Eating blueberries, grapes, apples and pears cuts the risk of type 2 diabetes but drinking fruit juice can increase it, a large study has found.
Experts from the UK, Singapore and a team from Harvard School of Public Health in the US have examined whether certain fruits impact on type 2, which affects more than 3,000,000 people in Britain.
The scientists found that blueberries, grapes, raisins, apples and pears were especially protective, while drinking fruit juice could increase the risk of developing the condition by as much as 8 percent.
People who ate three standard servings of blueberries a week had a 26 percent lower chance of developing the condition, they found.
Those who replaced fruit juices with three helpings of particular whole fruits a week, including apples and pears could expect a 7 percent drop in their risk of developing type 2.
Eating different fruits affected an individual's chances of developing the condition in different ways, the research suggests.
Those eating grapes and raisins had a 12 percent reduced risk. Prunes also had a protective effect, giving an 11 percent drop in the risk of developing type 2 diabetes.
Other fruits such as bananas, plums, peaches and apricots had a negligible impact but drinking fruit juice increased the risk by 8 per cent, according to the study.
For individual fruits, replacing three servings a week of fruit juice with blueberries cut the risk by 33 percent while replacing juice with grapes and raisins cut the risk by 19 percent.
The risk was also 14 percent lower if juice was replaced with apples and pears, 13 percent lower if replaced with bananas and 12 percent lower if replaced with grapefruit.
Qi Sun, one of the study's authors and an assistant professor at Harvard School of Public Health, said, in general, fruit juices contained less of the beneficial compounds found in whole fruits.
The relatively high glycaemic load of fruit juice along with "reduced levels of beneficial nutrients through juicing processes" may explain why juice increases the risk of type 2 diabetes, the authors suggest.
"Fluids pass through the stomach to the intestine more rapidly than solids even if nutritional content is similar. For example, fruit juices lead to more rapid and larger changes in serum levels of glucose and insulin than whole fruits," they said.
More research is needed, they added, but concluded: "Greater consumption of specific whole fruits, particularly blueberries, grapes and apples, is significantly associated with a lower risk of type 2 diabetes, whereas greater consumption of fruit juice is associated with a higher risk."
The research, published in the British Medical Journal, includes data on 187,382 people taken from three separate studies, of whom 12,198 developed type 2 diabetes.
Food questionnaires were used every four years to assess diet and they asked how often, on average, people consumed each food in a standard portion size.
Around 2.7 million people in the UK are diagnosed with type 2 and a further 850,000 are thought to have it but do not know it.
Another seven million people are estimated to be at high risk of developing the disease which is linked to obesity and inactive lifestyle.
Complications of type 2 include limb amputation, blindness, kidney failure, heart disease and stroke.
Dr Matthew Hobbs, head of research for Diabetes UK, said: "The best way to reduce your risk of developing Type 2 diabetes is to eat a balanced, healthy diet that includes a variety of fruits and vegetables and to be as physically active as possible.
"This research provides further evidence that eating plenty of whole fruit is a key part of the balanced diet that will help you to achieve a healthy weight and so minimise your risk of developing Type 2 diabetes.
"However, the associations between Type 2 diabetes and specific types or fruit or fruit drinks must be treated with much more caution.
"Some of the findings are based on a number of assumptions and models which may have distorted the results significantly.
"For example, the researchers used surveys to ask participants how often they ate certain foods.
"This type of survey can often be unreliable as people are more likely to remember certain types of food.
"In fact, the researchers tried to adjust for this by asking a small subset of participants to complete daily food diaries and comparing the results to the surveys.
"For a number of fruits, including blueberries, the numbers were not big enough to allow the researchers to correct their findings in this way."