Access data via the Application Programming Interface
The National Center for Health Statistics (NCHS) provides various ways to access their data, including some through Application Programming Interfaces (APIs). Here's a breakdown of how you can generally access NCHS data programmatically:
1. CDC Open Technology APIs:
| • | The CDC, which NCHS is a part of, offers a range of APIs through its Open Technology initiative. While not exclusively NCHS, some of these may provide access to relevant health data. You can explore them here: https://open.cdc.gov/apis.html |
| • | WONDER: CDC WONDER is a valuable resource. It allows you to access data in its online databases with automated data queries in XML format over HTTP. This can be used in your own web pages or widgets. |
| o | Endpoint: https://wonder.cdc.gov/controller/datarequest/[database ID] |
2. National Vital Statistics System (NVSS) API:
| • | NCHS is actively modernizing its national collection and exchange of mortality data. They have developed an API for vital records jurisdictions to submit mortality data and receive responses (acknowledgments, errors, coded data). |
| • | This API leverages modern health standards like HL7's Fast Healthcare Interoperability Resources (FHIR). |
| • | While primarily for data exchange with jurisdictions, there are public-facing components or reference implementations that might be useful for understanding the structure and accessing certain public datasets derived from NVSS. |
3. NCHS Mortality Data (via Delphi Epidata API):
| • | Carnegie Mellon University's Delphi Epidata API provides access to NCHS Mortality Data, specifically national provisional death counts. This is a public source. |
| • | Source Name: nchs-mortality |
| • | This is particularly useful for researchers and forecasters interested in provisional death count data, which is updated frequently. |
4. Public-Use Data Files and R Packages:
| • | NCHS makes many of its public-use datasets (PUFs) available directly on their survey and data system websites. These do not contain sensitive information and do not require special applications or Research Data Center (RDC) access. |
| • | You can find a list of current and historic NCHS data collections with public-use datasets here:https://www.cdc.gov/rdc/public-nchs-data/index.html |
| • | nchsdata R package: This R package (CDCgov/nchsdata on GitHub) provides selected public-use files from NCHS, making it easier to load and analyze this data in R. |
5. NCHS Data Query System (DQS):
| • | While not strictly an API in the traditional sense for programmatic access with code, the NCHS Data Query System (DQS) is an online tool that allows users to dynamically generate statistical tables, charts, and graphs from NCHS public-use data. It's a user-friendly interface for querying data without needing programming skills. |
6. Restricted Data Access (Research Data Center - RDC):
| • | For access to more detailed or sensitive NCHS data that cannot be publicly released due to privacy concerns, researchers must apply to use the NCHS Research Data Center (RDC). This involves submitting a research proposal and adhering to strict confidentiality agreements. This is not API-based access but rather secure, on-site access to restricted files. |
Key Considerations for Using NCHS Data:
| • | Data Use Agreements: Always review the specific data use agreements and licenses for any NCHS data you access. Some datasets have restrictions on how they can be used, particularly regarding re-identification of individuals. |
| • | Data Structure and Documentation: Familiarize yourself with the data dictionaries and documentation provided for each dataset to understand the variables, coding, and any limitations. |
| • | Data Updates and Lag: Be aware that some NCHS data, especially provisional data like mortality counts, may have a lag in reporting and are subject to revisions as more complete data becomes available. |
To determine the best approach for your specific needs, consider:
| • | What specific NCHS data are you trying to access? (e.g., mortality, survey data, specific health topics) |
| • | What is the volume and frequency of data you need? |
| • | What are your technical capabilities? (e.g., comfortable with API calls and programming, or prefer a GUI) |
| • | Do you require access to restricted or public-use data? |
