ICD-10-CM

Last updated: 05-21-2025

Data Dictionary

Column	Data Type	Description

Download CSV

What is ICD-10-CM?

ICD-10-CM stands for International Classification of Diseases, 10th Revision, Clinical Modification. It is the U.S. clinical adaptation of the ICD-10 system developed by the World Health Organization (WHO).

• Maintained by: National Center for Health Statistics (NCHS), part of the CDC.
• Purpose: Enables the standardized coding of diagnoses across all healthcare settings in the U.S.
• Usage: Medical billing, claims processing, clinical research, quality reporting, risk adjustment, and public health tracking.

Who Maintains ICD-10-CM?

• The National Center for Health Statistics (NCHS), under the Centers for Disease Control and Prevention (CDC), is responsible for the development and maintenance of ICD-10-CM.
• Updates are published annually, typically effective October 1 of each year.
• Coordination occurs through the ICD-10 Coordination and Maintenance Committee, jointly managed by:
  • NCHS (diagnoses)
  • CMS (Centers for Medicare & Medicaid Services – procedures via ICD-10-PCS)

📎[ICD-10-CM Updates and Information] (https://www.cms.gov/medicare/coding-billing/icd-10-codes)

📎 ICD-10-CM Official Guidelines for Coding and Reporting

Code Structure

Each ICD-10-CM code:

• Is alphanumeric, typically 3 to 7 characters long.
• The first character is always a letter.
• The next 2 characters are numbers.
• The 4th–7th characters (if present) provide greater specificity, such as:
  • Laterality (e.g., left vs. right)
  • Episode of care (e.g., initial vs. subsequent encounter)
  • Severity, etiology, or anatomical site

Example:

S52.521A
→ Displaced fracture of shaft of right radius, initial encounter for closed fracture

Diagnosis Headers and Code Categories

ICD-10-CM codes are structured hierarchically, organizing thousands of detailed diagnostic codes under broader diagnosis headers. These headers serve as category groupings that help users navigate the code system and group related diagnoses.

What is a Diagnosis Header?

A diagnosis header is a non-billable, non-specific entry in the ICD-10-CM tabular list. It typically corresponds to a three-character category, such as E11 for Type 2 diabetes mellitus, under which billable subcodes like E11.9 (Type 2 diabetes mellitus without complications) reside.

Diagnosis headers:

• Represent categories of related conditions
• Provide structure to the ICD-10-CM hierarchy
• Are not valid for billing or reporting purposes on their own
• Help organize the code set for analytic, clinical, and navigational use

Code Categories

Each ICD-10-CM code falls into one of two high-level categories:

• Billable codes
  These are complete, valid diagnosis codes that may be used in clinical documentation and claims (e.g., E11.9, I10).

• Non-billable headers
  These are category headers used to group related billable codes (e.g., E11, C50, I50). They cannot be used alone for billing or reporting.

The header_flag Field

To distinguish billable diagnosis codes from non-billable headers, CMS’s official ICD-10-CM order files include a column Tuva has named "header_flag" indicating whether that code is a header or not:

• 1 = Billable code
  A specific, valid code that may be submitted on claims and used in clinical systems

• 0 = Non-billable header
  A grouping code for organizational purposes only

This field is crucial for applications that need to filter for valid billing codes or build clinical user interfaces.

Why Use Headers?

Although headers are not directly used in claims, they serve multiple purposes:

• Navigation: Help users locate and drill into detailed codes
• Grouping: Facilitate roll-up reporting and cohorting (e.g., for diabetes or cancer types)
• Standardization: Provide a predictable structure to ICD-10-CM across clinical tools and systems
• Data modeling: Improve interoperability across EHRs, analytics platforms, and decision support tools

Key Use Cases for ICD-10-CM Codes

• Billing & Reimbursement: Required on all claims submitted to payers.
• Clinical Documentation Improvement (CDI): Ensures precise representation of patient conditions.
• Analytics & Reporting: Tracks trends, utilization, outcomes, risk stratification.
• Quality Measures & Risk Adjustment: Used in programs like HEDIS, Medicare Advantage, and ACOs.

📌 Notes for Data Analysts

• ICD-10-CM codes are often joined to hierarchical groupers like CCSR or HCC for summarization.
• Strings may need normalization (e.g., removing periods or standardizing case). Tuva's source for this data does not currently contain periods, but some sources may have them and they are typically removed.

Tuva Seed File Update Process

Note: This is the maintenance process used by Tuva to maintain the current codeset in the Tuva package. Tuva users do not need to complete this step unless they are leveraging a different version of codes or are no longer updating to the current version of the project, but need an updated terminology set.

1. Navigate to the CMS ICD 10 website
2. Go to the ICD-10 Files section, click the section for ICD-10-CM & PCS files of the current fiscal year (e.g. 2025 ICD-10 CM & PCS)
3. Within the ICD-10-CM Files section, select the hyperlink titled "Code Descriptions in Tabular Order" to download the associated ZIP file
4. Unzip the downloaded file and open "icd10cm_order_{year}"
5. Save the "icd10cm_order_{year}" as a text file
6. Load the text file into the below python script and run:

Since the text file contains fixed-length fields, the following Python script can be used to convert it into a CSV file.

import pandas as pd

#This function converts the text file into a dataframe with single field 'Text'
def text_to_dataframe(input_file):
    with open(input_file, 'r', encoding='utf-8') as file:
        lines = file.readlines()
    
    df = pd.DataFrame({'Text': lines})
    df['Text'] = df['Text'].str.strip('\n')
    
    return df


def main():
    input_filename = "your text file path"
    output_filename = "output csv file path"   
    
    df = text_to_dataframe(input_filename)
    print(df.head())

    #Slices the single column dataframe into dataframe with required columns
    df['SN'] = df['Text'].str.slice(0,6)
    df['icd_10_cm'] = df['Text'].str.slice(6,14)
    df['header_flag'] = df['Text'].str.slice(14,16)
    df['short_description'] = df['Text'].str.slice(16, 77)
    df['long_description'] = df['Text'].str.slice(77, )

    df2 = df[['icd_10_cm', 'header_flag', 'short_description',    'long_description']]
    # Remove leading/trailing whitespace from all string columns
    df2 = df2.apply(lambda x: x.str.strip() if x.dtype == "object" else x)
    
    print(df2.head())
    #save the final dataframe as a csv file
    df2.to_csv(output_filename, index=False)

if __name__ == "__main__":
    main()

Note: You might need to adjust the slicing indexes according to the length of your data field

7. Import the CSV file into any data warehouse and upload the CSV file from the data warehouse to S3 (credentials with write permissions to the S3 bucket are required)

-- example code for Snowflake
copy into s3://tuva-public-resources/terminology/icd_10_cm.csv
from [table_created_in_step_7]
file_format = (type = csv field_optionally_enclosed_by = '"')
storage_integration = [integration_with_s3_write_permissions]
OVERWRITE = TRUE;

8. Create a branch in docs. Update the last_updated column in the table above with the current date
9. Submit a pull request

The below steps are only required if the headers of the file need to be changed. The Tuva Project does not store the contents of the ICD-10-CM file in GitHub.

1. Create a branch in The Tuva Project
2. Alter the headers as needed in ICD-10-CM file
3. Submit a pull request