By Dr. William M. London and Dr. John W. Morgan
Posted: Sunday, October 1, 1995

ARTICLES
Publication Date: October 1, 1995

As observed by the noted geneticist J.B.S. Haldane; "We are all prisoners of life under sentence of death." Contrary to the views of radical life extensionists, in any 120-year period for any human population, the all-cause mortality rate is 100 percent. This means that if you don't die one way, you'll die some other way before the 13th decade following your birth. So the question isn't whether we're going to die, but how long (and how well) we get to live before the inevitable end.

Here in the United States relatively many people die of cancer. But this is not an indication that we are being poisoned by chemicals in our environment or that we are hurt by the technology surrounding us. Rather, the most important explanation is that where many people live long lives, many people die of cancer.

Infant Mortality and Life Expectancy at Birth

According to the most recent available data from the Population Reference Bureau, Inc., for every 1,000 live births in the United States, there are eight deaths during the first year of life. Life expectancy at birth is 75 years for males and 79 years for females. Let's compare those current figures with the figures for 1900. Life expectancy was much lower for newborns in 1900 than it is today: only 46 years for males and 48 years for females. Although some people born in 1900 are still alive today, many more children born in that year died in the first year of life (about 100 deaths per 1,000 live births) than will newborns from 1995. As Kathleen Newland, a former research associate with the World Watch Institute, has noted, "No cold statistic expresses more eloquently the difference between a society of sufficiency and a society of deprivation than infant mortality." American society provides far more sufficiency now than it did in 1900.

Societies of sufficiency, with their low infant mortality and long life expectancies at birth (see Table 1), happen to be those in which many people die of cancer as opposed to other causes. This is not to say, as many antitechnology ideologues imply, that societal sufficiency causes cancer. It is to say that societal sufficiency provides a large proportion of the population with the opportunity to live long enough to develop cancer.

All-Cause Mortality and Cancer Mortality Vary by Age

In the U.S., death is close to 20 times more common in the first year of life than it is from ages one through four. The chance of dying then declines by half for children aged five to 14. After age 14, as age increases, death becomes more common; but not until age 55 to 64 is it as common as death among infants.

According to the National Center for Health Statistics, of the 2,148,463 deaths in the U.S. in 1990, 15,367, or less than one percent, were among children ages one to 14. Of these childhood deaths, 1,632 were due to cancer. Although, tragically, cancer accounts for more than 10 percent of all deaths among children, cancer deaths among children represent less than one percent of all cancer deaths. Terminal cancers are predominantly diseases of middle-aged and elderly persons. Cancer surfaces as an important cause of death in populations that have escaped other causes of death. Where many people die of cancer, many people are fortunate to live long lives.

The Cancer Slice of the Mortality Pie

The pie chart in Figure 1 shows mortality in the U.S. for 1990, with slices indicating causes of death. In 1990 the two leading causes of death, heart disease and cancer, accounted for more than half of all U.S. deaths. Heart disease produced about 32 percent of all deaths, cancer about 25 percent and all other causes the remaining 43 percent.

Figure 1: 1990 Mortality All-cause age-adjusted mortality rate: 702.4 per 100,000 (100%) Based on data from the National Center for Health Statistics	Figure 2: Projected Mortality for 2005 All-cause age-adjusted mortality rate: 574.2 per 100,000 (100%) Based on assumptions discussed in the text

It's easy to misinterpret the mortality data presented in a mortality pie chart. Each slice of the pie indicates a proportion of mortality, or the percentage of total mortality accounted for by each cause of death. In each pie, all the slices together must add up to 100 percent. If one slice gets bigger, the combined total of the other slices in the pie must automatically get smaller.

In 1900 cancer was the eighth-leading cause of death in the United States, representing less than four percent of the mortality pie. Cancer was a smaller piece of the pie in 1900 largely because deaths from infectious diseases like influenza, pneumonia, tuberculosis and gastroenteritis were much more common.

In the mortality pie of developing nations today, the cancer slice is also relatively small. Infant diarrhea, starvation, tuberculosis, malaria and other infectious diseases represent much larger slices of the pie for a developing nation than does cancer. Where many people die of infectious diseases and famine as infants, fewer people will live long enough to develop and die of cancer.

Mortality Rates

Figure 1 shows how many people in 1990 died from cancer relative to deaths from other causes, but the percentages alone do not provide information in absolute terms about cancer mortality and mortality overall. A pie chart that shows only that cancer accounts for 25 percent of total deaths is incomplete. We also need to know the actual number of cancer deaths. Counts of cancer deaths can be misleading, however.

In his 1994 revised version of Cancer & Nutrition, Charles B. Simone, M.D., misuses counts of cancer deaths so as to leave the alarming impression that we're in the midst of a general cancer epidemic. Simone's book begins with a graph showing a dramatic rise in the number of cancer deaths in the U.S. since 1900. But the graph fails to take into account a dramatic rise in the U.S. population: from less than 77 million in 1900 to almost 250 million in 1990.

To account for changes in population size, epidemiologists normally use a mortality proportion, commonly called a mortality rate. A mortality rate expresses the proportion of living persons who die during a particular time period. Typically, a mortality rate is expressed as the number of deaths in one year per 100,000 persons in the population.

It's easy to misuse mortality rates to alarm the public — especially if we ignore how mortality varies markedly with age. We must be careful when comparing mortality rates, whether they are the rates of two separate populations or of the same population at different points in time. Any differences in the numbers of people of different age groups represented in the populations to be compared must be taken into account. Epidemiologists apply mathematical methods for this purpose and then present their data as "age-adjusted" to reflect an arbitrary standard population (e.g., the U.S. population of 1970).

Over the past 20 years, the crude cancer mortality rate (unadjusted for age) in the U.S. has been rising. This is not cause for alarm. The rise largely reflects the rising proportion of Americans who are elderly — the so-called graying of America. With the primary exceptions of true increases in cancer mortality attributable to cigarette smoking and AIDS, the age-adjusted cancer mortality rate overall has been relatively stable. Furthermore, age-adjusted stomach and cervical cancer mortality rates in the U.S. have decreased greatly during the last half century.

How Good News Can Sound Bad

A rise in the proportion of total deaths attributable to cancer can sound scary. But such a rise can occur even as the age-adjusted cancer mortality rate declines. This is precisely what has been happening in the U.S. during the 1990s. A comparison of the 1990 mortality pie shown in Figure 1 with the projected mortality pie for 2005 shown in Figure 2 illustrates this.

According to the National Center for Health Statistics, in 1990 cancer was the second-leading cause of death in the U.S., with a mortality rate (age-adjusted to reflect the 1970 population) of 174.0 per 100,000. The leading cause of death in 1990 was heart disease, with an age-adjusted mortality rate of 225.8 per 100,000. The age-adjusted mortality rate attributable to all other causes in 1990 was 302.6 per 100,000.

The age-adjusted mortality rate for heart disease has declined rapidly and steadily since the 1960s. Let's assume, as we project ahead to 2005, that this decline will continue; that there will be a steady decline in the age-adjusted heart disease mortality rate of about 2.5 percent per year.

The most recent available statistics show a net decline in the overall cancer mortality rate, with a marked decline for persons under 55 years more than offsetting the slight rise in cancer mortality for older persons. Let's assume, then, over our 15-year projection, that a modest decline of less than a third of a percent per year in age-adjusted mortality rates for cancer will continue; and let's use this rate for other (non- heart-disease-related) causes as well. These rates of decline are more modest than the decline in heart disease.

Using these assumptions, over 15 years from our known 1990 data to the year 2005 we project an increase in the percentage of total deaths attributable to cancer from 25 percent to 29 percent. Thus, in the year 2005 cancer, not heart disease, will be the leading cause of death. Fear-mongers may try to alarm the public with this projected trend; but, in fact, the news indicates that we can expect improved public health.

Although the slice of the mortality pie represented by cancer becomes larger relative to the other slices, what really counts as a measure of public health is the absolute size both of the entire pie and of each slice. Our comparison of Figure 1 with Figure 2 shows how each slice, along with the entire pie, is projected to become smaller in absolute terms — an encouraging trend. After taking the graying of America into account, dying becomes less common overall — and so does death due to cancer.

Preventing the Major Deadly Forms of Cancer

Given the association between increasing age and cancer, efforts to prevent cancer mean delaying the development of some types and reducing the incidence of avoidable types such as lung cancer and the many other cancers caused by cigarette smoking.

In Figure 3, we've taken the cancer mortality slice from the 1990 U.S. mortality pie and reformed it into its own pie. The figure shows that four cancer sites (lung, colon/rectum, breast and prostate) accounted for more than half of all cancer deaths in 1990. Cancer mortality could decline much more than we project with improvements in preventing just these four cancers.

 Figure 3: U.S. Mortality, 1990 Estimates
Total cancer deaths=538,000
Based on National Cancer Institute — Surveillance, Epidemiology and End Results Data

The challenge for prostate cancer prevention is the most daunting. Prostate cancer is currently the second-leading cause of cancer death among males and the fourth-leading cause of cancer death overall in the U.S. At present, however, there is inadequate evidence to support recommendations of lifestyle changes to prevent prostate cancer.

Breast cancer is the second-leading cause of cancer death among U.S. women and the third-leading cause of cancer death overall. Yet, its public health impact is far greater than these rankings suggest. Breast cancer kills even more women before age 54 than does lung cancer, the leading cause of cancer death among women. In general, risk factors for breast cancer — including family history, early age at menarche and late age at menopause — are difficult to control. Avoiding high levels of body fat (which can be accomplished through reduced caloric intake and/or increased physical activity) appears to lower risk, though not to a large extent.

Colorectal cancer, the second-leading cause of cancer death for men and women combined, is the most frequent cause of cancer death among nonsmoking Americans. Diets low in fat or rich in vegetables, fiber and calcium appear to offer some, but not strong, protection against colon cancer. Physical activity also appears to provide modest protective effects. The associations of these factors with colon cancer are rather weak, but these factors are common among Americans. Thus, the potential public health impact of ongoing efforts to promote dietary improvements and exercise is significant. There is also mounting evidence that regular aspirin use may provide protective effects. Avoidance of excessive alcohol intake appears to offer moderate protection against rectal and some other cancers.

The leading cause of cancer death in the U.S. among both men and women is lung cancer. Over 84 percent of the 163,000 lung cancer deaths in the U.S. in 1993 were attributable to cigarette smoking. The potential for lung cancer prevention through smoking prevention and cessation is enormous. Moreover, we recommend avoidance of cigarette smoking for preventing other cancers such as laryngeal, pancreatic and bladder cancers; liver and brain cancer; coronary heart disease; stroke; and a long list of other tobaccosis health problems.

Widespread adoption of prevention practices, particularly by not smoking and eating moderately of a broad variety of foods, would reduce the age-adjusted, cancer-specific mortality rates in the U.S. Even then, many people would die of cancer, but most of those cancers would occur at the end of long, productive lives.

William M. London, Ed.D., M.P.H., is Director of Public Health at the American Council on Science and Health.

John W. Morgan, Dr.P.H., is Associate Professor of Epidemiology and Preventive Medicine at Loma Linda University.

Table 1: International Comparison of Infant Mortality and Life Expectancy at Birth Infant Mortality Rate per 1,000 live births Life Expectancy at Birth in years United States 8 76 More Developed Nations (including U.S.) 10 74 Less Developed Nations 67 64 Less Developed Nations (excluding China) 72 62 Source: Population Reference Bureau, Inc., "1995 World Population Data."

(From Priorities Vol. 7, No. 4, 1995)