CDC WONDER (Wide-ranging Online Data for Epidemiologic Research) is a free, public database maintained by the US Centers for Disease Control and Prevention — and one of the most accessible ways for a doctor anywhere in the world to produce an original, publishable epidemiological study without needing institutional data access or a research grant.
What CDC WONDER Actually Contains
WONDER isn't one dataset — it's a portal to several, each covering a different slice of US public health data, all free to query online without an application process.
| Dataset | What It Covers | Typical Use |
|---|---|---|
| Multiple Cause of Death | US death certificate data by cause, demographics, and location | Mortality trend studies, cause-specific death rate analysis |
| Natality | US birth certificate data | Birth outcome and maternal health studies |
| Cancer Statistics | Cancer incidence by type, demographics, and region | Cancer epidemiology and disparity research |
| Provisional Mortality Statistics | More recent, provisional death data | Near-real-time trend monitoring |
Step 1: Define Your Question Before You Touch the Query Tool
Just like a meta-analysis, a good WONDER-based study starts with a specific, answerable question — a particular cause of death, a specific age group, a specific time range, a specific comparison (by state, by sex, by year-over-year trend). Opening the query tool without a defined question first is the fastest way to end up with an unfocused dataset and no clear paper to write from it.
Step 2: Running a Query, Step by Step
- Go to the CDC WONDER homepage and select the dataset relevant to your question (for example, "Multiple Cause of Death").
- Read the dataset's documentation page first — each dataset has its own rules about suppressed small counts and coding conventions that will directly affect your results.
- Use the "Request Form" to set your grouping variables (for example: year, state, age group) and your measures (crude rate, age-adjusted rate, deaths).
- Apply ICD-10 code filters for your specific cause of interest — WONDER lets you filter by ICD-10 code ranges, not just broad category names.
- Export the results as a tab-delimited file for analysis in R, Excel, or your statistics package of choice.
Mentor tip
ICD-10 coding nuances matter more than they first appear. A cause of death you assume maps to one ICD-10 range sometimes spans several, or overlaps with a related code your search would otherwise miss — always check the actual code list in WONDER's documentation rather than relying on a remembered code range.
Step 3: Understand What the Data Can and Can't Tell You
WONDER data is aggregated and ecological — it describes populations, not individuals. This matters for how you can validly interpret your results.
Watch out
Avoid the ecological fallacy: a trend visible at the population level (for example, a state with a higher rate of a given cause of death) cannot be used to make claims about individual-level causation. Reviewers will flag this immediately if your discussion section overreaches beyond what aggregated data can actually support.
Small counts are also suppressed in WONDER's public output for privacy reasons, which affects analyses of rare outcomes or small subgroups — check each dataset's documentation for its specific suppression threshold before designing a study around a narrow subgroup.
- Specific, answerable research question defined before opening the query tool
- Correct dataset selected and its documentation actually read
- ICD-10 codes (or equivalent) verified against the dataset's own code list
- Suppression rules for small counts checked for your specific subgroup
- Results exported and cross-checked against a second query for consistency
- Ecological nature of the data reflected accurately in your discussion section
Step 4: From Query Results to Publication
Once you have clean exported data, the analysis itself is usually straightforward descriptive or trend statistics — rate comparisons over time, across regions, or across demographic groups — well within reach using R, Stata, or even well-organized Excel. Structure the write-up as a standard observational study: background, methods (including the exact WONDER dataset, query parameters, and date accessed — reviewers will ask for this level of detail since WONDER data can be updated), results, and a discussion that stays within the bounds of what ecological, aggregated data can support.
Because it's public, free, and well-documented, CDC WONDER remains one of the most accessible entry points into original epidemiological research for a doctor without institutional data access — the main skill it actually demands is a well-defined question and careful attention to the dataset's own documentation, not advanced statistics.
Written by
Dr. Nadir Akhtar
Ready to turn a CDC WONDER query into a publication?
The CDC Specialist Card is a focused pathway for original research using the CDC WONDER database, start to submission.