Incidence and Death Estimates by Congressional District
Cancer death rates and counts for 2011–2016 were estimated for the 436 federal congressional districts according to the boundaries for the 116th Congress of the United States.1, 2 Cancer incidence rates and counts were estimated for 424 federal congressional districts. Incidence data are not included for Kansas (4 congressional districts) and Minnesota (8 congressional districts) because county-level incidence data were not available. Illinois opted not to present congressional district-specific estimated case counts and incidence rates (18 congressional districts). Therefore, estimated rates and counts are presented from 406 congressional districts.
Methods for Creating Congressional District Estimates
A brief description of the methods for estimating congressional district rates and counts is provided below. For specific inquiries, please e-mail the U.S. Cancer Statistics team at email@example.com.
- Eight congressional districts follow state or federal district boundaries: Alaska, Delaware, District of Columbia, Montana, North Dakota, South Dakota, Vermont, and Wyoming. Those districts were estimated according to the state rates and counts.
- For the remaining districts, rates were estimated by assigning the county-level age-adjusted rates (age-adjusted to the 2000 U.S. standard population using 15 age groups — 0–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49, 50–54, 55–59, 60–64, 65–69, 70–74, 75–79, 80–84, and ≥85) to the census block and weighting those by the block population proportion of the congressional district. Those weighted rates were then aggregated over the blocks within the congressional district to estimate the district rate. More specifically, the following steps were taken—
- Population estimates within each census block by race and sex and assigned to congressional districts were determined from the 2010 U.S. Census Summary File 1.
- The age-adjusted county-level rates by race and sex were calculated using SEER*Stat and merged with the block-level population estimates by county.
- The county rates assigned to the census blocks were weighted by the proportion of the block population within the congressional district and then aggregated over the blocks within the congressional district.
- To calculate counts for congressional districts, the county counts were weighted by the proportion of the county population in the congressional district to the overall county population. The weighted counts were then aggregated over the counties in the congressional district. This gives the same estimates as weighting at the block level similar to the rate calculations, but is a more efficient calculation in terms of computer time. Estimates for both sexes combined were obtained by summing the male estimate and female estimate.
Estimates are presented by sex (both sexes, male, and female) and race/ethnicity (all races, non-Hispanic white, black, and Hispanic). Block-level population data were not available by ethnicity for races other than white. As a result, the estimates for black race include both Hispanics and non-Hispanics. Data are presented for all cancers combined and 20 leading cancers. Data are suppressed for cells with fewer than 16 estimated cases. Data for specific race groups may be suppressed at the state’s request.
Since the congressional district estimates require county-level data, if any county data are missing, then the overall state counts presented in the Congressional Districts section will not match the counts in the U.S. Cancer Statistics Data Visualizations Tool’s State section. Instead, the counts in the Congressional Districts section will match the state counts calculated by aggregating across the U.S. Cancer Statistics county-level data.
- Hao Y, Ward EM, Jemal A, Pickle LW, Thun MJ. U.S. congressional district cancer death rates.external icon International Journal of Health Geographics 2006;5:28.
- Siegel RL, Sahar L, Portier KM, Ward EM, Jemal A. Cancer death rates in U.S. congressional districts.external icon CA: A Cancer Journal for Clinicians 2015;65(5):339–344.