Food consumption and dietary factors involved in health and disease in Nikkeis: systematic review

Tanabe, Fabiana Hitomi; Drehmer, Michele; Neutzling, Marilda Borges

doi:10.1590/S0034-8910.2013047003377

OBJECTIVE:

To analyze food consumption and dietary factors involved in the Nikkei population's health and disease processes

METHODS:

A systematic review of observational studies that described Nikkeis' food intake was carried out in electronic databases Lilacs, SciELO and PubMed/Medline databases, from 1997 to 2012. Initially, 137 titles and abstracts were analyzed, excluding intervention studies, those which only presented metabolite and vitamin plasma levels and those which did not meet the objective of this study. Of these, 38 studies were selected and evaluated using a method based on Downs & Black (1998), adapted for observational studies, leaving 33 studies to be analyzed.

RESULTS:

Few studies about Nikkei food intake were found outside of Hawaii, in the United States, and Sao Paulo (mainly in the city of Bauru) in Brazil. The total energy intake of Japanese-Brazilians had an elevated fat contribution, decreasing carbohydrate and protein intake. In the United States, the prevalence of Japanese-Americans who consumed high density energy food was elevated. The Niseis (children of immigrants) presented, on average, higher intake of Japanese food products, while the Sanseis (grandchildren of immigrants) showed more Westernized dietary habits.

CONCLUSIONS:

Although some Japanese food habits have been maintained, the Nikkeis' dietary intake reveals a high prevalence of typical Westernized food intake (high in processed food, fat and sodium, and poor in fiber), that may be contributing to the increasing development of chronic disease in this population.

Food Consumption; Food Habits; Acculturation; Emigrants and Immigrants; Japan; Observational Studies; Review Literature as Topic

[1] The authors declare that there are no conflicts of interest.

Authors e score	Location and subjects	Year	Methodology (Study Design and Evaluation Instrument for Food Consumption)	Key findings

de Salvo et al ³⁷ score: 15	Bauru, SP, Brazil 1,165 Japanese-Brazilians with and without macrovascular disease, aged 30 and over.	2009	Cross-sectional study FFQ validated for Japanese-Brazilians	Caloric distribution of Nikkei diet: 54% carbohydrate, 14% protein and 32% fat. In Japanese-Brazilians with macrovascular disease increased consumption of red meat, protein, trans fat, sodium, and low intake of fiber and calcium were observed.
Yamashita et al ⁴⁸ score: 18	Bauru, SP, Brazil 1,009 married Japanese-Brazilians individuals.	2009	Cross-sectional study FFQ validated for Japanese-Brazilians	Inter-ethnic marriage (Nikkei man and non-Nikkei woman, Nikkei woman and non-Nikkei man): higher average values of total calories, carbohydrates, proteins, lipids and saturated fats Intra-ethnic marriage (Nikkei man and Nikkei woman) in both sexes, higher intake of carbohydrates, vegetables and miso soup. Increased intake of fruits and fruit juices in men. Increased consumption of cereal and protein in women.
Massimino et al ²⁶ score: 14	Bauru, SP, Brazil 647 Isseis and Nisseis aged 40-79	2007	Cohort study FFQ adapted by Tsunehara et al	In individuals with peripheral arterial disease fiber intake of whole grains and linoleic acid was lower and saturated fat intake was higher than in individuals without the disease. Increased mortality among individuals with lower consumption of carbohydrates and cholesterol.
Bertolino et al ¹ score: 19	Bauru, SP, Brazil 328 men and women first (Issei) and second generation (Nisei), 40-79 years of age	2006	Comparison of transversal data (from two surveys of health and nutrition) FFQ adapted by Tsunehara et al (1993) and FFQ validated for Japanese-Brazilians (2000)	Lack of association between changes in the consumption of trans fatty acids and changes in serum lipid profile. Level of trans fat intake was above that recommended by WHO (up to 1% of total calories).
De Castro et al ⁵ score: 19	328 individuals aged 40-79	2006	Comparison of transversal data FFQ adapted by Tsunehara et al (1993) and FFQ created and validated for Japanese-Brazilians (2000)	There was a decrease in cholesterol and increased consumption of linoleic and oleic fatty acids and fruit in the diet of Japanese-Brazilians. However, there is also an increased intake of total fat. Japanese-Brazilian women had an increased caloric value of their diets, with reduced contribution candy. In both sexes there was a decrease in caloric contribution of protein and increase in fats, dairy, fruits and fruit juices
Damião et al ⁴ score: 19	Bauru, SP, Brazil 218 individuals, aged de 40-79 without DM or hypertension or dyslipidemia	2006	Cohort study FFQ validated for Japanese-Brazilians	After 7 years of follow-up, Japanese-Brazilian men and women decreased protein intake and total cholesterol and increased intake of total fat, particularly monounsaturated and polyunsaturated. Intake of red meat and whole milk and skim milk decreased increased. The usual intake of red meat was associated with metabolic syndrome, especially among men
Freire et al ⁸ score: 19	Bauru, SP, Brazil 877 individuals aged 30 and over	2005	Cross-sectional study FFQ validated for Japanese-Brazilians	The total fat intake may be a risk factor for metabolic syndrome (MS) in a population of Japanese-Brazilians. Men with MS: significantly higher intake of total fat, oleic acid and cholesterol. Consumption of fried foods was associated with an increased risk of MS in Japanese-Brazilians.
Freire et al ⁹ score: 18	Bauru, SP, Brazil 1.283 individuals aged between 30-90	2003	Cross-sectional study FFQ validated for Japanese-Brazilians	A common dietary pattern is shared between genders and generations of Japanese-Brazilians, with high risk of metabolic syndrome. The main difference in food consumption between generations studied was miso soup, being most commonly consumed among those of the first generation. The individuals of the second generation consumed more fats and oils. The consumption of traditional Japanese food was rare in both the first and second generation.
Ferreira et al ⁷ score: 10	Bauru, SP, Brazil 530 individuals, aged 40-79	2002	Cross-sectional study FFQ adapted by Tsunehara et al	Obese individuals and those with central adiposity had a greater proportion of the consumed energy from fat and carbohydrates lower (p <0.05). The second generation showed higher energy consumption than the first generation (p <0.05). After adjustment, protein intake was the only variable significantly associated with BMI.
Hamada et al ¹⁵ score: 14	São Paulo, SP, Brazil 96 cases and 192 controls Japanese-Brazilians aged between 38 and 89	2002	Case-control study Non-validated FFQ with 30 items	The introduction of the daily habit of meat consumption among Japanese immigrants and their descendants may be associated with stomach cancer. The protective effect of fruit consumption was confirmed in this population.
Costa et al ³ score: 19	Bauru, SP, Brazil 530 Japanese-Brazilians individuals, aged between 40-79	2000	Cross-sectional study FFQ adapted by Tsunehara et al	Results of the study do not support a significant association between nutritional factors and glucose intolerance. However, they suggest that consumption of foods rich in cholesterol and protein may be markers of increased BMI (which can lead to insulin resistance, preceding diabetes).
Cardoso et al ² score: 13	São Paulo, SP, Brazil 166 individuals aged between 40-69	1997	Comparison of cross-sectional data FFQ adapted by Tsunehara et al. (1 ^st stage in 1989), 3-day-food-diaries and repetition of the same FFQ (2 ^ndstage in 1995).	Japanese-Brazilians born in Brazil have a higher intake of fats and oils, vegetables, chicken and red meat, dairy products and coffee Nikkeis born in Japan report high consumption of seaweed, green teat, fish and Miso soup. The two generations of Japanese-Brazilians report infrequent consumption of soya, pickled vegetables and green tea.

Authors and score	Location and subjects	Year	Methodology (Study Design and Evaluation Instrument for Food Consumption)	Key findings

Sharma et al ⁴² Score: 20	Hawaii and California, USA 186,916 individuals 45-75 years of age living in Hawaii and Los Angeles 1993-1996	2012	Multi-ethnic cohort study Validated FFQ	Main foods consumed by Americans of Japanese origin: rice, bread, cereals, chicken, turkey, nuts, fish, salad dressing, butter, fried meat with vegetables; orange juice or grape grapefruit; bananas, tropical fruits, juices and soft drinks artificial fruit
Steinbrecher et al ⁴³ score: 20	Hawaii and California, USA 29,759 Caucasians, 35,244 Japanese-Americans and 10,509 Native Americans, 45-75 years of age at baseline	2011	Multi-ethnic cohort study Validated FFQ	Red meat consumption was positively associated with the risk of diabetes in men (fifth quintile vs. first quintile: HR: 1.43) and women (fifth versus first quintile: HR: 1. 30) in adjusted models. For men, there were significant interactions of ethnicity with the consumption of red meat and processed red meats: Caucasians had slightly higher risks than Japanese-Americans.
Hopping et al ¹⁸ score: 14	Hawaii and California, USA 75,512 Caucasians, Japanese Americans and Native Hawaiians	2010	Multi-ethnic cohort study FFQ	High intake of fiber from grains significantly reduced (by 10%) the risk of diabetes in men and women. High fiber intake (vegetable) decreased the risk by 22% in males but not in females. The intake mg reduced the risk (HR = 0.77 and 0.84) for men and women, respectively.
Hu et al ²⁰ score: 20	Massachusetts, Illinois, Michigan, California, New Jersey, EUA 2,025 African-American, Chinese, Caucasian, Japanese-American and Hispanic women, between 46-58 years of age	2009	Population based cohort study SWAN dietary questionnaire – modified version of the Block FFQ, 1995	The glycemic load was highly correlated with caloric intake and total CHO. Rice was the largest source of glycemic load among Japanese-American, contributing about 35% of the total, followed by bread and orange juice.
Henderson et al ¹⁶ score: 17	Hawaii and California, USA Subjects: 139,406 Japanese-American individuals, African-Americans, Latinos, whites and Hawaiians	2007	Multi-ethnic cohort study Questionnaire used not specified.	The highest intake of saturated fat was associated with higher mortality from cardiovascular disease. In Japanese-American men, this consumption was dramatically lower compared with other ethnic groups.
Pierce et al ³⁵ score: 18	Washington, EUA 496 Japanese-Americans	2007	Cross-sectional study FFQ with 40 items	Japanese-Americans of the second generation consumed more fish, tofu and tsukemonos, while the third ate more cheese, red meat and soft drinks. In the latter, the Western dietary pattern was significantly associated with the prevalence of diabetes.
Stram et al ⁴⁴ score: 18	Hawaii and California, USA 82,486 white, African-American, Japanese-American, Latino and Hawaiian men	2006	Multi-ethnic cohort study Food frequency questionnaire, calibrated	No protective effect arising from the consumption of fruits and vegetables was found in relation to prostate cancer in any of the studied groups.
Howarth et al ¹⁹ score: 18	Hawaii and California, USA 191,023 Latino, African-American, Caucasian, Japanese-American and Hawaiian men and women	2006	Multi-ethnic cohort study Calibrated FFQ	The mean energy density (ED) and BMI was lower among Japanese Americans than any other ethnic group studied. After adjusting for the amount of food consumed daily, age, current smoking, physical activity, chronic disease and education, an increase of 1 kJ / g ED was associated with an increase in BMI of 1 kg/m2 in each ethnic group.
Gold et al ¹⁴ score: 18	Pennsylvania, Massachusetts, Michigan, California, New Jersey, EUA 3,302 African-American, Latino, Japanese-American, Caucasian and Chinese women aged 42-52	2004	Cross-sectional study SWAN dietary questionnaire – modified version of the Block FFQ, de 1995, including Japanese and Chinese food items	Although no nutrient was significantly associated with vasomotor symptoms, in Japanese-Americans a positive relationship was indicated between intake of genistein, a soya phytoestrogen, with reported symptoms.
Willcox et al ⁴⁶ score: 19	Hawaii, USA 1,915 Japanese-American men between 45-68 years of age at recruitment	2004	Cohort study 24-hour recalled food intake and 7-day-food-diary in a baseline subsample FFQ validated in the second monitoring	Lower mortality in individuals who were in the second quintile of energy intake, suggesting that men who consumed 15% below the average of the group had a lower risk of mortality from all causes. Increased consumption of mortality was below 50% of the average for the group
Laurin et al ²⁵ score: 19	Hawaii, USA 2,459 Japanese-American men aged 45-68 years old in 1965-1968	2004	Prospective community – based study 24-hour recalled food intake	Intake of beta-carotene, flavonoids, vitamins E and C were not associated with risk of dementia or its subtypes. This study suggests that dietary intake of antioxidants in middle age does not modify the risk of dementia or its subtypes prevalent in a period later in life.
Wu et al ⁴⁷ score: 18	California, EUA 501 cases and 594 controls: women of Japanese, Chinese and Filipino origin, aged 25-74	2002	Case-control study FFQ with 14 food items rich in soya	Soy consumption, particularly in early childhood, can have a protective effect in later life on the risk of breast cancer.
Rice et al ³⁶ score: 16	Seattle, EUA 274 Japanese-American women between 65-93 years of age	2001	Cross-sectional study Nikkei soy food frequency questionnaire (NSFFQ), with 14 food items rich in soya	Foods rich in soy were more consumed: tofu (soybean), miso (fermented soybean paste) and aburaage (fried tofu). The intake of soy isoflavones in the diet was positively associated with the Japanese language, the consumption of traditional Japanese dishes (kamaboko, manju and mochi), skim milk and yellow / red vegetables, vitamin E supplement use and walking several blocks per day. Intake of soy isoflavones in the diet was negatively associated with the consumption of butter.
Kolonel et al ²³ score: 11	Hawaii and California, USA 215,251 men and women from five different ethnicities, aged 45-75 at the time of recruitment	2000	Cohort study Validated FFQ 24-hour recalled food intake	Japanese-Americans intakes were significantly lower for cholesterol, fiber, lycopene, calcium and folate compared to other ethnic groups. And were the only group in which the relationship of polinsturados fatty acids / saturated fat was greater than 1.