Data Sources
Jiro uses de-identified, aggregated data from multiple sources to give you an accurate picture of your practice, your patients, and the latest clinical evidence.
- Where Jiro's Data Comes From
- Data by Feature
- Complete and Recent Toggles
- Reference Data
- Data Coverage and Limitations
- How Data Is Cleaned and Prepared
- Frequently Asked Questions
- Related Articles
Jiro Health pulls from several underlying data sources to power practice Metrics, Insights, clinical chat, and literature recommendations. Each feature draws from a different mix of data depending on its purpose. This article explains what data is used, where it comes from, and what you should know about how it is delivered.
Where Jiro's Data Comes From
De-identified, Aggregated Claims Data
This is the foundation of the Practice Dashboard, Insights, and Spotlights. Jiro uses a national claims database that includes medical and pharmacy claims submitted to payers across the country. The data is de-identified and aggregated before it reaches the platform.
- Medical claims: diagnoses (ICD-10 codes), procedures (CPT/HCPCS codes), place of service, facility information, and provider identifiers. These include open claims, closed claims, and QE closed claims (Medicaid/Medicare).
- Pharmacy claims: prescription fills, drug codes (NDC), drug categories, and fill dates
- Patient demographics and payer eligibility
NPI Registry (NPPES)
Jiro uses the CMS National Plan and Provider Enumeration System to attribute claims to the right clinicians and organizations. Your NPI is the key that links claims data to your Jiro account, determines your specialty, and places you in the appropriate peer group for benchmarking.
Clinical Code Reference Tables
Jiro uses standard reference tables to interpret claims data. These include ICD-10 diagnosis codes, CPT procedure codes, NDC drug codes, and DRG groupings. These mappings allow Jiro to categorize conditions, procedures, and medications consistently across your patient population.
Peer-Reviewed Medical Literature
Consult and Discover pull from a curated body of peer-reviewed journals and clinical guidelines. This content is personalized to your specialty and is the source for responses in Consult and article recommendations in Discover. Each response and article is attributed to its original source.
Data by Feature
|
Feature |
Primary Data Sources |
What You See |
| Metrics | De-identified claims data, clinical code reference tables, NPI registry | Metrics across your attributed patient panel, benchmarked against peers at the same facility and specialty |
| Referrals | De-identified claims data, NPI registry | Derived from shared patient claims across providers; data lag and completeness handling may differ from other Metric types |
| Spotlights | De-identified claims data, peer-reviewed literature | Automatically surfaced highlights of notable patterns in your claims data |
| Insights | De-identified claims data, peer-reviewed literature | Practice insights that compare your patterns to peers and to evidence in the literature |
| Consult | Peer-reviewed literature and clinical guidelines | Clinical responses with cited sources, linked inline in every response |
| Discover | Peer-reviewed journals and clinical literature | Curated article recommendations personalized to your specialty; CME-eligible articles are labeled |
Complete and Recent Toggles
Complete
Shows data from periods where claims delivery is substantially complete (approximately 90% or more). More reliable for identifying real trends, but reflects a longer time lag.
Recent
Includes open claims for a fresher snapshot. Useful for near-term visibility, but some claims are still arriving and the data is less complete.
Every Metric card shows its applicable date range so you always know what time period your numbers reflect. The toggle exists in the Practice Dashboard. Spotlights and Insights have their own quality gates and are not affected by this toggle.
Reference Data
Provider and Facility
Jiro links claims data to curated reference information about providers and facilities. This supports questions about referral relationships, populations being managed, and how care differs by practice site.
- Practice, clinic, and facility attributes
- Clinician identifiers and specialty information
- Groupings used to attribute care to a practice or clinician
Data Coverage and Limitations
Jiro Does Not Capture Your Full Patient Panel
Jiro only sees patients whose claims flow through payers that are connected to the underlying data source. Patients who are self-pay, whose payer is not represented, or who were seen but not billed are not visible in the platform. The database covers approximately 60–70% of U.S. medical encounters and 50–60% of pharmacy encounters in any given year.
Claims Reflect Billing, Not All Clinical Care
Claims are generated when a service is billed to a payer. Some clinical activities — such as phone calls, informal consultations, or services bundled into another claim — may not appear in the data. Jiro's Metrics are derived from what was billed and adjudicated.
Consult Does Not Use Your Patient Data
When you ask Consult a clinical question, it draws from medical literature and guidelines — not your individual patient records or claims history. If you ask Consult about your patient population specifically, it will note that it does not have access to that data.
How Data Is Cleaned and Prepared
Before any data is used in Jiro, it undergoes a series of steps to ensure accuracy, consistency, and clinical usefulness.
Ingestion and validation: New records are validated for structure and completeness. Duplicate files or overlapping feeds are identified and reconciled.
Standardization and normalization: Codes and categories are mapped to consistent internal representations. Dates and identifiers are normalized. Services are grouped into clinically meaningful categories.
De-duplication and record linkage: Duplicate claim lines and encounters are identified and merged. Related codes, payments, and records from a single visit are linked. Encounters, prescriptions, and providers are connected to construct a coherent patient journey.
Aggregation into Metrics: Data is aggregated at the patient, clinician, and practice levels. Transparent, documented logic defines each Metric. Clinical rules have been reviewed by clinicians and validated against real-world scenarios.
Ongoing quality monitoring: New data loads are monitored for unexpected shifts. Validation checks and sampling are run on an ongoing basis. Metrics and logic are updated as coding practices, guidelines, or practice patterns evolve.
Frequently Asked Questions
Where does the data in Jiro come from? Jiro uses de-identified healthcare utilization and medication data, combined with provider and facility reference data and information generated within the platform.
Are my patients' names or identifiers visible in Jiro? No. All data in Jiro is de-identified and aggregated. You see counts, rates, and patterns across your attributed patient population. Individual patients are never identified by name, date of birth, or any other direct identifier.
Does Jiro show raw claim detail? No. Jiro does not expose raw claim forms or line-level billing detail in the clinician interface. Metrics and Insights reflect defined, documented logic applied to the underlying data.
Related Articles
- Time & Refresh Cadence
- Common Reasons Data May Appear Inaccurate
- Benchmarking Methodology
- Data Sources
Version History
Updated: April 17, 2026
Reviewed by: Claude, Help Scout Docs Reviewer
Approved by: [Name, Title]