Search for a command to run...
Cancer is an important public health concern in the United States and around the world. In order to provide an up-to-date perspective on the occurrence of cancer, we present an overview of cancer incidence, mortality, and survival statistics for 1997. Estimated New Cancer Cases: We estimated the number of new cancer cases that we expect to be diagnosed in the coming year using population data collected by the US Bureau of the Census and cancer incidence data collected by the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program.1-3 Estimates were calculated using a two-step procedure. First, we multiplied cancer incidence rates for 1979 through 1993 by US Census Bureau population projections for the same years to estimate the number of cancer cases diagnosed annually from 1979 to 1993.1-3 Next, we fitted these annual case estimates to a quadratic function using the SAS procedure PROC FORECAST to project the number of cancer cases expected to be diagnosed in 1997.4, 5 Some additional adjustments were made for sites with recently changing incidence rates or with widely varying year-to-year estimates. These sites included rectum, pancreas, and other leukemia in males and females; larynx, prostate, chronic lymphocytic leukemia, and anus, anal canal, and anorectum in males; and lung and bronchus, corpus and other uterus, eye and orbit, and Hodgkin's disease in females. Because cancer incidence rates for 1979 through 1993 were not available for many states, we used state-specific data on cancer deaths to estimate the number of new cancer cases occurring in each state. First, using methods described in the section below, we estimated the number of 1997 cancer deaths expected to occur in each state and in the United States as a whole. Next, we used these US and state estimates to calculate the proportion of cancer deaths expected to occur in each state. Finally, we estimated the number of cancer cases for each state by multiplying the proportion of cancer deaths expected to occur in each state in 1997 by the estimated number of new cancer cases for the United States for the same year. Estimated Cancer Deaths: We estimated the number of US cancer deaths expected to occur in 1997 using data on underlying cause of death gathered from death certificates by the National Center for Health Statistics (NCHS).6 Data on the number of cancer deaths occurring annually from 1979 to 1993 were fitted to a quadratic model using the SAS procedure PROC FORECAST. This model was used to forecast the number of cancer deaths expected to occur in 1997.4, 5 Some estimates were adjusted slightly to compensate for the effect of recently changing mortality rates or large year-to-year variations in estimates. These sites included rectum, pancreas, larynx, bones and joints, other nonepithelial skin, breast, testis, urinary bladder, other endocrine, and acute myelocytic leukemia in men; and stomach, cervix uteri, corpus and other uterus, and thyroid in women. The estimated number of cancer deaths for each state was calculated with the same modeling procedure that was used to estimate cancer deaths for the United States as a whole. Other Statistics: Statistics on cancer and noncancer mortality, the probability of developing cancer, and relative survival are also presented in this report (Figs. Figure 3-Figure 6, Tables 5–14). These statistics have been assembled from several sources, and methods used to calculate them have been previously described.7 Expected Numbers of New Cancer Cases: In 1997, we estimate that about 1,382,400 new cases of invasive cancer are expected to be diagnosed in the United States (Table 1). This estimate does not include carcinoma in situ of any site but bladder, nor does it include basal and squamous cell cancers of the skin. Over 900,000 cases of basal and squamous cell skin cancers, 36,400 cases of breast carcinoma in situ, and 20,100 cases of melanoma carcinoma in situ are expected to be diagnosed in 1997. Among women, we estimate that in 1997 the three most commonly diagnosed cancers will be cancers of the breast, lung and bronchus, and colon and rectum (Fig. 1). There will be about 324,800 new cancers occurring at these sites, accounting for over 50 percent of new cancer cases. Breast cancer alone will account for 30 percent of new cancer cases with about 180,200 new cases to be diagnosed in 1997. Among men, the most common cancers in 1997 will be cancers of the prostate, lung and bronchus, and colon and rectum (Fig. 1). Prostate is the leading cancer site, accounting for 43 percent of new cancer cases. Between 1990 and 1997, our estimates for prostate cancer cases increased sharply; this increase reflects the dramatic increase in prostate cancer incidence rates between 1988 and 1993. Expected Numbers of Cancer Deaths: In 1997, we estimate that about 560,000 Americans will die of cancer — more than 1,500 people a day (Table 2). Preliminary data for 1995 suggest that overall cancer mortality rates have recently begun to decline.9 Numbers of deaths, however, have continued to increase due to the aging population. Among women, we estimate that cancers of the lung and bronchus, breast, and colon and rectum will account for over half of all cancer deaths in 1997 (Fig. 2). Lung is the leading cancer site, accounting for 25 percent of all cancer deaths. Since 1987, lung cancer has been the leading cause of cancer death in women. Prior to that time, more deaths were caused by breast cancer, which for over 40 years was the major cause of cancer death in women.1, 6 In 1997, breast cancer is expected to be responsible for 17 percent of female cancer deaths. The overall breast cancer mortality rate has been declining since 1989. Between 1989 and 1993, breast cancer mortality declined among white women aged 80 years and younger and among African-American women aged 70 years and younger.1, 6 Among men, most 1997 cancer deaths are expected to be caused by cancers of the lung and bronchus, prostate, and colon and rectum (Fig. 2). Like women, men are more likely to die of lung cancer than cancer of any other site. In 1997 alone, lung cancer is expected to be responsible for 98,300 deaths in men (32 percent). Responsible for about half as many deaths as lung cancer, prostate cancer is expected to cause about 14 percent of cancer deaths in 1997. Recent increases in prostate cancer mortality rates have been much less dramatic than increases in prostate cancer incidence rates. Consequently, our estimated number of prostate cancer deaths increased only 39 percent between 1990 and 1997 compared to over 200 percent for our estimated number of prostate cancer cases. Our estimated numbers of new cancer cases and cancer deaths should be interpreted with caution when used to study trends in cancer incidence and mortality. Numbers can vary considerably from year to year, particularly for less common cancers and for states with small populations. NCHS mortality rates and SEER and state cancer incidence rates are generally more informative statistics to use when tracking cancer trends. Changes in incidence or mortality trends that have occurred since 1993 are not reflected in this year's estimates. For example, regional reports suggest that the rapid increase in prostate cancer incidence rates observed between 1988 and 1992 may have begun to slow or change in 1993.1, 8 If this is correct, the estimates of new prostate cases that we have published in recent years will be higher than the actual number of new cases. In addition, our estimates may not reflect fluctuations in trends occurring between 1979 and 1993. For example, although breast cancer incidence rates increased about one percent per year between 1979 and 1982 and four percent per year between 1982 and 1987, rates were about constant between 1987 and 1993. Despite the stabilization of rates during the latter time period, our estimates for new breast cancer cases continued increasing between 1988 and 1986. Despite these limitations, our estimates do provide an indication of current patterns of cancer in the United States. Continuing efforts in the areas of prevention, screening, and treatment are necessary to reduce the burden of cancer in the US and world populations as we approach the 21st century. Estimated New Cancer Cases* 10 Leading Sites by Sex, United States, 1997 Estimated Cancer Deaths* 10 Leading Sites by Sex, United States, 1997 Age-Adjusted Cancer Death Rates* Females by Site, United States, 1930–1993 Age-Adjusted Cancer Death Rates* Males by Site, United States, 1930–1993 Percent Distribution of Cancer Cases By Race and Stage at Diagnosis, United States, 1986–1992 Five-Year Relative Survival Rates By Race and Stage at Diagnosis, United States, 1986–1992
Published in: CA A Cancer Journal for Clinicians
Volume 47, Issue 1, pp. 5-27