Written and maintained by CASRAI Editorial Board
Last updated
The Surveillance, Epidemiology, and End Results (SEER) Program is the National Cancer Institute’s system of population-based cancer registries. This guide covers what it collects, what its coverage figure does and does not let you conclude, which data tier to request, and how to compute and report an age-adjusted rate from registry data.
Definition in one paragraph
SEER is a cancer-surveillance program run by the National Cancer Institute (NCI) and supported by the Surveillance Research Program in NCI’s Division of Cancer Control and Population Sciences. Its registries routinely collect patient demographics, primary tumor site, tumor morphology and stage at diagnosis, the first course of treatment, and follow-up for vital status. NCI describes it as the only comprehensive source of population-based information in the United States that includes stage of cancer at diagnosis and patient survival data. SEER began collecting data on cancer cases on January 1, 1973, and the data are updated annually. For the underlying measures, see the guide to prevalence vs. incidence and the overview of epidemiology.
What the SEER registries collect, and where the other data come from
Registry records alone give counts of diagnoses. Rates need a denominator, and cause-of-death information comes from outside the registries. According to the SEER overview page:
- From the registries: patient demographics, primary tumor site, tumor morphology, stage at diagnosis, first course of treatment, and follow-up for vital status.
- Mortality data: provided by the National Center for Health Statistics (NCHS).
- Population data: provided by the US Census Bureau.
The program started in 1973 with registries in Connecticut, Iowa, New Mexico, Utah and Hawaii, plus the Detroit and San Francisco-Oakland metropolitan areas, and has since expanded to cover additional areas. Some registries are co-funded with the Centers for Disease Control and Prevention’s National Program of Cancer Registries and with state funds.
Coverage: what “45.9 percent of the US population” means for your analysis
SEER reports that its registries cover approximately 45.9 percent of the US population. The coverage differs by group:
| Group | Share of that group’s US population covered by SEER registries |
|---|---|
| All groups | Approximately 45.9% |
| Whites | 39.6% |
| African Americans | 43.5% |
| Hispanics | 64.9% |
| American Indians / Alaska Natives | 59.3% |
| Asians | 68.2% |
| Hawaiian / Pacific Islanders | 69.9% |
Source: NCI SEER, “Overview of the SEER Program”, checked 2026-10-29.
Two consequences follow directly from these figures. First, case counts from the SEER registries describe the registry areas, not the whole country, so they are not national totals; rates, which divide cases by the population of the same areas, are the appropriate comparison. Second, the registries cover different shares of different groups, so a comparison across groups rests on differently composed samples of each group. When you generalize from SEER to the whole US, say so as an assumption; the guide to generalizability in research explains how to frame it, and sampling bias covers the failure mode.
Which SEER data to request: access tiers
The SEER data page lists several products with different access requirements. Data years on the page run from 1975 to 2023, and the access software is SEER*Stat.
| Product | Stated access requirement |
|---|---|
| SEER Research Data | Open to any user with a valid email address; registration and data-use agreements are required. |
| SEER Research Plus and NCCR Data | Requires an eRA Commons or HHS account and a supervisor or mentor. |
| Specialized Databases | Requires Research Plus access first, then a separate application. |
Source: NCI SEER, “Data and Software”, checked 2026-10-29.
Start with the lowest tier that answers your question. If the data-use agreements and registration are all you need, Research Data avoids the account and sponsor requirements of the higher tiers.
Formula: the age-adjusted incidence rate, with every symbol explained
SEER is not a formula, but its most common output is an age-adjusted incidence rate, calculated by the direct method:
Age-adjusted rate = Σ rᵢ × (pᵢ / P), summed over i = 1 to n
- rᵢ: the incidence rate in age group i of your population (cases divided by person-population, usually expressed per 100,000).
- pᵢ: the number of people in age group i in the standard population.
- P: the sum of all pᵢ, the total standard population.
- pᵢ / P: the weight of age group i; the weights sum to 1.
- n: the number of age groups.
SEER publishes its standard populations at https://seer.cancer.gov/stdpopulation. The full method, including the choice of standard, is in the guide to age standardization.
Worked example with the arithmetic shown
Illustrative composite. The numbers below are invented to show the arithmetic; they are not SEER data. Real analyses use the standard population’s published weights and finer age groups.
Two hypothetical regions have the same age-specific incidence rates but different age structures. Three age groups are used.
| Age group | Region A cases / population | Region B cases / population | Rate per 100,000 (both regions) | Illustrative standard weight |
|---|---|---|---|---|
| 0-39 | 20 / 400,000 | 10 / 200,000 | 5.0 | 0.5 |
| 40-64 | 300 / 300,000 | 150 / 150,000 | 100.0 | 0.3 |
| 65+ | 600 / 100,000 | 900 / 150,000 | 600.0 | 0.2 |
| Total | 920 / 800,000 | 1,060 / 500,000 | 1.0 |
- Age-specific rates. Region A, 65+: 600 ÷ 100,000 × 100,000 = 600.0. Region B, 65+: 900 ÷ 150,000 × 100,000 = 600.0. The other groups work out the same way (5.0 and 100.0).
- Crude rates. Region A: 920 ÷ 800,000 × 100,000 = 115.0. Region B: 1,060 ÷ 500,000 × 100,000 = 212.0.
- Age-adjusted rate (both regions). (5.0 × 0.5) + (100.0 × 0.3) + (600.0 × 0.2) = 2.5 + 30.0 + 120.0 = 152.5 per 100,000.
Region B’s crude rate is almost twice Region A’s, yet the age-adjusted rates are identical because the regions differ only in age structure.
Interpretation: how to read the result
| If you see | Reading |
|---|---|
| Crude rates differ, adjusted rates are equal | The difference is explained by age structure (as in the example above). |
| Adjusted rates differ | The difference is not explained by age structure; other factors, or differences in registration and detection, remain candidates. |
| Rate from the SEER registries | A rate for the covered areas, not a count of cases nationally. |
| Rate for one racial or ethnic group | Based on registries that cover that group to a different extent than other groups (see the coverage table). |
An age-adjusted rate is a comparison index. It depends on the chosen standard population, so rates computed with different standards cannot be compared with one another. Report uncertainty alongside the rate; see confidence intervals explained.
Assumptions and common mistakes
- Treating registry counts as national counts. Coverage is approximately 45.9 percent, so a count is for the registry areas.
- Comparing rates standardized to different populations. Use one standard throughout and state which.
- Assuming equal coverage across groups. Coverage varies from 39.6% to 69.9% across the groups listed above.
- Mixing up the data tier. Specialized Databases cannot be requested before Research Plus access is in place.
- Using stale data. The data are updated annually; record the release and data years you used.
How to report it (APA example)
State the registry program, the data years, the case definition, the standard population and the software. For a sentence in the Methods section:
“Incidence data were obtained from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute, covering diagnosis years 2000-2020 (National Cancer Institute, 2026). Rates were age-adjusted by the direct method to the 2000 US standard population and are expressed per 100,000.”
Reference-list entry in APA 7th edition format (group author, web page):
National Cancer Institute. (2026). Surveillance, Epidemiology, and End Results (SEER) Program. https://seer.cancer.gov/
The years in the sentence above are placeholders for your own extract; replace them with the years you used. See also how to report confidence intervals.
How to calculate an age-adjusted rate in Excel, R or Python
Each snippet reproduces the worked example above. The result is 152.5 per 100,000.
Excel
Put the age-specific rates in B2:B4 and the standard weights in C2:C4, then enter =SUMPRODUCT(B2:B4,C2:C4).
R
rates <- c(5, 100, 600) # per 100,000
weights <- c(0.5, 0.3, 0.2) # sums to 1
sum(rates * weights) # 152.5Python
rates = [5, 100, 600]
weights = [0.5, 0.3, 0.2]
print(round(sum(r * w for r, w in zip(rates, weights)), 1)) # 152.5With SEER data, SEER*Stat is the access software listed on the SEER data page; the arithmetic above is what an age-adjusted rate means regardless of which software computes it.
FAQ
What does SEER stand for?
Surveillance, Epidemiology, and End Results. It is a National Cancer Institute program of population-based cancer registries.
When did SEER start collecting data?
On January 1, 1973, with registries in Connecticut, Iowa, New Mexico, Utah, Hawaii, and the Detroit and San Francisco-Oakland metropolitan areas.
What share of the US population does SEER cover?
Approximately 45.9 percent, with coverage by group ranging from 39.6% (Whites) to 69.9% (Hawaiian/Pacific Islanders).
Is SEER data free?
SEER Research Data is available to any user with a valid email address who registers and signs the data-use agreements. Higher tiers require an eRA Commons or HHS account and a supervisor or mentor.
Where do SEER’s mortality and population data come from?
Mortality data come from the National Center for Health Statistics; population data come from the US Census Bureau.
Sources
All pages below were checked on 2026-10-29.
- National Cancer Institute, SEER. “Overview of the SEER Program.” https://seer.cancer.gov/about/overview.html
- National Cancer Institute, SEER. “SEER Registries.” https://seer.cancer.gov/registries/
- National Cancer Institute, SEER. “Data and Software.” https://seer.cancer.gov/data/
- National Cancer Institute, SEER. “Data Access.” https://seer.cancer.gov/data/access.html
- National Cancer Institute, SEER. “Standard Populations.” https://seer.cancer.gov/stdpopulation
