This week, we showed you what's possible when healthcare data is transformed instead of just accessed. Sarah's fragmented records became a complete health picture. NPI numbers became real people. Billing codes became clinical insights. XML documents became queryable FHIR resources.
But Sarah's story is just one example. These transformations — NPI expansion, condition classification, claims-to-clinical conversion, and C-CDA parsing — are a small subset of what Flexpa does. The platform applies hundreds of normalization rules across every data source, every payer, every record.
Today: the philosophy behind data transformation and why it matters more than access.
Don't feel like reading? Watch the video summary:
The Access vs. Utility Gap
The past decade brought major regulatory wins for healthcare data access. CMS mandated FHIR APIs for payers in 2020, TEFCA connected over 72,000 (and counting!) facilities in 2026, and payer-to-payer portability rules are launching next year. Patients can now request their complete medical history from any insurer or provider. Developers have standardized APIs to build on.
But access and usability are different problems. Query a payer's FHIR API for insurance claims and you get ExplanationOfBenefit resources filled with 10-digit NPI numbers (not provider names), ICD-10-CM codes like "E11.65" (not "chronic diabetes with hyperglycemia"), and CPT billing line items (not structured encounters showing when care happened). Request clinical records through TEFCA and you receive an 847-page C-CDA XML document with medications buried in <substanceAdministration> tags, lab results nested six levels deep in <organizer> sections, and allergies scattered across multiple <observation> elements. None of it is queryable. None of it is indexed. You'd need to write a custom parser just to extract the patient's active prescriptions.
Flexpa transforms everything automatically. Every API response includes enriched provider names, chronicity classifications, clinical timelines, and queryable FHIR resources.
Transformation Across Every Use Case
These aren't hypothetical scenarios. This is what developers are building with Flexpa today, across industries where healthcare data was previously too fragmented or too complex to use effectively.
Legal Tech: Medical Chronologies
Personal injury attorney receives 847-page C-CDA XML from 12 hospital visits spanning 2 years. Without transformation, they'd need to manually extract procedures, identify treating physicians, and build a litigation timeline from unstructured XML.
With Flexpa: Procedure?date=ge2024-01-01 returns 156 CPT-coded procedures automatically extracted from C-CDA and EOB line items. Every Procedure.performer resolves to enriched Practitioner resources with full names, specialties, and facilities (NPI 1234567890 becomes "Dr. Sarah Chen, MD, Orthopedic Surgery, Mission Bay Surgical Center"). ICD-10 diagnoses are grouped by CCSR body system (musculoskeletal, neurological). Query Encounter?_sort=-date&_revinclude=Procedure:encounter to build the complete chronology.
Benefits Navigation: Cost Estimation
Employee switching from PPO to HDHP needs to estimate out-of-pocket costs but doesn't know provider NPIs or procedure codes.
Condition resources automatically include CCIR chronic classification, identifying E11.9 (diabetes), I10 (hypertension), J45.909 (asthma) as ongoing care requirements. Encounter shows 18 visits across 4 practitioners, cross-referenced with plan network to check in-network status. MedicationRequest lists 6 active RxNorm codes filled monthly. Calculate: 18 visits × $150 + 72 prescriptions × $45 + 2 MRIs = $6,840 pre-deductible.
Care Navigation: Gap Detection
Patient with CKD Stage 3 (N18.3) hasn't seen a nephrologist in 14 months despite declining eGFR (48 to 42 mL/min over 8 months).
Observation?code=http://loinc.org|33914-3 (eGFR LOINC code) shows the declining trend automatically. Encounter resources reveal all 4 recent visits were with PCP (NPI 1234567890, Dr. James Liu, Family Medicine), none with nephrology taxonomy 207RN0300X. Flag the care gap: CKD Stage 3 guidelines require specialist visit every 6 months.
Pharmacy: Adherence Tracking
Metformin 1000mg (RxNorm 861004) prescribed twice daily. Last fill 127 days ago for 90-day supply. Prescription ran out 37 days ago.
MedicationRequest from TEFCA C-CDA includes dosing frequency (repeat.frequency=2, period=1, periodUnit=d). ExplanationOfBenefit filtered by NDC shows last pharmacy claim 2024-10-15. Math: 2024-10-15 + 90 days = depletion on 2025-01-13, current date 2025-02-19 = 37-day gap. Cross-reference with CCIR-tagged chronic diabetes (E11.9) to confirm continuous medication requirement.
Health AI: Feature Engineering
30-day readmission model needs chronic disease burden, utilization, comorbidity, and provider continuity across 50,000 patients.
Extract from FHIR: count Condition with CCIR chronic extension, aggregate CCSR categories (4+ = high complexity), compute encounter frequency from Encounter?_count=10&_sort=-date, calculate provider continuity from participant.individual.reference mode. Example patient: 7 chronic conditions across 5 CCSR categories, 12 encounters in 90 days, 67% visit continuity = readmission risk 0.83.
Wearables: Clinical Correlation
Consumer smartwatch detects 3 AFib episodes, elevated resting HR (92 bpm average). Needs correlation with clinical diagnoses and medications.
Condition?code=I48.91 (AFib) dated 2024-08-22, CCSR "circulatory system", CCIR chronic. MedicationRequest shows active Eliquis 5mg (RxNorm 1364445). Observation?code=8867-4 (heart rate LOINC) shows clinical measurements 88 bpm (Dec) and 85 bpm (Sep). Correlate with wearable trend to alert: "AFib episodes detected. Diagnosed 8/22/24, prescribed Eliquis. HR increasing 85→92 bpm. Schedule cardiology follow-up (NPI 1847562910)."
Provider Platforms: Quality Gaps
Primary care group managing 15,000 diabetic patients needs to identify who's overdue for A1C testing (6 months) for HEDIS reporting.
Filter Condition?code:text=E11 identifies all diabetes patients. For each, query Observation?code=http://loinc.org|4548-4&_count=1&_sort=-date (HbA1c) and filter where effectiveDateTime < 2024-08-19. Result: 4,237 overdue. Cross-reference Encounter to prioritize: 2,814 need appointment scheduling (last visit >90 days), 1,423 need lab order (visited recently). Export CSV with MRN, last A1C date/value, assigned PCP, days overdue.
Payers: HCC Risk Adjustment
Medicare Advantage plan calculating HCC risk scores for 250,000 members requires chronic condition validation and invalid diagnosis exclusion.
Filter Condition by CCIR extension.valueCoding.code=C for chronic conditions eligible for HCC mapping. Map ICD-10-CM to 2024 CMS-HCC V28: E11.65→HCC 37 (diabetes with hyperglycemia/hypoglycemia), N18.4→HCC 327 (CKD Stage 4), I50.23→HCC 223 (heart failure). Validate persistence via recordedDate and encounter.reference (minimum 2 face-to-face encounters per CMS). For member 987654321: 4 valid HCCs, 2 filtered (J20.9 acute bronchitis per CCIR NC, R07.9 rule-out excluded). Risk score: community non-dual female 72 + coefficients = 1.847.
Get Started
Flexpa Flux is live for all customers. Every API call includes automatic transformation with no configuration required.
- Read the transformation documentation
- View all FHIR resources
- Get started with the quickstart guide
- Pull your own health records
- Schedule a demo
Transform healthcare data. Transform healthcare.
This is Day 5 of Flexpa Flux Launch Week. Catch up on the full series: NPI Expansion, CCIR/CCSR Classification, Claims to Clinical, CCDA to FHIR.
