Skip to main content

24-01-2010 | Diabetes | Article

Dietary soy products, isoflavones ‘do not modify diabetes risk’


Free abstract

MedWire News: Dietary intake of soy products and isoflavones does not modify the risk for Type 2 diabetes in Japanese adults, a large cohort study has found.

But subgroup analysis identified a possible protective effect of soy and isoflavones in overweight women, which the study authors say deserves further investigation.

Akiko Nanri (International Medical Center of Japan, Tokyo) and colleagues investigated the relationship between dietary intake of soy products and isoflavones (genistein and daidzein) with the risk for incident diabetes.

“Animal studies have found that isoflavones, a major phytoestrogen found in these foods, improve glucose tolerance and exert an antidiabetic effect,” they explain. “However, human data on this issue are limited.”

The team studied 25,872 men and 33,919 women, aged 44–75 years, who participated in the second survey of the Japan Public Health Center-Based Prospective Study. All participants were free of diabetes at enrolment in 1993, but 1114 were newly diagnosed with the condition during the 5-year follow-up.

Writing in the Journal of Nutrition, Nanri et al reveal that intakes of soy products and isoflavones, measured at baseline using a 147-item food-frequency questionnaire, were not associated with the risk for Type 2 diabetes in either men or women.

In an analysis stratified by body mass index (BMI), however, elevated intakes of soy products and isoflavones were associated with a decreased incidence of Type 2 diabetes in overweight women (BMI ≥25.0 kg/m2). For this subgroup, those in the highest quintile of intake had multivariable- and energy-adjusted odds ratios for Type 2 diabetes of 0.89 for soy products, 0.92 for daidzein, and 0.91 for genistein, compared with women in the lowest quintile of intake.

There was no clear dose-response relationship, however, and trends in ORs for lowest through highest quintiles were non-significant for all three food types, raising questions about the robustness of the findings.

Nanri et al propose a number of potential explanations for the putative antidiabetic effect of isoflavones, such as their weak estrogenic properties, inhibitory effects on lipogenesis, favorable impact on insulin resistance, and activation of nuclear peroxisome proliferator-activated receptors.

They conclude: “The present study found no evidence to support the hypothesis that higher intakes of soy product and isoflavones prevent Type 2 diabetes in either men or all women.

“However, we did observe associations suggestive of a protective role of these food factors in overweight women or, to a lesser extent, postmenopausal women. Our findings warrant further investigation.”

MedWire ( is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010

By Joanna Lyford