Kara Smigel (NCI)
Joann Schellenbach (ACS)
Sandra Smith (CDC)
This report provides an update on the trends in cancer death rates in the United States and presents information about trends in cancer incidence (new cases reported) that has not been published before. The American Cancer Society (ACS), the National Cancer Institute (NCI), and the Centers for Disease Control and Prevention (CDC), including the National Center for Health Statistics (NCHS), collaborated to create this report.
2. Where do the data come from?
Data on cancer incidence come from the NCI's Surveillance, Epidemiology, and End Results (SEER) Program. The SEER program collects cancer incidence data from nine registries: five state-wide registries (Connecticut, Hawaii, Iowa, New Mexico, and Utah) and four metropolitan-area registries (Atlanta, Detroit, San Francisco-Oakland, and Seattle-Puget Sound). These nine registries are population-based (collect information on every cancer in a geographic area) and include 9.5 percent of the U.S. population.
Data on cancer mortality comes from NCHS. Death certificates filed in every state are processed and consolidated into the NCHS database, so 100 percent of the U.S. population is represented. Separate mortality information on Hispanics was not available from Connecticut, Louisiana, New Hampshire, and Oklahoma.
3. What is happening with cancer rates overall?
Overall cancer incidence and death rates for all sites combined declined from 1990 to 1995 and almost all racial and ethnic groups are included in this downturn.
After increasing 1.2 percent per year from 1973 to 1990, incidence rates for all cancer sites combined decreased 0.7 percent per year during 1990 to 1995. The turning point was 1992 and from 1992 to 1995 the rate decreased 2.7 percent per year. People ages 35 to 44 and those age 75 years and older had the largest decreases in incidence.
Cancer death rates increased 0.4 percent per year from 1973 to 1990. From 1990 to 1995, the annual percent rate for cancer death decreased 0.5 percent per year. The death rate decline is greater in males (0.9 percent per year) than females (0.1 percent per year). Persons younger than age 65 are having the greatest drops in cancer death rates.
4. What is happening with lung cancer rates?
The overall lung cancer incidence rate increased 1.9 percent per year from 1973 to 1990, but decreased 1.1 percent per year from 1990 to 1995 — due mainly to decreased rates for men.
Lung cancer incidence for white men declined 2.5 percent per year from 1990 to 1995.
The decline in lung cancer deaths from 1990 to 1995 was also confined primarily to men: rates for white males declined 1.5 percent per year, for black males declined 1.9 percent per year, and for Hispanics males declined 1.9 percent per year.
In contrast, only women of Hispanic origin had a decline in lung cancer mortality from 1990 to 1995 (dropping 0.9 percent per year), while annual rates increased for white women (1.7 percent per year).
5. Why are lung cancer rates changing?
Decreased incidence and mortality from lung cancer is most likely due to the overall decreased rate of smoking over the past 30 years — fewer than 25 percent of adults now smoke (1995 data) compared to 42 percent in 1965. However, the pattern of smoking in women lags behind that of men, so we do not see the impact of decreased smoking among women. The decline in adult tobacco use is slowing and tobacco use in youth is again on the rise.
6. What is happening with breast cancer rates in women?
The incidence rate for female breast cancer increased 1.8 percent per year from 1973 to 1990, but was level from 1990 to 1995. After increasing 0.2 percent per year from 1973 to 1990, breast cancer death rates decreased 1.7 percent per year from 1990 to 1995. This decrease was confined to white and Hispanic women, each of whom had a 1.9 percent per year decrease in death rates from 1990 to 1995.
7. Why are breast cancer rates changing?
The widespread incorporation of breast cancer screening into routine medical care and the increased use of adjuvant treatments may account for the overall decreased mortality.
8. What is happening with prostate cancer rates?
After increasing 3.4 percent per year from 1973 to 1990, prostate cancer incidence rates declined 1.0 percent per year from 1990 to 1995. Prostate cancer death rates increased 1.0 percent per year from 1973 to 1990, but have declined 1.1 percent per year from 1990 to 1995. White men had a 1.2 percent per year decrease in death rates from 1990 to 1995.
9. Why are prostate cancer rates changing?
Changes in the incidence and mortality of prostate cancer are being carefully analyzed, and while possible explanations have been suggested, the changes are likely due to several interacting factors. The sudden increased use of prostate-specific antigen testing in the late 1980s and early 1990s caused an increase in diagnoses of previously undetected cancers. Now that these prevalent cancers have been detected, the incidence rate may be dropping to an equilibrium that better reflects the true incidence in the population.
Another part of the decrease in incidence may be that because prostate cancer diagnoses and treatments are taking place in outpatient settings, there may be incomplete or delayed reporting of the diagnoses. Also, decreased use of PSA screening tests may be occurring, precipitated by organizations discouraging its use.
10. What is happening with colon and rectum cancer rates?
Colon and rectum cancer incidence declined 2.3 percent per year from 1990 to 1995.
Men are having the largest declines in colon and rectum cancer rates with annual percent changes for 1990 to 1995 declining 3.3 percent for white men and 2.5 percent for Asian and Pacific Islander men. White women had a 1.8 percent per year decline in incidence from 1990 to 1995.
Death rates for cancers of the colon and rectum declined 0.9 percent per year during 1973 to 1990, and they continued to decline from 1990 to 1995 by 1.5 percent per year. Rates for white males declined 2.0 percent per year and for white females declined 1.5 percent per year from 1990 to 1995.
11. Why are colon and rectum cancer rates declining?
The declines in colorectal cancer rates are not well understood. The use of fecal occult blood testing and sigmoidoscopy screening tests is low, signaling that factors other than screening tests are at play. Increased removal of polyps (which often become cancerous), better treatments, and other factors may be contributing.
12. Are there other findings about incidence that are important?
The incidence of bladder cancer increased 0.8 percent per year from 1973 to 1990 and declined at a slower rate from 1990 to 1995.
The incidence of non-Hodgkin's lymphoma and melanoma have continued to increase from both 1973 to 1990 and from 1990 to 1995. Non-Hodgkin's lymphoma increased 3.5 percent per year from 1973 to 1990 and continued to increase from 1990 to 1995 at a slower rate. Melanoma increased 4.3 percent per year from 1973 to 1990 and continued to increase 2.5 percent per year from 1990 to 1995.
Uterine cancer incidence rates decreased 2.5 percent per year from 1973 to 1990, but stopped declining from 1990 to 1995.
13. What are the other important findings about cancer death rates?
Deaths from cancer of the bladder, uterus, ovary, and leukemia decreased from 1973 to 1990 and at a slower rate from 1990 to 1995. From 1973 to 1990, the bladder cancer death rate declined 1.8 percent per year, the uterine cancer death rate declined 1.7 percent per year, and the ovarian cancer death rate declined 0.6 percent per year. All continued to decline from 1990 to 1995, but at a slower rate (not statistically significant). Leukemia rates declined 0.4 percent per year from 1973 to 1990 and continued to decline
0.3 percent per year from 1990 to 1995.
Deaths due to non-Hodgkin's lymphoma increased 1.8 percent per year from 1973 to 1990 and 1.9 percent per year from 1990 to 1995.
14. How is this report different from the report from November 1996?
This report is different in three ways: First, this report includes trends in cancer incidence while the November 1996 report included only cancer death data. The new report also includes information on cancer rates trends in four racial and ethic groups: white, black, Hispanic, and Asian/Pacific Islander. Finally, this report includes final death statistics from 1995, whereas the previous report was only preliminary data from 1995.
15. Is there preliminary data from 1996 that can be reported?
Preliminary death data for 1996 for all cancer sites combined shows that the decline in mortality is continuing. The mortality rates for such major sites as lung, female breast, prostate, and colorectal cancer also declined. As in the 1990 to 1995 period, non-Hodgkin's lymphoma deaths showed an increase. There was no change in 1996 for uterine cancer.
How to Read the Report
16. How is progress against cancer being measured in this report?
This report includes two different measures, the annual percent change in cancer rates and incidence and death rates alone. Annual percent change has been calculated for two time periods, 1973 to 1990 and 1990 to 1995, and by racial and ethnic group. Only statistically significant annual percent changes are discussed in the paper and given by numbers in the press materials. "Statistically significant" means that the rate calculated is unlikely to have occurred by chance alone.
17. What is an annual percent change or APC?
The annual percent change is the averaged rate of change in a cancer rate per year in a given time frame, i.e. how fast or slow a cancer rate has increased or decreased each year over a period of years. Annual percent change, sometimes abbreviated as APC, were calculated for both incidence and death rates. The number is given as a percent — such as the 0.7 percent per year decrease in incidence of all cancers from 1990 to 1995.
18. How are cancer incidence and death rates presented?
Cancer incidence rates and cancer death rates are measured as a number per 100,000 people. When a cancer affects only one gender, such as prostate cancer, the number is per 100,000 people of that gender.
In this report, cancer incidence and death rates are given for the period 1990 to 1995 for 23 cancer sites, including separate figures for lung cancer in men and women because the rate for men is much higher. The 23 sites are cancers of the prostate, female breast, lung, colon/rectum, uterus, bladder, ovary, oral cavity, kidney, pancreas, uterine cervix, brain, stomach, thyroid, larynx, esophagus, liver, and gallbladder and non-Hodgkin's lymphoma, melanoma, leukemia, multiple myeloma, and Hodgkin's disease.
19. What racial, ethnic and cultural groups are reported?
Cancer incidence and death rates are given for whites, blacks, Asian and Pacific Islanders, and Hispanics. Hispanic is not mutually exclusive from whites, blacks, and Asian and Pacific Islanders. Information for all races combined also includes data from Native Americans, who are not reported separately.
20. Where is this report being published?
The report is published in the March 15, 1998 issue of Cancer, and is titled, "Cancer incidence and mortality, 1973-1995: A report card for the U.S." The authors are Phyllis A. Wingo, Ph.D. (ACS), Lynn A. G. Ries (NCI), Harry M. Rosenberg, Ph.D. (NCHS), Daniel S. Miller, M.D. (CDC), and Brenda K. Edwards, Ph.D.(NCI).
21. What Internet sites have more information on cancer?
American Cancer Society: http://www.cancer.org
National Cancer Institute: http://rex.nci.nih.gov/ for patients, the public, and mass media or NCI's main website, http://www.nci.nih.gov
CDC's Division of Cancer Prevention and Control: http://www.cdc.gov/nccdphp/dcpc
CDC's National Center for Health Statistics: http://www.cdc.gov/nchswww/