INDEPENDENT THINKING. WHOLE-FOOD LIVING.NO ANIMALS / NO OIL
THENANODIET
← Metabolism / Health Desk

Whole fruit and juice had different diabetes associations

Three cohorts found variation among individual fruits rather than a single uniform fruit effect.

Whole-food ingredients prepared without added oil

Illustrative food image, not a photograph of the study intervention.

Follow people before diabetes develops

This analysis combined three US health-professional cohorts: 66,105 women in the Nurses’ Health Study, 85,104 women in Nurses’ Health Study II, and 36,173 men in the Health Professionals Follow-up Study. Participants were free of major chronic disease at baseline. Investigators examined incident type 2 diabetes, identified by self-report and confirmed using supplementary questionnaires. More than 3.4 million person-years yielded 12,198 cases. The design addresses associations with future diagnosis, not short-term glucose responses after eating fruit and not treatment of people already living with diabetes.

Muraki I, Imamura F, Manson JE, Hu FB, et al.. BMJ (Clinical research ed.), 2013. PMID 23990623 · DOI

Total whole fruit and juice diverged

After adjustment for personal, lifestyle, and dietary factors, each additional three weekly servings of total whole fruit was associated with a hazard ratio of 0.98 for type 2 diabetes. Its corrected confidence interval was 0.97 to 0.99. The same intake increment for fruit juice was associated with a hazard ratio of 1.08, with an interval from 1.05 to 1.11. These are relative associations for reported consumption, not the results of randomly replacing juice with fruit. The contrast is informative without becoming proof that a specific swap prevents a diagnosis.

Muraki I, Imamura F, Manson JE, Hu FB, et al.. BMJ (Clinical research ed.), 2013. PMID 23990623 · DOI

Individual fruits did not share one estimate

For three additional weekly servings, blueberries had a pooled hazard ratio of 0.74, grapes and raisins 0.88, and apples and pears 0.93. Several other fruit estimates were closer to no association, with confidence intervals spanning one. The abstract reports statistically significant variation among individual fruits. That is more nuanced than either claiming all fruit was equally protective or declaring fruit harmful because it contains sugar. The estimates belong to their specific fruit categories and adjusted cohort models; they do not establish a ranking of guaranteed individual benefits from different fruits.

Muraki I, Imamura F, Manson JE, Hu FB, et al.. BMJ (Clinical research ed.), 2013. PMID 23990623 · DOI

The recipe is not the prevention experiment

The investigators mutually adjusted individual fruit intakes, but participants still selected their own diets. That leaves a clear distinction between an observed association and a causal effect of adding a specific fruit. The linked blueberry breakfast and apple-containing slaw match foods discussed in the cohort, not tested diabetes-prevention interventions. The paper does not establish a recipe-specific glucose response, an optimal personal serving amount, or a treatment plan. Its supported message is that whole-fruit categories and juice showed different associations with future diabetes, with substantial variation among the individual fruits examined.

Muraki I, Imamura F, Manson JE, Hu FB, et al.. BMJ (Clinical research ed.), 2013. PMID 23990623 · DOI