Objective: The progression from impaired glucose tolerance (IGT)/impaired fasting glucose (IFG) to type 2 diabetes can be prevented or delayed through intensive lifestyle changes. How to translate this to implementation across whole communities remains unclear. We now describe the results to a pilot of a personal trainer (Maori Community Health Worker, MCHW) approach among Maori in New Zealand. Design, setting and subjects: A randomised cluster-controlled trial of intensive lifestyle change was commenced among 5240 non-pregnant Maori family members without diabetes from 106 rural and 106 urban geographical clusters. Baseline assessments included lifestyle questionnaires, anthropometric measurements and venesection. A pilot study (Vanguard Study) cohort of 160 participants were weighed before and during MCHW intervention, and compared with fiftytwo participants weighed immediately before intervention and with 1143 participants from the same geographical area. Interactions between participants and the MCHW were reported using personal digital assistants with a programmed detailed structured approach to each interview. Results: During the Vanguard Study, participants and MCHW found the messages, toolkit and delivery approach acceptable. Those with IGT/IFG diagnosed (n 27) experienced significant weight loss after screening and during the Vanguard Study (5?2 (SD 6?6) kg, paired t test P , 0?01). Significant weight loss occurred during the Vanguard Study among all participants (21?3 (SD 3?6) kg, P , 0?001). Conclusions: Comparable initial weight loss was shown among those with IGT/ IFG and those from existing trials. Community-wide prevention programmes are feasible among Maori and are likely to result in significant reductions in the incidence of diabetes.

The prevalence of retinopathy at diagnosis was lower than in previous studies, yet that of microalbuminuria/proteinuria remained high. The retinopathy data suggest that case detection for diabetes in the community may be improving, but that other strategies among those at risk of diabetes, including those promoting smoking cessation, will be needed to reduce the risk of renal disease among Maori with diabetes.

POCT significantly underestimated the true blood glucose at diagnostic levels for diabetes. POCT cannot be recommended as a means of screening for or diagnosing diabetes or pre-diabetes.

Objective: Uptake of advice for lifestyle change for obesity and diabetes prevention requires access to affordable 'healthy' foods (high in fibre/low in sugar and fat). The present study aimed to examine the availability and accessibility of 'healthy' foods in rural and urban New Zealand. Design: We identified and visited ('mapped') 1230 food outlets (473 urban, 757 rural) across the Waikato/Lakes areas (162 census areas within twelve regions) in New Zealand, where the Te Wai O Rona: Diabetes Prevention Strategy was underway. At each site, we assessed the availability of 'healthy' foods (e.g. wholemeal bread) and compared their cost with those of comparable 'regular' foods (e.g. white bread). Results: Healthy foods were generally more available in urban than rural areas. In both urban and rural areas, 'healthy' foods were more expensive than 'regular' foods after adjusting for the population and income level of each area. For instance, there was an increasing price difference across bread, meat, poultry, with the highest difference for sugar substitutes. The weekly family cost of a 'healthy' food basket (without sugar) was 29?1 % more expensive than the 'regular' basket ($NZ 176?72 v. $NZ 136?84). The difference between the 'healthy' and 'regular' basket was greater in urban ($NZ 49?18) than rural areas ($NZ 36?27) in adjusted analysis. Conclusions: 'Healthy' foods were more expensive than 'regular' choices in both urban and rural areas. Although urban areas had higher availability of 'healthy' foods, the cost of changing to a healthy diet in urban areas was also greater. Improvement in the food environment is needed to support people in adopting healthy food choices. Keywords Food environment Availability AccessibilityIt is estimated that nutrition-related risk factors accounted for 70 % of stroke and heart disease mortality and more than 80 % of diabetes in 1997 in New Zealand (1) . However, nutrition recommendations are often not met. For instance, in the 1990s, fat intake contributed about 35 % of energy in the diet of New Zealanders, with saturated fat comprising 15 % of total energy intake; while daily intake of dietary fibre was lower than the recommended 25-30 g/d, at 23 g/d in men and 18 g/d in women (2) . Although both intervention trials and population studies have demonstrated that type 2 diabetes is preventable by adopting a healthy lifestyle (3) , lifestyle change requires a 'healthy' food environment since food choices are driven by taste, cost and convenience, rather than health and variety (4)

We sought to identify the sex-specific cut-off in waist circumference which best identifies those with metabolic abnormalities consistent with the metabolic syndrome (MS) among Maori, the indigenous people of New Zealand of Polynesian origin. In 3816 self-identified Maori (2742 women, 1344 men) a 75 g oral glucose tolerance test, fasting lipid, anthropometric and blood pressure measurements were made. MS components were defined by Adult Treatment Panel (ATP) III criteria. Waist cut-off was defined using receiver operating characteristic (ROC) curve analysis to define the presence of at least two of the other MS components ($ 2MS). Prevalence of $2MS was high (42·1 %). In males and females, waist was as good, or better, a predictor of $2MS (area under ROC 0·73 women, 0·68 men) as waist:hip ratio (0·66, 0·67), BMI (0·72, 0·68) or percentage body fat (0·70, 0·68). The prediction of dysglycaemia using anthropometric variables followed a similar pattern to $2MS. Waist circumference to predict $2MS or dysglycaemia in Maori women and men was 98 cm and 103 cm. Applying this cut-off to the International Diabetes Federation (IDF) criteria would identify 27·8 % (34·0 % males, 25·5 % females) with the MS with an OR for $2MS (adjusted for sex, smoking and age) of 3·5 (95 % CI 3·1, 4·0). Age . 48 years, smoking and being male increased the odds of the MS. These waist cut-offs should be considered in both clinical practice and to optimise the definition of the MS for Maori. The validity of these criteria in other Polynesian groups should be confirmed.