Association of dietary diversity with total mortality and major causes of mortality in the Japanese population: JPHC study

• Published in: European journal of clinical nutrition Vol. 74; no. 1; pp. 54 - 66
• Main Authors: Kobayashi, Minatsu, Sasazuki, Shizuka, Shimazu, Taichi, Sawada, Norie, Yamaji, Taiki, Iwasaki, Motoki, Mizoue, Tetsuya, Tsugane, Shoichiro
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.01.2020; Nature Publishing Group
• Subjects: 692/308/174; 692/499; Alcoholic beverages; Animals; Beverages; Cardiovascular Diseases; Clinical Nutrition; Cohort Studies; Confidence intervals; Coronary artery disease; Developed countries; Diet; Epidemiology; Female; Fish; Food; Fruits; Guidelines; Health aspects; Heart diseases; Humans; Internal Medicine; Ischemia; Japan; Japan - epidemiology; Male; Meat; Medicine; Medicine & Public Health; Metabolic Diseases; Mortality; Observations; Population studies; Proportional Hazards Models; Prospective Studies; Public Health; Regression analysis; Regression models; Risk Factors; Statistical analysis; Surveys and Questionnaires; Japan
• ISSN: 0954-3007; 1476-5640
• DOI: 10.1038/s41430-019-0416-y

Background/objectives Eating a variety of foods has been recommended in the dietary guidelines of many developed nations, as well as global dietary guidelines. We investigated the potential effect of dietary diversity on the risk of total mortality and major causes of mortality in a large-scale cohort study involving the Japanese population. Subjects/methods From 1995 to 2012, 79,904 participants (37,240 men and 42,664 women) aged 45 to 74 years, without histories of ischemic heart disease, stroke, or cancer were followed-up for a median of 14.9 years. We used 133 food and beverage items listed on a 5-year follow-up food frequency questionnaire, and the daily frequency of consumption for each item (excluding five alcoholic beverages) was counted. Multivariable-adjusted Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) to compare the highest and lowest quintiles. Results An inverse association was identified between total dietary diversity and mortality rates in women (highest quintile HR: 0.81; 95% CI: 0.71, 0.92; p for trend = 0.002) but the same trend was not observed in men (highest quintile HR: 0.96; 95% CI: 0.87, 1.10; p for trend = 0.266). In addition, fruit diversity was associated with lower mortality rates in men (highest quintile HR: 0.87; 95% CI: 0.79, 0.95; p for trend = 0.006), whereas soy diversity was associated with lower mortality rates in women (highest quintile HR: 0.89; 95% CI: 0.79, 1.00; p for trend = 0.004). With regard to men, meat and fish diversity were associated with higher mortality rates (highest quintiles of meat and fish diversity HR: 1.15 [95% CI: 1.06, 1.25] and 1.12 [95% CI: 1.02, 1.22], respectively). Conclusions These findings indicate that consuming a greater diversity of total foods and increasing the diversity of fruit and soy consumed have considerable public health implications.