NPI Providers

Full schema reference, querying examples, aggregate views, and data quality notes for the NPI Healthcare Provider dataset.

The npi_providers table contains ~9 million active NPI-registered healthcare providers from the CMS NPPES Registry, with related tables for taxonomy codes, alternate names, and secondary practice locations.

Source

Property	Value
Source	CMS NPPES Data Dissemination
Total records	~9,034,000
Individuals (Type 1)	~7,140,000 (79%)
Organizations (Type 2)	~1,894,000 (21%)
Last CMS update	February 2026
Refresh strategy	Periodic (monthly full file from CMS)

Sample Record

A complete JSON object returned by the API (with select=*):

{
  "npi": 1234567890,
  "entity_type": 1,
  "replacement_npi": null,
  "organization_name": null,
  "last_name": "SMITH",
  "first_name": "JOHN",
  "middle_name": "A",
  "name_prefix": "Dr.",
  "name_suffix": null,
  "credential": "M.D.",
  "practice_address_1": "1234 MEDICAL CENTER DR",
  "practice_address_2": "STE 200",
  "practice_city": "HOUSTON",
  "practice_state": "TX",
  "practice_zip": "770304552",
  "practice_phone": "7135551234",
  "enumeration_date": "2005-06-15",
  "last_update_date": "2024-08-01",
  "deactivation_date": null,
  "reactivation_date": null,
  "gender": "M",
  "sole_proprietor": "N",
  "is_subpart": null,
  "parent_organization_name": null,
  "authorized_official_last_name": null,
  "authorized_official_first_name": null,
  "authorized_official_middle_name": null,
  "authorized_official_title": null,
  "authorized_official_phone": null,
  "authorized_official_credential": null,
  "authorized_official_prefix": null,
  "authorized_official_suffix": null,
  "primary_taxonomy_code": "207R00000X",
  "primary_taxonomy_display": "Internal Medicine"
}

The primary_taxonomy_code value 207R00000X maps to "Internal Medicine" — see the NUCC Taxonomy Code Reference for the full hierarchy and common codes.

Schema

Identity & Type

Column	Type	Description
`npi`	bigint	10-digit National Provider Identifier (unique, primary key)
`entity_type`	smallint	`1` = Individual, `2` = Organization
`replacement_npi`	bigint	Replacement NPI if this one was replaced

Provider Name

Column	Type	Description
`organization_name`	text	Legal business name (Type 2 only)
`last_name`	text	Provider last name (Type 1 only)
`first_name`	text	Provider first name (Type 1 only)
`middle_name`	text	Provider middle name (Type 1 only)
`name_prefix`	varchar(5)	Ms., Mr., Dr., Prof., etc.
`name_suffix`	varchar(5)	Jr., Sr., II, III, etc.
`credential`	varchar(20)	M.D., D.O., FNP, DDS, PhD, etc.

Practice Location Address

Column	Type	Description
`practice_address_1`	text	Street address line 1
`practice_address_2`	text	Suite, floor, building, etc.
`practice_city`	text	City name (stored in ALL CAPS)
`practice_state`	varchar(2)	2-letter state code (e.g., CA, NY, TX)
`practice_zip`	varchar(20)	ZIP code (5 or 9 digits, no hyphen)
`practice_phone`	varchar(20)	Phone number (10 digits, no formatting)

Dates & Status

Column	Type	Description
`enumeration_date`	date	Date NPI was first assigned
`last_update_date`	date	Date provider last updated their record
`deactivation_date`	date	Set only for previously-deactivated-then-reactivated providers
`reactivation_date`	date	Set only if reactivated after deactivation

Individual-Specific Fields (Type 1)

Column	Type	Description
`gender`	varchar(1)	`M`, `F`, `U` (undisclosed), `X` (undisclosed)
`sole_proprietor`	varchar(1)	`Y`, `N`, `X` (not answered)

Organization-Specific Fields (Type 2)

Column	Type	Description
`is_subpart`	varchar(1)	`Y`/`N`/`X` — indicates org is a division of a parent
`parent_organization_name`	text	Legal name of parent org (when `is_subpart=Y`)
`authorized_official_last_name`	text	Organization's official contact last name
`authorized_official_first_name`	text	Official contact first name
`authorized_official_middle_name`	text	Official contact middle name
`authorized_official_title`	varchar(35)	CEO, Administrator, Owner, CFO, etc.
`authorized_official_phone`	varchar(20)	Direct phone for the official
`authorized_official_credential`	varchar(20)	M.D., MBA, etc.
`authorized_official_prefix`	varchar(5)	Dr., Mr., etc.
`authorized_official_suffix`	varchar(5)	Jr., etc.

Taxonomy / Specialty

Column	Type	Description
`primary_taxonomy_code`	varchar(10)	NUCC taxonomy code for primary specialty (e.g., `207R00000X`)
`primary_taxonomy_display`	text	Human-readable specialty name (e.g., "Internal Medicine")

Search

Column	Type	Description
`search_vector`	tsvector	Full-text search vector (auto-generated from provider/org name)

The NPI dataset includes four related tables that extend the main provider records.

npi_taxonomy_codes (Lookup)

Static reference table mapping ~873 NUCC taxonomy codes to human-readable names.

Column	Type	Description
`taxonomy_code`	varchar(10)	NUCC code (primary key)
`provider_type`	text	Top-level grouping (e.g., "Allopathic & Osteopathic Physicians")
`classification`	text	Specialty (e.g., "Internal Medicine")
`specialization`	text	Sub-specialty (e.g., "Cardiovascular Disease"), nullable
`display_name`	text	Human-readable label
`slug`	text	URL-safe slug (e.g., `internal-medicine`)

npi_taxonomies (Join Table)

Maps providers to their taxonomy codes (a provider can have up to 15).

Column	Type	Description
`npi`	bigint	Foreign key to `npi_providers`
`taxonomy_code`	varchar(10)	NUCC taxonomy code
`is_primary`	boolean	True if this is the provider's primary specialty
`seq`	smallint	Original slot position (1-15)

Composite primary key: (npi, taxonomy_code). ~11.2M rows.

npi_other_names

DBA names and former legal names for organizations. ~709K rows.

Column	Type	Description
`npi`	bigint	Foreign key to `npi_providers`
`other_name`	text	The alternate name
`name_type_code`	smallint	`1`=Former Name, `2`=Professional Name, `3`=DBA, `4`=Former Legal Business Name, `5`=Other

npi_practice_locations

Secondary practice locations (the primary is in the main table). ~1.15M rows.

Column	Type	Description
`id`	bigserial	Auto-increment primary key
`npi`	bigint	Foreign key to `npi_providers`
`address_1`	text	Street address
`address_2`	text	Suite/floor
`city`	text	City
`state`	varchar(2)	State code
`zip`	varchar(20)	ZIP code
`phone`	varchar(20)	Phone number

Fast-Lookup Columns

These columns are optimized for fast filtering and sorting:

npi (unique)
practice_state
practice_state + practice_city (composite)
practice_state + practice_city + entity_type (composite)
entity_type
primary_taxonomy_code
primary_taxonomy_code + entity_type (composite)
practice_state + primary_taxonomy_code (composite)
practice_state + primary_taxonomy_code + entity_type (composite)
practice_state + practice_city + primary_taxonomy_code + entity_type (composite)
search_vector (full-text search, GIN index)

At least one of these columns must appear as a filter in table queries. practice_city requires practice_state as a co-filter (composite index). Guardrail violations return a 400 response.

Querying Examples

For general query syntax (operators, pagination, sorting), see the Query Syntax Reference.

Look up a provider by NPI number

curl "https://newbizbot.ai/api/data/npi_providers?npi=eq.1234567890" \
  -H "Authorization: Bearer YOUR_API_KEY"

[
  {
    "npi": 1234567890,
    "entity_type": 1,
    "last_name": "SMITH",
    "first_name": "JOHN",
    "credential": "M.D.",
    "practice_city": "HOUSTON",
    "practice_state": "TX",
    "primary_taxonomy_code": "207R00000X",
    "primary_taxonomy_display": "Internal Medicine"
  }
]

Search providers by name

# Full-text search for "Mayo Clinic"
curl "https://newbizbot.ai/api/data/npi_providers?search_vector=fts.mayo%20clinic&limit=5" \
  -H "Authorization: Bearer YOUR_API_KEY"

[
  {
    "npi": 1093707622,
    "entity_type": 2,
    "organization_name": "MAYO CLINIC",
    "practice_city": "ROCHESTER",
    "practice_state": "MN",
    "primary_taxonomy_display": "Clinic/Center"
  }
]

Find providers by state

# Individual providers in California
curl "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.CA&entity_type=eq.1&limit=10&order=npi" \
  -H "Authorization: Bearer YOUR_API_KEY"

Filter by specialty (taxonomy code)

# Cardiologists (taxonomy 207RC0000X) in New York
curl "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.NY&primary_taxonomy_code=eq.207RC0000X&limit=10&select=npi,first_name,last_name,credential,practice_city,primary_taxonomy_display" \
  -H "Authorization: Bearer YOUR_API_KEY"

[
  {
    "npi": 1003000126,
    "first_name": "ROBERT",
    "last_name": "JOHNSON",
    "credential": "M.D.",
    "practice_city": "NEW YORK",
    "primary_taxonomy_display": "Cardiovascular Disease"
  }
]

Find organizations in a city

# Healthcare organizations in Miami, FL
curl "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.FL&practice_city=eq.MIAMI&entity_type=eq.2&limit=10&order=npi&select=npi,organization_name,practice_city,primary_taxonomy_display,authorized_official_first_name,authorized_official_last_name,authorized_official_title" \
  -H "Authorization: Bearer YOUR_API_KEY"

[
  {
    "npi": 1003002411,
    "organization_name": "MIAMI CARDIAC AND VASCULAR INSTITUTE",
    "practice_city": "MIAMI",
    "primary_taxonomy_display": "Clinic/Center",
    "authorized_official_first_name": "MARIA",
    "authorized_official_last_name": "GARCIA",
    "authorized_official_title": "ADMINISTRATOR"
  }
]

Find sub-organizations with parent info

# Organizations that are subparts of a parent, in Texas
curl "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.TX&entity_type=eq.2&is_subpart=eq.Y&limit=10&select=npi,organization_name,parent_organization_name,practice_city,primary_taxonomy_display" \
  -H "Authorization: Bearer YOUR_API_KEY"

[
  {
    "npi": 1003045678,
    "organization_name": "BAYLOR SCOTT & WHITE MEDICAL CENTER - TEMPLE",
    "parent_organization_name": "BAYLOR SCOTT & WHITE HOLDINGS",
    "practice_city": "TEMPLE",
    "primary_taxonomy_display": "General Acute Care Hospital"
  }
]

Combine entity type with specialty

# Individual family medicine providers in Chicago, IL
curl "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.IL&practice_city=eq.CHICAGO&entity_type=eq.1&primary_taxonomy_code=eq.207Q00000X&limit=20&select=npi,first_name,last_name,credential,practice_city,primary_taxonomy_display" \
  -H "Authorization: Bearer YOUR_API_KEY"

Count providers

# Count all providers in Ohio
curl -I "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.OH&limit=0" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Prefer: count=exact"
# Read the Content-Range response header for the count

Natural Language → API Translation

These examples show how to translate common questions into API calls.

"Find orthopedic surgeons in Los Angeles"

Thinking: "Orthopedic surgeons" = taxonomy code 207X00000X. "Los Angeles" = city filter requires state co-filter (CA).

?practice_state=eq.CA&practice_city=eq.LOS%20ANGELES&primary_taxonomy_code=eq.207X00000X&limit=50&select=npi,first_name,last_name,credential,practice_city,primary_taxonomy_display

"Show me all hospitals in Texas"

Thinking: "Hospitals" = organization entity type. Hospital taxonomy codes start with 282N (General Acute Care Hospital). "Texas" = state TX.

?practice_state=eq.TX&entity_type=eq.2&primary_taxonomy_code=eq.282N00000X&limit=100&select=npi,organization_name,practice_city,primary_taxonomy_display,authorized_official_first_name,authorized_official_last_name

"How many dentists are there in Florida?"

Thinking: "Dentists" = taxonomy codes starting with 1223 (Dentist). "How many" = count query. "Florida" = state FL.

curl -I "https://newbizbot.ai/api/data/npi_providers?practice_state=eq.FL&primary_taxonomy_code=like.1223*&limit=0" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Prefer: count=exact"

"Find nursing facilities that are part of a larger health system"

Thinking: "Nursing facilities" = entity_type 2 with nursing taxonomy. "Part of a larger system" = is_subpart=eq.Y.

?entity_type=eq.2&is_subpart=eq.Y&primary_taxonomy_code=like.314*&limit=50&select=npi,organization_name,parent_organization_name,practice_city,practice_state

"Search for Dr. Patel in New Jersey"

Thinking: "Dr. Patel" = full-text search for "patel". "New Jersey" = state NJ. Individual provider.

?practice_state=eq.NJ&entity_type=eq.1&search_vector=fts.patel&limit=50&select=npi,first_name,last_name,credential,practice_city,primary_taxonomy_display

Translation Cheat Sheet

Human concept	API filter
Geographic region	`practice_state=in.(STATE1,STATE2,...)`
Specific city	`practice_state=eq.XX&practice_city=eq.CITY` (state required)
Individual vs organization	`entity_type=eq.1` (individual) or `entity_type=eq.2` (organization)
Specialty	`primary_taxonomy_code=eq.XXXXXXXXXX` (exact) or `like.XXXX*` (prefix)
Provider by NPI	`npi=eq.XXXXXXXXXX`
Provider by name	`search_vector=fts.name`
Sub-organizations	`entity_type=eq.2&is_subpart=eq.Y`
Parent org info	Add `parent_organization_name` to `select`
Decision maker	Add `authorized_official_*` fields to `select` (Type 2 only)
"How many"	Use `Prefer: count=exact` header + `limit=0`
"Biggest / most providers"	Use aggregate views (see below)

Aggregate Views

The NPI dataset uses pre-computed materialized views for aggregate queries. These views are refreshed periodically and return results in milliseconds.

Stats by State

The npi_stats_by_state_mv view provides aggregate counts grouped by state.

Column	Type	Description
`state`	text	2-letter state code
`total_providers`	bigint	Total NPI providers in the state
`type1_count`	bigint	Individual providers (Type 1)
`type2_count`	bigint	Organizations (Type 2)

Stats by City

The npi_stats_by_city_mv view provides aggregate counts grouped by state and city.

Column	Type	Description
`state`	text	2-letter state code
`city`	text	City name
`total_providers`	bigint	Total providers in the city
`type1_count`	bigint	Individual providers
`type2_count`	bigint	Organizations

Stats by Specialty

The npi_stats_by_specialty_mv view provides aggregate counts grouped by primary taxonomy code.

Column	Type	Description
`taxonomy_code`	varchar(10)	NUCC taxonomy code
`display_name`	text	Human-readable specialty name
`slug`	text	URL-safe slug
`total_providers`	bigint	Total providers with this specialty

Stats by State + Specialty

The npi_stats_by_state_specialty_mv view provides aggregate counts grouped by state and specialty.

Column	Type	Description
`state`	text	2-letter state code
`taxonomy_code`	varchar(10)	NUCC taxonomy code
`display_name`	text	Human-readable specialty name
`slug`	text	URL-safe slug
`total_providers`	bigint	Total providers

Data Quality Notes

This is CMS NPPES data loaded as-is. Known characteristics:

Issue	Scope	Details
Self-reported data	All records	Provider information is self-reported and not independently verified
City name casing	All records	City names are stored in ALL CAPS (e.g., `HOUSTON`)
ZIP code format	All records	Stored as raw digits without hyphen (e.g., `770304552` for `77030-4552`)
Phone number format	All records	Stored as 10 digits without formatting (e.g., `7135551234`)
Deactivated providers	Excluded	~333K permanently deactivated records are excluded from the dataset
Foreign providers	Excluded	~5.7K non-US providers are excluded
Missing gender	~0.4% of Type 1	Some individuals have `U` or `X` for gender
Multiple specialties	~19%	Most providers have one taxonomy code; ~1.73M have 2+ (see `npi_taxonomies` table)
Sub-organization data	~8% of Type 2	Only ~154K organizations have `parent_organization_name` populated
Stale records	Varies	Some providers haven't updated their NPI record in years; check `last_update_date`

On this page