Robert T. Greenlee, PhD, MPH
Mary Beth Hill-Harmon, MSPH
Taylor Murray
Michael Thun, MD, MS

Abstract

Each year the American Cancer Society compiles estimates of the number of new cancer cases and deaths expected in the US in the current year and the most recent data on cancer incidence, mortality, and survival. An estimated 1,268,000 new cases of cancer will be diagnosed in the year 2001 and an estimated 553,400 Americans will die from cancer. Overall cancer incidence and death rates have continued to decrease in men and women since the early 1990s, and the decline in overall cancer mortality has been greater in recent years. Despite reductions in age-adjusted rates of cancer death, the total number of recorded cancer deaths in the US continues to increase, due to an aging and expanding population. Large disparities in cancer incidence and mortality across racial/ethnic groups continue. Black men and women experience higher incidence of cancer and poorer survival than white men and women. The disparity in survival reflects both diagnosis of cancer at later disease stages, and poorer survival within each stage of diagnosis. (CA Cancer J Clin 2001;51:15-36.)

Introduction

Despite the decrease in overall cancer incidence and mortality rates in the US since the early 1990s, cancer remains a major public health problem. To provide an up-to-date perspective on the occurrence of cancer, the American Cancer Society presents an overview of cancer frequency, incidence, mortality, and survival statistics for the year 2001.

Methods

Estimated New Cancer Cases

Because the US does not have a nationwide cancer registry, the exact number of new cases of cancer diagnosed each year in the US and in all individual states is not known. Consequently, we first estimated the number of new cancer cases occurring annually in the US from 1979 through 1997 (Table 1) using population data reported by the US Bureau of the Census and age-specific cancer incidence rates collected by the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program.¹ We fit these annual cancer case estimates with an autoregressive quadratic model to forecast the number of cancer cases expected to be diagnosed in the US in the year 2001.²

The observed trend in prostate cancer incidence was not compatible with the selected forecasting model, as rates increased markedly between 1988 and 1992, declined sharply between 1992 and 1995, and leveled off from 1995 to 1997.³ This trend likely reflects extensive use of prostate-specific antigen (PSA) screening in a previously unscreened population and the subsequent increase in diagnoses at an early stage.⁴ We therefore assumed that the number of prostate cancer cases is approaching the pattern in effect prior to widespread use of PSA screening, and estimated new cases of prostate cancer for 2001 using a linear projection based on data from 1979 to 1989 and 1995 to 1997 only.

Because cancer incidence rates and case counts for 1979 through 1997 were not available for many states, we could not use the methods mentioned earlier to estimate new cases for individual states. To derive these estimates, we relied on state cancer death data and assumed that the ratio of cancer deaths to cancer cases for each state was the same as the ratio for the US.²

Estimated Cancer Deaths

We estimated the number of cancer deaths expected to occur in the US and in each state in the year 2001 using underlying cause-of-death data from death certificates, as reported to the National Center for Health Statistics.⁵ The recorded numbers of cancer deaths occurring annually from 1979 to 1998 in the US and in each state were fit with autoregressive quadratic models to forecast the number of cancer deaths expected to occur in 2001.

Mortality statistics for the leading causes of death, the leading causes of death from cancer, and age-adjusted cancer mortality rates for 1930 to 1997 were obtained using data from the National Center for Health Statistics.⁵ Age-adjusted cancer incidence rates, the probability of developing cancer, and five-year relative survival rates were obtained from the SEER program.^3,6

Selected Findings

Expected Numbers of New Cancer Cases

In the year 2001, we estimate that about 1,268,000 new cases of invasive cancer will be diagnosed in the US (Table 1). This estimate does not include carcinoma in situ of any site except urinary bladder, and it does not include basal and squamous cell cancers of the skin. More than a million cases of basal and squamous cell skin cancers, 46,400 cases of breast carcinoma in situ, and 31,400 cases of in situmelanoma are expected to be newly diagnosed in 2001. Estimated numbers of new cancer cases by state are shown in Table 2.

Among men, the most common cancers in 2001 are expected to be cancers of the prostate, lung and bronchus, and colon and rectum (Fig. 1). The prostate is the leading site for cancer incidence, accounting for 31% of new cancer cases in men. This year 198,100 new cases of prostate cancer are expected to be diagnosed.

Among women, the three mostly commonly diagnosed cancers are expected to be cancers of the breast, lung and bronchus, and colon and rectum (Fig. 1). Cancers occurring at these sites are expected to account for over 50% of new cancer cases in women. Breast cancer alone is expected to account for 192,200 new cancer cases (31%) in 2001.

Lifetime Probability of Developing Cancer

The lifetime probability of developing cancer is higher in men (43.48%) than in women (38.34%) (Table 3), but women have a higher probability than men of developing any cancer before age 60. In general, however, the probability of developing specific cancers is higher in men than women at all ages, with the exception of breast cancer and cancers specific to women.

Trends in Cancer Incidence

For all sites combined, age-adjusted cancer incidence rates declined an average of 1.3% per year from 1992 to 1997, reversing increasing trends in earlier years (Fig. 2).⁷ However, the overall decline has been limited to incidence rates among men, which were heavily influenced by trends in prostate cancer incidence. Declines have also been seen recently for several other leading cancer sites (Fig. 3 and Fig. 4).

Breast cancer incidence remained approximately level during the 1990s, but may appear to be decreasing in younger women.³ Colon and rectum cancer incidence began to decline in 1985, on average 1.6% per year through 1997. Decreases in colon and rectum cancer incidence rates have been observed among males and females in all racial/ethnic groups, (with the exception of American Indian women for whom data were not sufficient to make a determination as to the direction of the trend.)³

A significant downturn in the incidence of lung and bronchus cancer in males began in the 1980s; between 1992 and 1997, incidence rates decreased 3.2% per year. Overall incidence rates of female lung and bronchus cancer have been stable since 1991, but rates have begun to decline among women aged 40 to 59.⁸

Prostate cancer incidence was essentially level from 1995 to 1997, following large annual increases of 17.5% from 1988 to 1992 and a sharp decline of 10.3% per year from 1992 to 1995.³

Expected Numbers of Cancer Deaths

In 2001, an estimated 553,400 Americans are expected to die of cancer—more than 1,500 people a day (Table 4). The estimated numbers of cancer deaths in 2001 by state are shown in Table 5.

Also shown in Table 5 is the reported death rate by state for the years 1993 to 1997 for all cancers combined. The rate varied considerably by state, with the lowest rate in Utah (119 per 100,000 per year) and the highest rate in Washington, DC (208 per 100,000 per year). The average age-adjusted cancer death rate across the entire US was 168 per 100,000 per year.

Cancers of the lung and bronchus, prostate, and colon and rectum combined are expected to contribute 52% of cancer deaths among men in the year 2001 (Fig. 5). Among women, cancers of the lung and bronchus, breast, and colon and rectum are expected to account for 51% of all cancer deaths in 2001 (Fig. 5). Lung cancer has surpassed breast cancer as the leading cause of cancer death in women since 1987 and is expected to account for 25% of all female cancer deaths in 2001.

Trends in Cancer Death Rates

The death rate for all cancers combined peaked in 1991 and then decreased on average 0.6% per year until 1995. The decline was even more marked between 1995 and 1997, at 1.7% per year (Fig. 6).³ Significant decreases in cancer death rates have occurred among males of all ages, and among females younger than 75 years old.⁷

The death rates associated with many of the common cancer sites have also been decreasing (Figs. 7 and Fig. 8). The breast cancer death rate among females decreased an average of 2.2% per year between 1990 to 1997; decreases were more pronounced among white women and among younger women. Colon and rectum cancer death rates have been decreasing 1.8% per year on average since 1984, with equally strong declines among men and women.³ Similar to trends in incidence, significant decreases in death rates for lung and bronchus cancer have occurred only among males (on average, 1.7% per year between 1990 and 1997); however, the increase in lung cancer death rates among femails has begun to slow recently. Prostate cancer death rates stopped increasing in 1991, and decreased an average of 4.4% annually from 1994 through 1997.³
Recorded Numbers of Deaths from Cancer and Other Causes

Among fatal conditions in the US, all cancers combined rank second highest, following only heart disease (Table 6). Cancer accounted for 23.2% of all deaths in 1998. Upon examination of causes of death by age and gender, cancer is by far the leading cause of death among women aged 40 to 59 and also ranks first among women aged 60 to 79 (Table 7). The lowest that cancer ranks as a cause of death for any age-gender group is fifth, among men ages 20 to 39.

Table 8 and Table 9 present the leading site-specific causes of cancer death for males and females according to age. Leukemia is the most common cause of cancer death among men under age 40, while lung and bronchus cancer is the leading cause of cancer death for men 40 years and older. Prostate cancer is the second most common cause of cancer death among men 60 years and older. Among women, leukemia is the leading cause of cancer death before age 20, but breast cancer and uterine cervical cancer each cause more deaths among women between 20 and 39 years old. Breast cancer is the leading cause of cancer death among womenaged 40 to 59. Lung cancer causes the most cancer deaths among women 60 years and older.

The decrease in the total number of cancer deaths that occurred between 1996 and 1997 among men in the US was not sustained. The 282,065 deaths recorded in 1998 represented an increase of 955 deaths from the previous year (Table 10). . The number of prostate cancer deaths continued to decline, however, from a peak of 34,902 in 1994 to 32,203 in 1998. The 91,399 lung and bronchus cancer deaths among men in 1998 remain lower than in the peak year of 1993. Colon and rectum cancer deaths among men were highest in 1990, and were slightly lower in 1998 at 28,024.

Among women, the recorded number of total cancer deaths continues to increase, with 259,467 deaths recorded in 1998, although the rate of increase has diminished recently (Table 10). The upward trend among females is primarily due to sustained increases in the number of deaths from lung and bronchus cancer. The numbers of female deaths from breast cancer, however, have begun to decline. Breast cancer deaths were highest in 1995 at 43,844 and have declined to 41,737 in 1998. The number of colorectal cancer deaths among females has remained fairly constant in recent years.

Cancer Occurrence by Race/Ethnicity

Overall rates of cancer incidence vary considerably among racial and ethnic groups (Table 11). African Americans have the highest incidence rates of cancer. They are about 60% more likely to develop cancer than Hispanics and Asian/Pacific Islanders and more than twice as likely to develop cancer than American Indians. Between 1990 and 1997, incidence rates for all cancers combined decreased among Caucasians (1.0% per year), Hispanics (1.6% per year), American Indians (0.6% per year), and blacks (0.5% per year), but remained relatively stable among Asian/Pacific Islanders.³

White women are more likely to develop breast cancer than are women of other racial and ethnic groups, and black women are more likely to develop cancers of the colon and rectum.³ Incidence rates for lung and bronchus cancer are similar among white and black women. Black men have the highest incidence rates for cancers of the colon and rectum and lung and bronchus, and incidence rates of prostate cancer among black men are at least 50% higher than rates for men of other racial and ethnic groups.

African Americans are about 33% more likely to die of cancer than are whites, and more than twice as likely to die from cancer as are Asian/Pacific Islanders, American Indians, and Hispanics. Between 1990 and 1997, mortality rates decreased significantly among whites
0.7% per year), blacks (1.0% per year), Hispanics (0.9% per year), and Asian/Pacific Islanders (0.8% per year), but may be increasing among American Indians.³

Black women are more likely to die of breast and colon and rectum cancers than are women of any other racial or ethnic group, and lung and bronchus cancer death rates are particularly high for both black and white women compared with other racial or ethnic groups. As was seen with incidence rates, black men have the highest death rates of colon and rectum, lung and bronchus, and prostate cancer.³

Cancer Survival

Contributing to the higher death rates among black men and women is a poorer probability of survival once diagnosed with cancer. Blacks are less likely than whites to be diagnosed with cancer at a localized stage, when the disease may be more easily and successfully treated, and more likely to be diagnosed with cancer at a regional or distant stage of disease. This is true for most of the common cancer sites (Fig. 9). Furthermore, for nearly every cancer site, blacks have lower five-year relative survival rates than whites at each stage of diagnosis, suggesting possible influences of differences in treatment, tumor pathology, and comorbid conditions (Fig. 10).

There have been notable improvements over time in the probability of survival from most common cancers and from all cancers combined (Table 12). This is true for both whites and blacks. Survival has not significant-ly improved for cancers of the uterine cervix, larynx, and oral cavity in the past 25 years.

Cancer in Children

Cancer is the second leading cause of death among children between one and 14 years of age in the US; accidents are the most frequent cause of death in this age group (Table 13). The most commonly occurring cancers in children are leukemias in particular, acute lymphocytic leukemia), tumors of the central and sympathetic nervous systems, lymphomas, soft-tissue sarcomas, and renal tumors.³ Over the past 25 years, there have been significant improvements in the five-year relative survival rate for many childhood cancers, especially acute lymphocytic and acute myeloid leukemia, non-Hodgkin's lymphoma, and Wilms' Tumor (Table 14). Between 1974/1976 and 1989/1996, five-year relative survival rates among children for all cancer sites combined improved from 56% to 75%.³

Estimates of the expected numbers of new cancer cases and cancer deaths should be interpreted with caution when tracking trends over time. These estimates may vary considerably from year to year, particularly for less common cancers and for states with smaller populations. We discourage the use of these estimates to track year-to-year changes in cancer occurrence and death. The recorded number of cancer deaths and cancer death rates from the National Center for Health Statistics, and SEER cancer incidence rates are generally more informative statistics for tracking cancer trends. For example, breast cancer incidence rates increased about 1% per year between 1979 and 1982, increased 4% per year between 1982 and 1987, and were approximately constant between 1987 and 1996. Despite the stabilization of incidence rates during the latter time period, the estimates of new breast cancer cases continued to increase between 1988 and 1996, partly due to the residual effects of the strong rate increases through 1987.

Our estimates are based on the most currently available cancer mortality and incidence data; however, these data are three and four years old, respectively, at the time that the estimates are calculated. As such, the effect of large changes occurring in the three or four-year interval between 1997 or 1998 and 2001 cannot be captured by our modeling efforts. Finally, our estimates of new cancer cases are based on incidence rates for the geographic locations that participate in the SEER program and, therefore, may not be representative of the total US.

Despite these limitations, the American Cancer Society estimates do provide evidence of the current patterns of cancer incidence and mortality in the US. Such estimates will assist our continuing efforts to reduce the public health burden of cancer in 21st century.

Ms. Hill-Harmon is an Epidemiologist in the Department of Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA.

Mr. Murray is Manager, Surveillance Data Systems, in the Department of Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA.

Dr. Thun is Vice-President of Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA.

The authors thank Cheryll Cardinez, Vilma Cokkinides, PhD, April Harris, Elyse Luke, and Kate O'Brien for their assistance in preparation of this manuscript.

References

1. National Cancer Institute. SEER Cancer Incidence Public-Use Database, 1975-1997, August 1999 Submission. Bethesda, MD, US Department of Health and Human Services, Public Health Service, 2000.

2. Wingo PA, Landis S, Parker S, et al. Using cancer registry and vital statistics data to estimate the number of new cancer cases and deaths in the United States for the upcoming year. J Reg Management 1998;25:43-51.

3. Ries LAG, Eisner MP, Kosary CL, et al. (eds). SEER Cancer Statistics Review, 1973-1997. Bethesda, MD, National Cancer Institute. 2000.

4. Wingo PA, Landis S, Ries LAG. An adjustment to the 1997 estimate for new prostate cancer cases. CA Cancer J Clin 1997;47:239-242.

5. National Center for Health Statistics, Division of Vital Statistics. Multiple Cause-of-Death (inclusive of Underlying Cause of Death) for ICD-9 1997 Data Public-Use Documentation. (Web site) http://www.cdc.gov/nchs/about/ major/dvs/mcd/1997mcd.htm. 2000.

6. Feuer EJ, Wun LM. DEVCAN: Probability of Developing or Dying of Cancer (Software), version 4. Bethesda, MD, National Cancer Institute, 2000.

7. Ries LAG, Wingo PA, Miller DS, et al. Annual report to the nation on the status of cancer, 1973-1997, with a special section on colorectal cancer. Cancer 2000; 88:2398-2424.

8. Wingo PA, Ries LAG, Giovino GA, et al. Annual report to the nation on the status of cancer, 1973-1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999; 91:675-690.