1603 Environmental Causes of Cancer among Native Americans Nathaniel Cobb, M.D. Cancer Prevention and Control Program, Indian Health Service, Albuquerque, New Mexico. Age-adjusted mortality from cancer is steadily increasing among Native Americans, raising concern about environmental causes. Toxic waste dumps, mining activities, paper mills, military activities, agricultural chemicals, and urban pollution are common sources of carcinogen exposure on reservations and among the urban poor. Despite documented hazards, we do not see a pattern of cancer excess that might result if Native Americans were exposed to these carcinogens more than o t h e r h e r i c a n s . Abuse of tobacco can be shown to be the most important contributor to cancer mortality among Native Americans. Although there are good reasons for environmental regulation and clean-up, the most cost-effective method of cancer prevention is tobacco control. Cancer 1996; 781603-6. 0 1996 American Cancer Society. KEYWORDS neoplasms, environmental pollution, North American Indians, tobacco. t the turn of the century, cancer was a rare disease among Native mericans.' This might be attributed to a rural life with little exposure to industrial pollution, a healthy diet of home-grown fruits and vegetables and wild game, limited use of tobacco, and a tendency to die young from infectious diseases and injuries. Native Americans still enjoy some advantage over other groups in the United States, with age-adjusted cancer incidence and mortality one third less than the rate for all races combined.' Unfortunately, cancer incidence seems to be steadily increasing among American Indian and Alaska Native population^,^ and there is growing concern in native communities that environmental exposures may be the cause. Paper mills pollute the water with dioxin, while mining and smelting activities may cause exposure to arsenic, lead, radioactive materials, and air pollution. Toxic waste dumps pollute water and soil with a variety of poisons. Jobs in mining and industry can be a source of exposure to dangerous chemicals, mineral dust, and radon gas. People who work and live in agricultural areas may be exposed to pesticides and herbicides. Many urban Indians live in poverty, which may be associated with increased exposure to air pollution, lead, and other hazardous materials. Inadequate housing can be a cause of exposure to vaporized chemicals such as formaldehyde as well as smoke and fumes from cooking and heating sources. In recognition of the hazards faced by minority populations and the poor, President Clinton recently signed an executive order mandating federal actions to address environmental j ~ s t i c e . ~ The food we eat can be an important source of hazardous exposure and can provide protection from cancer as well. Fish, wild game, and the fat of sea mammals, which are an important part of the diet of many Native groups, have been shown to contain pesticide resi- A, Presented at the Native American Cancer Conference Ill: Risk Factors, Outreach and Intervention Strategies, Seattle, Washington, June 1619, 1995. Address for reprints: Nathaniel Cobb, M.D., Director, Cancer Prevention and Control Program, Indian Health Service, 5300 Homestead Road NE, Albuquerque, NM 87110. The opinions expressed in this paper are those of the author and do not necessarily reflect the views of the Indian Health Service. Received May 31, 1996; accepted June 18, 1996. iij 1996 American Cancer Society 1604 CANCER Supplement October 1, 1996 / Volume 78 / Number 7 dues, PCBs, and heavy metals.' Radioactive isotopes from nuclear testing may be concentrated in caribou and other wild food sources.6 The traditional diet of corn, beans, squash, and game has been largely replaced by processed meats and foods that are higher in fat and lower in fiber and antioxidants. A high-fat diet is suspected of contributing to cancer of the breast, colon, and prostate, whereas antioxidants and other naturally occurring substances in fresh fruits and vegetables may prevent a variety of cancers. In parts of the world where the standard diet is low in fat and high in fiber, colon cancer is very rare. Personal or life style exposures, such as the abuse of tobacco and alcohol. are important causes of cancer among Native Americans. In the Southwest, where smoking rates have been very low, lung cancer is extremely rare except among uranium miners.' In other parts of the United States, however, more than 45% of Native American persons smoke or chew tobacco.* Cigarette smoking is the cause of most lung cancer deaths and contributes to cancer of the larynx, bladder, pancreas, stomach, kidney, and cervix. Smokeless tobacco use can cause cancer of the mouth. Alcohol acts with tobacco and other risk factors to increase the risk of cancer of the throat, liver, and pancreas. How much cancer mortality is actually caused by environmental pollution? Which exposures should we worry about the most? It is rather difficult to reach a realistic estimate of the proportion of cancers caused by ambient or industrial exposures. In an important and widely quoted work, Doll and Pet0 estimate that environmental exposures play a significant role in 7080% of cancer cases in the United Statesg This figure, however, includes life style exposures such as diet, tobacco, and alcohol use. They estimate that only 15% of all cancers are caused by pollution and industrial products. How do we protect ourselves from exposure to these cancer-causing agents? The prevention of disease from personal exposures is usually based in the health system, through education and medical intervention, whereas control of ambient environmental exposures is usually the responsibility of engineers and sanitarians in an environment department. Many tribes have active environmental protection agencies that, with assistance from the U.S. Environmental Protection Agency, Agency for Toxic Substances, and Disease Registry (among others), are actively trying to clean up some of these hazards. These tribes must establish priorities for their limited funds; environmental clean-up is often extremely expensive. Many tribes ask whether the cancers they are experiencing were caused by environmental pollution. Table 1 lists some known carcinogens and the cancers TABLE 1 Some Carcinogens and Associated Cancer Site Carcinogen Cancer site Alcohol Arsenic Asbestos Benzene Ionizing radiation Nickel dust Polycyclic hydrocarbons Tobacco Liver, esophagus, mouth, pharynx, larynx Lung, skin Lung, pleura, peritoneum, pericardium Acute myelocytic leukemia Lung, leukemia, breast, brain, thyroid, other organs Lung, nasal sinuses Lung, scrotum, skin Lung, mouth, pharynx, larynx, esophagus, pancreas, bladder, kidney, renal pelvis, cervix Angiosarcoma of liver ~~~ Vinvl chloride they cause. If American Indians and Alaska Natives were being exposed to chemicals and radioactive materials at higher rates than other U.S. races, we would expect to see that reflected in higher rates of leukemia, breast cancer, lung cancer, lymphoma, and brain and thyroid cancer. This is not the pattern of cancers we see, however. Table 2 lists the top 11 causes of cancer mortality among Native Americans. At the top of the list is lung cancer, which is also the largest cause of cancer death for all U.S. races. The lung is one of the most sensitive organs to carcinogenic agents; lung cancer has been demonstrated to follow exposure to asbestos, arsenic, diesel exhaust, radon, coal tar products, and many industrial chemicals and heavy metals. Can we blame the large number of lung cancer deaths on environmental pollution? Probably not; it has been demonstrated that 87% of all lung cancer cases in the US. are caused by cigarette smoking," and smoking rates among American Indians and Alaska Natives are high enough to explain the current lung cancer rate. For comparison, U.S. adult male smoking rates peaked in 1964 at nearly 50% and have been dropping steadily since then. Twenty-five years later, the lung cancer death rate for that group has leveled off at approximately 74 deaths per 100,000 each year." We do not have good information on Native American smoking rates from 25 years ago, but the 1991 smoking rates for Native American males ranged from 20% in the Southwest to as high as 70% in the Northern Plains and Alaska; the age-adjusted lung cancer death rate was only 40 per 100,000 and rising steadily." The nonsmoking Southwest tribes have been heavily exposed to uranium mining and milling and the residual mine and mill tailings, yet the lung cancer death rate in northeastern Arizona is 7.6 per 100,000, less than one sixth of the US. all-races rate. Although Southwestern tribes have maintained low lung cancer rates for some time, there is evidence that smoking rates are increasing rapidly among the young. Unfortunately, based on Environmental Causes of CancerKobb TABLE 3 Costs of Prevention TABLE 2 Cancer Mortality Among American Indians and Alaska Natives, 1984-88 Cancer site 1. Lung and bronchus 2. Colon and rectum 3. Stomach 4. Breast 5. Pancreas 6. Prostate 7 . Gallbladderlother biljary 8. All leukemias 9. Cervixuteri 10. Kidney 11. Liver 81 intrahepatic ducts No. of deaths (5 years) 510 175 162 134 128 123 99 94 89 82 81 1605 Ratea US rate' 32.7 11.2 9.9 8.4 8.3 7.7 5.1 4.4 47.1 20.1 5.1 5.1 5.5 1.7 3.3 2.6 5.0 15.3 8.4 9.1 0.8 6.4 Does not include Oklahoma, Portland, and California Areas because of known inaccuracies of racial idenrificatioii in those areas. Bold = Greater than IJS rate. * /\I1 rates per 100,000 population, agti-adiusted to 1970 US population. current trends in smoking behavior, we expect to see rising lung cancer rates among all Native American groups well into the next decade. Of the other top cancers on the list, age-adjusted Native American death rates exceed U.S. rates for cancer of the stomach, gallbladder, cervix, kidney, and liver. Could any of these be related to excessive exposure to environmental pollution? We will go through the list. Stomach cancer is generally found in societies in which access to refrigeration is a problem and in which foods are preserved by salting, smoking, and pickling. It may also be influenced by chronic H. pylori infection. A hundred years ago stomach cancer was far more common than lung cancer in the United States, but the incidence has been steadily dropping since then. If this cancer was caused by environmental pollution, we should expect to see a steady increase. Gallbladder cancer is much higher in Native Americans than other races, probably because of the same genetic factors that cause the high prevalence of gallstone~.'~ Cervical cancer is most commonly caused by infection with human papillomavirus. Smoking has been found to be a minor contributor as well.14 Kidney cancer is also related to smoking, genetic factors, and possibly environmental exposures. The excess of liver cancer is probably caused by high rates of hepatitis B infection'5 and cirrhosis from chronic alcohol abuse. Intervention Cost (per life-year saved) Smoking cessation advice Alcohol safety programs for drunk drivers Mammography every 3 years, age 50-65 Annual mammography age 55-64 Annual Pap screening women >20 Radon remediation in homes: >21 pCilL >a pCi/L >4 pCilL (EPA recominended) Radionuclide emission control during disposal of uranium mill tailings piles Benzene emission control at bulk gasoline terminals $<O-9,79ga $20,968 $2,706 $108,401 $50,348-1,477,24gb $6.107 $34,990 $141,301 $39,589,443 $155,467,742 $<O means an overall cost savings. Ranee of values are from different sources. It seems that much of the excess cancer among Native Americans can be explained by genetic, infectious, and life style influences. The fact that Native American cancer rates were so extraordinarily low until quite recently seems to be evidence for a life style and environment that is (or was) relatively free from harmful exposures. One current measure of this is that only 18% of Native Americans live in counties with significant air pollution, compared with 23% of whites, 25% of African Americans, and 42% of Hispanics.16 This is only small comfort, of course, because we know that the U.S. population in general is exposed to more harmful chemicals than are good for us. The problem of setting priorities remains a difficult one. A recent publication attempted to compare the costs of prevention for a variety of health and environmental problems." The authors conducted an extensive review of the literature on costs of prevention, converting cost-effectiveness estimates to a common measure of "cost per year of life saved." Although the assumptions used in different studies varied widely, at least this approach allows us to look at a range of cost estimates made by different researchers. Table 3 shows some representative costs of life-saving interventions from this article. These numbers should be used with some caution, as the assumptions made by different researchers may or may not match your situation. In a comparison of 587 interventions, the authors found that the median medical intervention cost $19,000 per life-year saved, the median for injury reduction cost $48,000 per life-year, and the median toxin control cost $2,800,000 per life-year. In contrast, the simple prevention measure of smoking cessation advice can result in a net savings, and in the least 1606 CANCER Supplement October 1, 1996 / Volume 78 / Number 7 favorable analysis costs less than half of the median medical intervention. This article demonstrates that environmental clean-up and regulation are often extremely costly when calculated on the basis of human illness prevented. When considering a multimillion dollar remediation action, we need to ask the following questions. Does this situation really represent a possible threat to human health or to the environment? Might the money be better spent on other interventions that will have a known positive impact? An example from Alaska (this is the short version): Some years ago, small amounts of radionuclide tracers were placed in the ground in northwestern Alaska to study the movement of radioactivity through permafrost. Recent publicity led to an expensive remediation effort, despite dose calculations and expert opinions that found no risk to the population from this radiation source. The same millions could have been used to provide the Alaska Native population with comprehensive cancer screening and prevention services, with far more positive effect on the health of the people. This is not to say that we should not regulate and clean up our environmental pollution. Clearly, there are hazardous materials out there that are causing human illness and death as well as harming plant and animal ecosystems. Studies of the massive industrial pollution in Eastern Europe and the former Soviet Republics have demonstrated that human health effects become evident only after fish, wildlife, and forests have suffered tremendous damage and the more susceptible species have become extinct. When allocating our scarce resources, however, we must balance the cost of various interventions against the good that we expect to come of each intervention. In conclusion, the evidence indicates that environmental exposures are not the major source of cancer in the American Indian and Alaska Native populations, but there are certainly many good reasons to continue to regulate and clean up sources of pollution. When it comes to preventing cancer, however, the most cost-effective thing we can do is control the abuse of tobacco. Achieving a more healthy diet is probably the next most effective intervention. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Hrdlicka A. Physiological and medical observations. Washington, DC: US Government Printing Office, 1908. Indian Health Service. Regional differences in Indian health. USPHS, DHHS, 1995. Indian Health Service. Trends in Indian health. USPHS, DHHS, 1994. Executive Order 12898 of February 11, 1994: Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations. Leifer A, et al. Environmental transport and transformation of PCBS. USEPA 560/5-83-025, 1983. Hanson WC, Watson DG, Perkins RW. Concentration and retention of fallout radionuclides in Alaskan Arctic ecosystems. Aberg B, Hungate FP, editors. Radioecological concentration processes. Oxford: Pergamon Press, 1967233-45. Samet JM, Kutvirt DM, Waxweiler RJ, Key CR. Uranium mining and lung cancer in Navajo men. N Engl J Med 1984; 31011481-4. Behavioral Risk Factor Surveillance System, United States, 1988-92 aggregate data. Doll R, Pet0 R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the Untied States. Oxford: Oxford University Press, 1981. U S . Department of Health and Human Services. Reducing the health consequences of smoking: 25 years of progress. A Report of the Surgeon General. DHHS publication no. (CDC) 89-8411, 1989. Miller BA, Ries LAG, Hankey BF, Kosary CL, Harras A, Devesa SS, et a1 editors. SEER Cancer statistics review: 1973-1990. NIH publication no. 93-2789, 1993. Washington, DC: National Cancer Institute, 1994. Valway S, Kileen M, Paisano R, Ortiz E. Cancer mortality among Native Americans in the United States. US Public Health Service, Indian Health Service, 1992. Lowenfels AB. Gallstones and gallbladder cancer in southwestern Native Americans. Ani Indian Culture Res I 1992; 16(3):77-86. Vousden KH. Human papillomavirus and cervical carcinoma. Cancer Cells 1989;1:43-50. Heyward WL, Lanier AP, McMahon BJ, Bender TR, Francis DP, Maynard JE. Serological markers of hepatitis B virus and alpha-fetoprotein levels preceding primary hepatocellular carcinoma in Alaskan Eskimos. Lancet 1990;8304:889-91. Plepys C, Klein R. Health status indicators: differentials by race and Hispanic origin. Healthy People 2000 statistical note 10. DHHS publication no. (PHS) 95-1237. Washington, DC: National Center for Health Statistics, 1995. Tengs TO, Adams ME, Pliskin JS, Safran DG, Siege1 JE, Weinstein MC, et al. Five-hundred life-saving interventions and their cost-effectiveness. Risk Anal 1995;15(3):369-90.