An estimated 41 million Americans have 'pre-diabetes', or impaired glucose tolerance, in which blood sugar levels are higher than normal and lead to high risk of developing type 2 diabetes.
Around 25 per cent of UK adults are also thought to have this condition and the number of people with diabetes, now 3 per cent of the population, will continue to rise as the population ages and becomes more overweight.
While it would cost a lot to give all of those at risk intensive help with diet and exercise, doing nothing to try to prevent the condition developing would cost significantly more, write the authors in the March issue of the Annals of Internal Medicine (vol 142, no 5, pp 323-332).
The researchers at the University of Michigan Health System used sophisticated computer modelling of data from a large national clinical trial completed in 2001 called the Diabetes Prevention Project, to show the benefits from intervention.
One group of participants followed a lifestyle intervention including brisk walking for 30 minutes five days a week, lowered fat and calorie intake, and a weight-reduction goal of 7 per cent of body weight.
Another group took the medication metformin, and a placebo group received information on exercise and diet. More than 3,200 Americans participated.
In just three years, the risk of developing type 2 diabetes was reduced by 58 per cent among those in the lifestyle change group, and 31 per cent in the metformin group.
The computer model showed that a lifestyle-change programme could delay the onset of diabetes by an average of 11 years and reduce the risk of developing diabetes by 20 per cent, when compared with no intervention.
Twice-daily doses of 850 milligrams of metformin would delay the onset by 3 years, on average, and lead to an 8 per cent reduction in the overall risk of diabetes.
"By projecting the trial's findings into the future, and factoring in all costs including the future cost of diabetes complications, we were able to show cost-effectiveness on a societal basis, and in some age groups, cost savings compared with no action," said lead author William Herman, director of the Michigan Diabetes Research and Training Center at UMHS.
The study supports marketing of foods that can help lower blood sugar levels and improve the pre-diabetes condition.
The authors note that while the results of the DPP trial were published three years ago, it has not had the major impact on clinical practice that the researchers would have hoped for because of cost concerns. They hope the new study will change this.
"The bottom line is, we shouldn't be asking if we can afford to reach out to every at-risk person and help them reduce their risk," said Herman. "The real question is, in the face of today's epidemic of obesity, can we afford not to?"
Neither intervention would prevent every case of diabetes. But both approaches would spare many individuals, and society, the costs of long-term blood-sugar monitoring and medications, and the cost of treating the expensive complications of diabetes that may occur later in life, including blindness, kidney failure, disabling nerve damage and heart disease.
The researchers conclude that American health policy should immediately begin promoting diabetes prevention in high-risk people.
This view is also supported in an accompanying editorial by Jaakko Tuomilehto from the University of Helsinki, Finland, who has led a large Finnish diabetes study.