DATAWATCH 121 Increasing The Odds For Cancer Survival Conquering cancer, the nation's second leading cause of death, is a high priority in the United States. Although the federal government's "war on cancer," declared in 1971, has proven to be an intractable battle, statisticians continue to report that the number of people diagnosed as having cancer who then survive is increasing. 1 Experts predict that data scheduled for release at the end of 1984 will show the five-year survival rates for cancer surpassing 50 percent for the first time. The most promising improvements in survivability have come among children afflicted with cancer. These statistics, however, are subject to different interpretations, as The New York Times reported September 18 in an article entitled, "Cancer Progress: Are the Statistics Telling the Truth?" Reporter Phillip M. Boffey canvassed the views of researchers, some of whom argue that cancer survival statistics show inflated gains in the war against cancer. These analysts claim that the statistics have been manipulated to show artificial progress by three main factors: (1) diagnosis techniques have improved to the point where tumors are being diagnosed today that have the characteristics of cancer and are included in the cancer counts, but which actually would not result in death, thereby increasing the survival counts; (2) earlier detection of cancer shows more people with some forms of cancer (particularly breast cancer) who appear to survive five years after diagnosis since the diagnosis was determined at an earlier point in the cancer's development; and () improvements in recording nonfatal cancer cases as compared to cancer deaths may also be a factor in the increasing survival rates. Vincent T. DeVita, director of the National Cancer Institute discounts the skeptics' claims, saying in the Times article, "I think it's a bunch of nonsense. We're saving thousands of lives today that weren't saved twenty years ago. To me, that's pretty damn exciting." Although other leaders in the cancer field acknowledge some validity to these claims, they assert that the error in the reported statistics is not so great as to reverse the charted trends. A combination of increased funding and improved research efforts, stemming in part from enactment of the National Cancer Act of 1971, and some key discoveries about cancer genes through recombinant DNA technologies, have led to the improvements in cancer survival rates. Congress increased appropriations for the National Cancer Institute (NCI) from $20 million in 1971 to $1 billion by 1980. 2 The number of medical oncologists practicing in the United States increased from 100 in the late 1960s to about 2,800 by 1980. Dr. Henry C. Pitot, chairman of the National Cancer Advisory Board, has noted that, "Improved cancer survival rates, and decreases in cancer deaths for patients under age forty-five, occurred in part because of the community-based oncologists who see many can-
122 HEALTH AFFAIRS cer patients themselves, and extend their expertise by consulting with general practitioners who treat still other patients." 4 The rapidity of progress in solving the mysteries of cancer has surprised even optimistic members of the research community. Lewis Thomas, President Emeritus of the Memorial Sloan-Kettering Cancer Center, recently wrote in Health Affairs: "In the late 1960s, if any young post-doctoral student or M.D. had asked me about the advisability of going into cancer research, my advice would have been to stay away, and instead to pick a field such as immunology or molecular biology, in which things are clearly moving along And then, still in the early 1970s, things began to change at a great rate, and they have been changing with stunning speed ever since... Cancer is beginning to look like an approachable problem, even (although at this point everyone tends to speak softly) a soluble problem." 5 Data released in November 198 estimate that 48 percent of all cancer patients are now curable. These statistics, shown in Exhibit 1, are from the National Cancer Institute's Surveillance, Epidemiology, and End-Results (SEER) Program. This program, begun in 197, monitors the annual occurrence and survival rates of cancer in the United States. Data for the SEER program are collected for registries in eleven geographic areas, which, in 1984, included: Connecticut, Iowa, New Mexico, Utah, Hawaii, Atlanta, New Jersey, Detroit, San Francisco-Oakland, Seattle-Puget Sound, and Puerto Rico. These registries represent 12 percent of the total U.S. population. The SEER program data, which are Exhibit 1 General Five-Year Cancer Relative Survival Rates Source: Surveillance, Epidemiology, and End Results (SEER) Program, National Cancer Institute, Biometry Branch, 1982 and 198 reports; Cancer Patient Survival Report No. 5, 1976; American Cancer Society, "Cancer Incidence, Survival and Mortality for Children Under 15 Years of Age," September 1978. a Races for 197-1980 data include: Anglos, Hispanics, Blacks, American Indians, Chinese, Japanese, Filipinos, and Hawaiians. Data from 1967-7 include only Blacks and Caucasians.
DATAWATCH 12 updated annually, are considered the major source of information on the incidence of cancer in the United States. Exhibit 2 documents the wide range of survival rates for the various types of cancers. For some major cancers, such as breast, prostate, melanoma, and testis, more than two-thirds of the patient population is expected to survive five years after diagnosis. The five-year mark is a generally accepted indicator of curability. Except for cancers of the breast, prostate, and kidney, most patients who survive five years after being diagnosed as having cancer are considered cured. The SEER rates are calculated by an actuarial, or life-table, method and thus include information on patients observed for less than five years. A complete five-year follow-up was conducted for those patients diagnosed during the years 197 to 76. The survival rates released in November 198 were revised upward because patients diagnosed in 1980 were included and an additional year of follow-up was completed 1981. As shown in Exhibit 1, the survival rates for black cancer patients lag considerably behind those of whites. Exhibit breaks down the sur- Exhibit 2 Five-Year Relative Survival Rates For Selected Sites For White Cancer Patients, Diagnosed During The Years Site Endometrium Testis Melanoma of the skin Breast (females) Bladder Hodgkins Cervix Prostate Colon Kidney Rectum Non-Hodgkins Ovary Leukemia Brain Stomach Lung Esophagus Pancreas 1960-6 a 7 6 60 6 5 40 58 50 4 7 8 1 2 18 11 8 4 1 1970-7 a 81 72 61 67 64 6 49 45 41 6 22 20 1 10 4 2 197-80 b 88 82 79 74 7 70 51 50 48 Source: Vincent T. DeVita, "Update: Cancer Patient Survival Statistics," (Bethesda, Md.: National Cancer Institute, Office of Cancer Communications, November 198). a Rates are based on data from a series of hospital registries and one population-based registry summarized in Cancer Patient Survival Experience. b Rates are from the SEER Program. They are based on data from population-based registries in Connecticut, New Mexico, Utah, Iowa, Hawaii, Atlanta, Detroit, Seattle-Puget Sound, and San Francisco-Oakland. They include patients diagnosed through 1980 and follow-up on all patients through 1981. 7 2 21 12 5
124 HEALTH AFFAIRS Exhibit Comparative Five-Year Relative Survival Rates For Selected Sites For White And Black Cancer Patients, Diagnosed During 197-80 Site Endometrium Testis Melanoma of the skin Breast (female) Bladder Hodgkins Cervix Prostate Colon Kidney Rectum Non-Hodgkins Ovary Leukemia Brain Stomach Lung Esophagus Pancreas Black 58 67 * 62 48 70 62 58 45 52 8 9 27 2 10 White 88 82 79 74 7 70 51 50 48 7 2 21 12 5 Source: Rates are from the SEER Program and reported in Vincent T. De Vita, "Update: Cancer Patient Survival Statistics," (National Cancer Institute, Office of Cancer Communications, November 198.) *Rates could not be calculated due to insufficient number of cases. vival rates for a number of major cancers for patients diagnosed during 197-80. One of the major success stories in cancer, however, has been the great improvement in survivability for children with cancer. Exhibit 4 charts the upward trends for a variety of children's cancers. Dr. Guilio D'Angio, Director of the Cancer Center at Children's Hospital of Philadelphia has said, "Now more than half of the children with a variety of cancers can expect to live a normal life free of disease." 6 Advances in surgical techniques and radiation therapy in combination with new cancer drugs, or chemotherapy, have fueled the increases in cures for childhood cancer victims. Fine-tuning to decrease the side effects of cancer treatments is still needed, though. Many of cancer's mysteries remain unsolved, and a glimpse at some cancer mortality data shows higher death rates. Data released in September 198 in U.S. Cancer Mortality Rates and Trends: 1950-1979 by the Environmental Protection Agency and NCI indicate that total cancer death rates have increased slightly over the past three decades, up 16 percent for white males and 9 percent for nonwhite males. For females, mortality rates fell slightly for whites and changed negligibly for non-
DATAWATCH 125 Exhibit 4 Five-Year Relative Survival Rates For Selected Sites For White Children Under Age 15 Source: Cancer Patient Survival Experience-, and SEER Program data reported in Vincent T. De Vita, "Update: Cancer Patient Survival Statistics" (National Cancer Institute, November 198). whites. Mortality rates for lung cancer have shown the most sizeable increases, 116 percent for while males, 199 percent for white females, and 185 and 188 percent for nonwhite males and females, respectively. 7 The increases in lung cancer mortality rates for women are particularly alarming. Exhibit 5 shows these increases of the past fifteen years as compared to breast cancer. By the end of 1984, lung cancer is expected to surpass breast cancer as the leader of cancer deaths in women, according to NCI statisticians. Also, two researchers working with the SEER program reported in The Journal of the National Cancer Institute that the total incidence rates of lung cancer for women may even equal that of men by the year 2000, if current trends continue. 8 Nevertheless, cancer statistics on the whole are encouraging as new discoveries about the treatment and diagnosis of cancer are made and new programs of prevention are promoted. Although a recent study predicts that by 200 the number of U.S. cancer cases each year may reach 1.5 million, double today's rate of 750,000 per year due to the aging of the U.S. population, advances in prevention and treatment are being developed to increase the chances of survival. 9 Secretary of Health and Human Services Margaret Heckler announced a new prevention campaign on March 6, 1984, in which she estimated that changes in
126 HEALTH AFFAIRS Exhibit 5 Age-Adjusted Mortality Rates For Female Lung Cancer And Female Breast Cancer In The U.S., 1969-1984 Year 1969 1970 1971 1972 197 1974 1975 1976 1977 1978 1979 1980 1981 1982 198 1984 Average annual percent change ( 1969-79) Female Lung Cancer 10. 11.1 12.0 12.7 1.. 15.2 16.4 17. 18.5 19.4 20.9 22.2' 2.7 25.2 26.9 6.2 Female Breast Cancer 26.7 26.9 27.0 26.7 26.2 26.5 27.0 26.5 26.0 26.5-0.1 Source: National Cancer Institute, 1984. Note: The rates from 1980 through 1984 are estimates only. All rates are age-adjusted to 1970. life-style and behavior could save 95,000 lives per year by the year 2000. 10 The National Cancer Institute has, for the first time, set an overall goal for the reduction of cancer: a reduction of 50 percent in cancer deaths compared with today's level by the year 2000. Jane K. White Health Affairs
DATAWATCH 127 NOTES 1. Wilson B. Riggen, John-Van Bruggen, John F. Acquavella, Jeff Beaubier, Thomas J. Mason, U.S. Cancer Mortality Rates and Trends: 1950-1979 (Washington, D.C.: Environmental Protection Agency, National Cancer Institute, 198), iv-x. 2. Henry C. Pitot, "Forward," in Decade of Discovery: Advances in Cancer Research 1971-1981 (Washington, D.C.: National Cancer Institute, 1981), 1.. Ibid., 1-2. 4. Ibid., 2. 5. Lewis Thomas, "Biomedical Research and the Future of Public Health," Health Affairs (Winter 198): 6. 6. Decade of Discovery, 8. 7. U.S. Cancer Mortality Rates, ix. 8. John W. Horm and Ardyce J. Asire, "Changes in Lung Cancer Incidence and Mortality Rates Among Americans: 1969-1978," The Journal of the National Cancer Institute (October 1982): 87. 9. Lawrence K. Altman, "Study Predicts Cancer Cases in U.S. May Double by 200," The New York Times, 22 June 1984. 10. Margaret Heckler, Statement, 6 March 1984.