1612

Nutrition and Cancer among American Indians and Alaska Natives

Tim Byers, M.D., M.P.H.

Department of Preventive Medicine and Biomet- rics, University of Colorado School of Medicine, Denver, Colorado.

Supported in part by grant #CA46934-09 from the National Cancer Institute to the University of Colorado Cancer Center.

Presented at the Native American Cancer Con- ference Ill: Risk Factors, Outreach and Interven- tion Strategies, Seattle, Washington, June 16- 19, 1995.

Address for reprints: Tim Byers, M.D., M.P.H., Department of Preventive Medicine and Biomet- rics, University of Colorado School of Medicine, Campus Box C 245, 4200 East Ninth Avenue, Denver, CO 80262.

Received May 31, 1996; accepted June 18, 1996.

0 1996 American Cancer Society

BACKGROUND. More than a third of all cancers in the United States are thought to be attributable to various nutritional factors, and major changes in nutrition have been experienced by American Indians and Alaska Natives in the past century. METHODS. The published literature was reviewed to summarize the relationship between diet and cancer, to summarize what is known about the past and current diet of American Indians and Alaska Natives, and to consider whether nutrition might play a role in their current or future risk of cancer. RESULTS. Epidemiologic studies show a consistent pattern of lower risk for cancers of the colon and lung among those who eat larger amounts of fruits and vegetables and higher risk for cancers of the colon and prostate for those who eat larger amounts of fat. Limited data indicate that the diets of American Indians and Alaska natives are similar to the current average American diet, that is, high in fat and low in fruits and vegetables. CONCLUSIONS. Diet is an important factor for several major cancers. Although the cancer rates among many Native American groups are currently lower than among non-native Americans, they seem to be increasing. Cancer rates among Native Alaskans are already much higher than for non-native Americans. Nutrition-related cancers are likely to increase in the future among Native Americans as a result of past changes in diet. However, Native Americans may have a stronger cultural basis than do non-native Americans to adopt traditional principles of the value of good foods and physical activity to achieve future nutritional improvements to reduce cancer risk. Cancer 1996; 78:1612-6. Q 1996 Ainericmz Cancer Society.

KEYWORDS: North American Indians, Eskimos, diet, nutrition, neoplasms.

ajor changes in nutrition have accompanied the many social M and health changes experienced by American Indians and Alaska Natives in the past century. Coincident with these many changes have been increases in the incidence of chronic diseases, including diabetes, heart disease, and cancer.' However, virtually no studies of the relationship between diet and cancer have been conducted among the populations of American Indians or Alaska Natives. This article reviews what is known from studies in other populations about the relationship between nutrition and cancer risk, and comments on the possible relevance of nutrition to both current and future cancer patterns among American Indians and Alaska Natives.

Cancer incidence rates are well described for only a few of the various American Indian and Alaska Native tribes.' Both this limited data on cancer incidence and the more extensive data on cancer mortality indicate that cancer rates for most of the major organ sites are lower among American Indians than among non-Native Ameri- c a n ~ . ~ Exceptions are cancers of the stomach, cervix, and liver, for which infectious agents are the principal etiologic factors. In contrast,

Nutrition and Cancer among Indians/Byers 1613

cancer rates for Alaska Natives are the same or higher than for non-Native American~.~ Colorectal cancer, the organ site most strongly linked to diet, is considerably more common among Alaska Natives than among non-Native Americans. The reasons for this are not known, but may well be due to the high levels of fat and low levels of fruits and vegetables in the Alaska Native diet. Between one third and two thirds of all cancers in the United States are thought to be attribut- able to various nutritional factors, such as caloric over- load, high levels of fat, and low levels of fruits and vegetables in the diet.5 There are important limita- tions, however, in the methods that can be used to assess the role of diet in cancer among free-living peo- ple. I n particular, limitations in the methods available to estimate diet create considerable uncertainty, and often scientific controversy, about the role of specific constituents of the diet in the etiology of specific can- cers.

Researchers have used four types of studies to as- sess the relationship between nutrition and cancer: ecologic, case-control, cohort, and experimental. In ecologic studies, either (1) the diets of different popu- lations can be correlated to their different rates of can- cer at a point in time or (2) a single population with a changing diet can be compared with regard to its changing cancer rates over time. The most informative ecologic analyses have been the migrant studies that show increasing cancer rates for migrants moving from low-risk countries to high-risk countries. These studies paint a simple picture: something in the life experience of people in modern societies increases cancer risk. Case-control studies compare the recalled diets of people newly diagnosed with cancer (cases) to those from people without cancer (controls), and cohort studies follow large groups of healthy people to relate cancer incidence to diets as measured years before. Experimental studies are controlled trials in which, by random assignment, healthy people are given alternative diets, or alternative nutrient supple- ments, to see if they affect cancer risk years later. Very few experimental trials for cancer prevention have been completed.

METHODS The epidemiologic literature on nutrition and cancer pertaining to cancers of the lung, colon, breast, and prostate was reviewed by searching Medline. Review and commentary articles and studies presenting data on the diets of Native Americans and Alaska Natives were searched for in the same way. The findings were summarized in tables and figures.

FINDINGS The epidemiologic studies on diet and cancer are not reviewed in detail here, as several recent reviews have been published."" Figures 1 and 2 summarize the findings from those case-control and cohort studies that have tested the major dietary hypotheses related to the major cancer sites. (Full citations for the studies summarized in the figures can be found in the review articles). In these figures, the relative risks of cancer are plotted for those in the highest level of intake of a particular nutrient or food, relative to those with the lowest intake. In most studies, the extremes being compared are usually the highest versus the lowest quartiles or quintiles. Hence, relative risks lower than 1 .O imply a protective relationship between the nutri- ent or foods and cancer risk, whereas relative risks greater than 1.0 imply a risk-enhancing effect of a food or nutrient.

Figure 1 shows that the risk of both colon cancer and lung cancer has been found to be inversely associ- ated with the intake of vegetables in most studies. Risk is approximately 50% lower for those in the upper quartile or quintile of intake than those in the lowest quartile or quintile. In many studies, this contrast would be for those consuming five or more servings per day compared with those consuming less than two servings per day. These associations are independent of tobacco use.

Figure 2 shows that the risk of both colon cancer and prostate cancer have been found to be positively associated with the intake of dietary fats. Those in the upper quartile or quintile have approximately 50% higher risk than those in the lowest quartile or quintile in these studies. Dietary relationships with breast can- cer are much weaker than those for cancers of the colon, lung, and prostate (data not ~hown). '" '~ In sum- mary, for the major cancer sites, vegetables and fruits have been found to be associated with lower risk for cancers of the colon and lung, and dietary fat has been found to be associated with higher risk for cancers of the colon and prostate.

Much of the research on the diets of American Indians and Alaska Natives has been qualitative re- search done by anthropologists, with an intent to study the unique features of current diets as indicators of the characteristics of past diets, using oral histories and traditional recipes." In general, these studies have suggested that the traditional diets of many Native Americans were based on a variety of foods from plant sources, that meat was lean, fat intakes were low, and regular physical activity was a key element in the life style. Quantitative assessments of the current diets of American Indian and Alaska Native tribes have been much less common. American Indians and Alaska Na-

1614 CANCER Supplement October 1, 1996 / Volume 78 / Number 7

Colon Cancer Lung Cancer

1.5 1.5

I 1

0.5 0.5

n

FIGURE 1. Relative risks for colon and lung cancer from case-control and cohort studies that have presented findings for vegetables in the diet."' Some studies presented findings for fruits and vegetables together, and others for dietary carotene intake (a nutrient derived from fruits and vegetables). The relative risk is the risk of cancer for those in the highest quartile or quintile of dietary intake, relative to those in the lowest quartile or quintile. Relative risks under 1 .O therefore suggest that vegetables are associated with lower cancer risk.

2.5 1 Colon Cancer Prostate Cancer

2.5 1

FIGURE 2. Relative risks for colon and prostate cancer from case-control and cohort studies that have presented findings for total dietary fat.8~9~'1~12 The relative risk is the risk of cancer for those in the highest quartile or quintile of dietary fat intake, relative to those in the lowest quartile or quintile. Relative risks over 1.0 therefore suggest that dietary fat is associated with higher cancer risk.

tives have not been specifically sampled in the surveys of the National Nutrition Monitoring system.16 In re- cent years, however, dietary surveys have been con- ducted among several tribes (Zephier E. Personal com- munication, 1995; White L. Personal communication, 1995).'7-2' These surveys, summarized in Table 1, have generally shown the current diets of members of these tribes to be similar to that of the American population as a whole, with high levels of dietary fat intake and low levels of fruit and vegetable intake. Obesity has

also been found to be an increasingly common condi- tion among American Indians and Alaska Natives.'3324

CONCLUSIONS As American Indian and Alaska Native populations have developed in the 20th century, their diets have increased in fat and decreased in fruits, vegetables, grains, and legumes; in addition, their physical activity has decrea~ed."~'~ Traditional Native American diets were based on foods derived largely from plant

TABLE 1 Selected Characteristics of Quantitative Dietary Surveys of American Indian or Alaska Native populations

~~~ ~~ ~

Investigator (re0

Wolfe (17) Story (18) Teufel (191 Nobmann (20) Riissell (21) Brown (22) Zephier hhite

Tribe Dates Sample Diet method

Navajo Cherokee Havasupai Alaska Natives Mvskoke Hopi 10 tribes Navaio

1979-80 approx 1984 1985-86 1987-88 approx 1991

1988-91 1990-91

1991-92

107 women 277 teens 28 women 351 adults 180 adults 96 children 875 adults 985 adults

~~

24 hour recalls 24 hour recalls 24 hour recalls 24 hour recalls food frequency 3 day records 24 hour recalls 24 hour recalls

sources, complemented with fish and low-fat wild meats, all balanced by regular physical activity. This same nutritional transition has been experienced by peoples in developing countries elsewhere in the ~ o r l d . ~ ' The emergence of chronic diseases, including cancer, among populations from developing countries in recent years may have resulted in large part from this nutrition transition marked by increases in the levels of fat in the diet, decreases in foods coming from plants (grains, fruits, vegetables, and legumes), and decreases in the levels of physical activity.

Coincident with the well-documented increases in obe~i@'*'~ has been an increase in non-insulin- dependent diabetes mellitus in American Indian Iribes.'" Interestingly, although obesity is an estab- lished risk factor for cancers of the c ~ l o n ~ ~ ~ and breast, these cancers are not seen more frequently among persons with diabetes and they are not seen more frequently among Native Americans. This may suggest that the syndrome of insulin resistance, which leads to obesity and diabetes in American Indians, is unrelated to the energy balance factors that account for the obesity-cancer connection.

Migrant studies give us some sense about the time it might take for dietary change to create changes in cancer risk. Persons migrating from Japan (where rates of colon and breast cancers have historically been low) experience increased rates of breast cancer only after several decades or into subsequent generations, whereas the increases in colon cancer rates are seen much sooner, even among those who migrated during adulthood.29 Thus, we might expect to see different latencies in the changes in rates of cancers following dietary changes among populations such as American Indians and Alaska Natives.

Although there are many advantages to the tech- nologic revolution of the 20th century, the ready sup- ply of high-fat foods, the erosion of the dietary base of foods derived from plants, and increased physical sedentariness do not seem to offer any long-term

Nutrition and Cancer among IndianslByers

health advantages. Returning to historically traditional diets, with attention to practicality, palatability, eco- nomics, and convenience, is a worthy challenge for all of us?" Small-scale experimental studies have demon- strated the benefits of traditional diets for the modern- day problems of glucose i n t ~ l e r a n c e . ~ ' - ~ ~ Although re- turning to hunter-gatherer life styles is not a viable alternative, those with higher regard for historical cul- tural traditions, such as many American Indians and Alaska Natives, may be well prepared to integrate past principles of the value of good foods and physical ac- tivity to achieve future nutritional improvements to reduce cancer risk.

REFERENCES

1615

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

16.

17.

Rhodes ER, Hammond J , Welty TK, Handler AO, Amler RW. The Indian burden of illness and future health interventions. Public Health Rep 1987; 102:361-8. Nutting PA, Freeman WL, Helgerson SG, Risser DR. Cancer incidence in American Indians and Alaska Natives. Am J Public Health 1993;83:1589-98. Valway S, Kileen M, Paisano R, Ortiz E. Cancer mortality among Native Americans in the United States: regional dif- ferences in Indian health, 1984-1988 and trends over time. 1968-1987. Rockville, MD: Indian Health Service, 1992. Lanier AP, Bulkow LR, Ireland B. Cancer in Alaskan Indians, Eskimos, and Aleuts, 1969-83: implications for etiology and control. Public Health Rep 1989; 104:658-64. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States. J Nut1 Can- cer lnst 1981; 66: 1191-308. Byers T. Diet and lung cancer. Samet J, editor. Epidemiology of lung cancer. New York Marcel Dekker, 1994. Ziegler R, Taylor-Mayne S. Swanson C. Nutrition and lung cancer. Cancer Causes Control 1996; 7:157-77. Giovannucci E, Willett WC. Dietary factors and the risk of colon cancer. Ann Merl 1994;26:443-52. Potter 7. Nutrition and colorectal cancer. Cancer Causes Con- trol 1996;7:127-46. Byers T, Guerrero M. Epidemiologic evidence for vitamin C and vitamin E in cancer prevention. Am J CIin Nutr 1995;

Pienta K, Esper P. Is dietary fat a risk factor for prostate cancer? J Nut1 Cancer Irzst 1993;85:1583-90. Kolonel L. Nutrition and prostate cancer. Cancer Causes Conrrol 1996;7:83-94. Byers T. Nutritional risk factors for breast cancer. Cancer

Hunter D, Willett W. Nutrition and breast cancer. Cancer Causes Conrrol 1996;7:56-68. Burhansstipanov L, Dresser CM. Native American mono- graph no. 1: documentation of the cancer research needs of North American Indians and Alaska Natives. NIH publica- tion no. 94-3603. Washington, DC: National Cancer Insti- tute, 1994. Fanelli-Kuczmarski M, Moshfegh A, Briefel R. Update on nutrition monitoring activities in the United States. J An1 Diet Assoc 1994;94:753-76. Wolfe WS, Sanjur D. Contemporary diet and body weight of Navajo women receiving food assistance: an ethnographic and nutritional investigation. J Am Diet Assoc 1988; 88:822- 7

62:1385S-92S.

1994; 74288-95.

1616 CANCER Supplement October 1, 1996 / Volume 78 / Number 7

18.

19.

20.

21.

22.

23.

24.

25.

26

Story M, Tompkins RA, Bass MA, Wakefield LM. Anthropo- metric measurements and dietary intakes of Cherokee I n - dian teenagers in North Carolina. J Am Dier Assoc 1986; 86: 1555-60. Teufel NI, Dufour DL. Patterns of food use and nutrient intake of obese and non-obese Hualapai Indian women of Arizona. JAnz Diet Assoc 1990;90:1229-35. Nobmann ED, Byers T, Lanier AP, Jackson MY. The diet of Alaska Native adults: 1987-1988. Anz J Clin Nutr 1992;

Russell ME, Weiss KM. Buchanan AV, Etherton TD, Moore JH, Kris-Ethereton PM. Plasma lipids and diet of the Mvskoke Indians. An2 I Clin Nutr 1994; 59:847-52. Brown AC, Brenton B. Dietary survey of Hopi Native American elementary students. J Am Diet Assoc 1994;

Broussard B, Johnson A, Himes JH, Story M, Fichtner R, Hauck F, et al. Prevalence of obesity in American Indians and Alaska Natives. Am J C h i Nitfr 1991; 53:15358-428. Byers T. The epidemic of obesity in American Indians. Am

Boyce VL, Swinburn 84. The traditional Pima Indian diet. Diaberes Cure 1993; 16:369-71. Jackson MY. Nutrition i n American Indian health: past, pres- ent, and future. J Am Diet Assoc 1986;86:1561-5.

55:1024-32.

49:517-22.

J D ~ s Child 1992; 1461285-6.

27. Popkin BM. The nutrition transition in low income coun- tries: an emerging crisis. Nutr Rev 1994; x?:285-98.

28. Sugarman JR, Hickey M, Hall T, Gohdes D. The changing etiology of diabetes mellitus among Navajo Indians. West J Med 1990; 153:140-5.

29. Haenszel W, Kurihara M. Studies of Japanese migrants: part 1. mortality from cancer and other diseases among Japanese in the United States. J Natl Cancer Inst 1968;40:43-68.

30. U.S. Department of Agriculture and U.S. Department of Health and Human Services. Nutrition and your health: di- etary guidelines for Americans. 4th ed. Home and Garden bulletin no. 232. Washington, DC: USDA, 1995.

31. Adler Al, Schraer CD, Boyko EJ, Murphy NJ. Lower preva- lence of impaired glucose tolerance and diabetes associated with daily seal oil or salmon consumption among Alaska Natives. Diubefes Care 1994; 17:1498-501.

32. Ravussin E, Bennett PH, Valencia ME, Schulz LO, Esparza J. Effects of a traditional lifestyle on obesity in Pima Indians. Diabetes Cure 1994; 17:1067-74.

33. Swinbum BA, Boyce VL, Bergman RN, Howard BV, Bogardus C. Deterioration in carbohydrate metabolism and lipoprotein changes induced by modern, high-fat diet in Pima Indians and Caucasians. I Clin Eizdocnnol Metab 1991; 73:156-65.

34. Cowen R. Seeds of protection: ancestral menus may hold a message for diabetes-prone descendants. Science 1990; 137:350-1.