The prevalence of diabetes is increasing globally, and it is estimated to be 9·9 % by 2030( Reference Whiting, Guariguata and Weil 1 ). In Japan, the prevalence of diabetes has been increasing since the 1950s( Reference Kuzuya 2 ), and this increase has generally been ascribed to a Westernised diet that is characterised by high intakes of fat and meat. Dietary cholesterol intake in Japan increased from 57 mg/d in 1947 to approximately 350 mg/d in the 1970s( Reference Tsuji 3 ). To date, meta-analyses of the relevant prospective studies have shown that high consumption of meat is associated with an increased risk of type 2 diabetes( Reference Micha, Wallace and Mozaffarian 4 – Reference Pan, Sun and Bernstein 6 ). However, evidence regarding the association between dietary cholesterol intake and type 2 diabetes risk is scarce.
Most( Reference Djousse, Gaziano and Buring 7 – Reference Song, Manson and Buring 10 ), but not all( Reference Djousse, Kamineni and Nelson 11 ), prospective studies have reported that dietary cholesterol intake is associated with an increased risk of type 2 diabetes in Western populations in which dietary cholesterol is mainly derived from meat( Reference Cotton, Subar and Friday 12 ). However, most of these Western studies did not consider meat intake when examining the association between dietary cholesterol intake and type 2 diabetes risk( Reference Salmeron, Hu and Manson 8 – Reference Song, Manson and Buring 10 ), which may have confounded the results reported. Eggs are rich in cholesterol( 13 ), and the association between egg intake and type 2 diabetes risk has also been examined in Western countries; however, the results are inconsistent. Some studies have found no association( Reference Djousse, Kamineni and Nelson 11 , Reference Vang, Singh and Lee 14 , Reference Montonen, Jarvinen and Heliovaara 15 ), whereas one study has found egg intake to be associated with an increased risk of type 2 diabetes in both men and women( Reference Djousse, Gaziano and Buring 7 ).
To our knowledge, all the previous prospective studies on this topic have been conducted in Western populations. Importantly, Western and Asian populations consume similar sources of dietary cholesterol, but contributions of each source differ( Reference Cotton, Subar and Friday 12 , 16 ). In addition, Asian populations have relatively lower insulin secretory capacity( Reference Torrens, Skurnick and Davidow 17 ) and body masses( Reference Finucane, Stevens and Cowan 18 ) than Western populations. Thus, the effect of cholesterol and egg intakes on type 2 diabetes risk in Asian populations may differ from that in Western populations. In the present study, the association of dietary cholesterol and egg intakes with type 2 diabetes risk was investigated prospectively in a large-scale, population-based cohort of Japanese men and women. Furthermore, stratified analyses that included obesity level and menopausal status were conducted.
Materials and methods
Study design
The Japan Public Health Center-based Prospective (JPHC) Study was established in 1990 for cohort I and in 1993 for cohort II( Reference Tsugane and Sobue 19 ). Subjects in cohort I comprised residents of five Japanese public health centre areas (Iwate, Akita, Nagano, Okinawa and Tokyo) aged 40–59 years. Subjects in cohort II included residents of six public health centre areas (Ibaraki, Niigata, Kochi, Nagasaki, Okinawa and Osaka) aged 40–69 years. A questionnaire survey was conducted at baseline and at the 5-year and 10-year follow-ups. Information on medical histories and health-related lifestyles including smoking, drinking and dietary habits was obtained during each survey. Although written informed consent was not required, the subjects were informed of the objectives of the study, and those who responded to the questionnaire survey were considered to have consented to participating in the survey.
Of the potential baseline subjects (n 140 420), 113 403 responded to the questionnaire survey at baseline. Of these subjects, 89 947 (79·3 %) responded to the questionnaire survey at the 5-year follow-up (i.e. the second survey) and 76 901 (67·8 %) responded to the questionnaire survey at the 10-year follow-up (i.e. the third survey). Subjects (n12 872) who reported histories of type 2 diabetes, cancer, stroke, IHD or chronic liver disease during the baseline or second survey, those who reported kidney disease during the baseline survey and those with missing information on cholesterol and egg intakes were excluded from the study. Furthermore, subjects (n563) who reported extreme total energy intake (outside of the mean (sd 3) for each sex) were excluded. Ultimately, a total of 63 466 subjects (27 248 men and 36 218 women) were included in the study. The present study was approved by the Institutional Review Board of the National Cancer Center of Japan.
FFQ
The subjects completed a self-administered FFQ during the baseline, second and third surveys. Data on 147 food and beverage items and nine frequency categories( Reference Sasaki, Kobayashi and Ishihara 20 ) from the second survey were used as the baseline data for the present analysis because the questionnaire that was used during the second survey more comprehensively provided information on food intake than did the questionnaire that was used during the baseline survey. During the second survey, the subjects were asked details regarding typical consumption of eggs within the past year( Reference Sasaki, Kobayashi and Ishihara 20 ). For most of the food items, nine response options that described consumption frequencies were available; these options ranged from rarely ( < 1 time/month) to ≥ 7 times/d. A standard portion size was specified for each food, and the subjects were asked to indicate their typical portion size from among three options ( ≤ 0·5 times, standard or ≥ 1·5 times). Daily egg intake was calculated by multiplying the daily consumption frequency by the typical portion size and is expressed as g/d.
Dietary intake in terms of energy and selected nutrients, including cholesterol, were estimated based on the Standard Tables of Food Composition in Japan( 13 ). The validity and reproducibility of the FFQ were examined in a subsample of subjects in cohorts I and II of the JPHC Study. Details of the validation study have been given elsewhere( Reference Ishihara, Sobue and Yamamoto 21 – Reference Sasaki, Kobayashi and Tsugane 23 ). When examining the validity of the FFQ, energy-adjusted Spearman's correlation coefficients for the intake values of cholesterol and eggs derived from the FFQ and those derived from 28- or 14-d dietary records were found to be 0·33 and 0·25, respectively, for men in cohort I and 0·35 and 0·42, respectively, for women in cohort I( Reference Sasaki, Kobayashi and Tsugane 23 ) and 0·47 and 0·47, respectively, for men in cohort II and 0·47 and 0·45, respectively, for women in cohort II( Reference Ishihara, Sobue and Yamamoto 21 ). When examining the reproducibility of the FFQ, energy-adjusted Spearman's correlation coefficients for the intake values of cholesterol and eggs derived from the two FFQ that were administered 1 year apart were found to be 0·49 and 0·51, respectively, for men in cohort I and 0·46 and 0·54, respectively, for women in cohort I( Reference Sasaki, Ishihara and Tsugane 22 ) and 0·50 and 0·55, respectively, for men in cohort II and 0·53 and 0·57, respectively, for women in cohort II( Reference Ishihara, Sobue and Yamamoto 21 ).
Ascertainment of type 2 diabetes
Physician-diagnosed type 2 diabetes during the 5-year period after the second survey was ascertained via a self-administered questionnaire during the third survey. During the third survey, the study participants were questioned about their histories of major diseases, including diabetes. Because the second survey data were used as the baseline survey data in the present study, only subjects who were diagnosed after this survey were considered as incident cases during the follow-up. In a validation study in a subsample of JPHC Study participants( Reference Kato, Noda and Inoue 24 ), 94 % of the self-reported diabetes cases were confirmed by medical records.
Other variables
Information on other variables including weight, height, smoking status, alcohol consumption, physical activity level, menopausal status and history of hypertension used in the present analysis was derived from the second survey, whereas that on family history of diabetes was derived from the baseline survey. BMI was calculated as weight in kg divided by height in m2. Average daily ethanol intake from alcoholic beverages was calculated as drinking frequency multiplied by ethanol consumption per drinking day. Metabolic equivalents (MET) per d were estimated from the average amount of time spent per d engaged in three types of physical activities at work and during leisure time, and this measure has been validated previously( Reference Inoue, Yamamoto and Kurahashi 25 ).
Statistical analyses
Analyses were carried out separately for men and women. The subjects were divided into quartiles according to their intake of cholesterol or eggs. The following confounding variables were considered: age (years, continuous); study area (eleven areas); BMI ( < 21, 21–22·9, 23–24·9, 25–26·9, or ≥ 27 kg/m2); smoking status (lifetime non-smoker, former smoker, or current smoker smoking either < 20 or ≥ 20 cigarettes/d); alcohol consumption (non-drinker, occasional drinker, or drinker consuming < 150, 150–299, 300–499, or ≥ 450 g ethanol/week for men and < 150 or ≥ 150 g ethanol/week for women); total physical activity levels (MET-h/d, quartiles); history of hypertension (yes or no); family history of diabetes mellitus (yes or no); coffee consumption (almost never, < 1, 1, or ≥ 2 cups/d); total energy intake (kJ/d, continuous); Ca intake (mg/d, continuous); Mg intake (mg/d, continuous); rice intake (g/d, continuous); fish and shellfish intake (g/d, continuous); meat intake (g/d, continuous); vegetable intake (g/d, continuous); soft drink intake (g/d, continuous). The dietary factors considered in the present study either have been shown to be associated with type 2 diabetes risk in previous studies or have been found to be associated with type 2 diabetes risk in the present cohort. An indicator variable for missing data was created for each covariate. Age-adjusted means were calculated using ANOVA, and age-adjusted proportions were calculated using logistic regression. Trend associations between confounding factors and dietary cholesterol intake were examined using linear regression analysis for continuous variables and logistic regression analysis for categorical variables.
The association of the energy-adjusted (residual method( Reference Willett 26 )) intakes of cholesterol and eggs with diabetes risk was assessed by OR, which were estimated by multiple logistic regression. The 95 % CI of the OR were estimated using the Wald method. The first model was adjusted for age and study area, and the second model was further adjusted for BMI, smoking status, alcohol consumption, family history of diabetes mellitus, history of hypertension and total physical activity levels (multivariate model 1). The final model was further adjusted for total energy intake, coffee consumption, and intakes of Ca, Mg, rice, fish and shellfish, meat, vegetables and soft drinks (multivariate model 2). To avoid problems in effect estimation such as variance inflation, all variables of interest were tested for multicollinearity by calculating the variance inflation factor. The variance inflation factor for multivariate model 2 was not significant for multicollinearity (mean variance inflation factor was 1·6 in men and 1·5 in women). Tests for linear trends were conducted by assigning a median value for each category and treating that term as a single continuous variable in the model. In addition, dietary variables were included one at a time to multivariate model 1 to identify variables influencing the association between dietary cholesterol and egg intakes and type 2 diabetes risk. Data were also analysed according to BMI ( < 25 or ≥ 25 kg/m2) in both men and women and according to menopausal status (premenopausal or postmenopausal) in women. Interaction terms between dietary intake (continuous) and the above-mentioned stratifying variables (dichotomous) were created and included in the model to assess statistical interactions. Statistical significance was defined by two-sided P values below 0·05. All analyses were carried out with SAS software (version 9.2; SAS Institute, Inc.).
Results
During the 5-year period after the second survey, 1165 participants (672 men and 493 women) were newly diagnosed as having type 2 diabetes. At baseline (i.e. at the time of the second survey), it was found that men with higher dietary cholesterol intake were more likely to be older and physically inactive and less likely to consume alcohol (Table 1). Furthermore, it was found that women with higher dietary cholesterol intake were less likely to report history of hypertension and postmenopausal status. Both men and women with higher dietary cholesterol intake had higher intakes of protein, fat, Ca, eggs, meat, and fish and shellfish and lower intakes of carbohydrates and rice when compared with those with lower dietary cholesterol intake. Men with higher dietary cholesterol intake consumed higher amounts of Mg and vegetables.
Table 1 Baseline characteristics of the subjects according to quartiles (Q) of energy-adjusted cholesterol intake (Mean values and standard deviations; mean values with their standard errors; proportions)
MET, metabolic equivalents.
* On the basis of the logistic regression analyses for categorical variables and linear regression analysis for continuous variables, each category of intake was weighted against the median intake for that category.
† Subjects with missing information were excluded (BMI (kg/m2): 496 men and 794 women; smoking status: 579 men and 2161 women; total physical activity levels: 4400 men and 6086 women; coffee consumption: 1241 men and 1602 women).
‡ Age-adjusted means and standard errors.
§ Age-adjusted proportion.
∥ Energy-adjusted by the residual method except energy intake and coffee consumption.
Dietary cholesterol intake was found to be inversely associated with type 2 diabetes risk in women after adjustment for non-dietary covariates (multivariate model 1) (P trend= 0·007; Table 2). This association was attenuated after further adjustment for dietary covariates (multivariate model 2); dietary cholesterol intake was found to be associated with a 23 % lower odds of type 2 diabetes risk in women in the highest quartile of intake compared with those in the lowest quartile, albeit this difference was not statistically significant (OR 0·77, 95 % CI 0·57, 1·04; P trend= 0·08). Dietary variables that changed point estimates of multivariate model 1 by ≥ 3 % were rice (4·8 %), fish and shellfish (11·5 %), and meat ( − 7·7 %) in men and Ca (4·3 %), rice (7·1 %), fish and shellfish ( − 4·3 %), and meat (4·3 %) in women.
Table 2 Association between dietary cholesterol intake and type 2 diabetes risk according to quartiles (Q) of energy-adjusted cholesterol intake (Odds ratios and 95 % confidence intervals; medians, number of cases and subjects)
* Linear trends across quartiles of cholesterol intake were tested using the median consumption value for each quartile as an ordinal variable.
† Adjusted for age and public health centre area.
‡ Additionally adjusted for BMI, smoking status, alcohol consumption, total physical activity levels, history of hypertension and family history of diabetes.
§ Additionally adjusted for Mg intake, Ca intake, coffee consumption, rice intake, fish and shellfish intake, meat intake, vegetable intake, soft drink intake and energy intake.
∥ Subjects with missing information (496 men and 794 women).
In the analysis stratified by menopausal status, dietary cholesterol intake was found to be significantly associated with the odds of type 2 diabetes risk in the postmenopausal group (multivariate model 2); the multivariate-adjusted OR for type 2 diabetes risk for the lowest quartile through the highest quartile of intake were 1·00 (95 % CI reference), 0·87 (95 % CI 0·65, 1·16), 0·96 (95 % CI 0·71, 1·29) and 0·68 (95 % CI 0·47, 0·94) (P trend= 0·04). In contrast, no such trend association was found among premenopausal women (P trend= 0·78; P interaction= 0·83). In the analysis stratified by BMI with adjustment for non-dietary covariates (multivariate model 1), high dietary cholesterol intake was found to be significantly associated with a decreased odds of type 2 diabetes risk in both women with BMI < 25 and those with ≥ 25 kg/m2 (P trend= 0·03 in both groups; P interaction= 0·92). After further adjustment for dietary covariates (multivariate model 2), the significant inverse associations disappeared in both normal-weight (BMI < 25 kg/m2) and overweight (BMI ≥ 25 kg/m2) women (P trend= 0·15 and 0·10, respectively; P interaction= 0·94). No significant association between dietary cholesterol intake and type 2 diabetes risk was found in men.
The association between egg intake and type 2 diabetes risk is summarised in Table 3. In multivariate model 1 with adjustment for non-dietary covariates, type 2 diabetes risk was found to be 21 % lower among women (OR 0·79, 95 % CI 0·61, 1·02; P trend= 0·047) in the highest quartile of egg intake than among those in the lowest quartile. This association was attenuated after additional adjustment for dietary covariates (multivariate model 2); the multivariate-adjusted OR for type 2 diabetes risk for the lowest quartile through the highest quartile of intake were 1·00 (95 % CI reference), 1·01 (95 % CI 0·79, 1·29), 0·94 (95 % CI 0·73, 1·21) and 0·82 (95 % CI 0·63, 1·06) (P trend= 0·098). No significant association between egg intake and type 2 diabetes risk was found in men. Similar results were obtained when crude egg intake instead of the energy-adjusted intake was included in the model. Fully adjusted OR of type 2 diabetes risk for the lowest quartile through the highest quartile of crude egg intake were 1·00 (95 % CI reference), 0·96 (95 % CI 0·78, 1·18), 1·15 (95 % CI 0·88, 1·49) and 1·10 (95 % CI 0·89, 1·36), respectively, in men (P trend= 0·24) and 1·00 (95 % CI reference), 1·54 (95 % CI 1·05, 2·27), 0·88 (95 % CI 0·71, 1·11) and 0·91 (95 % CI 0·70, 1·17), respectively, in women (P trend= 0·49).
Table 3 Association between egg intake and type 2 diabetes risk according to quartiles (Q) of energy-adjusted egg intake (Odds ratios and 95 % confidence intervals; medians, number of cases and subjects)
* Linear trends across quartiles of egg intake were tested using the median consumption value for each quartile as an ordinal variable.
† Corresponded to approximately one egg.
‡ Adjusted for age and public health centre area.
§ Additionally adjusted for BMI, smoking status, alcohol consumption, total physical activity levels, history of hypertension and family history of diabetes.
∥ Additionally adjusted for Mg intake, Ca intake, coffee consumption, rice intake, fish and shellfish intake, meat intake, vegetable intake, soft drink intake and energy intake.
Discussion
In the present large-scale, population-based prospective study of Japanese adults, high dietary cholesterol intake was found to be not associated with an increased risk of type 2 diabetes in either men or women overall. Conversely, it was found to be associated with a reduced risk of diabetes in postmenopausal women. Egg intake was found to be not associated with type 2 diabetes risk in either men or women. This is the first prospective study to be carried out on this topic in Asia.
The null findings recorded for both men and women overall are inconsistent with those reported for US populations. Most( Reference Djousse, Gaziano and Buring 7 – Reference Song, Manson and Buring 10 ), but not all( Reference Djousse, Kamineni and Nelson 11 ), studies have consistently reported positive associations between dietary cholesterol intake and diabetes. This discrepancy may be due differences in the dietary sources of cholesterol and not due to differences in cholesterol intake levels. In the present study population, the median cholesterol intake of those in the highest quartile of intake (423·0 mg/d in men and 389·6 mg/d in women) was similar to that reported in the Women's Health Study carried out in the USA (382 mg/d)( Reference Song, Manson and Buring 10 ). Dietary cholesterol is mainly derived from eggs (47 %), shellfish (22 %) and meat (19 %) in Japan( 16 ), whereas it is mainly derived from meat (35 %), eggs (29 %), and cheese and milk (11 %) in the USA( Reference Cotton, Subar and Friday 12 ). Meat intake makes a notably higher contribution to cholesterol intake in the USA than in Japan, and three meta-analyses have consistently shown that high consumption of meat is associated with an increased risk of diabetes( Reference Micha, Wallace and Mozaffarian 4 – Reference Pan, Sun and Bernstein 6 ). Meat intake was not considered in three of the US studies( Reference Salmeron, Hu and Manson 8 – Reference Song, Manson and Buring 10 ), and other relevant studies that did adjust for meat intake reported inconsistent results( Reference Djousse, Gaziano and Buring 7 , Reference Djousse, Kamineni and Nelson 11 ). Thus, it is unclear whether dietary cholesterol per se is independently associated with the risk of type 2 diabetes.
An inverse association between dietary cholesterol intake and type 2 diabetes risk was found in postmenopausal women in the present study. This finding was unexpected and is not in agreement with those of the three previously mentioned US studies carried out in women aged ≥ 45 years, which reported that an increased risk of type 2 diabetes is associated with high dietary cholesterol intake( Reference Djousse, Gaziano and Buring 7 , Reference Meyer, Kushi and Jacobs 9 , Reference Song, Manson and Buring 10 ). However, one US study carried out in elderly women has reported a decreased risk of type 2 diabetes in women in the highest quartile of cholesterol intake (albeit this association was not statistically significant)( Reference Djousse, Kamineni and Nelson 11 ). There are some possible explanations for our finding of a protective association. In an animal study, ovariectomised mice on high-cholesterol diets had higher serum oestrogen concentrations than those on a control diet( Reference Li, Xiao and Yang 27 ). High intake of fat/cholesterol has been reported to be associated with higher serum oestrogen concentrations in Japanese postmenopausal women( Reference Nagata, Nagao and Shibuya 28 ). Furthermore, postmenopausal therapy with oestrogen and progestin has been found to reduce the incidence of diabetes in a large-scale intervention study( Reference Bonds, Lasser and Qi 29 , Reference Margolis, Bonds and Rodabough 30 ). These data suggest that dietary cholesterol in postmenopausal women could be beneficial in the prevention of diabetes via the induction of sex hormones.
No significant association between egg intake and type 2 diabetes risk was found in either men or women in the present study. This is in line with those of most( Reference Djousse, Kamineni and Nelson 11 , Reference Vang, Singh and Lee 14 , Reference Montonen, Jarvinen and Heliovaara 15 ), but not all( Reference Djousse, Gaziano and Buring 7 ), studies on this topic in both men and women in Western countries. Thus, the majority of these studies did not detect an association between an increased risk of type 2 diabetes and egg intake( Reference Djousse, Kamineni and Nelson 11 , Reference Vang, Singh and Lee 14 , Reference Montonen, Jarvinen and Heliovaara 15 ). Eggs not only contain SFA, which have been reported to be associated with an increased risk of type 2 diabetes( Reference Salmeron, Hu and Manson 8 – Reference Song, Manson and Buring 10 ), but also contain MUFA, PUFA, antioxidants, including lutein and zeaxanthin, and folate. MUFA, PUFA and antioxidants can improve insulin sensitivity( Reference Lichtenstein and Schwab 31 , Reference Ceriello and Motz 32 ). Taken together, egg intake may not be detrimental to glucose metabolism.
The major strengths of the present study include a large number of subjects comprising men and women, a population-based prospective design, the use of a validated FFQ, and the adjustments for potentially important confounding variables. Despite these strengths, the present study has several limitations that warrant mention. First, the incidence of diabetes was ascertained based on self-reported information. However, it was confirmed that 94 % of the cases of self-reported diabetes were correctly documented in medical records of the sample of a validation study( Reference Kato, Noda and Inoue 24 ). Furthermore, any under-reporting is unlikely to have varied with cholesterol or egg intake. Second, only data collected during the second survey were used to assess dietary intake. This could have led to non-differential misclassification of cholesterol and egg intakes and thus attenuated the risk estimates towards null. Third, the validity of the FFQ used in the present study was moderate for cholesterol and egg intakes (r 0·33–0·47 and r 0·25–0·47, respectively( Reference Ishihara, Sobue and Yamamoto 21 , Reference Sasaki, Kobayashi and Tsugane 23 )), and resulting measurement errors might have attenuated the associations. Fourth, circulating concentrations of oestrogen, which might play a role in the association between dietary cholesterol intake and type 2 diabetes risk, were not measured. Fifth, the follow-up period was relatively short (5 years). Sixth, participants with pre-diabetes at baseline might have been instructed to decrease their egg or cholesterol intake. If so, risk estimates would be attenuated towards null. Lastly, the possibility of bias due to residual confounding or unmeasured confounders cannot be entirely excluded. Specifically, the observed inverse association between dietary cholesterol intake and type 2 diabetes risk in postmenopausal women could be ascribed to the differences in unmeasured lifestyles between women in the lowest quartile of intake and those in the highest quartile of intake.
In conclusion, higher intake of cholesterol or eggs may not be associated with an increased risk of type 2 diabetes in Japanese populations. The observed association between decreased type 2 diabetes risk and higher dietary cholesterol intake in postmenopausal women warrants further investigation.
Acknowledgements
The present study was supported by Grants-in-Aid for Cancer Research (19shi-2) and a Health Sciences Research Grant (Research on Comprehensive Research on Cardiovascular Diseases H19-016) from the Ministry of Health, Labour and Welfare of Japan.
The authors’ contributions are as follows: S. T. was involved in the design of study as the principal investigator; S. T. and N. S. conducted the surveys; K. K., A. N., A. G., T. M., M. N. and S. O. drafted the plan for data analyses; K. K. conducted the data analyses; T. M. provided statistical expertise; K. K. drafted the manuscript; K. K. and T. M. had primary responsibility for the final content. All authors were involved in the interpretation of the results and revision of the manuscript and approved the final version of the manuscript.
Members of the JPHC Study (principal investigator: S. T.) Group include the following: S. T., M. Inoue, T. Sobue and T. Hanaoka, National Cancer Center, Tokyo, Japan; J. Ogata, S. Baba, T. Mannami, A. Okayama and Y. Kokubo, National Cardiovascular Center, Osaka, Japan; K. Miyakawa, F. Saito, A. Koizumi, Y. Sano, I. Hashimoto, T. Ikuta and Y. Tanaba, Iwate Prefectural Ninohe Public Health Center, Iwate, Japan; Y. Miyajima, N. Suzuki, S. Nagasawa, Y. Furusugi and N. Nagai, Akita Prefectural Yokote Public Health Center, Akita, Japan; H. Sanada, Y. Hatayama, F. Kobayashi, H. Uchino, Y. Shirai, T. Kondo, R. Sasaki, Y. Watanabe, Y. Miyagawa and Y. Kobayashi, Nagano Prefectural Saku Public Health Center, Nagano, Japan; Y. Kishimoto, E. Takara, T. Fukuyama, M. Kinjo, M. Irei and H. Sakiyama, Okinawa Prefectural Chubu Public Health Center, Okinawa, Japan; K. Imoto, H. Yazawa, T. Seo, A. Seiko, F. Ito, F. Shoji and R. Saito, Katsushika Public Health Center, Tokyo, Japan; A. Murata, K. Minato, K. Motegi and T. Fujieda, Ibaraki Prefectural Mito Public Health Center, Ibaraki, Japan; K. Matsui, T. Abe, M. Katagiri and M. Suzuki, Niigata Prefectural Kashiwazaki and Nagaoka Public Health Center, Niigata, Japan; M. Doi, A. Terao, Y. Ishikawa and T. Tagami, Kochi Prefectural Chuo-higashi Public Health Center, Kochi, Japan; H. Sueta, H. Doi, M. Urata, N. Okamoto and F. Ide, Nagasaki Prefectural Kamigoto Public Health Center, Nagasaki, Japan; H. Sakiyama, N. Onga, H. Takaesu and M. Uehara, Okinawa Prefectural Miyako Public Health Center, Okinawa, Japan; F. Horii, I. Asano, H. Yamaguchi, K. Aoki, S. Maruyama, M. Ichii and M. Takano, Osaka Prefectural Suita Public Health Center, Osaka, Japan; Y. Tsubono, Tohoku University, Miyagi, Japan; K. Suzuki, Research Institute for Brain and Blood Vessels, Akita, Japan; Y. Honda, K. Yamagishi, S. Sakurai and N. Tsuchiya, Tsukuba University, Ibaraki, Japan; M. Kabuto, National Institute for Environmental Studies, Ibaraki, Japan; M. Yamaguchi, Y. Matsumura, S. Sasaki and S. Watanabe, National Institute of Health and Nutrition, Tokyo, Japan; M. Akabane, Tokyo University of Agriculture, Tokyo, Japan; T. Kadowaki, Tokyo University, Tokyo, Japan; M. N. and T. M., National Center for Global Health and Medicine, Tokyo, Japan; Y. Kawaguchi, Tokyo Medical and Dental University, Tokyo, Japan; Y. Takashima and M. Yoshida, Kyorin University, Tokyo, Japan; K. Nakamura, Niigata University, Niigata, Japan; S. Matsushima and S. Natsukawa, Saku General Hospital, Nagano, Japan; H. Shimizu, Sakihae Institute, Gifu, Japan; H. Sugimura, Hamamatsu University, Shizuoka, Japan; S. Tominaga, Aichi Cancer Center Research Institute, Aichi, Japan; H. Iso, Osaka University, Osaka, Japan; M. Iida, W. Ajiki and A. Ioka, Osaka Medical Center for Cancer and Cardiovascular Disease, Osaka, Japan; S. Sato, Chiba Prefectural Institute of Public Health, Chiba, Japan; E. Maruyama, Kobe University, Hyogo, Japan; M. Konishi, K. Okada and I. Saito, Ehime University, Ehime, Japan; N. Yasuda, Kochi University, Kochi, Japan; and S. Kono, Kyushu University, Fukuoka, Japan.
None of the authors has any conflicts of interest to declare.
