Out-patient VLCD treatment proved safe and effective in overweight diabetic subjects but those who chose conventional diet and exercise had a slower but more sustained weight loss. Diabetic patients willing to attempt VLCD may safely lose sufficient weight to allow major surgery, but weight regain is inevitable. Patients willing to undertake a long-term group programme of conventional diet can sustain significant weight loss for 5 years, but still require antidiabetic medication.

'Alternative' medicines are becoming increasingly popular, and in this paper we describe our experience with alternative approaches to orthodox diabetes management. Four patients with insulin-dependent diabetes reduced or stopped their insulin in favour of therapeutic approaches including prayer, faith healing, unusual diets, and supplements of vitamins and trace elements. This resulted in ketoacidosis in three, in one case life-threatening; and weight loss and hyperglycaemia in the other. One patient developed serious retinopathy. Additionally, eight other types of alternative diabetic treatment are described, not as far as we know associated with such serious complications. These include homeopathy, reflexology, meditation, herbal treatment, 'cellular nutrition', 'subconscious healing', 'pearl therapy' (drinking milk in which pearls have been boiled) and 'astrotherapy' (typing pieces of coral around the arm). Diabetes is a chronic incurable disease, for which modern treatments remain somewhat unsatisfactory. It is therefore perhaps not surprising that some patients seek alternative treatments with more attractive claims. Diabetes health professionals need to be aware of the potential dangers associated with some of these treatments.

Macrovascular and microvascular complications of diabetes may be associated with different environmental factors. To investigate this further, a prevalence study of 503 Mexican type II diabetic subjects was carried out while their patterns of nutrition were constrained by government food subsidies. Average daily dietary intakes were 1866 kcal; 46.5% as carbohydrate, 13.7 mmol cholesterol, 8.7 g fiber, and a polyunsaturated/saturated fat ratio of 0.98. With respect to macrovascular disease, 49.3% of patients had evidence of peripheral vascular disease, and 21.6% myocardial ischemia, 6.0% angina, 10.8% EKG evidence of ischemia, 4.8% EKG evidence of myocardial infarction. Only 1.2% (six patients) had a clear history of completed stroke, and all were hypertensive. Six patients had also undergone amputations for diabetic gangrene. Tabulation of the means of clinical characteristics according to presence or absence of myocardial ischemia showed that higher cholesterol, calorie, and fat intake, higher mean blood pressure, higher serum cholesterol, and serum triglyceride levels were found in those with myocardial ischemia. Patients with peripheral vascular disease were more commonly smokers. Stepwise logistic regression revealed significant positive associations between myocardial ischemia and dietary cholesterol, serum cholesterol, and mean blood pressure. In contrast, the presence of peripheral vascular disease was significantly related only to smoking and retinopathy. There were no associations between macrovascular complications and duration of diabetes in the multivariate analysis, and they occurred with equal frequency in men and women. Prospective studies of atherosclerosis in maturity-onset diabetes should assess and seek to modify dietary cholesterol, serum cholesterol, and hypertension.

An assessment of short‐term feasibility, safety and efficacy of intensive conventional diet with life‐style advice (ICD) and very low calorie diet (VLCD) has been conducted in obese Type 2 diabetic subjects and obese non‐diabetic patients. The study will continue for three years and comprises 19 non‐diabetic subjects on VLCD, 14 diabetic subjects on VLCD, and 14 diabetic subjects on ICD. At three months both VLCD groups showed significant reductions in weight, BMI, waist measurement, waist/hip ratio, serum cholesterol and triglycerides and serum fructosamine, even though antidiabetic and hypolipidaemic treatments had been discontinued. All the patients had returned to a low fat conventional diet by six months, at which time weight losses were maintained at 11 kg for the diabetic and 21 kg for the non‐diabetic groups. Waist and waist/hip ratio remained significantly improved in the diabetic group, though median systolic and diastolic blood pressures, serum fructosamine, cholesterol and triglycerides were less than at baseline but not significantly so. In the non‐diabetic VLCD group, waist, systolic and diastolic blood pressures, serum cholesterol and triglycerides but not waist/hip ratio were all significantly less at six months compared with baseline. The ICD diabetic group showed significant reductions only in weight (2kg) and BMI (1 unit) at three months, though by six months there was a weight loss of a median of 3kg and a significant fall of 11 mmHg in diastolic blood pressure. One severe hypoglycaemic attack occurred in a diabetic on VLCD and one myocardial infarction in a diabetic in the ICD group. This analysis of the acute phase of the study shows favourable results, and such interventions could readily be incorporated into hospital or primary healthcare diabetes programmes. Long‐term studies of large groups of Type 2 diabetic patients will be required in order to assess effects on outcome.