components. Legacy connected to modern — without replacement.
Nothing was replaced. The legacy EHR continued to run exactly as before. We built everything around it — connecting it to the modern tools the clinical team needed.
Custom HL7 v2 connector
A custom-built HL7 v2 listener connecting to the legacy EHR via MLLP TCP interface — the only integration method the system exposed. Receives ADT (patient), ORM (order), and ORU (result) message types in real time. Parses and validates every message before transformation. The engineering foundation that makes everything else possible.
FHIR R4 translation layer
Transforms HL7 v2 messages to FHIR R4 resources — Patient, Encounter, DocumentReference, Observation, Claim — enabling every downstream modern system to consume structured, standards-compliant data from the legacy EHR. Bidirectional: data written in modern systems is translated back to HL7 and written to the legacy EHR.
Mobile field documentation app
iOS and Android app built for home health clinical workflows — visit notes, vital signs, wound assessments, medication reconciliation, and OASIS form completion. Offline mode stores all data locally and syncs automatically when connectivity is restored. Notes completed in the patient home and pushed to the legacy EHR without requiring office access or desktop login.
OASIS validation & CMS iQIES integration
OASIS forms completed in the mobile app are automatically validated against CMS accuracy rules before submission — catching errors that previously generated 29% rejection rates. Valid OASIS assessments submitted directly to CMS iQIES electronically, eliminating the fax workflow. Submission status tracked and alerts generated on any rejection requiring correction.
Payer portal integrations
Real-time eligibility verification through Availity — triggered automatically at start-of-care and at each 60-day episode renewal. Electronic claim submission through Waystar eliminating manual data entry. ERA/EOB payment posting automated. Prior authorization tracking for therapy services with renewal alerts before expiry.
PDGM grouping automation
Patient-Driven Groupings Model (PDGM) billing requires matching clinical characteristics to specific payment groupings — a calculation that was previously done manually by the billing team as a separate step after clinical documentation. The integration layer extracts the relevant clinical data from visit notes and OASIS assessments and calculates PDGM groupings automatically, triggering the correct billing classification without manual billing team action. Eliminates a significant source of billing errors and revenue underperformance.
From 71% to 94% first-pass acceptance — how we did it
OASIS — the Outcome and Assessment Information Set — is the federally mandated assessment that drives Medicare home health payment. A rejected OASIS doesn't just mean administrative rework. It delays Medicare payment for the entire 30-day episode. At 29% rejection rate, the agency was routinely waiting 5+ weeks to be paid for completed care.
Structured OASIS fields in mobile app
OASIS form built into the mobile app with clinical logic validating each field entry in real time — flagging impossible combinations, missing required fields, and illogical response patterns before the clinician submits.
Pre-submission CMS rules engine
Before transmission to CMS, every OASIS is run through a validation engine applying the complete CMS OASIS Data Specifications rules — catching the specific error codes that previously generated rejection notices from iQIES.
Clinical consistency cross-check
Cross-validates OASIS responses against visit note clinical content — flags when the OASIS functional status responses are inconsistent with the clinical observations documented in the visit note.
Direct CMS iQIES submission
Validated OASIS submitted electronically to CMS iQIES — replacing the fax workflow entirely. Submission confirmation received and logged. Rejection notices route back to the responsible clinician with specific correction guidance.
29% of OASIS submissions were previously rejected by CMS and required manual correction and resubmission — each rejection delaying Medicare payment for an entire 30-day episode by an average of 3.2 weeks. At 94% first-pass acceptance, that payment delay is now rare rather than routine.
Integrated in 12 weeks. Legacy system untouched throughout.
The integration was built entirely alongside the legacy system — never modifying the source system or disrupting its operation. Field clinicians started using the mobile app at pilot branches while the legacy system continued running normally.
System Audit
Reverse-engineered the legacy EHR's HL7 v2 interface — mapped all message types, data fields, and business rules. Documented the exact OASIS rejection error codes from the previous 12 months. Established baseline metrics for all KPIs. Obtained BAA and network access for staging work.
Build & Test
HL7 v2 connector built and tested against staging instance of legacy EHR. FHIR translation layer configured for all required resource types. Mobile app built for iOS and Android — OASIS form, visit notes, vitals. CMS iQIES integration built and tested with synthetic OASIS data. Availity and Waystar connections established.
Pilot — 4 Branches
Piloted at 4 branches across 2 geographic regions. Field clinicians trained on mobile app over one 2-hour session. OASIS submissions from pilot branches routed through validation engine and submitted electronically. First-pass acceptance rate at pilot branches measured weekly.
Full Rollout
Deployed to all 22 branches. All field staff trained. OASIS electronic submission active across the full organization. Eligibility verification automated at start-of-care. PDGM grouping automation live. Legacy EHR continuing to operate normally as the source of record.
Results Measured
OASIS first-pass acceptance: 94% (from 71%). Field documentation completed in patient home: confirmed by time-stamping in mobile app. After-hours documentation burden: reduced from 90 min/day to under 15 min/day (catching up notes from any offline periods). Medicare payment delay from OASIS rejections: effectively eliminated. Decision made to proceed with full legacy migration — see CS-05.
Four outcomes. All measured. All sustained.
The outcomes below were measured at 90 days post-full-deployment. The OASIS acceptance improvement was the most financially significant — each percentage point improvement in first-pass rate eliminates weeks of Medicare payment delay for dozens of episodes.
The field documentation change was the most operationally significant for clinical staff. Nurses went from spending 90+ minutes per day entering notes after their shift to completing notes in the patient home in real time — a change that dramatically improved both data accuracy and job satisfaction. Exit interview data in the following 12 months showed documentation burden as a reason for leaving dropped to near-zero from 34%.
The $180K IT cost reduction was the outcome that funded the decision to proceed with the full legacy migration — the subject of CS-05.
All metrics measured at 90 days post-full-deployment against baselines from week 1–2 audit. OASIS acceptance rate measured from CMS iQIES submission reports. Field documentation timing measured from mobile app timestamps. IT cost reduction calculated from infrastructure, licensing, and support cost comparison.
OASIS first-pass acceptance — Medicare payment delays effectively eliminated
Faster field documentation — notes completed in patient home, not hours later
Data accessible to modern systems — zero loss, zero disruption to legacy EHR operation
Annual IT cost reduction from retiring on-premise infrastructure — funded full migration (CS-05)
Built for Medicare home health's regulatory requirements
HIPAA
Full Privacy & Security Rule
BAA Executed
Before any data flows
OASIS Compliant
CMS Data Specification rules applied
PDGM Ready
Grouping automation from clinical data
FHIR R4
Modern interoperability standard
HL7 v2
Legacy system connectivity
AES-256 + TLS 1.3
Encryption at rest & transit
Offline-secure
Encrypted local storage on mobile devices
Explore the full capabilities
EHR & System Integration
Deep integration with 50+ EHR systems — FHIR R4, HL7 v2, and browser extension. Works with legacy systems that have no modern API.
Legacy System Modernization
This engagement led directly to a full legacy migration — see CS-05 for the complete migration story with zero data loss.
Home Health
See how Metricoid serves home health agencies — OASIS, PDGM, mobile field documentation, and legacy EHR connectivity.
Other case studies
CS-05: Full legacy EHR migration. 100% data. $180K IT savings.
Days in AR from 71 to 29 — $520K recovered annually
From 13.2% denial rate to 4.1% — 62% less documentation time
Get Started
Running a legacy EHR that's blocking your AI roadmap?
We will audit your current system stack — EHR version, available interfaces, integration requirements — and show you exactly what connectivity is possible. 15 minutes, no commitment.
- Works with legacy systems
- OASIS & PDGM compliance built in
- No rip-and-replace required
- BAA signed before any data flows
