Fat and Dietary Fiber Intake and Colon Cancer Mortality


Abstract: To estimate the role of dietary fiber (DF) and fat in the striking growth of colon cancer mortality in Japan after World War II, we analyzed relations between the above variables in comparison with those in the United States. In the United States, fat intake grew by only one-third over the past 70 years (from 124 g in 1909-1913 to 166 g in 1984), whereas colon cancer mortality increased fourfold (from 5 to 20 per 100,000). In Japan, although fat intake roughly doubled during the 40 years after World War II (from 20 to 38 g), colon cancer mortality grew 5.5-fold (from 2 to 11 per 100,000). It is difficult to give a consistent explanation for the growth patterns of colon cancer mortality in both countries on the basis of fat consumption as a cancer promoter. In the United States, DF intake continuously dwindled at a level always less than in Japan throughout this century. DF intake in Japan also declined rather steadily, except for war time, over the past 80 years. However, with regard to the growth pattern of colon cancer mortality, it began rising steeply around the period when the daily DF intake diminished below 20 g, suggesting the presence of a threshold level in this neighborhood in preventing the development of colon cancer.


Mortality from colon cancer is higher in the Western countries than in Japan, although its growth in recent years is also remarkable in Japan. Inasmuch as the incidence of colon cancer among immigrants converges to that in their adopting country, environmental factors, particularly the changes in dietary pattern, are considered decisively important for the development of colon cancer ( 1-7).

There are two major hypotheses regarding causes for developing colon cancer. Burkitt and co-workers ( 8-10) hypothesized that a diet low in dietary fiber (DF) prolongs the fecal retention time in the large intestine, also prolonging the contact time of carcinogens with the intestinal lumen and thus raising cancer risk. Hill and colleagues ( 11-13) proposed that the high-fat, high-protein diet commonly consumed in Western countries stimulates the discharge of bile, which is transformed to secondary bile acids, the promoter of colon cancer.

As Japanese dietary habits after World War II became increasingly Westernized, fat intake grew and DF intake diminished, thereby raising the incidence of colon cancer ( 14-17). However, because the composition of a Japanese meal based on rice as a staple food is considerably different from meals in Western countries, there remains a possibility that the effect of dietary factors on the incidence of colon cancer may be accordingly modified ( 18-23).

We analyzed the chronological changes of DF and fat intake in relation to the pattern of colon cancer mortality in the United States and Japan; the former has been characterized by high colon cancer mortality, a high-fat diet, and relatively low DF intake throughout the century, and the latter has been characterized by a recent surge of colon cancer mortality combined with rapid dietary changes since the end of World War II.


Statistical Data Concerning Colon Cancer Mortality

We used the vital statistics of Japan for the period from 1947 to 1988 to obtain the mortality rate from colon cancer (excluding rectal cancer) ( 24-26). Data before 1947 were not used, because they were limited to only a small number of prefectures and probably were not representative of the whole nation.

For the United States, the vital statistics between 1900 and 1986 were used ( 27, 28). Colon cancer was described as one of the cancers of the digestive tract and peritoneum between 1900 and 1945. Because it was not classified under an independent category until 1946, we calculated the mortality of colon cancer before 1946 according to the proportion indicated in the years from 1946 and 1947.

Data Concerning Fat and DF Intake

Plant-derived DF alone was included. With regard to data concerning dietary habits in Japan from 1911 to 1935, those published by the Science Council as the Current Status of National Diet (revised by the Science and Technology Agency) ( 29) were used. For postwar years, the National Nutrition Survey was used ( 30). The amount of DF was calculated on the basis of measurements reported by the Chikenkyo in 1989 and its food group substitution tables ( 31).

The US food intake according to "food group" between 1907 and 1975 was given by Page and Friend ( 32). Food composition by groups was based on the simulated American diet calculated by Anderson and associates (33; also see Reference 34). To estimate DF content, the measurements reported by the Chikenkyo were employed, except for the period between 1976 and 1980, for which the values given by Lanza and others ( 35) were employed.


Changes in Colon Cancer Mortality

The mortality rate from colon cancer in Japan was reported by the vital statistics as 2 in 100,000 for the first time in 1947. Colon cancer mortalities were much higher in other industrialized countries at that time, particularly in the United States, where its rate was about eight times higher than in Japan (Figure 1).

Chronological changes in colon cancer mortality in Japan starting from 1947 show initially slow and then increasingly rapid rises; it rose twofold over 20 years (1948-1968), threefold in eight years (1969-1976), fourfold in five years (1977-1981), and fivefold in four years (1982-1985). In 1988, the mortality was six times higher (12 of 100,000) than in 1947 (2 of 100,000).

The mortality rate from colon cancer in the United States was 5 of 100,000 during the period 1901-1905, then doubled in two decades (1926) and tripled in 17 years (1943), and the speed of its rise slowed and quadrupled 34 years later (1971) to 20 in 100,000.

Changes in Fat Intake

Fat intake in Japan remained unchanged at about 15 g from the early 20th century to 1950 (Figure 2). Between 1956 and 1970, the Japanese diet was considerably Westernized, and fat intake almost doubled (from 23 to 44 g). After 1971 the increment of fat intake leveled off after the peak at 58 g (1982-1984).

Fat intake in the United States grew from 125 g in the period 1909-1913 to 134 g in 1925-1929 (Figure 2). In the 1930s it temporarily dropped, perhaps because of the recession but turned upward again, reaching 152 g in 1975.

A comparison of apparent fat intakes in Japan and the United States (Figure 2) reveals that consumption was about 10 times greater in the United States than in Japan in the 1910s and about 9 times greater even in the 1930s. Although fat intake increased in postwar Japan starting in the latter half of the 1950s, in the United States it was still four times greater than in Japan in 1965 and still three times greater in 1975. Americans still consume far more fat than the Japanese, and the margin between these countries remains relatively unchanged. Fat intake in Japan remained fairly constant between 50 and 60 g for 15 years from the early 1970s.

Changes in DF Intake

Chronological changes in DF intake in Japan (Figure 3) show that it was >21 g from the early 20th century to 1955, including the years when Japan was engaged in the war. After 1951-1955, Japanese dietary habits diversified and food consumption increased as the economy boomed: It increased by 1.3 times from 1951-1955 to 1971-1975. However, DF intake decreased over these years: to 21 g in 1951-1955, to 19 g in 1956-1960, to 17 g in 1971-1975, and to 15 g in 1986-1987, which is almost 30% less than for 1951-1955.

The diminishing DF intake is explained by a reduced grain intake; grain accounted for 40% of the diet until 1955 but only 25% after 1966. This is due to diminishing staple food consumption and changes in breakfast ( 30, 36). The DF content per kilogram of food also declined from 19 g in 1951-1955 to about 12 g after 1971. The overall dietary pattern of the Japanese has shifted to a greater consumption of foods containing less DF.

DF intake in the United States was 22 g in 1909-1913 (Figure 3) and dwindled considerably from 1935-1939 to 1957-1959 (from 20 to 16 g), coinciding with the rise in colon cancer mortality (1931-1945). DF intake after 1965 leveled off until 1975. In the meantime, the DF content per kilogram of food was 11 g for 1909-1913 and declined considerably from 1935-1939 to 1947-1949 (from 10 to 8 g), reducing the DF level in 1970 to 72% of that of the initial period (19091913). Principal DF sources in the United States were grains ( 32-40%) for the period 1909-1939, followed by vegetables ( 21-26%). After 1947, grains and vegetables contributed approximately the same amount (30%).

A comparison of DF intakes between the two countries reveals that Japan's DF intake was 12-16% greater for the period 1956-1970, but the margin of difference between the two countries diminished thereafter (Figure 3).

Relationship Between Colon Cancer Mortality and Dietary Factors
Although in the United States colon cancer mortality and fat intake were already high from the outset, Japan had low colon cancer mortality, low fat intake, and relatively high DF intake at the time of starting statistical registration.

Japan seems to provide a uniquely suitable case to analyze the extent of contributions to colon cancer mortality made by the dietary factors. From 1947 to 1955, colon cancer mortality in Japan was still very low (2 of 100,000) and began an exponential upsurge in the next two decades (Figure 1). Therefore, it is expected that the colon cancer mortality may reflect either or both chronological changes in fat and DF intake.

With regard to fat contribution, the following analysis by staging the mortality of Japan and the United States is illustrative (Table 1): Whereas the shift in mortality for Japan from Stage 1 (6 of 100,000) to Stage 3 (11 of 100,000) took only 10 years, it took 25 years for the United States. Meanwhile, fat intake in Japan remained at only about 40% (56-58g) of that in the United States (124-133g); moreover, in terms of consumption rate, the increase for Japan was only 3.6% (2 g) compared with 5.6% (9 g) for the United States. Thus we suggest that fat, in terms of quantity and increment rate in consumption, contributed relatively little to the rapid growth of colon cancer mortality, which continued throughout these stages in Japan.

On the other hand, DF intake was 21 g during the immediate postwar period (1947-1955), when there was no notable increase in colon cancer mortality. However, DF intake subsequently diminished steadily (Figure 3) and rather linearly.

Suppose that the dwindling DF intake was mainly responsible for the increase in colon cancer mortality; its exponential upsurge starting around the late 1960s is better explained by assuming a threshold level of DF intake, which would give an effective protection against the development of colon cancer. Figure 3 shows that the DF intake for the period before the upsurge (1956-1965) was about 20 g, suggesting that the threshold exists in this vicinity.

In the United States, although a decrease in DF intake and an increase in fat intake undoubtedly raised colon cancer mortality, there is no clue in its growth pattern that would permit further analysis.

On the basis of the hypotheses that fat promotes development of colon cancer and DF inhibits it, we examined the relations among these factors concerning Japan and the United States. One characteristic finding was that Japan, starting from a very low level, had a steep rather than an exponential rise in mortality, which began somewhere in the late 1960s to early 1970s. This fact is difficult to explain in quantitative terms by an increase in fat intake during the postwar period but can be explained by assuming a threshold level for DF to exert an effective inhibition on the development of colon cancer.

Although it may be argued that the sensitivity toward carcinogenicity of fats in the colon may be different between Japanese and Americans, it is clear from the reports concerning immigrants that there are no such differences ( 1-7). With regard to immigrants from Japan living in the United States, the colon cancer incidence converged from the level in their mother country to that in the adopting country. Thus, in appreciating fats as the promoter of colon cancer, there is no substantive racial difference between the Japanese and other ethnic groups in the United States.

To estimate the threshold level concerning the protection conferred by DF against colon cancer in Japan, it is necessary to take into consideration a lead time before the clinical manifestation of cancer ( 10-20yr) ( 37-39). Because its mortality rate appears to begin rising in the 1960s-1970s, it is tempting to assume that the DF intake of around 19-20 g for the 1950s represents the threshold level.

Analyzing the increased incidence of colonic diverticulosis in recent years in Japan, Ohi and co-workers ( 40, 41) suggested the presence of a threshold level of DF intake at around 20 g. Interestingly enough, the critical DF intake to exert the inhibitory effect against colon cancer appears to lie at the same level, although no relations between the two diseases have been noted.

One substantial limitation of our study is that methodologies estimating nutrient intakes are different between Japan and the United States: in Japan, nutrient intakes are estimated on the basis of the individual diet sampled for study, whereas in the United States, food "disappearance data," which will obviously overestimate actual consumption, are used ( 42). However, despite this limitation, it is possible to compare the long-term trends of nutrient intakes between the two countries so long as observational biases are consistent throughout data collection. Another ambiguity is the fact there are not yet established theories regarding the types of the effective constituents of DF that inhibit colon cancer generation and the effectiveness of these foods ( 43-45).

Acknowledgments and Notes
The authors thank Dr. Tsuguyoshi Suzuki for his insightful discussion, Address reprint requests to Dr. Ichiro Kai, Dept. of Social Gerontology, School of Health Sciences and Nursing, Tokyo University, 7-3-1 Hongo, Bunkyo-Ku, Tokyo 113-0033, Japan. E-mail: G38378@simail.ne.jp.

Submitted 6 January 1998; accepted in final form 4 September 1998.

Table 1.
Fat and Dietary Fiber Intake When Death Rates From Colon Cancer Became Equal in Japan and the United States

Legend for Chart:

A - Death Rates From Colon Cancer,[a] per 100,000
B - Period
C - Time Between Stages, yr
D - Fat, g/capita/day
E - Dietary Fiber, g/capita/day


Stage 1

Japan 6 1976-1980
-- 56 17
United States 6 1906-1910
-- 124[b] 22[b]

Stage 2 -- --
1 right -- --
arrow 2

Japan 9 1981-1985
5 58 16

United States 9 1926-1930
20 135[c] 21[c]

Stage 3 -- --
2 right -- --
arrow 3

Japan 11 1986-1989
5 58 15

United States 11 1931-1935
5 133[d] 20[d]
a: Values are rounded up or down to nearest whole number.

b: 1909-1913.

c: 1925-1927.

d: 1935-1939.

GRAPH: Figure 1. Chronological changes in colon cancer mortality in Japan (filled circles) and United States (open circles). Dotted lines, estimated values. *1, 1947-1950; *2, 1986-1988.

GRAPH: Figure 2. Chronological changes in fat intake in Japan (filled circles) and United States (open circles). *1, 1909-1913; *2, 1925-1929; *3, 1935-1939; *4, 1947-1949; *'5,1957-1959; *6, 1967-1969; *7,1975; *8, 1980; *9, 1984.

GRAPH: Figure 3. Chronological changes in dietary fiber intake in Japan (filled circles) and United States (open circles). *1, 1909-1913; *2, 1925-1929; *3, 1935-1939; *4, 1947-1949; *5, 1957-1959; *6, 1965; *7, 1970; *8, 1975.

1. Matayoshi, Z: Statistical observation of deaths among Japanese in Hawaii Island. Tohoku Igaku Zasshi 78, 62-162, 1969. [In Japanese]

2. Haenszel, W, Berg, JW, Seji, M, Kurihara, M, and Locke, FB: Large-bowel cancer in Hawaiian Japanese. JNCI 51, 1765-1776, 1973.

3. Hirohata, T, Tomita, J, and Shibata, A: Nutrition, food and cancer as seen from epidemiology. Eiyo to Shokuryo 33, 1-7, 1980. [In Japanese]

4. Smith, RL: Recorded and expected mortality among the Japanese of the United States and Hawaii, with special reference to cancer. JNCI 17, 459-473, 1956.

5. Buell, P, and Dunn, JE: Cancer mortality among Japanese Issei and Nisei of California. Cancer 18, 656-664, 1965.

6. Haenszel, W, and Kurihara, M: Studies of Japanese migrants. 1. Mortality from cancer and other diseases among Japanese in the United States. JNCI 40, 43-68, 1968.

7. Locke, FB, and King, H: Cancer mortality risk among Japanese in the United States. JNCI 65, 1149-1156, 1980.

8. Burkitt, DP: Epidemiology of cancer of the colon and rectum. Cancer 28, 3-13, 1971.

9. Burkitt, DP, Walker, ARP, and Painter, NS: Effect of dietary fiber on stools and transit-times, and its role in the causation of disease. Lancet 2, 1408-1411, 1972.

10. Burkitt, DP, Walker, ARP, and Painter, NS: Dietary fiber and disease. JAMA 19, 1068-1074, 1974.

11. Hill, MJ, and Aries, VC: Faecal steroid composition and its relationship to cancer of the large bowel. J Pathol 104, 129-139, 1971.

12. Hill, MJ, Crowther, JS, Drasar, BS, Hauksworth, G, and Aries, V, et al.: Bacteria and etiology of cancer of large bowel. Lancet 2, 95-100, 1971.

13. Hill, MJ, Drasar, BS, Williams, REO, Meade, TW, and Cox, AG, et al.: Fecal bile-acids and Clostridia in patients with cancer of the large bowel. Lancet 1, 535-538, 1975.

14. Ohta, M, Ishiguro, M, Iwane, S, Nakaji, S, Sano, M, et al.: Review of dietary fiber intake in patients with the diseases of colon. J Jpn Soc Dig Tract 82, 51-56, 1985. [In Japanese]

15. Minowa, M, Bingham, S, and Cummings, JH: Dietary fiber intake in Japan. Hum Nutr 37A, 113-119, 1983.

16. Kuratsune, M, Honda, T, Englyst, HN, and Cummings, JH: Dietary fiber in the Japanese diet as investigated in colon cancer risk. Jpn J Cancer Res 77, 736-738, 1986. [In Japanese]

17. Willet, W: The search for the causes of breast and colon cancer. Nature 338, 389-394, 1989.

18. Phillips, RL: Role of life-style and dietary habits in risk of cancer among Seventh-Day Adventists. Cancer Res 35, 3513-3522, 1975.

19. Armstrong, B, and Doll, R: Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer 15, 617-631, 1975.

20. Knox, EG: Foods and diseases. Br J Prev Soc Med 31, 71-80, 1977.

21. Greenwald, P, Lanza, E, and Eddy, GA: Dietary fiber in the reduction of colon cancer risk. J Am Dietet Assoc 87, 1178-1188, 1987.

22. Edere, F, Leren, P, Turpeinen, O, and Frantz, ID: Cancer among men on cholesterol-lowering diets. Lancet 2, 203-206, 1971.

23. Bingham, SA, Williams, DRR, and Cummings, JH: Dietary fiber consumption in Britain: new estimates and their relation to large bowel cancer mortality. Br J Cancer 52, 399-402, 1985.

24. Statistics and Investigation Department, Ministry of Health and Welfare: Kousei no Sihyo (Cancer) 27, 1980. [In Japanese]

25. Ministry of Health and Welfare: Mortality Statistics on Malignant Neoplasms (Special Report on Population Studies and Statistics 19721984), 1986. [In Japanese]

26. Ministry of Health and Welfare: Vital Statistics (1947-1987). [In Japanese]

27. US Department of Health and Welfare: Vital Statistics Rates in the United States, 1940-1960, 1968.

28. National Office of Vital Statistics: Vital Statistics of the United States, 1937-1976.

29. Oda, S, Nakahara, S, and Inoue, K: 40 years of National Nutrition Survey. Kousei no Shihyo 34, 1987. [In Japanese]

30. Nutrition Division, Public Health Bureau, Ministry of Health and Welfare: Current Status of National Nutrition, 1950-1989. [In Japanese]

31. National Council of Municipal Laboratories of Public Health: Study on DF Determination in Principal Foods, 1989. [In Japanese]

32. Page, L, and Friend, B: The changing United States diet. Bioscience 28, 192-197, 1978.

33. Anderson, JW, Bridges, SR, Tietyen, J, and Gustafson, NJ: Dietary fiber content of a simulated American diet and selected research diets. Am J Clin Nutr 49, 352-357, 1989.

34. Agricultural Study Bureau, Department of Agriculture, USA: Food Composition Table of USA. Kenpakusha, 1981. [Translation to Japanese]

35. Lanza, E, Jones, DY, Block, G, and Kessler, L: Dietary fiber intake in the US population. Am J Clin Nutr 46, 790-797, 1987.

36. Uchino, S: Dietary Life and Population Migration. Tokyo: Daiichi Shuppan, 1977, pp 212-214. [In Japanese]

37. Cummings, JH: Dietary fiber and large bowel cancer. Proc Nutr Soc 40, 7-14, 1981.

38. Narisawa, T, Takamatsu, M, Masuda, T, and Niwa, M: Factors related to development and inhibition of colonic cancer. Gan to Rinsho 32, 674-679, 1986. [In Japanese]

39. Winawer, SJ: Neoplasms of the large and small intestine. In: Cecil Textbook of Medicine. Philadelphia, PA: Saunders, 1985, pp 761-771.

40. Ohi, G, Minowa, K, Oyama, T, Nagahashi, M, Yamazaki, N, et al.: Changes in dietary fiber intake among Japanese in the 20th century, a relationship to the prevalence of diverticular disease. Am J Clin Nutr 38, 115-121, 1983.

41. Nagahashi, S, Yamazaki, N, Ohi, G, Kai, I, Suzuki, H, et al.: Dietary fiber intake and diverticulosis. Case control study. Nihon Eiseigaku Zasshi 40, 781-788, 1985. [In Japanese]

42. Raper, NR, and Marston, RM: Levels and sources of fat in the US food supply. In: Dietary Fat and Cancer. New York: Liss, 1986, pp 127-152.

43. Committee on Diet, Nutrition and Cancer; Assembly of Life Sciences National Research Council. Diet, Nutrition and Cancer. Washington, DC: National Academy Press, 1982.

44. Bingham, SA, and Cummings, JH: Dietary starch, non-starch polysaccharides and large bowel cancer. In: Proceedings of the Kellogg International Symposium on Dietary Fiber. Tokyo, Japan: Center for Academic Publications Japan, 1990, pp 97-117.

45. Japan Association of Nutrition: Dietary Fiber, revised ed, S Innami and S Kiriyama (eds). Tokyo: Daiichi Shuppan, 1995. [In Japanese]


By Teruko Honda; Ichiro Kai and Gen Ohi

T. Honda is affiliated with Yamaguchi Prefectural University, Yamaguchi, Japan. 1. Kai is affiliated with Tokyo University, Tokyo, Japan. G. Ohi is affiliated with the National Institute for Environmental Studies, Tsukuba, Japan.


Share this with your friends