The National Cancer Institute's (NCI) annual update of cancer statistics, the "SEER Cancer Statistics Review 1973-1991," reports and summarizes the key measures of cancer's impact on the U.S. population for the years 1973 through 1991. NCI monitors these cancer statistics to assess progress and to identify population subgroups and geographic areas where cancer control efforts need to be concentrated.

The 1994 edition is the latest in a series that has been produced annually by the Surveillance, Epidemiology, and End Results (SEER) Program in NCI's Division of Cancer Prevention and Control. Data included in the book were compiled by the SEER Program, which has monitored occurrence of cancer and survival of patients since 1973.

The tables, charts, and graphs of the "SEER Cancer Statistics Review 1973- 1991" present cancer incidence rates (the number of cases per 100,000 persons), cancer mortality rates (the number of deaths per 100,000 persons), and five-year relative survival rates (the estimated percentage of patients surviving five years after diagnosis of cancer).

Data are given for all major cancer sites by age, sex, and race (all races, whites, and blacks). Cancer incidence and survival data are collected by the SEER Program, while mortality data come from the National Center for Health Statistics (NCHS). The report also provides data on trends in cancer incidence and mortality rates, American Cancer Society estimates of new cancer cases and cancer deaths in 1994, and measures of the years of life lost prematurely by those dying of cancer.

The book also contains estimates, by race, sex, and cancer site, of the probabilities of developing or dying of cancer. These include probabilities, given current age, of developing cancer before a specific later age, as well as lifetime probabilities.

Single copies of the "SEER Cancer Statistics Review 1973-1991" may be obtained by writing the Office of Cancer Communications, National Cancer Institute, 31 Center DR MSC 2580, Building 31, Room 10A19, Bethesda, Md. 20892-2580, or calling 1-800-4-CANCER (1-800-422-6237).

1. What is the "SEER Cancer Statistics Review, 1973-1991?"
2. What is the purpose of the CSR?
3. How can the CSR be used?
4. How does the 1994 CSR differ from the previous volume?

Background and Resources

5. Who prepared the CSR
6. What is the Surveillance, Epidemiology, and End Results (SEER) Program?
7. What is a cancer registry?
8. Which cancer registries report to SEER?
9. What other resources are used to prepare the CSR?
10. How is the 1994 CSR organized?

Interpretation of the Data in the CSR

11. What factors should be considered when interpreting the CSR data?
12. What are the definitions of statistical terms used in the CSR?

13. What are the highlights of the 1994 CSR?
14. Which cancer sites show the highest incidence rates for the period 1987-1991?
15. Which cancer sites show the greatest increases and decreases in incidence rates over the interval 1973-1991?
16. Which cancer sites show the highest mortality rates for the period 1987-1991?
17. Which cancer sites show the greatest increases and decreases in mortality rates over the interval 1973-1991?
18. What are the probabilities for developing the most common cancers in men and women?
19. What are the probabilities of dying of the most common causes of cancer death in men and women?
20. Which states had the highest and lowest cancer mortality rates for the period 1987-1991?
21. Which cancer sites show the highest and lowest five-year relative survival rates for patients diagnosed during 1983-1990?
22. Is there a difference in cancer incidence, mortality, and survival between sexes?
23. Is there a difference in cancer incidence, mortality, and survival between blacks and whites?
24. Does the CSR tell us anything about geographic patterns of cancer?

Questions and Answers

"SEER Cancer Statistics Review 1973-1991"

Description of the "SEER Cancer Statistics Review 1973-1991" (CSR)

1. What is the "SEER Cancer Statistics Review 1973-1991?"

The CSR is a book published annually by the National Cancer Institute (NCI) that provides data on cancer incidence, mortality, and patient survival for more than 20 different cancers and for all cancers combined. The 1994 CSR covers the years 1973 through 1991. Cancer incidence and survival data from nine geographic areas representing about 10 percent of the U.S. population were used in the book, along with mortality data from the entire United States. Data are reported by age, sex, and race (all races, whites, and blacks).

2. What is the purpose of the CSR?

The CSR reports and summarizes the key measures of cancer's impact on the U.S. population. The primary statistics reported are the rate at which new cases occur (incidence), the rate of death from cancer (mortality), and the likelihood that a patient will live a specific length of time after diagnosis (survival).

3. How can the CSR be used?

The CSR provides a cancer data resource for a variety of scientific, educational, business, and governmental activities. It is used by national, state, and local health agencies to help identify cancer problems. Data from the CSR are used to plan, implement, and evaluate cancer control efforts, and to design research studies.

Researchers use the CSR as the primary assessment of cancer status in the United States. Besides serving as a valuable reference, the CSR data provide a statistical baseline for comparison with investigational findings.

The CSR data may be linked with information from other agencies to address cancer-related issues. For example, CSR incidence data linked to county statistics on socioeconomic status have shown that for some cancers, socioeconomic status is more important than race in explaining differences in incidence. Also, CSR data linked to Medicare records at the Health Care Financing Administration are being used to help assess the costs of cancer tests and treatment by site and stage.

4. How does the 1994 CSR differ from the previous volume?

The 1994 CSR differs from the 1993 volume by the addition o f a single year's statistics to the SEER database. New inclusions are incidence and mortality data for 1991, and survival data for 1990. These are the most recent years for which complete data are available.

Background and Resources

5. Who prepared the CSR?

The CSR was prepared by NCI statisticians in the Surveillance Program, Division of Cancer Prevention and Control. NCI is part of the National Institutes of Health, U.S. Department of Health and Human Services.

6. What is the Surveillance, Epidemiology, and End Results (SEER) Program?

The SEER Program, started in 1973, is NCI's main tool for tracking, assembling, and reporting data on cancer incidence, mortality, and survival in the United States. Data come primarily from population-based cancer registries under contract to NCI in nine geographic areas across the country. Data are collected by age, sex, and race (all races, whites, and blacks). National trends in cancer incidence and patient survival are derived from this database. Mortality statistics come from the National Center for Health Statistics (NCHS).

The SEER Program was recently expanded in order to increase coverage of minority populations, especially Hispanics. The two areas added are in California: Los Angeles County and the four counties in the San Jose-Monterey area south of San Francisco. Data from the two new areas were not available for the 1994 CSR, but will be included in later editions.

Data from the report, and from SEER special studies, are used by NCI and others to design specific research studies to investigate causal factors. Examples of such research include epidemiologic studies designed to test dietary or hormonal hypotheses suggested by observed racial or sex differences. SEER data are also used by NCI and others to guide program decisions in cancer prevention and control. NCI has used these data to create special programs aimed at breast, prostate, and lung cancers, and to target cancer prevention and control programs at population groups with high incidence or mortality rates.

7. What is a cancer registry?

A cancer registry (sometimes known as a tumor registry) collects and stores data on cancers diagnosed either in a specific hospital or medical facility (hospital-based registry) or in a defined geographic area (population based registry). A population-based registry is generally composed of a number of hospital-based registries. Registries participating in the SEER Program arepopulation-based.

8. Which SEER cancer registries contribute data to the CSR?

Cancer incidence data presented in the 1994 CSR come from nine population- based cancer registries in four metropolitan areas and five states across the United States. The registries cover Atlanta, Georgia (five counties); Detroit, Michigan (three counties); San Francisco-Oakland, California (five counties); Seattle-Puget Sound, Washington (13 counties); and all counties in Connecticut, Iowa, Hawaii, New Mexico, and Utah.

9. What other resources are used to prepare the CSR?

Cancer mortality data on all deaths occurring in the United States as a whole, each state, and each SEER geographic area are obtained from the NCHS. The 1994 projections of the number of new cancer cases and cancer deaths in the United States are obtained from the American Cancer Society, which uses SEER and NCHS data to make the projections. Population data are obtained from the U.S. Census Bureau.

10. How is the 1994 CSR organized?

The CSR is divided into 27 sections and an appendix.

Section I is an overview of the report. It describes the data sources used, provides supplementary material for improved understanding of the data, and lists references cited. This section presents cancer incidence and mortality from 1987-1991, and survival rates for the period 1983-1990. It describes trends in incidence and mortality rates between 1973 and 1991, and summarizes long-term trends since 1950. It also includes data on cigarette smoking prevalence and median ages of cancer patients at diagnosis and death.

Section II includes incidence, mortality, and survival rates on all cancers combined. Sections III through XXVII contain data on specific cancers and cancer in children. The appendix provides tables showing the numbers of cancer cases in the SEER geographic areas, and numbers of cancer deaths in the United States (by site, race, and sex) during 1987-1991. It also includes population estimates, for U.S. states and SEER areas, that were used in calculating cancer rates.

Interpretation of the Data in the CSR

11. What factors should be considered when interpreting the CSR data?

The statistics outlined in the CSR must be considered in light of changes in factors that can affect cancer incidence or patient survival. Examples include risk factors such as smoking, screening procedures such as mammography, diagnostic procedures for which there have been technological advances, and the introduction of new treatment regimens.

There may be a time lag between the introduction of new treatments or technologies and their impact on cancer statistics.

12. What are the definitions of statistical terms used in the m CSR?

• Incidence - the number of newly diagnosed cases for a specific cancer or for all cancers combined during a specific time period. When expressed as a rate, it is the number of new cases "per standard unit of population" during the time period. Incidence rates can be calculated based on a number of factors such as age, race, or sex.

  Example: 9.0 cases of pancreatic cancer per 100,000 persons annually (all races, both sexes, for the period 1987-1991).

• Mortality - the number of deaths for a specific cancer or for all cancers combined during a specific time period. When expressed as a rate, it is the number of deaths "per standard unit of population" during the time period. Mortality rates can be calculated based on a number of factors such as age, race, or sex.

  Example: 2.9 deaths from multiple myeloma per 100,000 persons annually (all races, both sexes, for the period 1987-1991).

• Age-adjusted rate - an incidence or mortality rate that has been adjusted to reduce the effects of differences in the age distributions of the populations being compared.

  An age-adjusted rate is computed by weighing the age-specific rates in the population of interest by the proportions of persons in the corresponding age groups within a standard population. For the 1994 CSR, the 1970 U.S. standard population is used for most rates. In addition, the world standard population is used to age-adjust some rates, permitting comparisons of rates in SEER areas with those from other countries that have published rates adjusted to the world standard.

  Unless labeled "age-specific," all incidence and mortality rates in the 1994 CSR are age-adjusted rates.

• Relative survival rate - an estimate of the percentage of patients that would be expected to survive the effects of their cancer. This rate is calculated by adjusting the observed survival rate so that the effects of causes of death other than those related to the cancer in question are removed. (Observed survival is the actual percentage of patients still alive at some specified time after diagnosis of cancer. It considers deaths from all causes, cancer or otherwise.)
• Percent change - a measure of the change in incidence and mortality rates over a specified time interval. For the 1994 CSR, it is calculated as the percent change between the average rates for the periods 1973-1974 and 1990-1991. Percent changes are also provided for two 5-year periods, 1975 to 1979 and 1987 to 1991.
• Estimated Annual Percent Change (EAPC) - a measure of the estimated yearly percent change in incidence and mortality rates over a specified time interval. For the 1994 CSR, the EAPC is given for the entire interval 1973 to 1991, and for two 5-year periods, 1975 to 1979 and 1987 to 1991.
• Person Years of Life Lost (PYLL) - the sum of years of life lost by all persons in a population who died of a particular cancer. Actuarial (life-expectancy) tables are used to determine the years of life remaining if the persons had not died of cancer at that particular age.

  Example: In 1991, lung cancer had a PYLL of 2.15 million years for the United States (all races, both sexes).

• Average Years of Life Lost (AYLL) - the average years lost to a particular cancer among all persons who died of that cancer. It is calculated by dividing the PYLL for a particular cancer by the number of deaths from that cancer.

  Example: In 1991, melanoma had an AYLL of 19.9 years for the United States (all races, both sexes).

• Significance - when comparing rates, statistical significance means that differences between or among rates are unlikely to have occurred merely by chance.

  For example, the probabilities that differences occurred by chance may be given as less than 1 in 10 (<0.10) or less than 1 in 20 (<0.05).

• Standard error - the variability associated with a reported cancer statistic.
• Time trend - the change in a cancer incidence or mortality rate (increasing, decreasing, or not changing) over time and the magnitude of that change.

Findings in the CSR

13. What are the highlights of the 1994 CSR?

• Incidence (also see answers 14 and 15 for more data):

  For all cancers combined, increases in cancer incidence were seen between 1973 and 1991 for whites (23.0 percent), blacks (23.6 percent), men (31.5 percent), and women (13.6 percent). The overall annual incidence rate for all races, all ages, men and women (1987- 1991) is 390.4 cases per 100,000 persons.

• Mortality (also see answers 16, 17, and20 for more data):

  For all cancers combined, statistically significant declines in the mortality rate between 1973 and 1991 were seen for persons under age 55: men (16.2 percent), women (16.0 percent), and men and women combined (16.1 percent). In children (ages 0-14) the overall cancer mortality rate between 1973 and 1991 has dropped 42.1 percent.

  During the same time period, cancer mortality rates in persons aged 55-64 increased 1.9 percent in men and 5.8 percent in women. For persons aged 65 and older, the rates increased 14.6 percent in men and 20.2 percent in women.

  For all age groups combined, the U.S. cancer mortality rate for 1987-1991 was 172.8 deaths per 100,000 persons.

• Survival (also see answer 21 for more data):

  For all cancers combined, the five-year relative survival rate for all races, all ages, men and women (1983-1990) was 53.9 percent. This represents a 4.6 percentage point increase in survival from the 1974-1976 rate of 49.3 percent. The survival rate for persons diagnosed at ages under 65 was 57.5 percent, while the rate for those diagnosed at ages 65 and older was 50.3 percent.

• Lung cancer:

  1991 was the fourth consecutive year in which lung cancer incidence declined in men. Incidence rates for both white and black men have shown an overall decline during the last five years (1987-1991): 4.7 percent for whites and 3.9 percent for blacks. For both white and black women, the incidence rates have increased approximately 5 percent per year over the interval 1973-1991. This increase in rates has slowed, however, in recent years for white women (1987- 1991).

  Lung cancer remains the chief cause of cancer death in both sexes. It is estimated that 28 percent of all cancer deaths in 1994 will be from lung cancer (33 percent of cancer deaths in men and 23 percent of cancer deaths in women).

  During the interval 1973-1991, lung cancer mortality rates decreased for both white (22.1 percent) and black (16.2 percent) men under age 55. Mortality rates declined for white women under age 45 and for black women aged 35-44, but increased in all other age groups. For both white and black women at all ages, the lung cancer mortality rates have been increasing approximately 5 percent per year since 1973. However, the increases appear to be slowing in recent years (1987-1991) to about 3 percent per year. For the sixth consecutive year, lung cancer mortality for white women is higher than breast cancer mortality.

  Interpretation of trends:

  The decline in lung cancer mortality rates for men under age 55 and women under age 45 is due largely to reductions in cigarette smoking that have occurred since the 1964 Surgeon General's report on smoking and health.

  The steep rise in lung cancer mortality rates for women at all ages during the interval 1973-1991 reflects the increase in cigarette smoking by women since the 1960s. The log interval between initiation of smoking and lung cancer death indicates that lung cancer will remain the leading cause of cancer mortality in women for many more years.

• Breast cancer:

  The incidence rate for breast cancer increased slightly in 1991, from 109.4 cases per 100,000 women in 1990 to 110.2 in 1991. Breast cancer incidence increased rather dramatically through much of the 1980s, rising from 85.2 cases per 100,000 women in 1980 to 112.4 in 1987. Since 1987, however, incidence appears to have levelled off. From 1973-1991, incidence rates increased 8.5 percent for white women under 50 years old, 31.8 percent for white women 50 and older, 26.1 percent for black women under 50, and 32.0 percent for black women 50 and older. During the period 1987-1991, average annual incidence rates were 32.9 per 100,000 women under 50 and 345.8 per 100,000 women 50 and older.

  From 1973-1991, breast cancer mortality rates decreased 14.0 percent for white women under age 50, while increasing for white women 50 and older (4.0 percent), black women under 50 (2.0 percent), and black women 50 and older (26.3 percent). For women of all races and all ages combined, the mortality rate increased 1.8 percent from 1973-1991. For 1987-1991, average annual breast cancer mortality rates were 6.2 per 100,000 women under age 50 and 92.6 per 100,000 women 50 and older.

  Interpretation of trends:

  There is evidence that the marked increase in breast cancer incidence that began in the early 1980s was due primarily to a nationwide increase in mammography screening, particularly in women ages 50 and older. (Because mammography can detect tumors earlier than they would be discovered symptomatically or through clinical exams, increased mammography use results in increased case reporting.) The stabilization of incidence rates in the past three years is consistent with expectations, based on the assumption that excess cases detected earlier by mammography would reach a peak, then begin to decline. That peak may have occurred in 1987. The earlier detection of cases through mammography is expected to result in improvements in mortality by the mid-1991's, because breast cancer is more effectively treated when detected earlier.

  The more gradual, long-term increase in breast cancer incidence, seen over the past few decades, is more difficult to explain. About 60 percent of women diagnosed have no known risk factors such as family history of the disease or those associated with reproductive history (late child bearing, early menarche, late menopause, and possibly, use of exogenous hormones such as oral contraceptives and postmenopausal replacement estrogen). Other possible causes, including lifestyle and environmental risk factors, are under investigation.

• Colorectal cancer:

  Incidence rates increased during the interval 1973-1991 for white men (3.0 percent), black men (36.1 percent), and black women (20.0 percent). The incidence rate for white women decreased 6.6 percent over this period.

  Colorectal cancer mortality rates declined during the last five years (1987-1991) for white men (4.9 percent), white women (6.7 percent). The mortality rate increased 4.1 percent in black men and 0.7 percent in black women.

  Interpretation of trends:

  Declines in incidence rates since the mid-1980s suggest that colorectal cancer incidence in whites may have peaked, and if trends continue as expected, rates for white men and women and black women should continue to decline, and rates for black men should begin to decline. Part of the decrease in incidence may be due to the removal of benign colon polyps during endoscopic screening, which may prevent progression to invasive disease.

  Declines in colorectal cancer mortality rates are believed to reflect earlier detection and improvements in diagnostic tests and techniques. Refinements in treatment practices, making surgical treatment possible for more patients, are also considered a factor in the downward trends.

• Prostate cancer:

  Incidence rates increased during the interval 1973-1991 for both white (127.1 percent) and black (81.9 percent) men. Rates are significantly higher in blacks than in whites. The overall incidence rate increased 41.3 percent during the last five years (1987-1991).

  Mortality rates increased during the interval 1973-1991 for both white (21.3 percent) and black (39.2 percent) men. The increase in prostate cancer mortality from 1987-1991 is more than double the rise between 1975-1979. The mortality rate for blacks is more than twice the rate for whites.

  Interpretation of trends:

  The rise in prostate cancer incidence between 1973 and 1987 is believed to have been largely an incidental finding due to the increased use of transurethral resection of the prostate (TURP) to relieve urinary obstruction. As more tests were performed and specimens examined for cancer, more cancers have been detected, resulting in the increased incidence rates. Since 1987, however, prostate cancer rates have continued to rise sharply while TURP rates declined. Investigations are ongoing to determine the extent to which recent increases in incidence are due to wider use of tests to detect prostate cancer, such as transrectal ultrasound-guided needle biopsy, and in particular, blood testing for prostate- specific antigen (PSA). However, the concurrent increase in mortality rates suggests that some of the incidence increase may be a result of actual changes in the risk for developing prostate cancer. Investigations are under way to determine the extent to which increases in mortality are due to the increasing numbers of men surviving to advanced old age, when prostate cancer is most prevalent.

• AIDS-related cancers:

  The AIDS epidemic in the United States has resulted in sharp increases in a number of cancers associated with AIDS. Kaposi's sarcoma (KS) is often an early manifestation of AIDS. Before the advent of AIDS, this cancer was extremely rare among Americans under age 55. Incidence rates in this group rose from nearly zero in the early 1970s to a peak of 15.3 cases per 100,000 men in 1989, then declined slightly in 1990 and 1991.

  In San Francisco county, the SEER area with the highest AIDS rates, incidence of KS in this group reached a peak of 211.6 in 1988 and declined slightly thereafter.

  Non-Hodgkin's lymphoma (NHL) is also common among AIDS patients. The overall incidence of NHL increased 72.8 percent between 1973 and 1991. For men under age 65, incidence increased 104.4 percent, while for women in this age group the increase was 46.7 percent. Not all of the increase in NHL incidence is explained by AIDS, however, particularly in older age groups. Incidence has been rising gradually over 40 years for reasons that are largely unknown and currently under investigation. Some studies have suggested that exposures to pesticides and other chemicals may play a role.

14. Which cancers show the highest incidence rates (based on all ages combined) for the period 1987-1991?

(Rates are per 100,000 and are age-adjusted to the 1970 U.S. population. Unless otherwise specified, rates are given for all races, both sexes.)

                                  Age <65         Age 65+         All Ages

     All Sites                      198.1           2,145.4            390.4

     SITE
     Prostate (men)                  26.7            1002.2            123.0
     Breast (women)                  72.8             444.7            109.5
     Lung and Bronchus               28.4             329.9             58.2
     Colon and Rectum                18.1             323.1             48.2

15. Which cancers show the greatest increases and decreases in incidence rates (based on all ages combined) over the interval 1973-1991?

                                           EAPC*            EAPC* / %Change**

                                   Age <65        Age 65+           All Ages

     All Sites                       +0.7          +1.6           1.2 / +22.4

     INCREASES BY SITE
     Prostate (men)                  +4.8          +3.8          +3.9 / +126.3
     Melanoma                        +3.5          +5.2          +3.9 / +94.0
     Non-Hodgkin's Lymphoma          +3.5          +3.2 /        +3.3 / +72.8
     Testis (men)                    +2.4          -1.7          +2.2 / +42.7
     Kidney and Renal Pelvis         +1.8          +2.4          +2.1 / +35.4
     Lung and Bronchus               +0.8          +2.6          +1.7 / +34.4
     Brain and Nervous System        +0.7          +2.5          +1.2 / +24.6
     Breast (women)                  +1.2          +2.6          +1.7 / +23.9


     DECREASES BY SITE
     Cervix Uteri (women)            -2.3          -3.6          -2.6 / -36.4
     Corpus Uteri (women)            -4.0          +0.2          -2.4 / -27.7
     Stomach                         -1.3          -1.6          -1.5 / -26.0
     Hodgkin's Disease               +0.2          -2.2          -0.2 / -11.7

*Estimated Annual Percent Change

**Percent change for interval 1973-1991 derived using average rates for 1973- 1974 and 1990-1991.

16. Which cancers show the highest mortality rates (based on all ages combined) for the period 1987-1991?

(Rates are per 100,000 of the entire U.S. population and are age-adjusted to the 1970 U.S. population. Unless otherwise specified, rates are given for all races, both sexes.)

                                  Age <65           Age 65+          All Ages
     All Sites                     74.9             1,066.3           172.8

     SITE
     Lung and Bronchus             22.9               290.0            49.3
     Breast (women)                16.6               125.8            27.3
     Prostate (men)                 2.9               232.3            25.6
     Colon and Rectum               6.4               135.3            19.1
     Ovary (women)                 10.0                58.6            14.8
     Pancreas                       3.1                57.2             8.4

17. Which cancers show the greatest increases and decreases in mortality rates (based on all ages combined) over the interval 1973-1991?

                                             EAPC*           EAPC* / %Change**

                                   Age <65          Age 65+          All Ages
     All Sites                      -0.3             +0.9         +0.4 / +6.9

     INCREASES BY SITE
     Lung and Bronchus              +0.9             +3.0         +2.1 / +47.7
     Melanoma                       +0.8             +3.0         +1.7 / +35.2
     Non-Hodgkin's Lymphoma         +0.7             +2.8         +1.9 / +33.2
     Multiple Myeloma               +0.5             +1.9         +1.5 / +31.1
     Kidney and Renal Pelvis        +0.3             +1.5         +0.9 / +16.9
     Brain and Nervous System       -0.5             +2.8         +0.7 / +15.8
     Esophagus                      +0.2             +1.3         +0.8 / +15.6


     DECREASES BY SITE
     Testis (men)                   -6.4             -5.8         -6.3 / -66.8
     Hodgkin's Disease              -4.7             -4.3         -4.5 / -56.5
     Cervix Uteri (women)           -2.9             -3.6         -3.1 / -41.9
     Stomach                        -1.9             -2.5         -2.3 / -32.1
     Corpus Uteri (women)           -3.0             -0.9         -1.7 / -24.6
     Thyroid Gland                  -1.8             -1.8         -1.8 / -23.4
     Urinary Bladder                -2.1             -1.6         -1.7 / -22.5

*Estimated Annual Percent Change

**Percent change for interval 1973-1991 derived using the average rates for 1973-1974 and 1990-1991.

18. What are the lifetime probabilities of developing the most common cancers in men and women? (based on 1989-1991 incidence rates)

     Men                                      Women

     Si                 Probability (%)       Site             Probability (%)

     All races

     All sites                   44.84        All sites                39.26
     Prostate                    15.44        Breast                   12.30
     Lung and Bronchus            8.49        Colon and Rectum          5.92
     Colon and Rectum             6.14        Lung and Bronchus         5.17
     Bladder (includes in situ)   3.29        Corpus Uteri and
                                                 Uterus,NOS*            2.64
     Non-Hodgkin's Lymphoma       1.82        Ovary                     1.79


     Whites

     All sites                   45.54        All sites                40.25
     Prostate                    15.57        Breast                   12.86
     Lung and Bronchus            8.61        Colon and Rectum          5.99
     Colon and Rectum             6.28        Lung and Bronchus         5.34
     Bladder (includes in situ)   3.64        Corpus Uteri and
                                                 Uterus, NOS*           2.79
     Non-Hodgkin's Lymphoma       1.93        Ovary                     1.90


     Blacks

     All sites                   37.47        All sites                32.01
     Prostate                    13.11        Breast                    8.80
     Lung and Bronchus            8.69        Colon and Rectum          5.21
     Colon and Rectum             4.38        Lung and Bronchus         4.81
     Oral Cavity and Pharynx      1.53        Pancreas                  1.40
     Stomach                      1.43        Cervix Uteri              1.29


     *not otherwise specified

19. What are the probabilities of dying of the most common causes of cancer death in men and women? (based on 1989-1991 mortality rates)

     Men                                      Women

     Site             Probability (%)        Site             Probability (%)

     All races

     All sites               23.35            All sites               20.39
     Lung and Bronchus        7.09            Lung and Bronchus        4.09
     Prostate                 3.41            Breast                   3.59
     Colon and Rectum         2.60            Colon and Rectum         2.65
     Pancreas                 1.08            Pancreas                 1.18
     Stomach                  0.86            Ovary                    1.07


     Whites

     All sites               23.59            All sites               20.81
     Lung and Bronchus        7.22            Lung and Bronchus        4.22
     Prostate                 3.36            Breast                   3.73
     Colon and Rectum         2.68            Colon and Rectum         2.68
     Pancreas                 1.09            Pancreas                 1.17
     Non-Hodgkin's Lymphoma   0.92            Ovary                    1.15


     Blacks

     All sites               22.98             All sites              19.01
     Lung and Bronchus        7.17             Lung and Bronchus       3.82
     Prostate                 4.17             Breast                  3.26
     Colon and Rectum         2.04             Colon and Rectum        2.63
     Stomach                  1.09             Pancreas                1.28
     Esophagus                0.97             Stomach                 0.66

Lifetime probabilities are calculated by assuming current age-specific incidence and mortality rates will hold in the future. The probability estimates are therefore very sensitive to current rates. They do not take into account future changes in rates which are likely to occur. In evaluating cancer risk for a cancer-free individual at a given age, age-specific "interval probabilities" are more relevant than lifetime probabilities. Interval probabilities for specific cancers and for all cancers combined are included in the CSR. For example, a 50-year-old cancer-free black woman has a 6.4 percent chance of developing breast cancer before age 80.

For some cancers, age-specific incidence rates are higher for blacks than whites, while lifetime probabilities are higher for whites than blacks. This apparent contradiction is explained by the fact that blacks have higher mortality rates from other causes, with the result that fewer blacks survive to the older ages at which cancer rates rise rapidly.

20. Which states had the highest and lowest cancer mortality rates for all cancer sites combined for the period 1987-1991?

(Rankings range from highest, 1, to lowest, 51.)

(Rates are per 100,000 of the entire U.S. population and are age-adjusted for both sexes, all races.)

                                            Rate            Rank
          Total U.S.                       172.8             --

          STATE

          Delaware                         195.7              2
          Louisiana                        192.6              3
          Maryland                         192.2              4
          Kentucky                         190.5              5

          Colorado                         147.7             48
          New Mexico                       146.0             49
          Hawaii                           137.1             50
          Utah                             125.4             51

21. Which cancer sites show the highest and lowest five-year relative survival rates (based on all ages combined) for patients diagnosed during the period 1983-1990?

                                       % Survival


                                Age <65             Age 65+           All Ages

        All Sites                57.5                50.3               53.9

        SITE

        Testis (men)             93.4                80.2               93.3
        Melanoma                 86.2                82.4               85.1
        Corpus Uteri (women)     88.0                78.1               83.9
        Breast (women)           79.2                82.2               80.4
        Urinary Bladder          85.9                75.4               79.8
        Hodgkin's Disease        83.3                40.1               78.9


        Chronic Myeloid          32.7                12.4               23.7
                Leukemia
        Stomach                  20.3                17.3               18.5
        Lung and Bronchus        15.5                11.3               13.4
        Acute Myeloid            17.3                 1.9               10.4
              Leukemia
        Esophagus                10.3                 8.2                9.2
        Liver                     8.8                 3.4                6.0
Lowest   Pancreas                 4.9                 2.3                3.2

22. Are there differences in cancer incidence, mortality, and survival between sexes?

For all cancers combined, the cancer incidence rate (1987-1991) is higher in men than women: 465.7 cases per 100,000 men and 342.5 cases per 100,000 women. Men have higher incidence rates for nearly every cancer common to both sexes, except breast cancer where the rate in women is 120 times higher. Large differences in incidence between the sexes are seen in cancer of the lung and bronchus (men twice as high) and cancer of the urinary bladder (men 4 times higher).

For all cancers combined, the U.S. cancer mortality rate (1987-1991) is higher in men than women: 220.2 deaths per 100,000 men and 141.1 deaths per 100,000 women. Men have higher mortality rates for nearly every cancer common to both sexes. Large differences in mortality between the sexes are seen in cancer of the lung and bronchus (men 2.5 times higher), cancer of the esophagus (men 4 times higher), cancer of the oral cavity (men nearly 3 times higher), and cancer of the urinary bladder (men 3 times higher).

The five-year relative survival rate (for cancers diagnosed between 1983-1990) for all cancers combined is 49.0 percent for men and 58.6 percent for women.

23. Are there differences in cancer incidence, mortality, and survival between blacks and whites?

While the age-adjusted incidence rate for all cancers combined increased similarly for whites and blacks between 1973 and 1991, the overall incidence rates (1987-1991) are different: 422.1 cases per 100,000 blacks and 392.0 cases per 100,000 whites. These rates indicate that blacks bear a disproportionate burden of cancer.

Mortality rates for all cancers and all ages combined during the interval 1973-1991 show an increase of 16.4 percent for blacks, compared with an increase of 6.2 percent for whites. Mortality rates (1987-1991) are 226.2 deaths per 100,000 blacks and 169.0 deaths per 100,000 whites.

For the period 1983-1990, whites had more than a 10 percentage point advantage in five-year survival for cancers of the breast, cervix uteri, colon, corpus uteri, larynx, oral cavity, prostate, rectum, urinary bladder, and melanoma, as well as all cancer sites combined. The five-year relative survival rates (for cancers diagnosed between 1983-1990) for all cancers combined are 55.5 percent for whites and 40.4 percent for blacks.

24. Does the CSR tell us anything about geographic patterns of cancer?

Yes, cancer mortality rates by state for all cancers combined and for specific cancers are presented in the 1994 CSR.

NCI has been publishing a series of cancer atlases since 1975, containing color maps of U.S. mortality data that visually represent cancer patterns. The most recent volumes are the "Atlas of U.S. Cancer Mortality Among Whites," 1950-1980, published in 1987, and the "Atlas of U.S. Cancer Mortality Among Nonwhites, 1950-1980," published in 1991.