In general, the cancer death rates by site for males and females parallel the incidence rates (Fig- ures 1.4 and 1.5), especially with cancers of the stomach, lung, pancreas, brain, liver, ovary, and several other sites. On the other hand, for certain types of cancer, the mortality rates have fallen dramatically during the past several decades. This is true for cancer of the uterus and cervix, where the overall death rate has decreased more than 70% over the past 40 years. This is proba- bly due to the widespread use of the Pap test (Chapter 2) and early detection of cervical cancer. Mortality from colorectal cancer has fallen 29% for women and 6% for men during the past 30 years, while—as pointed out earlier—the mortality rate for cancer in children has declined 60% since 1950 primarily because of better methods of treatment. In some cancers, such as cancer of the stomach, there has been a steady decrease in both incidence and mortality in this country and in many others since the 1930s, although in Japan the decrease in mortality did not begin to occur until after 1955 (Coggon and Acheson, 1984). In contrast, not only in this country (Fig- ures 1.3 to 1.5) but throughout the world, cancer of the lung has been increasing in both sexes.
Since 1978, in the age group 75 to 84, trends of cancer incidence have revealed significant increases in myeloma, a cancer of the bone marrow, as well as cancers of the breast and the brain (see above) (Davis et al., 1990). The reasons for such specific changes in incidence of the cancer mentioned—with the exception of tobacco abuse—are not entirely clear. However, such factors as diet, workplace conditions, and infectious disease may have played a role in this group during their younger years as well as in the population in general. These factors will be considered later in the text (Chapter 10).
Trends in relative survival rates in the United States for white males and females over the quarter century beginning in 1960 may be seen in Table 1.4. Survival rates in a number of differ- ent cancers have increased significantly even over the past 10 years; however, interpretation of cancer survival rates is difficult and subject to many variables, as pointed out by Enstrom and Austin (1977). Improvements in survival rates may be due to a number of factors including:
1. An increase in the proportion of cancers diagnosed at a controllable stage of develop- ment as a result of improved diagnostic techniques and of a better-informed public alerted to the danger signals of cancer and the benefits of early diagnosis
2. Improvements in surgical and supportive techniques, including the control of infec- tious diseases that complicate cancer
3. Improvements in radiotherapy, endocrine therapy, and chemotherapy
Specifically, for patients with lung cancer, Feinstein et al. (1985) pointed out that the en- hanced survival from 1960 to 1977 was apparent rather than real, being the result of the use of new diagnostic imaging procedures to classify patients into specific “stages” of the disease more accurately. This effect has been called the “Will Rogers phenomenon” by these investigators. Cairns (1985) also pointed out that the apparent increase in survival for patients with cancer of the prostate was the result of better diagnosis, resulting in an actual increase in the number of patients reported with the disease but with no change in the actual death rate from the disease. Thus, a number of factors must be considered in determining the absolute or real survival rates of patients with cancer to obtain a clear picture of the efficacy of cancer therapy. Despite such factors, it is clear that there has been a steady though slow improvement in the survival rates of most types of cancers in the human (Table 1.4). Therefore, although the absolute numbers of cancers and cancer deaths are increasing in our society, owing primarily to one or two major types of cancer, significant advances have been made in the treatment of many types of cancer, as listed in Table 1.4.
Social and Economic Costs Resulting from Cancer
Especially significant and important to society in general in relation to this dread disease are the number of years of life lost and the cost to the individual and to society of the mortality and morbidity resulting from cancer. If a person is struck down by this disease in early or middle life, the remaining productivity of that individual’s life is lost to society. When measured in dollars
Table 1.4 5-Year Relative Survival Rates (Percentage) By Selected Sites, White Males and Females
aRates are based on End Results Group data from a series of hospital registries and one population-based registry. bRates 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. Rates are based on follow-up of patients through 1986.
cThe difference in rates between 1974–76 and 1980–85 is statistically significant (p < 0.05).
alone, such a loss in the United States was estimated to be nearly $25 billion per year about a decade ago (Hodgson and Rice, 1982). In 1990, a decade later, the National Cancer Institute estimated that the overall annual cost for cancer was $104 billion and in 2000, a total of $180.2 billion. Of this $105.2 billion, or almost three times the amount estimated in 1982 represented the mortality costs, which include the loss of productivity because of premature death (Brown, 1990). In 1985, it was estimated that cancer accounts for a little more that 10% of the total cost of disease in the United States and almost 21% of the cost of premature death or mortality in this country (Hodgson and Rice, 1995).
Although it is difficult to estimate the economic impact of cancer on society with a high degree of accuracy, it is possible through the use of statistical data compiled for the United States to determine the years of life lost as a result of premature death because of cancer.
Tables 1.5 and 1.6 show the person-years of life lost (PYLL) in males and females because of prema- ture death from cancer in the United States (Horm and Sondik, 1989) for the years 1970 and 1984. A comparison of the data between the two years indicates the dramatic increase in the social and economic burden of all cancers, several in particular. In 1984 the most costly cancer for women was cancer of the breast, resulting in over 760,000 PYLL among almost 40,000 women; this was an increase of almost 28% over the 1970 figure. These women died an average of 19.3 years earlier than expected. Some studies (Shapiro et al., 1982; Tabar et al., 1985) have indicated that the mortality from breast cancer can be reduced by 30% through mass screening programs. This would mean the prevention of almost 12,000 breast cancer deaths per year, or nearly 230,000 PYL that could be saved. Similarly, the PYLL to lung cancer in women in 1984 was almost triple that in 1970. In men, the toll of this cancer is considerably higher. In both
Table 1.5 Person-Years of Life Lost (PYLL) Due to Premature Deaths from Cancer in the United States, All Races, Males
Table 1.6 Person-Years of Life Lost (PYLL) Due to Premature Deaths from Cancer in the United States, All Races, Females
sexes, prevention of the disease by the cessation of tobacco abuse could save well over a million
PYL every year (Horm and Sondik, 1989).
These statistics on cancer incidence, mortality, survival, and economic and social impact are critical to our understanding of the impact of this disease on our society. Obviously, these figures cannot take into account the morbidity and suffering, both mental and physical, caused by cancer. However, they do point out a variety of differences in the effects of this disease on males and females and in the types of cancers that are most common in our society and in other countries. Later in this text (Chapter 10), a consideration of the epidemiology of cancer includes explanations for the various statistics relative to specific neoplasms.