Q&A

Is the sugar in fruit bad for you? What the diabetes data show

Whole fruit and fruit juice behave oppositely for type 2 diabetes risk, and a trial that told newly diagnosed patients to eat less fruit found no benefit at all.

Type 2 diabetes hazard ratio per three servings/week, whole fruits vs juiceBlueberries: 0.74; Apples and pears: 0.93; Total whole fruit: 0.98; Fruit juice: 1.08012Blueberries0.74Apples and pears0.93Total whole fruit0.98Fruit juice1.08
Type 2 diabetes hazard ratio per three servings/week, whole fruits vs juice
GroupValue (value)
Blueberries0.74 (0.66 to 0.83)
Apples and pears0.93 (0.9 to 0.96)
Total whole fruit0.98 (0.97 to 0.99)
Fruit juice1.08 (1.05 to 1.11)
Type 2 diabetes hazard ratio per three servings/week, whole fruits vs juice Pooled hazard ratios from three US cohorts; a value of 1.0 would mean no difference. Whiskers show 95% confidence intervals. Source: BMJ

"Fruit is just sugar" is one of the more durable nutrition myths, and the evidence does not support it. The sugar in whole fruit arrives wrapped in fibre, water and polyphenols, and in the largest studies whole fruit is associated with slightly lower — not higher — risk of type 2 diabetes, while fruit juice runs the other way [s1]. And when researchers actually told people with diabetes to eat less fruit, it made no difference to their blood sugar [s2]. The form of the fruit matters far more than the fact that it contains sugar.

Whole fruit and juice pull in opposite directions

The most-cited evidence comes from a 2013 BMJ analysis pooling three large US cohorts: 66,105 women from the Nurses' Health Study, 85,104 from the Nurses' Health Study II, and 36,173 men from the Health Professionals Follow-up Study, all free of major chronic disease at baseline [s1]. Over 3,464,641 person-years, 12,198 people developed type 2 diabetes [s1].

After adjusting for lifestyle and dietary factors, every three servings a week of total whole fruit carried a pooled hazard ratio of 0.98 for type 2 diabetes (95% confidence interval 0.97 to 0.99) — a small reduction in risk [s1]. Individual fruits varied sharply. Blueberries showed the strongest association, a hazard ratio of 0.74 per three weekly servings (95% CI 0.66 to 0.83), followed by grapes and raisins at 0.88 and apples and pears at 0.93 [s1]. Fruit juice went the wrong way: the same increment in juice consumption carried a hazard ratio of 1.08 (95% CI 1.05 to 1.11) [s1]. Swapping juice for whole fruit, the authors estimated, was associated with lower risk — the same divergence we cover in fruit juice versus whole fruit.

Why the split, when both contain fructose? Juicing strips the intact fibre and delivers the sugar in a fast, liquid form, closer to a sugary drink than to the fruit it came from. Whole fruit's fibre and structure slow that release.

The test that mattered: telling patients to eat less fruit

Cohort studies can only show association. The sharper question — does cutting fruit help people who already have diabetes? — was put to a randomised trial. A 2013 study in Nutrition Journal enrolled 63 men and women with newly diagnosed type 2 diabetes and gave everyone standard medical nutrition therapy, then randomised them to advice to eat at least two pieces of fruit a day or no more than two [s2].

The groups did change their intake: the high-fruit group increased fruit by 125 grams a day (95% CI 78 to 172) and the low-fruit group cut it by 51 grams (95% CI -18 to -83) [s2]. Yet after 12 weeks, HbA1c — the standard marker of long-term blood sugar — fell in both groups with no difference between them (difference 0.19 percent, 95% CI -0.23 to 0.62) [s2]. Both groups also lost weight and trimmed their waistlines, again with no advantage to the low-fruit arm [s2]. The authors' plain conclusion: "the intake of fruit should not be restricted in patients with type 2 diabetes" [s2].

What this does and does not settle

Keep two limits in view. The cohort findings are observational, so residual confounding cannot be ruled out and people who eat more whole fruit tend to have healthier habits overall [s1]. The randomised trial was small and short — 63 people over 12 weeks — so it speaks to blood-sugar control, not to long-term complications [s2]. Neither study licenses unlimited fruit juice or dried fruit eaten by the handful, which concentrate sugar and, in juice's case, are tied to higher risk [s1].

What the evidence does support is narrower and more useful than the myth. Whole fruit is not a diabetes risk to be rationed; if anything it is faintly protective, and telling patients to cut it back achieves nothing but less fruit [s1][s2]. The sugar in an apple is not the problem — the sugar in what the apple gets turned into can be.

So, is fruit sugar bad for you?

No, not as whole fruit. The best cohort data associate whole fruit with slightly lower type 2 diabetes risk, the best trial found no benefit from restricting it, and the clear loser in both is juice rather than fruit [s1][s2]. Eat the fruit; be warier of the glass.

This article is informational and is not medical advice.

Sources

Sources

  1. Fruit consumption and risk of type 2 diabetes: results from three prospective longitudinal cohort studies — BMJ , August 29, 2013
  2. Effect of fruit restriction on glycemic control in patients with type 2 diabetes—a randomized trial — Nutrition Journal , March 5, 2013

More on

Related coverage
WHAT THE STUDY ACTUALLY SAYS

Glycaemic index: does choosing low-GI foods actually help?

In a tightly controlled feeding trial, swapping high-GI for low-GI carbohydrate inside a healthy diet did not improve insulin sensitivity, cholesterol or blood pressure. The wider evidence rates the GI case as weak.