# CDC WONDER MCP > CDC WONDER (Epidemiologic Data) connects your AI agent directly to the massive public health databases of the CDC WONDER system. You can query raw data on mortality rates, birth statistics, and vaccine adverse events using natural language prompts. It bypasses complex web forms, giving you structured access to critical epidemiological research data. ## Overview - **Category:** data-analytics - **Price:** Free - **Tags:** epidemiology, public-health, cdc-wonder, mortality-data, medical-research ## Description This MCP lets you treat huge public health datasets like a simple conversation. Instead of spending hours navigating the CDC WONDER website and filling out confusing forms—the kind that make you question your career choice—you just talk to your agent. It handles the complex query structure for you. You can ask it for mortality rates in specific years, look at birth statistics across different states, or pull raw vaccine safety data from VAERS. The system pulls structured results directly into your workflow, no CSV downloads required. If you're working with public health policy or academic research, this is huge. When you connect to the Vinkius catalog, you get access to tools like this one and thousands of others, keeping all your specialized services in one place. ## Tools ### query_wonder_database Runs complex queries against CDC WONDER databases using specific ID and parameter inputs (B_, M_, V_, etc.). ## Prompt Examples **Prompt:** ``` Query the CDC WONDER D76 database for mortality rates in 2021 with parameters for age groups. ``` **Response:** ``` I've initiated a request to the D76 Detailed Mortality database. The query is processing with your specified age group filters and 2021 timeframe. I will present the resulting epidemiologic table once the CDC system responds. ``` **Prompt:** ``` Fetch VAERS reports related to vaccine code 'COVID19' from the CDC WONDER database. ``` **Response:** ``` Accessing the VAERS database... I am executing `query_wonder_database` with the 'VAERS' ID and filtering for the 'COVID19' vaccine code. I'll summarize the reported adverse events and counts for you. ``` **Prompt:** ``` Analyze birth statistics for the state of California using database D10. ``` **Response:** ``` Querying the D10 Natality database for California. I'm applying the necessary location filters (F_ prefixes) to extract birth rates and maternal demographics. One moment while I retrieve the data. ``` ## Capabilities ### Query Mortality Statistics Retrieve detailed records on causes of death (D76) for specific regions and timeframes. ### Analyze Birth Rates Generate reports on natality data, allowing you to track maternal health metrics by location. ### Check Vaccine Safety Reports Extract raw adverse event data from the VAERS database for safety monitoring research. ### Run Advanced Queries Execute complex, ad-hoc queries using standard CDC parameters to refine data extraction precision. ## Use Cases ### Analyzing Outbreak Patterns A public health official needs to see how mortality rates changed in three specific metro areas during the last five years. Instead of manually running separate reports for each region and year, they prompt their agent: 'Compare D76 data for these three cities over this period.' The MCP executes multiple queries and returns a consolidated dataset. ### Academic Research on Birth Trends An epidemiologist is writing a paper on maternal health. They use the tool to pull D10 natality data for California, cross-referencing specific location prefixes (F_) and time periods to build a precise statistical model. ### Monitoring Vaccine Safety A researcher needs an immediate snapshot of adverse event reports. They instruct their agent: 'Fetch VAERS data for the last quarter, filtering by vaccine code X.' The MCP handles the specific database query and delivers summarized counts. ### Comparative State Studies A policy analyst needs to compare how different states handled a public health crisis. They ask the agent to run a comparative analysis across several states, using structured queries for consistency, saving hours of cross-platform data compilation. ## Benefits - Stop navigating complex web interfaces. Instead of clicking through dozens of filters across multiple tabs, your agent builds and runs the perfect query in a single step. - Get structured JSON output instead of PDFs or raw CSVs. Data Scientists can pull clean, machine-readable records for immediate integration into analytical tools. - Cover multiple data types (mortality, births, vaccine safety) from one connection point. You don't need three different databases; you just need the CDC WONDER MCP. - Speed up research cycles. Quickly compare mortality rates between distinct timeframes and regions without needing manual data collection or clean-up. - Access raw source material. The tool lets you pull foundational, raw epidemiologic records that are necessary for deep academic analysis. ## How It Works The bottom line is, you talk naturally, and the system handles all the complicated database logic. 1. Subscribe to this MCP and provide your necessary API access token or accept the required Data Use Agreement. 2. Send a natural language request through your AI client, specifying the type of data you need (e.g., 'Mortality rates for 2021 in Texas'). 3. The MCP executes the complex query against the CDC WONDER databases and returns the structured epidemiologic data directly to your agent. ## Frequently Asked Questions **How do I use the query_wonder_database tool with multiple parameters?** You provide all necessary filters (like location, year, and age groups) in a single prompt. The agent structures this into the required JSON object for the underlying database call. **Does CDC WONDER MCP handle data from different sources like D10 and VAERS?** Yes. You can query multiple distinct datasets—mortality, births, vaccine safety reports—using the same connection point via your agent's prompt. **Is this better than just searching Google for CDC data?** Absolutely. Google gives you links and summaries; this MCP executes precise database calls to retrieve structured records, giving you usable data, not just articles. **What prefixes are required when calling query_wonder_database?** The tool requires specific parameters like B_, M_, V_, F_, or O_ prefixes. Your agent handles the correct JSON structure for these inputs based on your request. **How do I handle authentication when calling query_wonder_database?** You must provide a valid CDC WONDER API Access Token or ensure the required Data Use Agreement is accepted first. Your AI client will prompt you for these credentials if they aren't already configured in your MCP environment. **What happens if query_wonder_database encounters rate limits?** If a query fails due to excessive usage, the system recommends implementing an exponential backoff strategy. You should adjust your agent workflow to pause and retry the request after increasing time intervals. **What format does query_wonder_database return its results in?** The tool returns structured JSON data, making it ready for immediate ingestion into analytical pipelines. This clean output allows your agent to process and interpret the epidemiologic records without manual parsing. **Can I use query_wonder_database for real-time health event monitoring?** No, this MCP accesses historical and published CDC WONDER datasets. It is designed for deep academic research and trend analysis, not continuous, live stream reporting of current health events. **Which databases can I access using this server?** You can query any database supported by the CDC WONDER API by providing its ID to the `query_wonder_database` tool. Common IDs include D76 (Detailed Mortality), D10 (Natality), and VAERS (Vaccine Adverse Event Reporting System). **How should I format the parameters for a query?** Parameters should be provided as a JSON object using the standard CDC prefixes: B_ for by-variables, M_ for measures, V_ for values, F_ for filters, and O_ for other options. The `query_wonder_database` tool handles the XML conversion for you. **Do I need to include the 'accept_datause_restrictions' parameter?** No. The `query_wonder_database` tool is designed to handle the data use restrictions agreement internally. You only need to provide the specific database ID and the analytical parameters for your search.