According to Dr Wolfe, there are three reasons pharmaceutical companies, doctors and patients.

The magic bullet

When these 'oral hypoglycaemic agents' first became available, they were meant to serve as insulin substitutes for those few type 2 diabetics who needed a treatment of both diet and insulin to control their diabetes.

"Instead, the pills became substitutes for the diet," says Dr Wolfe. "With the availability of oral drugs, experts stopped stressing the role of diet in controlling the disease, mostly in those very people whose diabetes could have been controlled by an appropriate diet."

According to Dr Wolfe, as it is assumed that older people won't change their diet or lose weight, doctors won't even suggest an initial trial weight loss period, opting instead to begin treatment with a diabetic pill.

"Patients, handed a complex diet by their doctor and referred to a dietitian for ins tructions on weighing food portions and memorising food choices, often find it easier to take a pill," continues Dr Wolfe. "However, it is foolhardy to increase the already present risk of heart and blood vessel disease for the convenience of popping a pill when proper instruction, limited dietary changes and a little encouragement can help you to reach optimal weight, better health and normal blood sugar."

Weight loss, restricted diets and exercise have all been advocated for the treatment of type 2 diabetes. Exercise as an adjunct to diet leads to greaterweight loss and prevention of weight gain among patients with type 2 diabetes. Although there are some inconsistencies, most studies have demonstrated the effectiveness and feasibility of exercise over the long term in treating adult onset diabetes (Diabetologia, 1987; 30: 930-3; Diabetes Care, 1992; 15 [Suppl]: 1800-10).

In one trial, 577 people with impaired glucose tolerance (a milder form of hyperglycaemia) were randomly allocated into diet, exercise and control groups. During the six year trial period, 67 per cent of the control group, but only 41 per cent and 43 per cent of the diet and exercise groups, respectively, developed type 2 diabetes, a risk reduction of approximately 25 per cent (Diabetes Care, 1997; 20: 537-44).

In several large scale studies with follow up periods of 6-14 years, there was a decrease of 30-50 per cent in the development of type 2 diabetes among those who exercised regularly compared with those who were sedentary (N Engl J Med, 1991; 325: 147-52; Lancet, 1991; 338: 774-8; Am J Epidemiol, 1995; 41: 360-8). This result was found in both obese and non obese men and women.

However, the evidence of longer term trials suggests that diet alone does not always improve glucose control or reduce deaths due to type 2 diabetes (Ann Intern Med, 1996; 124: 136-45).

So, which drug treatments have been shown to lower blood glucose?

Preliminary results from a large trial examining glucose control and risk of diabetes complications in type 2 diabetics show an improvement in hemoglobin A1c (HbA1c) levels in patients who received treatment whether with sulphonylurea, metformin or insulin (Ann Intern Med, 1998; 128: 165-75). Levels of HbA1c, which has links to sugar, are raised in poorly controlled diabetics.

A trial of 2520 patients comparing diet alone with diet plus sulphonylurea (chlorpropamide or glyburide [glibenclamide]), insulin or metformin, found that all of the drugs were equally good at lowering glucose and were better than diet alone (BMJ, 1995; 310: 83-8).