Global estimates of the prevalence of diabetes for 2010 and 2030

AIM

We estimated the number of people worldwide with diabetes for the years 2010 and 2030.

METHODS

Studies from 91 countries were used to calculate age- and sex-specific diabetes prevalences, which were applied to national population estimates, to determine national diabetes prevalences for all 216 countries for 2010 and 2030. Studies were identified using Medline, and contact with all national and regional International Diabetes Federation offices. Studies were included if diabetes prevalence was assessed using a population-based methodology, and was based on World Health Organization or American Diabetes Association diagnostic criteria for at least three separate age-groups within the 20-79 year range. Self-report or registry data were used if blood glucose assessment was not available.

RESULTS

The world prevalence of diabetes among adults (aged 20-79 years) will be 6.4%, affecting 285 million adults, in 2010, and will increase to 7.7%, and 439 million adults by 2030. Between 2010 and 2030, there will be a 69% increase in numbers of adults with diabetes in developing countries and a 20% increase in developed countries.

CONCLUSION

These predictions, based on a larger number of studies than previous estimates, indicate a growing burden of diabetes, particularly in developing countries.

Differences in height explain gender differences in the response to the oral glucose tolerance test- the AusDiab study

AIM

To determine the extent of gender-related differences in the prevalence of glucose intolerance for the Australian population and whether body size may explain such differences.

METHODS

Cross-sectional data were collected from a national cohort of 11 247 Australians aged > or = 25 years. Glucose tolerance status was assessed according to both fasting plasma glucose (FPG) and 2-h plasma glucose (2hPG) levels following a 75-g oral glucose tolerance test (OGTT). Anthropometric and glycated haemoglobin measurements were also made.

RESULTS

Undiagnosed diabetes and non-diabetic glucose abnormalities were more prevalent among men than women when based only on the FPG results (diabetes: men 2.2%, women 1.6%, P = 0.02; impaired fasting glycaemia: men 12.3%, women 6.6%, P < 0.001). In contrast 16.0% of women and 13.0% of men had a 2hPG abnormality (either diabetes or impaired glucose tolerance, P = 0.14). Women had a mean FPG 0.3 mmol/l lower than men (P < 0.001), but 2hPG 0.3 mmol/l higher (P = 0.002) and FPG-2hPG increment 0.5 mmol/l greater (P < 0.001). The gender difference in mean 2hPG and FPG-2hPG increment disappeared following adjustment for height. For both genders, those in the shortest height quartile had 2hPG levels 0.5 mmol/l higher than the tallest quartile, but height showed almost no relationship with the FPG.

CONCLUSIONS

Men and women had different glycaemic profiles; women had higher mean 2hPG levels, despite lower fasting levels. It appeared that the higher 2hPG levels for women related to lesser height and may be a consequence of using a fixed glucose load in the OGTT, irrespective of body size.

Electrocardiographic abnormalities amongst Melanesian and Indian men of Fiji: prevalence and associated factors

Electrocardiograms were performed on 427 Melanesian and 391 Indian males aged 30-69 years surveyed in 1980. The age-adjusted prevalence of abnormalities in Minnesota coding suggesting coronary arterial disease was: Indians - 17.1%, Melanesians - 9.1%. This difference was significant at P less than 0.001. For the Melanesians body mass index, plasma uric acid, cholesterol and triglyceride and systolic blood pressure showed significant positive associations with electrocardiographic abnormalities in a univariate analysis, but only the plasma uric acid level was independently associated with such abnormalities following adjustment for these other factors in a multivariate analysis (P less than 0.001). For the Indians body mass index, systolic blood pressure, the presence of diabetes, urban residence and physical inactivity were positively associated with electrocardiographic abnormalities in the univariate analysis, but only body mass index (P less than 0.01), systolic blood pressure (P less than 0.01), and the 2-hour glucose level (P less than 0.01) were independently associated with such abnormalities when adjustment was made for the levels of the other factors.

Weight change amongst Nauruans over 6.5 years: extent, and association with glucose intolerance

We are reporting here the extent of weight change over 6.5 years of follow-up amongst 366 randomly selected Nauruans initially surveyed in 1975-1976. Although the initial level of obesity was high (mean body mass index: males 30.8 kg/m2, females 33.0 kg/m2), further increase in body mass index subsequently occurred in 285 subjects (78%). The weight loss that occurred amongst the other 81 subjects was generally mild (mean = 5.3 kg), and only 12 persons lost more than 10 kg. Weight loss was far more common amongst the older subjects: 40% of those older than 50 years compared with 10% of those less than 30 years showed a decrease in body mass index (P less than 0.001). After age adjustment, greater initial obesity was associated with subsequent weight loss (P less than 0.001), and higher baseline plasma glucose (P less than 0.001), and triglyceride levels (P less than 0.05) and lower plasma uric acid levels (P less than 0.01) were also predictive of some weight loss. Obesity and diabetes remained the major predictors of weight loss in the multivariate analysis. Subjects who lost weight showed similar deterioration in glucose tolerance over the 6.5 years (mean increase = 1.8 mmol/l) to those not losing weight (mean = 1.0 mmol/l) and weight loss was not associated with glucose tolerance improvement (or less deterioration) for either those diabetic, or those not diabetic initially. This result suggests that weight loss is not invariably associated with improvement in glucose tolerance amongst Nauruans.

Plasma insulin response among Nauruans. Prediction of deterioration in glucose tolerance over 6 yr

A longitudinal study of 266 randomly selected nondiabetic Nauruans [215 with normal tolerance and 51 with impaired glucose tolerance (IGT)] over 6 yr showed that deterioration in glucose tolerance status had occurred in 61 subjects. Of the subjects with initially normal tolerance, 34 (16%) progressed to IGT and 14 (6.5%) progressed to diabetes. Thirteen of the subjects with IGT (25%) progressed to diabetes. Subjects were examined in 1975 through 1976, and follow-up examinations were performed in 1982. After age, a high 2-h plasma insulin response to a glucose load was the factor most predictive of progression from normal tolerance to both diabetes (P less than .001) and IGT (P less than .01). Both a high 2-h glucose level and greater obesity independently predicted progression from IGT, and a diminished 2-h insulin response just failed to significantly improve the model (P less than .06). The negative parameter of the insulin response associated with deterioration from IGT differed significantly (P less than .01) from the positive-parameter estimate of the response associated with progression to diabetes from normal tolerance (P less than .01), implying a qualitative difference between these nondiabetic subgroups. The use of a glucose-insulin interaction term to predict progression to diabetes for all nondiabetic subjects confirmed this difference; this term's addition improved the model (P less than .01), and progression to diabetes was associated with a high insulin response for 2-h glucose less than 7.8 mM but a low response for 2-h glucose greater than 7.8 mM.

The association of non-insulin-dependent diabetes with parity and still-birth occurrence amongst five Pacific populations

Previous reports have shown the prevalence of non-insulin-dependent diabetes to be high amongst several populations living in the South and Central Pacific region, and a number of factors including a sedentary life-style, urban residence, obesity and genetic factors have been implicated in its aetiology. Amongst some populations increasing parity has been shown to be associated with abnormal glucose tolerance, but the cross-sectional data available did not suggest any such association amongst the five Pacific Island populations surveyed. Still-birth rates are high amongst all these populations, particularly so amongst the older women, and several methods of analysis suggest that abnormal glucose tolerance is associated with a significantly increased risk of still-birth. This association appears more marked amongst younger women in all populations, with the overall relative risk of a still-birth occurrence for diabetic women aged less than 45 years being 2.6 (95% confidence interval being 1.4-4.8), and for women aged less than 45 years with impaired glucose tolerance being 2.2 (95% confidence interval being 1.3-3.7). As many of the women diabetic at the time of surveys would not have been during their pregnancies, these risk estimates are probably underestimates. Longitudinal studies may suggest even higher figures. Nonetheless the results of these surveys suggest that the high prevalence of abnormal glucose tolerance in these populations may be at least partly responsible for their high levels of still-birth occurrence.

Mortality and health service utilization amongst Melanesian and Indian diabetics in Fiji

A population-based survey of the biracial population of Fiji conducted in 1980 showed non-insulin dependent diabetes (NIDDM) prevalence to be high amongst urban Melanesians and Asian Indians. Follow-up of the cohort of urban residents has been conducted at the major sites of health service delivery in Suva, and surveillance now encompasses 4 years of such attendances. Age-adjusted mortality rates for diabetic subjects were increased compared with normal subjects (relative risks for 4 year mortality being 4.6 for Indians, P less than 0.01, and 1.5 for Melanesians, P greater than 0.1). Inpatient admission rates were also increased amongst diabetic subjects, but only significantly for the females. The relative risk of admission was 3.1 for Melanesian and 2.6 for Indian females (both significant at P less than 0.05). These results suggest that NIDDM in these populations is associated with several adverse health outcomes, and confirms for developing country populations the association of NIDDM with excess mortality noted amongst developed country populations.