CA: A Cancer Journal for Clinicians, Vol 29, 352-361, Copyright
© 1979 by American Cancer Society
Cancer Mortality Among Low-Risk Populations
James E. Enstrom Ph.D.1
1 Cancer Epidemiology Researcher, School of Public Health and Program Director for Cancer Control Epidemiology, Jonsson Comprehensive Cancer Center, University of California, Los Angeles, California.
It is clear that several nonsmoking populations experience relatively low cancer rates and low total mortality rates. The precise reasons for the low-risk are not clear at this time. The most plausible explanation is that lack of smoking se reduces the total cancer rate by a substantial amount. Furthermore, selection associated with being a questionnaire respondent apparently reduces the cancer rate by another 20 percent in men and 10 percent in women, based on the American Cancer Society and U.S. veteran cohort studies. The respondents in these cohort studies could be healthier than normal for several reasons: they probably are health conscious and follow several good habits in addition to not smoking; also, it is well known that very sick persons do not usually respond to questionnaire surveys, thereby raising the average level of health of the respondents.
If all Americans did not smoke, the mortality reduction that would occur has been estimated to be 80,000 lung cancer deaths plus 22,000 other cancer deaths of the 1978 total of 390,000 cancer deaths—a reduction of 26 percent. Another estimation is that about 33 percent of cancer deaths are preventable by applying all current knowledge, e.g., eliminating human exposure to known carcinogenic factors like cigarette smoking, high alcohol intake, excess sunlight, and certain occupational and industrial exposures.
Table 2 shows that representative U.S. white nonsmokers (males and females averaged together) have a total cancer rate of 24 percent less than that of all U.S. whites. This is in good agreement with the predicted 26 percent reduction above. The nonsmoking healthy American Cancer Society, Mormon and Adventist cohorts of males and females combined, as shown in Table 2, have an average total cancer rate reduction of 39 percent relative to U.S. whites. This is slightly more than the 33 percent reduction predicted above. What else are these groups doing, in addition to not smoking, to diminish their risk of cancer death?
It may be possible to lower the cancer mortality rate even below the Table 2 rates by minimizing several risk factors simultaneously. Some indication of this is the fact that 282 adult men and 386 adult women in Alameda County, California who followed seven good health habits experienced over a nine and one half year period only 52 percent of the total mortality rate of a representative sample of 6,928 Alameda County adults. (This 52 percent is based on only 48 deaths, with 95 percent confidence limits from 38 percent to 69 percent.)
The actual preventability of cancer remains to be demonstrated. There is certainly an overwhelming amount of evidence pointing to the benefits of being a nonsmoker. However, the precise effects of smoking cessation on current smokers are still unknown. One major study showed that British physicians who reduced their cigarette smoking as a whole by more than 50 percent over a 20-year period experienced a reduction in lung cancer mortality relative to the mortality of the general British population: from a ratio of 65 percent in 1955 to 35 percent in 1972. On the other hand, a recent randomized controlled trial of smoking [see Table 2 in source PDF] cessation among middle-aged British men surprisingly showed no difference in total mortality rates between intervention group and control group after eight years of followup.
In view of these data, it is both important to realize the benefits of not smoking and to ascertain what factors, in addition to lack of smoking, account for the low cancer mortality experience among various low-risk populations, and to discover how to apply these findings to persons at higher risk of cancer. Some carcinogens that have been given a great deal of attention by the media in recent years are not likely to have an important impact on reducing cancer mortality. These factors include hair dyes, food additives, saccharin, menopausal estrogens, and low-level ionizing radiation. Even if these factors caused most of the types of cancer with which they have been most strongly linked, primarily cancers of the bladder and endometrium and childhood leukemia, they would have little impact on total cancer mortality. This is because cancers of the bladder and endometrium and childhood leukemia comprise less than four percent of all cancer deaths. As of now, there is no good evidence that the above factors have caused an increase in the cancer death rate in the general population.
For the purposes of public health, it is imperative that research be directed first and foremost at factors that are likely to have the greatest impact on cancer mortality. These factors now appear to include personal health habits, diet, socioeconomic status, and certain host factors related to the aging process. The potential for identifying lower risk exists if sufficiently large and healthy population groups can be studied with respect to the important risk factors. Interpreting low risk for personal health is genuine cancer control.
