Written and maintained by CASRAI Editorial Board
Last updated
CDC WONDER (Wide-ranging Online Data for Epidemiologic Research) is a free web query system run by the Centers for Disease Control and Prevention (CDC). Its Underlying Cause of Death database lets researchers tabulate U.S. death certificate data by cause, year, place and demographic group. It is a routine data source for public health surveillance.
Each death certificate yields exactly one underlying cause, defined by CDC as the disease or injury that initiated the train of events leading directly to death. Analysts often pair underlying-cause counts with baseline comparisons such as excess deaths, report observational studies with the STROBE checklist, and place the work within epidemiology vs. public health or the CDC program evaluation framework.
What It Is Used for in Research
Researchers use CDC WONDER mortality data to count deaths and compute rates by cause, geography, year and demographic group. Two concepts matter most:
- Underlying cause of death (UCD): the single condition that started the chain of events leading to death. Because every record has exactly one, the counts for different causes do not overlap.
- Multiple cause of death (MCOD): a record can carry up to twenty additional multiple causes alongside the underlying cause, so MCOD queries can count one death under several conditions. Choose it when the question is about conditions that appear together on certificates.
Install or Access (Official Source Only)
CDC WONDER needs no installation. It is a free web tool at https://wonder.cdc.gov/. The Deaths section of the home page lists the Underlying Cause of Death and Multiple Cause of Death databases, among others. Before using the data you must accept the data use restrictions published on the database page; CDC’s documentation includes a rule that death counts of 9 or fewer must not be published. For general thinking about data use terms, see our data use agreement template guide.
Step-by-Step Task: Querying Underlying Cause of Death
The query form is organised in numbered sections. Menu labels change between releases, so check them against the database’s own help page.
- Open the database and accept the terms. Choose the Underlying Cause of Death database from the Deaths section of the home page and agree to the data use restrictions.
- Group results. Choose how to group the table, for example by year, state or cause.
- Select measures. Pick the columns you want, such as deaths, population, crude rate and age-adjusted rate, with confidence intervals and standard errors where offered.
- Set filters. Restrict by place, age, sex, race and cause, using 4-digit ICD-10 codes or groups of codes.
- Send and export. Run the query and, to use the results elsewhere, tick the export option to save a tab-delimited file.
Output and How to Read It
- Deaths: the count of deaths matching the query. Counts of 1 to 9 are suppressed, and the data use rules forbid publishing counts of 9 or fewer.
- Crude rate: deaths divided by population, per 100,000 people. It reflects the overall burden but is affected by the age mix of the population.
- Age-adjusted rate: a rate weighted to a common age distribution so that populations with different age structures can be compared.
- Unreliable rates: WONDER flags a rate as Unreliable when it rests on fewer than 20 deaths. CDC documents the relative standard error as 100 × √(1/D), where D is the number of deaths, so small counts give wide uncertainty.
Troubleshooting
- Suppressed cells: a table split into many small groups produces cells with 1 to 9 deaths. Group more coarsely (fewer categories or wider age bands) so fewer cells fall below the threshold.
- Rate marked “not applicable”: rates for records with age or Hispanic origin “Not stated” are not calculated because no population denominator exists for them.
- Counts that exceed total deaths in MCOD queries: one certificate can list several conditions, so adding counts across causes counts some deaths more than once. Use the underlying cause database for totals.
- Race categories: for decedents with multiple races, NCHS assigns one of four single races, and rates use Census Bureau population estimates. Differences between certificate and census reporting of race or Hispanic origin can bias rates.
Free Alternatives
| Platform | Agency | Focus |
|---|---|---|
| CDC WISQARS | CDC Injury Center | Injury data |
| NCI SEER*Stat | National Cancer Institute | Cancer statistics |
| WHO Global Health Observatory | World Health Organization | Global health statistics |
| IHME GBD Compare | Institute for Health Metrics and Evaluation | Global burden of disease estimates |
Frequently Asked Questions
What is the difference between underlying cause of death and multiple cause of death in CDC WONDER?
The underlying cause is the single condition that initiated the train of events leading to death, and each certificate has one. A certificate can also carry up to twenty additional multiple causes, which the Multiple Cause of Death database includes.
What does “Unreliable” mean in CDC WONDER?
A rate is flagged Unreliable when it is based on fewer than 20 deaths, because its relative standard error is large.
Why are some death counts suppressed?
Counts of 1 to 9 deaths are suppressed under CDC’s data use restrictions, and users must not publish counts of 9 or fewer.
Which years does the CDC WONDER Underlying Cause of Death database cover?
CDC’s Underlying Cause of Death documentation page is titled for 1999-2020. Check the WONDER home page for the databases and years currently offered.
Sources
All sources checked 2026-11-07.
- CDC, National Center for Health Statistics. Underlying Cause of Death, 1999-2020, CDC WONDER. https://wonder.cdc.gov/ucd-icd10.html
- CDC WONDER. Underlying Cause of Death help. https://wonder.cdc.gov/wonder/help/ucd.html
- CDC WONDER home page. https://wonder.cdc.gov/
