# NLM Clinical Tables

> Search and look up the medical code sets the U.S. National Library of Medicine serves: ICD-10-CM, ICD-11 and ICD-9-CM diagnosis codes, HCPCS codes, LOINC lab tests, questions and forms with their answer lists, RxTerms drugs with their strengths and RxNorm ids, drug ingredients, everyday names for conditions and surgeries with their codes, NPI providers and organizations, UCUM units, HPO phenotypes, HGNC genes and ClinVar variants. Check a batch of codes from a claim or a lab feed in one step, turn a code into its description or a term into its code, list a questionnaire's questions and answers, or find out who a provider is from their NPI. Every action reads: nothing here changes anything. No account or API key required.

App · 20 actions. Web page: https://docs.lodol.com/docs/integrations/nlm-clinical-tables

## Actions

### Get LOINC answer list

Get the answers a LOINC question offers, in order, each with its LOINC answer code and score: 'Not at all' to 'Nearly every day' for a PHQ-9 question, for example. Every answer list the question has comes back, with the forms each is used in. A question that takes a free value comes back with no answer list. No API key required.

**Inputs**

- LOINC number (`loinc_number`, text, required): The question's LOINC number, such as 44250-9. Find it with 'Search LOINC'.

**Outputs**

- `answer_count` (number): How many answers 'answers' holds.
- `answer_lists` (list): Every answer list the item has, with the forms each one is used in.
- `answers` (list): The answers of the item's first answer list, in order: the list NLM gives for the item. 'answer_lists' says which forms it is used in.
- `data_type` (text): The kind of answer the item takes, e.g. 'CNE'.
- `has_answer_list` (true/false): Whether the item has any answer list.
- `loinc_number` (text): The LOINC number read.
- `long_common_name` (text): LOINC's long common name for the item.
- `name` (text): The item's text as NLM shows it.

**Example output**

```json
{
  "answer_count": 4,
  "answer_lists": [
    {
      "answers": [
        {
          "code": "LA6568-5",
          "score": 0,
          "sequence": 1,
          "text": "Not at all"
        },
        {
          "code": "LA6569-3",
          "score": 1,
          "sequence": 2,
          "text": "Several days"
        },
        {
          "code": "LA6570-1",
          "score": 2,
          "sequence": 3,
          "text": "More than half the days"
        },
        {
          "code": "LA6571-9",
          "score": 3,
          "sequence": 4,
          "text": "Nearly every day"
        }
      ],
      "forms": [
        "44249-1",
        "52744-0",
        "52745-7"
      ],
      "list_id": "LL358-3",
      "list_name": "PHQ"
    },
    {
      "answers": [
        {
          "code": "LA33-6",
          "score": null,
          "sequence": 1,
          "text": "Yes"
        },
        {
          "code": "LA32-8",
          "score": null,
          "sequence": 2,
          "text": "No"
        }
      ],
      "forms": [],
      "list_id": "LL361-7",
      "list_name": "Y/N"
    }
  ],
  "answers": [
    {
      "code": "LA6568-5",
      "score": 0,
      "sequence": 1,
      "text": "Not at all"
    },
    {
      "code": "LA6569-3",
      "score": 1,
      "sequence": 2,
      "text": "Several days"
    },
    {
      "code": "LA6570-1",
      "score": 2,
      "sequence": 3,
      "text": "More than half the days"
    },
    {
      "code": "LA6571-9",
      "score": 3,
      "sequence": 4,
      "text": "Nearly every day"
    }
  ],
  "data_type": "CNE",
  "has_answer_list": true,
  "loinc_number": "44250-9",
  "long_common_name": "Little interest or pleasure in doing things in last 2 weeks",
  "name": "Little interest or pleasure in doing things"
}
```

### Get LOINC form

Get every question of a LOINC form or panel, such as the PHQ-9 or the vital signs panel, in the form's order: each with its LOINC number, text, kind of answer, units and answers with their scores, and the sections that group them. Also returns the form's coding instructions and copyright notice. No API key required.

**Inputs**

- LOINC number (`loinc_number`, text, required): The form or panel's LOINC number, such as 44249-1 (PHQ-9). Find it with 'Search LOINC' searching in forms and panels.

**Outputs**

- `coding_instructions` (text): How to fill in or score the form.
- `copyright_notice` (text): The form's copyright notice, when it has one.
- `loinc_number` (text): The form's LOINC number.
- `name` (text): The form's name.
- `question_count` (number): How many questions the form has, not counting sections.
- `questions` (list): Every question and section, in the form's order, sections followed by their questions.
- `url` (text): The form's page on loinc.org.

**Example output**

```json
{
  "coding_instructions": "The Patient Health Questionnaire is a self-administered version of the PRIME-MD diagnostic instrument for common mental disorders. The PHQ-9 is the depression module, which scores each of the 9 DSM-IV criteria as \"0\" (not at all) to \"3\" (nearly every day).\r\nFor initial diagnosis:\r\n1. Patient completes PHQ-9 Quick Depression Assessment on accompanying tear-off pad.\r\n2. If there are at least 4...",
  "copyright_notice": "Copyright \u00a9 Pfizer Inc. All rights reserved. Developed by Drs. Robert L. Spitzer, Janet B.W. Williams, Kurt Kroenke and colleagues, with an educational grant from Pfizer Inc. No permission required to reproduce, translate, display or distribute.",
  "loinc_number": "44249-1",
  "name": "PHQ-9 quick depression assessment panel [Reported.PHQ]",
  "question_count": 3,
  "questions": [
    {
      "answers": [
        {
          "code": "LA6568-5",
          "label": "0",
          "score": 0,
          "text": "Not at all"
        },
        {
          "code": "LA6569-3",
          "label": "1",
          "score": 1,
          "text": "Several days"
        },
        {
          "code": "LA6570-1",
          "label": "2",
          "score": 2,
          "text": "More than half the days"
        },
        {
          "code": "LA6571-9",
          "label": "3",
          "score": 3,
          "text": "Nearly every day"
        }
      ],
      "data_type": "CNE",
      "is_section": false,
      "level": 1,
      "local_code": "PHQ-9.1",
      "loinc_number": "44250-9",
      "question": "Little interest or pleasure in doing things",
      "section": null,
      "units": []
    },
    {
      "answers": [
        {
          "code": "LA6568-5",
          "label": "0",
          "score": 0,
          "text": "Not at all"
        },
        {
          "code": "LA6569-3",
          "label": "1",
          "score": 1,
          "text": "Several days"
        },
        {
          "code": "LA6570-1",
          "label": "2",
          "score": 2,
          "text": "More than half the days"
        },
        {
          "code": "LA6571-9",
          "label": "3",
          "score": 3,
          "text": "Nearly every day"
        }
      ],
      "data_type": "CNE",
      "is_section": false,
      "level": 1,
      "local_code": "PHQ-9.2",
      "loinc_number": "44255-8",
      "question": "Feeling down, depressed, or hopeless",
      "section": null,
      "units": []
    },
    {
      "answers": [
        {
          "code": "LA6568-5",
          "label": "0",
          "score": 0,
          "text": "Not at all"
        },
        {
          "code": "LA6569-3",
          "label": "1",
          "score": 1,
          "text": "Several days"
        },
        {
          "code": "LA6570-1",
          "label": "2",
          "score": 2,
          "text": "More than half the days"
        },
        {
          "code": "LA6571-9",
          "label": "3",
          "score": 3,
          "text": "Nearly every day"
        }
      ],
      "data_type": "CNE",
      "is_section": false,
      "level": 1,
      "local_code": null,
      "loinc_number": "44259-0",
      "question": "Trouble falling or staying asleep, or sleeping too much",
      "section": null,
      "units": []
    }
  ],
  "url": "https://loinc.org/44249-1"
}
```

### Look up codes

Check up to 100 codes from one code set at once, such as the diagnosis codes on a claim or the test codes in a lab feed, and get each one's name. Works with ICD-10-CM, ICD-11, ICD-9-CM, HCPCS, LOINC, RxNorm ingredient ids, HPO and UCUM. Codes are taken as people write them (E119 finds E11.9), and a code the set does not have comes back marked not found rather than failing the step. No API key required.

**Inputs**

- Code set (`code_system`, text, required): The code set the codes belong to. One of: `icd10cm` (ICD-10-CM diagnoses), `icd11` (ICD-11), `icd9_diagnosis` (ICD-9-CM diagnoses), `icd9_procedure` (ICD-9-CM procedures), `hcpcs` (HCPCS Level II), `loinc`, `rxnorm` (RxNorm ingredients (RxCUI)), `hpo` (HPO phenotypes), `ucum` (UCUM units).
- Codes (`codes`, text, required): One or more codes, separated by commas, spaces or new lines, or a list from an earlier step. Up to 100 at once.

**Outputs**

- `code_system` (text): The code set read, e.g. 'icd10cm'.
- `code_system_name` (text): Its name, e.g. 'ICD-10-CM'.
- `codes` (list): One row per code, in the order given.
- `count` (number): How many codes were looked up.
- `found_count` (number): How many were found.
- `not_found` (list): The codes the code set does not have, as given.

**Example output**

```json
{
  "code_system": "icd10cm",
  "code_system_name": "ICD-10-CM",
  "codes": [
    {
      "code": "E11.9",
      "discontinued": null,
      "found": true,
      "name": "Type 2 diabetes mellitus without complications",
      "requested": "E119",
      "short_name": null
    },
    {
      "code": "I10",
      "discontinued": null,
      "found": true,
      "name": "Essential (primary) hypertension",
      "requested": "I10",
      "short_name": null
    },
    {
      "code": "E11.65",
      "discontinued": null,
      "found": true,
      "name": "Type 2 diabetes mellitus with hyperglycemia",
      "requested": "E11.65",
      "short_name": null
    },
    {
      "code": null,
      "discontinued": null,
      "found": false,
      "name": null,
      "requested": "ZZZ9",
      "short_name": null
    }
  ],
  "count": 4,
  "found_count": 3,
  "not_found": [
    "ZZZ9"
  ]
}
```

### Look up providers by NPI

Read the full NPI record of one or more providers, people or organizations, from their 10-digit NPI: name, credential, every taxonomy (specialty) with its licence, practice and mailing addresses, phone and fax, other ids such as Medicaid numbers, the day the NPI was issued and an organization's authorized official. An NPI NLM does not have comes back marked not found, with whether its check digit is valid, rather than failing the step. No API key required.

**Inputs**

- NPI numbers (`npis`, text, required): One or more 10-digit NPIs, separated by commas, spaces or new lines, or a list from an earlier step. Up to 100 at once.

**Outputs**

- `count` (number): How many NPIs were looked up.
- `found_count` (number): How many were found.
- `not_found` (list): The NPIs NLM has no provider for, as given.
- `providers` (list): One record per NPI, in the order given. One that was not found has 'found' false and its other fields empty.

**Example output**

```json
{
  "count": 3,
  "found_count": 2,
  "not_found": [
    "1234567890"
  ],
  "providers": [
    {
      "authorized_official": {
        "credential": "NP",
        "name": "SANDRA MUSE",
        "phone": "(617) 643-0336",
        "title": "NURSE PRACTITIONER"
      },
      "check_digit_valid": true,
      "credential": null,
      "entity_type": "organization",
      "enumeration_date": "2007-05-01",
      "fax": null,
      "first_name": null,
      "found": true,
      "gender": null,
      "last_name": null,
      "last_updated": "2020-08-22",
      "mailing_address": {
        "city": "BOSTON",
        "country": "US",
        "full": "55 FRUIT ST, BOSTON, MA 02114",
        "line1": "55 FRUIT ST",
        "line2": null,
        "postal_code": "02114",
        "state": "MA"
      },
      "middle_name": null,
      "name": "MASSACHUSETTS GENERAL HOSPITAL",
      "npi": "1992926992",
      "organization_subpart": false,
      "other_identifiers": [],
      "other_name": null,
      "parent_organization": null,
      "phone": "(617) 643-0336",
      "practice_address": {
        "city": "BOSTON",
        "country": "US",
        "full": "55 FRUIT ST, BOSTON, MA 02114",
        "line1": "55 FRUIT ST",
        "line2": null,
        "postal_code": "02114",
        "state": "MA"
      },
      "provider_type": "Hospital-General",
      "replacement_npi": null,
      "requested": "1992926992",
      "sole_proprietor": null,
      "taxonomies": [
        {
          "classification": "General Acute Care Hospital",
          "code": "282N00000X",
          "grouping": "Hospitals",
          "license_number": "213663",
          "license_state": "MA",
          "medicare_types": [
            "Hospital-General",
            "Hospital-Acute Care",
            "Hospital-Short-Term (General and Specialty)"
          ],
          "primary": true,
          "specialization": null
        }
      ]
    },
    {
      "authorized_official": null,
      "check_digit_valid": true,
      "credential": "MD",
      "entity_type": "individual",
      "enumeration_date": "2005-12-30",
      "fax": "(434) 296-1036",
      "first_name": "ELIZABETH",
      "found": true,
      "gender": "female",
      "last_name": "WILLIAMS",
      "last_updated": "2026-07-13",
      "mailing_address": {
        "city": "BATESVILLE",
        "country": "US",
        "full": "PO BOX 215, BATESVILLE, VA 22924",
        "line1": "PO BOX 215",
        "line2": null,
        "postal_code": "22924",
        "state": "VA"
      },
      "middle_name": null,
      "name": "WILLIAMS, ELIZABETH",
      "npi": "1720065618",
      "organization_subpart": null,
      "other_identifiers": [
        {
          "identifier": "267257",
          "issuer": "MAMSI/ALLIANCE",
          "state": "VA",
          "type": "Other"
        },
        {
          "identifier": "333844",
          "issuer": "ANTHEM",
          "state": "VA",
          "type": "Other"
        }
      ],
      "other_name": null,
      "parent_organization": null,
      "phone": "(434) 296-9161",
      "practice_address": {
        "city": "CHARLOTTESVILLE",
        "country": "US",
        "full": "1011 E JEFFERSON ST, CHARLOTTESVILLE, VA 22902",
        "line1": "1011 E JEFFERSON ST",
        "line2": null,
        "postal_code": "22902",
        "state": "VA"
      },
      "provider_type": "Physician/Pediatric Medicine",
      "replacement_npi": null,
      "requested": "1720065618",
      "sole_proprietor": false,
      "taxonomies": [
        {
          "classification": "Pediatrics",
          "code": "208000000X",
          "grouping": "Allopathic & Osteopathic Physicians",
          "license_number": "0101041370",
          "license_state": "VA",
          "medicare_types": [
            "Physician/Pediatric Medicine"
          ],
          "primary": true,
          "specialization": null
        }
      ]
    },
    {
      "authorized_official": null,
      "check_digit_valid": false,
      "credential": null,
      "entity_type": null,
      "enumeration_date": null,
      "fax": null,
      "first_name": null,
      "found": false,
      "gender": null,
      "last_name": null,
      "last_updated": null,
      "mailing_address": null,
      "middle_name": null,
      "name": null,
      "npi": null,
      "organization_subpart": null,
      "other_identifiers": [],
      "other_name": null,
      "parent_organization": null,
      "phone": null,
      "practice_address": null,
      "provider_type": null,
      "replacement_npi": null,
      "requested": "1234567890",
      "sole_proprietor": null,
      "taxonomies": []
    }
  ]
}
```

### Search ClinVar variants

Find genetic variants in NCBI's ClinVar by HGVS name, dbSNP id, gene or condition words, optionally within one gene. Each comes back with its clinical significance (pathogenic, benign...), review status, gene, coding and protein changes, location, dbSNP id, the conditions it is reported with and a ClinVar link. No API key required.

**Inputs**

- Search for (`query`, text): An HGVS change, dbSNP id, gene or condition words, e.g. 'c.68_69del' or 'rs80357914'. Leave empty to list a gene's variants.
- Gene (`gene`, text): Only variants in the gene with this symbol, e.g. BRCA1.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `gene` (text): The gene the search was limited to, if any.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.
- `variants` (list): The matching ClinVar variants, best match first.

**Example output**

```json
{
  "count": 2,
  "gene": "BRCA1",
  "has_more": true,
  "offset": 0,
  "query": "c.68_69del",
  "total": 4,
  "variants": [
    {
      "amino_acid_change": "p.Glu23fs",
      "assembly": "GRCh37",
      "chromosome": "17",
      "clinical_significance": "Pathogenic",
      "conditions": [
        {
          "code": "C2676676",
          "name": "Breast-ovarian cancer, familial, susceptibility to, 1"
        },
        {
          "code": "C3280442",
          "name": "Pancreatic cancer, susceptibility to, 4"
        },
        {
          "code": "C0677776",
          "name": "Hereditary breast ovarian cancer syndrome"
        }
      ],
      "cytogenetic_location": "17q21.31",
      "dbsnp_id": "rs80357914",
      "gene_id": "672",
      "gene_symbol": "BRCA1",
      "hgvs_c": "NM_001407581.1:c.66_67AG[1]",
      "hgvs_p": "NP_001394510.1:p.Glu23Valfs",
      "name": "NM_007294.4(BRCA1):c.68_69del (p.Glu23fs)",
      "nucleotide_change": "c.68_69del",
      "review_status": "reviewed by expert panel",
      "start": 41276045,
      "stop": 41276046,
      "url": "https://www.ncbi.nlm.nih.gov/clinvar/variation/17662/",
      "variant_type": "Microsatellite",
      "variation_id": "17662"
    },
    {
      "amino_acid_change": "p.Cys24fs",
      "assembly": "GRCh37",
      "chromosome": "17",
      "clinical_significance": "Pathogenic",
      "conditions": [
        {
          "code": "C2676676",
          "name": "Breast-ovarian cancer, familial, susceptibility to, 1"
        }
      ],
      "cytogenetic_location": "17q21.31",
      "dbsnp_id": "rs886040312",
      "gene_id": "672",
      "gene_symbol": "BRCA1",
      "hgvs_c": "NM_001407625.1:c.69_70GT[1]",
      "hgvs_p": "NP_001394554.1:p.Cys24Serfs",
      "name": "NM_007294.4(BRCA1):c.71_72del (p.Cys24fs)",
      "nucleotide_change": "c.71_72del",
      "review_status": "reviewed by expert panel",
      "start": 41276042,
      "stop": 41276043,
      "url": "https://www.ncbi.nlm.nih.gov/clinvar/variation/266572/",
      "variant_type": "Microsatellite",
      "variation_id": "266572"
    }
  ]
}
```

### Search drug ingredients

Find the active ingredients prescribable drugs are made of, by name, each with its RxNorm id (RxCUI): use it to turn 'metformin' into RxCUI 6809. No API key required.

**Inputs**

- Search for (`query`, text, required): An ingredient's name, or the start of it.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `ingredients` (list): The matching ingredients, best match first.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 1,
  "has_more": false,
  "ingredients": [
    {
      "name": "metformin",
      "rxcui": "6809"
    }
  ],
  "offset": 0,
  "query": "metf",
  "total": 1
}
```

### Search drugs

Find prescribable drugs by brand or generic name in RxTerms, NLM's list for prescribing, such as 'metformin' or 'lipitor'. Each drug comes back with its route (oral pill, injectable...), its RxNorm id (RxCUI) and every strength and form it comes in, each with its own RxCUI. No API key required.

**Inputs**

- Search for (`query`, text, required): A drug's brand or generic name, or the start of it.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `drugs` (list): The matching drugs, best match first.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "drugs": [
    {
      "drug": "metFORMIN",
      "name": "metFORMIN (Oral Pill)",
      "route": "Oral Pill",
      "rxcui": "1161611",
      "strengths": [
        {
          "rxcui": "861007",
          "strength": "500 mg Tab"
        },
        {
          "rxcui": "861021",
          "strength": "625 mg Tab"
        },
        {
          "rxcui": "2703582",
          "strength": "750 mg Tab"
        }
      ],
      "synonyms": []
    },
    {
      "drug": "metFORMIN XR",
      "name": "metFORMIN XR (Oral Pill)",
      "route": "Oral Pill",
      "rxcui": "1161611",
      "strengths": [
        {
          "rxcui": "860975",
          "strength": "500 mg 24 HR XR Tab"
        },
        {
          "rxcui": "1807915",
          "strength": "500 mg Modified 24 HR XR Tab"
        },
        {
          "rxcui": "1807917",
          "strength": "500 mg Osmotic 24 HR XR Tab"
        }
      ],
      "synonyms": []
    }
  ],
  "has_more": true,
  "offset": 0,
  "query": "metformin",
  "total": 20
}
```

### Search genes

Find human genes in HGNC's approved nomenclature by symbol, name, alias or HGNC id, such as 'BRCA1' or 'breast cancer'. Each comes back with its approved symbol and name, chromosomal location, alias and previous symbols and names, RefSeq accession and a link. No API key required.

**Inputs**

- Search for (`query`, text, required): A gene symbol, name, alias or HGNC id, e.g. 'BRCA1'.
- Match on (`match`, text): What the words are matched against. Gene symbols only puts the gene with that symbol first; the default also searches names and aliases. One of: `anything` (Symbols, names and aliases), `symbol` (Gene symbols only). Default: `anything`.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `genes` (list): The matching genes, best match first.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `match` (text): What the words were matched against: 'anything' or 'symbol'.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "genes": [
    {
      "alias_names": [
        "BRCA1/BRCA2-containing complex, subunit 1",
        "protein phosphatase 1, regulatory subunit 53"
      ],
      "alias_symbols": [
        "RNF53",
        "BRCC1",
        "PPP1R53"
      ],
      "hgnc_id": "HGNC:1100",
      "location": "17q21.31",
      "name": "BRCA1 DNA repair associated",
      "previous_names": [
        "breast cancer 1, early onset",
        "breast cancer 1"
      ],
      "previous_symbols": [],
      "refseq_accession": "NM_007294",
      "symbol": "BRCA1",
      "url": "https://www.genenames.org/data/gene-symbol-report/#!/hgnc_id/HGNC:1100"
    },
    {
      "alias_names": [
        "BRCA1/BRCA2-containing complex, subunit 2"
      ],
      "alias_symbols": [
        "FAD",
        "FAD1",
        "BRCC2"
      ],
      "hgnc_id": "HGNC:1101",
      "location": "13q13.1",
      "name": "BRCA2 DNA repair associated",
      "previous_names": [
        "Fanconi anemia, complementation group D1",
        "breast cancer 2, early onset"
      ],
      "previous_symbols": [
        "FANCD1",
        "FACD",
        "FANCD"
      ],
      "refseq_accession": "NM_000059",
      "symbol": "BRCA2",
      "url": "https://www.genenames.org/data/gene-symbol-report/#!/hgnc_id/HGNC:1101"
    }
  ],
  "has_more": true,
  "match": "symbol",
  "offset": 0,
  "query": "BRCA",
  "total": 4
}
```

### Search HCPCS codes

Find HCPCS Level II codes, the codes Medicare and insurers use for supplies, equipment, drugs and services outside CPT (wheelchairs, ambulance rides, injections), by words or by code. Each comes back with its short and full descriptions, the day it was added and, for a discontinued code, the day it ended. No API key required.

**Inputs**

- Search for (`query`, text, required): Words or part of a code, e.g. 'wheelchair' or 'E01'.
- Include discontinued codes (`include_discontinued`, true/false): Also return codes CMS has discontinued, each with the day it ended. Default: `false`.
- Not otherwise classified codes (`noc_codes`, text): What to do with 'not otherwise classified' codes, the catch-all codes used when no specific code fits. One of: `include` (Include them), `exclude` (Leave them out), `only` (Only them). Default: `include`.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `codes` (list): The matching HCPCS codes, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "codes": [
    {
      "added_on": "2006-01-01",
      "code": "E2210",
      "description": "Wheelchair accessory, bearings, any type, replacement only, each",
      "discontinued": false,
      "effective_on": "2006-01-01",
      "not_otherwise_classified": false,
      "short_description": "Wheelchair bearings",
      "terminated_on": null
    },
    {
      "added_on": "1994-01-01",
      "code": "K0001",
      "description": "Standard wheelchair",
      "discontinued": false,
      "effective_on": "1994-01-01",
      "not_otherwise_classified": false,
      "short_description": "Standard wheelchair",
      "terminated_on": null
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "wheelchair",
  "total": 324
}
```

### Search ICD-10-CM codes

Find U.S. diagnosis codes (ICD-10-CM, the codes on claims and in medical records) by words in their description or by the start of a code, such as 'tuberculosis lung', 'E11' or a code without its dot like 'E119'. Each code comes back with its description, with and without the dot. No API key required.

**Inputs**

- Search for (`query`, text, required): Words or part of a code. Words match the start of words in the description, and every word must match.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `codes` (list): The matching ICD-10-CM codes, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "codes": [
    {
      "code": "A18.39",
      "code_without_dot": "A1839",
      "name": "Retroperitoneal tuberculosis"
    },
    {
      "code": "P37.0",
      "code_without_dot": "P370",
      "name": "Congenital tuberculosis"
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "tuberculosis",
  "total": 57
}
```

### Search ICD-11 codes

Find WHO's ICD-11 codes by words in their title or index terms, or by code, optionally within one chapter, or list a chapter's codes. Each comes back with its title, definition, chapter, other index terms and a link to WHO's ICD-11 browser. No API key required.

**Inputs**

- Search for (`query`, text): Words or part of a code, e.g. 'diabetes' or '5A1'. Leave empty to list a chapter's codes.
- Chapter (`chapter`, text): Only codes in this ICD-11 chapter. Chapter X holds the extension codes that add detail to other codes. One of: `01` (01: Certain infectious or parasitic diseases), `02` (02: Neoplasms), `03` (03: Diseases of the blood or blood-forming organs), `04` (04: Diseases of the immune system), `05` (05: Endocrine, nutritional or metabolic diseases), `06` (06: Mental, behavioural or neurodevelopmental disorders), `07` (07: Sleep-wake disorders), `08` (08: Diseases of the nervous system), `09` (09: Diseases of the visual system), `10` (10: Diseases of the ear or mastoid process), and 18 more.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `chapter` (text): The chapter the search was limited to, if any.
- `codes` (list): The matching ICD-11 codes, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "chapter": "05",
  "codes": [
    {
      "browser_url": "https://icd.who.int/browse/2026-01/mms/en#1697306310",
      "chapter": "05",
      "chapter_name": "Endocrine, nutritional or metabolic diseases",
      "code": "5A14",
      "code_type": "stem",
      "definition": null,
      "foundation_uri": "http://id.who.int/icd/entity/1697306310",
      "index_terms": [
        "Diabetes mellitus, type unspecified",
        "DM - [diabetes mellitus] NOS",
        "Diabetes mellitus, type unspecified without complications"
      ],
      "title": "Diabetes mellitus, type unspecified"
    },
    {
      "browser_url": "https://icd.who.int/browse/2026-01/mms/en#1156295616",
      "chapter": "05",
      "chapter_name": "Endocrine, nutritional or metabolic diseases",
      "code": "5A24",
      "code_type": "stem",
      "definition": "Brittle diabetes mellitus is a term used to describe particularly hard-to-control Type 1 or Type 2 diabetes mellitus. It results in frequent, extreme swings in blood glucose levels, causing hyperglycaemia that could lead to ketoacidosis or hypoglycaemia.",
      "foundation_uri": "http://id.who.int/icd/entity/1156295616",
      "index_terms": [
        "Uncontrolled or unstable diabetes mellitus",
        "brittle diabetes",
        "inadequate control of diabetes"
      ],
      "title": "Uncontrolled or unstable diabetes mellitus"
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "diabetes",
  "total": 31
}
```

### Search ICD-9-CM diagnosis codes

Find ICD-9-CM diagnosis codes, the U.S. codes used before ICD-10-CM (until October 2015), by words or by code with or without the dot. Use it to read older records and claims. Each code comes back with its short and full names. No API key required.

**Inputs**

- Search for (`query`, text, required): Words or part of a code, e.g. 'diabetes' or '250.00'.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `codes` (list): The matching ICD-9-CM diagnosis codes, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "codes": [
    {
      "code": "250.00",
      "code_without_dot": "25000",
      "name": "Diabetes mellitus without mention of complication, type II or unspecified type, not stated as uncontrolled",
      "short_name": "DMII wo cmp nt st uncntr"
    },
    {
      "code": "250.01",
      "code_without_dot": "25001",
      "name": "Diabetes mellitus without mention of complication, type I [juvenile type], not stated as uncontrolled",
      "short_name": "DMI wo cmp nt st uncntrl"
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "250.0",
  "total": 4
}
```

### Search ICD-9-CM procedure codes

Find ICD-9-CM procedure codes (volume 3, used for inpatient procedures until October 2015) by words or by code with or without the dot. Each code comes back with its short and full names. No API key required.

**Inputs**

- Search for (`query`, text, required): Words or part of a code, e.g. 'appendectomy' or '47.01'.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `codes` (list): The matching ICD-9-CM procedure codes, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "codes": [
    {
      "code": "47.01",
      "code_without_dot": "4701",
      "name": "Laparoscopic appendectomy",
      "short_name": "Lap appendectomy"
    },
    {
      "code": "47.09",
      "code_without_dot": "4709",
      "name": "Other appendectomy",
      "short_name": "Other appendectomy"
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "appendectomy",
  "total": 4
}
```

### Search individual providers

Find doctors, nurses, therapists and other individual health care providers in the U.S. NPI registry by name, NPI, specialty or address words, filtered by state, city, ZIP code, specialty and gender. Each comes back with their NPI, credential, specialty and primary taxonomy, practice address and phone. No API key required.

**Inputs**

- Search for (`query`, text): Words from the provider's name, NPI, specialty or practice address, e.g. 'john smith'. Leave empty to search by the filters alone.
- State (`state`, text): Only providers whose practice is in this state or territory. Leave empty for anywhere. One of: `AL` (Alabama (AL)), `AK` (Alaska (AK)), `AS` (American Samoa (AS)), `AZ` (Arizona (AZ)), `AR` (Arkansas (AR)), `AA` (Armed Forces Americas (AA)), `AE` (Armed Forces Europe (AE)), `AP` (Armed Forces Pacific (AP)), `CA` (California (CA)), `CO` (Colorado (CO)), and 49 more.
- City (`city`, text): Only providers whose practice is in this city, e.g. Boston.
- ZIP code (`postal_code`, text): Only providers whose practice is at this five-digit ZIP code, e.g. 02114. A ZIP+4 code is cut to its first five digits.
- Specialty (`specialty`, text): Only providers whose type or taxonomy (specialty) mentions these words, e.g. Cardiology or Pediatrics.
- Gender (`gender`, text): Only providers of this gender. Leave empty for any. One of: `female`, `male`.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `providers` (list): The matching individual providers, best match first.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "has_more": false,
  "offset": 0,
  "providers": [
    {
      "credential": "MD",
      "entity_type": "individual",
      "fax": "(434) 243-3300",
      "gender": "male",
      "name": "WILLIAMS, MICHAEL",
      "npi": "1215067236",
      "phone": "(434) 924-0123",
      "practice_address": {
        "city": "CHARLOTTESVILLE",
        "country": "US",
        "full": "1204 W MAIN ST, CHARLOTTESVILLE, VA 22903",
        "line1": "1204 W MAIN ST",
        "line2": null,
        "postal_code": "22903",
        "state": "VA"
      },
      "provider_type": "Physician/Pediatric Medicine",
      "taxonomy": {
        "classification": "Pediatrics",
        "code": "208000000X",
        "grouping": "Allopathic & Osteopathic Physicians",
        "specialization": null
      }
    },
    {
      "credential": "MD",
      "entity_type": "individual",
      "fax": "(434) 296-1036",
      "gender": "female",
      "name": "WILLIAMS, ELIZABETH",
      "npi": "1720065618",
      "phone": "(434) 296-9161",
      "practice_address": {
        "city": "CHARLOTTESVILLE",
        "country": "US",
        "full": "1011 E JEFFERSON ST, CHARLOTTESVILLE, VA 22902",
        "line1": "1011 E JEFFERSON ST",
        "line2": null,
        "postal_code": "22902",
        "state": "VA"
      },
      "provider_type": "Physician/Pediatric Medicine",
      "taxonomy": {
        "classification": "Pediatrics",
        "code": "208000000X",
        "grouping": "Allopathic & Osteopathic Physicians",
        "specialization": null
      }
    }
  ],
  "query": "williams",
  "total": 2
}
```

### Search LOINC

Find LOINC codes for lab tests, measurements and survey questions, or for whole forms and panels, by words or by LOINC number. Each comes back with its long common name, what it measures, its usual units, the kind of answer it takes, whether it has an answer list, and any third-party copyright. No API key required.

**Inputs**

- Search for (`query`, text, required): Words or a LOINC number, e.g. 'glucose serum' or '2345-7'.
- Search in (`item_type`, text): Which LOINC items to search: single questions and tests, whole forms and panels, forms with the sections inside them, or everything. One of: `question` (Questions, tests and measurements), `form` (Forms and panels), `form_and_section` (Forms, panels and their sections), `any` (Everything). Default: `question`.
- Leave out third-party copyrighted items (`exclude_copyrighted`, true/false): Leave out items another organization holds the copyright to, such as some published questionnaires. Every item is still under LOINC's own terms of use. Default: `false`.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `item_type` (text): The kind of items searched: 'question', 'form', 'form_and_section' or 'any'.
- `items` (list): The matching LOINC items, best match first.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "has_more": true,
  "item_type": "question",
  "items": [
    {
      "component": "Glucose.protein bound",
      "consumer_name": null,
      "data_type": "REAL",
      "external_copyright_notice": null,
      "has_answer_list": false,
      "has_external_copyright": false,
      "loinc_number": "6883-3",
      "long_common_name": "Glucose.protein bound [Mass/volume] in Serum or Plasma",
      "method": null,
      "name": "PB Glucose SerPl-mCnc",
      "property": "MCnc",
      "question_text": null,
      "short_name": "PB Glucose SerPl-mCnc",
      "units": [
        "mg/dL"
      ],
      "url": "https://loinc.org/6883-3"
    },
    {
      "component": "Glucose^10 AM specimen",
      "consumer_name": null,
      "data_type": "REAL",
      "external_copyright_notice": null,
      "has_answer_list": false,
      "has_external_copyright": false,
      "loinc_number": "16165-3",
      "long_common_name": "Glucose [Mass/volume] in Serum or Plasma --10 AM specimen",
      "method": null,
      "name": "Glucose 10 AM SerPl-mCnc",
      "property": "MCnc",
      "question_text": null,
      "short_name": "Glucose 10 AM SerPl-mCnc",
      "units": [
        "mg/dL"
      ],
      "url": "https://loinc.org/16165-3"
    }
  ],
  "offset": 0,
  "query": "glucose serum plasma mass",
  "total": 330
}
```

### Search medical conditions

Find medical conditions by everyday or clinical names and synonyms, such as 'heart attack' or 'asthma', from a list of about 2,400 kept for patient forms. Each comes back with a name for patients, its synonyms, suggested ICD-10-CM codes, its ICD-9-CM code and links to MedlinePlus pages about it. No API key required.

**Inputs**

- Search for (`query`, text, required): A condition's name or a synonym, e.g. 'heart attack'.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `conditions` (list): The matching conditions, best match first.
- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "conditions": [
    {
      "consumer_name": "Diabetes mellitus (DM)",
      "icd10cm_codes": [
        {
          "code": "E11.9",
          "name": "Type 2 diabetes mellitus without complications"
        }
      ],
      "icd9_code": "250.00",
      "icd9_name": "type II diabetes mellitus [non-insulin dependent type] [NIDDM type] [adult-onset type] or unspecified type, not stated as uncontrolled, without mention of complication",
      "id": "2143",
      "info_links": [
        {
          "title": "Diabetes",
          "url": "http://www.nlm.nih.gov/medlineplus/diabetes.html"
        }
      ],
      "name": "Diabetes mellitus",
      "synonyms": [
        "DM"
      ]
    },
    {
      "consumer_name": "Diabetes - type 2 (adult, non-insulin-independent)",
      "icd10cm_codes": [
        {
          "code": "E11.9",
          "name": "Type 2 diabetes mellitus without complications"
        }
      ],
      "icd9_code": "250.00",
      "icd9_name": "type II diabetes mellitus [non-insulin dependent type] [NIDDM type] [adult-onset type] or unspecified type, not stated as uncontrolled, without mention of complication",
      "id": "10180",
      "info_links": [
        {
          "title": "Diabetes",
          "url": "http://www.nlm.nih.gov/medlineplus/diabetes.html"
        }
      ],
      "name": "Diabetes mellitus - non insulin dependent",
      "synonyms": [
        "NIDDM",
        "diabetes mellitus type II"
      ]
    }
  ],
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "diabetes",
  "total": 21
}
```

### Search phenotypes

Find terms in the Human Phenotype Ontology (HPO), the vocabulary genetics and rare disease work uses for signs and symptoms, by name, synonym or HPO id such as HP:0001250. Each comes back with its definition, synonyms (including words for patients), the more general terms it sits under and a link. No API key required.

**Inputs**

- Search for (`query`, text, required): A sign or symptom, a synonym, or an HPO id such as HP:0001250.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `phenotypes` (list): The matching HPO terms, best match first.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "has_more": true,
  "offset": 0,
  "phenotypes": [
    {
      "alternative_ids": [
        "HP:0001275",
        "HP:0001303",
        "HP:0002125"
      ],
      "comment": "A type of electrographic seizure has been proposed in neonates which does not have a clinical correlate, it is electrographic only. The term epilepsy is not used to describe recurrent febrile seizures. Epilepsy presumably reflects an abnormally reduced seizure threshold.",
      "definition": "A seizure is an intermittent abnormality of nervous system physiology characterized by a transient occurrence of signs and/or symptoms due to abnormal excessive or synchronous neuronal activity in the brain",
      "id": "HP:0001250",
      "name": "Seizure",
      "parents": [
        {
          "id": "HP:0012638",
          "name": "Abnormal nervous system physiology"
        }
      ],
      "synonyms": [
        {
          "relation": "RELATED",
          "term": "Epilepsy",
          "type": "layperson"
        },
        {
          "relation": "EXACT",
          "term": "Epileptic seizure",
          "type": null
        },
        {
          "relation": "EXACT",
          "term": "Seizures",
          "type": "plural_form"
        }
      ],
      "url": "https://hpo.jax.org/browse/term/HP:0001250"
    },
    {
      "alternative_ids": [
        "HP:0002124"
      ],
      "comment": "This term describes the initial semiology of the seizure without specifying whether the onset is focal or generalized. Thus it can be used for coding atonic seizures when the onset is not known.",
      "definition": "Atonic seizure is a type of motor seizure characterized by a sudden loss or diminution of muscle tone without apparent preceding myoclonic or tonic event lasting about 1 to 2 seconds, involving head, trunk, jaw, or limb musculature",
      "id": "HP:0010819",
      "name": "Atonic seizure",
      "parents": [
        {
          "id": "HP:0020219",
          "name": "Motor seizure"
        }
      ],
      "synonyms": [
        {
          "relation": "EXACT",
          "term": "Astatic seizure",
          "type": null
        },
        {
          "relation": "EXACT",
          "term": "Astatic seizures",
          "type": "plural_form"
        },
        {
          "relation": "EXACT",
          "term": "Atonic seizures",
          "type": "plural_form"
        }
      ],
      "url": "https://hpo.jax.org/browse/term/HP:0010819"
    }
  ],
  "query": "seizure",
  "total": 318
}
```

### Search provider organizations

Find hospitals, clinics, pharmacies, labs and other health care organizations in the U.S. NPI registry by name, NPI, type or address words, filtered by state, city, ZIP code and type. Each comes back with its NPI, type and primary taxonomy, practice address and phone. No API key required.

**Inputs**

- Search for (`query`, text): Words from the organization's name, NPI, type or practice address, e.g. 'massachusetts general hospital'. Leave empty to search by the filters alone.
- State (`state`, text): Only providers whose practice is in this state or territory. Leave empty for anywhere. One of: `AL` (Alabama (AL)), `AK` (Alaska (AK)), `AS` (American Samoa (AS)), `AZ` (Arizona (AZ)), `AR` (Arkansas (AR)), `AA` (Armed Forces Americas (AA)), `AE` (Armed Forces Europe (AE)), `AP` (Armed Forces Pacific (AP)), `CA` (California (CA)), `CO` (Colorado (CO)), and 49 more.
- City (`city`, text): Only providers whose practice is in this city, e.g. Boston.
- ZIP code (`postal_code`, text): Only providers whose practice is at this five-digit ZIP code, e.g. 02114. A ZIP+4 code is cut to its first five digits.
- Specialty (`specialty`, text): Only providers whose type or taxonomy (specialty) mentions these words, e.g. Pharmacy or General Acute Care Hospital.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `providers` (list): The matching organizations, best match first.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "has_more": true,
  "offset": 0,
  "providers": [
    {
      "credential": null,
      "entity_type": "organization",
      "fax": null,
      "gender": null,
      "name": "MASSACHUSETTS GENERAL HOSPITAL",
      "npi": "1225232366",
      "phone": "(617) 643-2013",
      "practice_address": {
        "city": "BOSTON",
        "country": "US",
        "full": "55 FRUIT ST YAWKEY BUILDING SUITE 7B, BOSTON, MA 02114",
        "line1": "55 FRUIT ST",
        "line2": "YAWKEY BUILDING SUITE 7B",
        "postal_code": "02114",
        "state": "MA"
      },
      "provider_type": "Hospital-General",
      "taxonomy": {
        "classification": "General Acute Care Hospital",
        "code": "282N00000X",
        "grouping": "Hospitals",
        "specialization": null
      }
    },
    {
      "credential": null,
      "entity_type": "organization",
      "fax": "(617) 726-3971",
      "gender": null,
      "name": "MASSACHUSETTS GENERAL HOSPITAL",
      "npi": "1184613721",
      "phone": "(617) 724-1935",
      "practice_address": {
        "city": "BOSTON",
        "country": "US",
        "full": "55 FRUIT ST CARDIOLOGY, GRB 800, MS 843, BOSTON, MA 02114",
        "line1": "55 FRUIT ST",
        "line2": "CARDIOLOGY, GRB 800, MS 843",
        "postal_code": "02114",
        "state": "MA"
      },
      "provider_type": "Hospital-General",
      "taxonomy": {
        "classification": "General Acute Care Hospital",
        "code": "282N00000X",
        "grouping": "Hospitals",
        "specialization": null
      }
    }
  ],
  "query": "massachusetts general hospital",
  "total": 45
}
```

### Search surgeries and implants

Find major surgeries and implants by everyday or clinical names, such as 'knee replacement' or 'pacemaker', from a list of about 280 kept for patient histories. Each comes back with a name for patients, its synonyms, its ICD-9-CM code when it has one and links to MedlinePlus pages about it. No API key required.

**Inputs**

- Search for (`query`, text, required): A surgery's or implant's name or a synonym, e.g. 'knee'.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `procedures` (list): The matching procedures, best match first.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.

**Example output**

```json
{
  "count": 2,
  "has_more": true,
  "offset": 0,
  "procedures": [
    {
      "consumer_name": "Knee replacement",
      "icd9_code": null,
      "icd9_name": null,
      "id": "X144",
      "info_links": [
        {
          "title": "Knee Replacement",
          "url": "http://www.nlm.nih.gov/medlineplus/kneereplacement.html"
        }
      ],
      "name": "Knee replacement",
      "synonyms": []
    },
    {
      "consumer_name": "Below knee amputation (BKA)",
      "icd9_code": "897.0",
      "icd9_name": "Traumatic amputation of leg(s) (complete) (partial), unilateral, below knee, without mention of complication",
      "id": "11202",
      "info_links": [
        {
          "title": "Leg or foot amputation",
          "url": "http://www.nlm.nih.gov/medlineplus/ency/article/007365.htm"
        }
      ],
      "name": "Below knee amputation",
      "synonyms": [
        "BK amputation",
        "BKA"
      ]
    }
  ],
  "query": "knee",
  "total": 4
}
```

### Search units of measure

Find UCUM units of measure, the codes lab results and LOINC use for units, by name, synonym or code, such as 'milligram per deciliter' or 'mg/dL'. Each comes back with its exact code, name, category, synonyms and the LOINC property it measures. Use it to turn the unit a person typed into the code a lab system expects. No API key required.

**Inputs**

- Search for (`query`, text): A unit's name, a synonym or its code, e.g. 'mg/dL' or 'inch'. Leave empty to list a category's units.
- Category (`category`, text): Only units in this category. Clinical units are the ones used in health care. One of: `clinical`, `nonclinical`, `constant` (Constants), `obsolete`.
- Simple units only (`simple_only`, true/false): Only units with no operators, such as 'g' rather than 'mg/dL'. Default: `false`.
- How many to return (`limit`, number): The most records to return, from 1 to 500. The best matches come first. From 1 to 500. Default: `20`.
- Skip (`offset`, number): How many matches to skip before the first one returned, to read the next page. Skip and How many to return together can reach at most 7,500 matches into a search. From 0 to 7499. Default: `0`.

**Outputs**

- `count` (number): How many records this step returned.
- `has_more` (true/false): Whether more matches can be read, by raising the limit or the offset.
- `offset` (number): How many matches were skipped before the first one returned.
- `query` (text): The words or code searched for, as sent to NLM (accented letters are sent without their accents). Empty when only filters were used.
- `total` (number): How many records match, as NLM counts them. NLM caps the count (usually at 10,000), and for a broad search it can be below the true number.
- `units` (list): The matching units, best match first.

**Example output**

```json
{
  "count": 2,
  "has_more": true,
  "offset": 0,
  "query": "milligram per",
  "total": 36,
  "units": [
    {
      "category": "Clinical",
      "code": "g/mg",
      "guidance": null,
      "is_simple": false,
      "loinc_property": "MCnt; MRto",
      "name": "gram per milligram",
      "source": "LOINC",
      "synonyms": [
        "g per mg",
        "grams per milligram"
      ]
    },
    {
      "category": "Clinical",
      "code": "nmol/mg",
      "guidance": null,
      "is_simple": false,
      "loinc_property": "SCnt",
      "name": "nanomole per milligram",
      "source": "LOINC",
      "synonyms": [
        "nmol per mg",
        "nanomoles per milligram"
      ]
    }
  ]
}
```
