Fig. 2. Change in prevalence of diabetes 1981-2000 in Australia (36).

Some of the earliest signs of the modern diabetes epidemic were found in the Pacific islands. The tiny island of Nauru became one of the richest countries in the world (on a per capita basis), as its phosphate deposits were mined. The island underwent major environmental changes after destruction of the reef to allow ships to approach the island, resulting in loss of fishing enterprises, and destruction of agricultural land for mining. Consequently, life became sedentary, and food sources shifted to packaged and canned food supplied from overseas. The traditionally active, healthy, and lean population became markedly obese and was found to have a prevalence of diabetes of 11% in those aged 25-34 yr old, rising to 56% in those aged 55-64 yr old (40). Other island populations in the Pacific have also experienced major social and environmental changes, with concurrent increases in the prevalence of diabetes.

North America

Data from the National Health and Nutrition Examination Survey (NHANES) series of surveys have provided a clear picture of the epidemiology of diabetes within the US. The rise in diabetes prevalence over the last 30 yr has occurred in parallel with a rise in the prevalence of obesity. The most recent estimates put the prevalence of diabetes at 9.3% for adults aged 20 yr and older (41). Of the largest population groups in the US, the highest prevalence is seen in African Americans (11.0%) followed by those of Mexican origin (10.4%), with Europids (non-Hispanic whites) being at the lowest risk (5.2%) (41). Figure 3 shows the prevalence of diabetes according to ethnicity within the US and demonstrates that as the proportion of people of Hispanic ethnicity increases, so the national prevalence of diabetes will rise.

However, Native American populations have by far the highest diabetes risk in the US, with the highest diabetes prevalence in the world being recorded in the Pima Indian population of Arizona. More than 20 yr ago, the prevalence in this population was found to be 50% in middle-aged adults, with an incidence that was 19 times greater than the predominantly white population of Rochester, Minnesota (42). Similar, though not quite so spectacularly high diabetes prevalences have been recorded in other Native American groups, in the United States, as well as in Canada.

The diabetes prevalence in the US is one of the highest in the developed world, reflecting not only the high prevalence of obesity, but also the significant proportion of the population belonging to high-risk ethnic groups. If migration and differential birth rates lead to a further increase in the size of these groups, particularly in the Hispanic population, further rises in diabetes prevalence can be expected.

Recent data from very large, national studies in Mexico revealed a prevalence of 8.2% (43) and a similar prevalence of about 9.0% (44). Although these figures are marginally lower than that reported for the US, the Mexican population age structure is much younger than that in the US; hence, the age-specific prevalences are higher in Mexico. For example, among those aged 40-59 yr, the prevalence of diabetes was 7.9% for non-Hispanic whites in the US (41), but approx 15.3% for Mexican Americans, which was similar to the prevalence for this age group in the 2 Mexican studies (43,44).

Fig. 3. Prevalence of diabetes according to ethnicity in the USA (41).

South and Central America, and Caribbean

Data from this region are relatively limited, but a study from Jamaica found diabetes in 13.4% of adults (45), with those in the top quartile of BMI having a 3.3-5.4-fold higher risk of having diabetes than those in the lowest quartile. In Brazil, a study from 9 large cities found that the prevalence of diabetes was 7.6% with no differences observed between the prevalence in whites and in nonwhites (46). Very similar findings were reported from Argentina, where the prevalence in a population drawn from 4 cities was 6.5-7.7% (47), and from Colombia, with a prevalence of 7% (48).

Changes in the Prevalence of Diabetes Over Time

Estimates of the global burden of diabetes have frequently concluded that the prevalence of diabetes is rising, but these conclusions are not always based on studies that can be directly compared. However, a number of pairs or series of studies do allow a more accurate assessment of changes over time. Between 1976 and 1988, the prevalence of diabetes among people age 40-74 yr rose from 11.4% to 14.3% in the US (49). More recently the 1988-1994 (NHANES III) and NHANES 1999-2000 were compared, indicating a statistically nonsignificant increase from 8.2% to 9.3% for diabetes in the 20 year and older population (41), suggesting that the rate of rise of diabetes prevalence may be slowing. The changing prevalence over time in the US illustrates not only the effects of increasing obesity and aging over time, but also the impact of a changing ethnic mix. In Australia, an estimated 7.4% of adults in the year 2000 had diabetes, compared to an estimated 3.4% in 1981 (36). A report from Denmark directly compared the prevalence of diabetes and impaired glucose tolerance (IGT) (50) between 2 cohorts of 60 year olds: the former in 1974/5, and the latter in 1996/1997. Overall, the rates of abnormal glucose tolerance (either diabetes or IGT) had increased by 55% (p < 0.001). Data from 2 population-based surveys in the south Indian city of Chennai revealed a diabetes prevalence rising from 8.2% in 1988/1989 to 11.6% in 1994/1995 (51), with a further survey in the same city carried out in 2003/2004 reporting a diabetes prevalence of 14.3% (31). A series of 3 surveys conducted in the Indian Ocean island of Mauritius has shown the prevalence of diabetes to have risen from 12.8% in 1987 to 15.2% in 1992, and 17.9% in 1998 (52).

In China, national surveys assessing the prevalence of diabetes were conducted in 1994/1995 and 2000/2001 (33,34). The 1994/5 survey involved more than 200,000 participants, and based the prevalence on the 2-h plasma glucose value following the oral glucose tolerance test, or on previously diagnosed diabetes. The national prevalence for the 25-64 year old population was 2.5%. The 2001 survey used the fasting criterion recommended by the American Diabetes Association (ADA) (53) (fasting plasma glucose > 7.0 mmol/L), and was conducted on an older subgroup (35 - 74 yr), but even among those in the 35-64 range, the overall prevalence was about 50% higher than detected previously. In both surveys only about one third of persons with diabetes had been previously diagnosed; the others having diabetes detected at the examination. There was little gender difference, but urban prevalence was higher than rural, when analyzed for the 2001 survey.

Although there is little doubt that there is a major genetic component to the etiology and development of type 2 diabetes, the rapid rise in the prevalence of type 2 diabetes witnessed in recent decades indicates the importance of environmental influences. The time period is far too short to have seen any significant shift in the gene pool, but huge changes in lifestyle, with increasing mechanization of manual tasks and of transport, and a rise in caloric intake has led to increasing obesity, and an epidemic of diabetes. The intertwining of the effects of genes and the environment is illustrated by the higher prevalences of diabetes in people of Indian compared to Europid (white European) origin that is so frequently observed, within urban settings. For example, Indians from the city of Chennai have a diabetes prevalence of 11-14% (31,51), whereas many European countries have a prevalence of under 8% (1,3). Several direct comparisons, within the same country or region have also shown that Hispanics and ethnic groups originating from India have a higher diabetes prevalence than do Europids (21,49).

Rising Prevalence: Owing to Increasing Incidence or Better Survival?

The above data show an increase in diabetes prevalence (i.e., the percentage of a population that has diabetes at a given point in time), occurring in almost all countries, which is usually assumed to be primarily owing to increasing incidence (i.e., an increase in the number of new cases of diabetes developing each year), but could also be a consequence of reduced mortality. Thus, it is possible that with no change in the rate of new cases developing, the total number of individuals with diabetes within a population could rise if diabetes mortality were to fall (as a result of improved treatment). The burden of diabetes within a population depends on the prevalence and is undoubtedly climbing, but understanding the reason for the rising prevalence is important for understanding how to reverse the rise in prevalence.

Rising incidence generally results from a worsening of the risk factor profile of a population (in the case of diabetes, this would include increasing obesity, age and sedentary habits), whereas reduced mortality reflects either the disease becoming less harmful, or an improvement in the care provided for those with the disease. There have been a number of opinions as to the main reasons for the increasing prevalence of diabetes (54-57), particularly as to whether it is based on an increase in incidence. Green et al analyzed Danish data examining the numbers of people commencing pharmacological treatment for diabetes, and the mortality of those individuals (55). The undoubted rise of approx 50% over 10 yr in the numbers of people with drug treated diabetes was explained by an almost constant incidence of drug treated diabetes over the 10 yr, which exceeded a slowly falling mortality rate.

Based on modeling with different age and prevalence patterns for westernized and developing region populations, Colagiuri et al (54) concluded that demographic changes were insufficient to explain the documented rises in prevalence, and that there was good evidence of rising incidence. Unfortunately, although there are many cross-sectional studies of prevalence, there are very few true incidence studies, and so analyses on this important issue often use surrogates such as the incidence of drug-treated diabetes, which clearly can vary for many reasons other than a change in the actual incidence of diabetes. Wareham and Forouhi (57) highlighted this need for better data to establish which are the principal factors underpinning the rising prevalence. What is clearly needed is age-specific incidence data for the same populations, separated in time, but likely to have experienced life style and/or other risk factor changes. Ideally, this should be part of a formal surveillance program, rather than ad hoc research studies.

Type 2 Diabetes in Children and Adolescents

One of the most alarming consequences of the diabetes epidemic is the appearance of type 2 diabetes in children and adolescents (58,59). Until a decade or so ago, type 2 diabetes was regarded as a disease of the middle aged and elderly. Although it still is true that this age group maintains a higher relative risk (in relation to younger adults), there is accumulating and disturbing evidence that onset in the 20 to 30 yr of age group is increasingly seen (59,60). Now, even children are becoming caught up in the type 2 diabetes epidemic. Although type 1 diabetes remains the main form of the disease in children worldwide, it is more than likely that within 10 yr type

2 diabetes will be the more prevalent form in many ethnic groups, potentially including Europid groups (61). There are now numerous reports of type 2 diabetes in children from countries including Japan, the United States, Pacific Islands, Hong Kong, Australia, the United Kingdom and Taiwan (59,60,62-64). Dabelea and coworkers have reported on changes in rates of diabetes in Pima Indian children over a 30 year period (65). They have demonstrated rising rates of glucose intolerance with time and age, as well as a female preponderance. From 1967-76 to 1987-96 the prevalence of type 2 diabetes in children markedly increased from 2.4% in males and 2.7% in females to 3.8% in males and 5.3% for females.

Precise estimates of the prevalence of type 2 diabetes in children and adolescents remain few and far between, but nevertheless some indication of the magnitude of this growing problem is available. Data from a survey of

3 million children in Taiwan (66) found the annual rate of newly identified diabetes to be 9.0/100,000 boys and 15.3/100,000 girls. In the US, national data from 1988-1994 (67), and data from a single school district (68) collected approx 10 yr later showed diabetes prevalences of 0.13% and 0.4% respectively. Clinic studies from the US (69), Thailand (70) and New Zealand (71) have shown that, of all new referrals to clinical diabetes services, the proportion that are for type 2 diabetes has risen markedly over recent years such that, by the end of each observation period, type 2 diabetes accounted for 18-35% of the new cases presenting to these clinics. However, not all populations are witnessing such a marked rise in type 2 diabetes among children and adults. Well-designed studies from Germany, Austria, France and the UK (72-74), reporting data from diabetes registers and from multiple diabetic clinics show type 2 diabetes accounting for only 1-2% of all cases of diabetes. Nevertheless, even in these lower-risk European populations, where most of the cases of type 2 diabetes have occurred in children from high-risk ethnic groups, a small number of cases of type 2 diabetes have occurred in Europid children.

The emergence of type 2 diabetes in children brings a serious new aspect to the diabetes epidemic and heralds an emerging public health problem of major proportions in the pediatric area. The rise of type 2 diabetes in this age group is mainly owing to the increase in time spent on sedentary activities such as television and computer usage, either for games or school-work, with consequent reduction in sports. The additional effects of diets high in energy, carbohydrate, and fat simply add to the risk of developing diabetes and obesity.

This fall in the age of onset of type 2 diabetes is an important factor influencing the future burden of the disease. Onset in childhood heralds many years of disease and an accumulation of the full range of both micro-and macrovascular complications (61). The American Diabetes Association (ADA) and the American Academy of Pediatrics have published a consensus statement on the problem (62). A key area raised in this report is the issue of poor compliance with diet and pharmacological therapies. Recently, a number of pharmaceutical companies have embarked on clinical trials of oral hypoglycaemic agents to check their safety and efficacy in this age group as they may face up to 40-50 yr of therapy.

Another worrying aspect is the high risk of, and early appearance of long term micro- and macrovascular complications in the adolescent and early adult years. As with adults, it is expected that youth with type 2 diabetes will also develop diabetes related micro- and macrovascular complications. This was reported recently in a study from Canada, where subjects who developed type 2 diabetes as children were then surveyed as young adults, aged between 18 and 33 yr. Of the 51 subjects reviewed, 9% had died, 6% were on dialysis, while one had a toe amputation and one was blind (75).

Another follow-up study from Japan compared those with type 1 and type 2 diabetes diagnosed under 30 yr of age for the development of nephropathy (76). After 30 yr of diabetes, 44% of those with type 2 and 20.2% of those with type 1 had nephropathy. Yet another study (77) looked at the incidence of retinopathy and nephropathy among Pima Indians diagnosed with type 2 diabetes under 20 yr of age (youth), 20-39 yr (young adults) and 40-59 yr of age (older). At less than 5 yr duration of type 2 diabetes, nephropathy had developed at a similar rate in all age groups (incidence/1,000 person years: 13/1,000 youth, 8/1,000 young adults and 7/1,000 older). However, retinopathy was not apparent in those with youth onset diabetes for less than 5 yr, and only appeared among this group after 5-10 yr duration (incidence/1,000 person years: 10/1,000 youth, 29/1000 young adults and 35/1,000 older). A study of New Zealand Maori diagnosed with diabetes before the age of 30 compared the prevalence of several diabetic complications between those with type 1 and those with type 2 diabetes (78). Not only was type 2 more common among this population, but the prevalences of nephropathy and retinopathy were higher in those with type 2 diabetes, and the prevalence of hypertension also greater. Data from Taiwan indicate that compared to children with normal glucose tolerance, those with type 2 diabetes have a 70% increased risk of having hypertension and an 80% increased risk of having an elevated serum cholesterol (66).

These studies have important implications in that they highlight the risk of complications occurring at a relatively young age and, as in the case of the Pima Indian study, that these complications can occur relatively soon after diagnosis. The data on complications confirm that type 2 diabetes in children and adolescents is not a mild and benign elevation of blood glucose. Rather, it carries at least as high a risk of microvascular complications as is seen in type 1 diabetes, and predisposes to premature vascular disease in the form of hypertension and dyslipidemia. In this population, complications of diabetes occur as these people enter their peak working and earning capacity, potentially increasing the burdens on health budgets and society as a whole. Early detection and intervention is therefore essential to reduce the risk of future complications.

Type 2 Diabetes in the Elderly

As seen in Fig. 2, the risk of developing diabetes rises sharply with increasing age, rising in Australia from 0.3% in the 25-34 year old age group to 23.6% in those over 75. Among the over 75s, when the prevalence of impaired glucose tolerance and impaired fasting glucose is added to the figure for diabetes, the prevalence of abnormal glucose metabolism is 53% (36). In this age group, it is clearly normal to be abnormal. It should be noted that, in some populations, there is a reduction in the prevalence of diabetes in the oldest age group, compared to the prevalence in the middle-aged. This is likely to be owing to a survivor effect, in which those with diabetes are less likely to survive into old age, and so the prevalence of diabetes among those who do survive into old age is slightly lower than in younger age groups.

The elderly would also be expected to suffer significantly from the morbidity associated with diabetic complications, as their age and other comorbidities provide additional risk. However, the effect of 'competing morbidities' may also mean that the impact of any single disease in the elderly is less than in younger people. Furthermore, in considering the impact of diabetes on morbidity and mortality in the elderly, it may be important to differentiate between those elderly people who have had diabetes for many years and those who only develop diabetes when they are older.

A recent meta-analysis of studies on mortality among people developing diabetes over the age of 60 has, in fact, confirmed that the impact of diabetes on total mortality seems to fall with increasing age of onset of diabetes (79). In comparison to nondiabetic populations, the relative risk (with 95% CI) of mortality for men diagnosed between the ages of 60 and 70 was 1.38 (1.08-1.76) and for men diagnosed aged 70 yr or older was 1.13 (0.88-1.45). The findings for women were similar, with relative risks of 1.40 (1.10-1.79) and 1.19 (0.93-1.52) for the 2 age groups respectively.

Etiological Factors in the Development of Type 2 Diabetes Environmental Factors

Obesity. There is an enormous amount of evidence implicating obesity in the development of diabetes. This includes population studies comparing rates of obesity and of diabetes across different populations, cross-sectional and longitudinal studies within populations, and intervention studies assessing the impact of weight loss. Those populations with the highest rates of diabetes, such as the Pima and Nauruans, also have very high rates of obesity. Similarly, populations with low rates of obesity tend to have low prevalence of diabetes.

More significantly, studies within populations tend to show a gradation of diabetes prevalence, with diabetes being markedly less common at all ages among the leanest members of the population. This association has been demonstrated in most ethnicities and populations. Longitudinal studies also show increasing likelihood of development of diabetes according to obesity level. Data from the Nurses Health Study (80) demonstrate that, with increasing body mass index (BMI), the risk of developing diabetes increases. It is interesting to note that in this large, prospective study, the excess risk is not restricted to those in the obese category (Fig. 4). Indeed the risk of developing diabetes appears to be related to BMI in a continuous fashion, such that even a BMI of 24 kg/m2, which is usually considered to be within the "normal" range, carries a greater risk of developing diabetes than does a lower BMI.

The duration of obesity is also important. Data from a study from Israel (81) show that, for any current BMI, a greater BMI 10 yr previously increased the risk of developing diabetes. Randomized controlled clinical trials (RCT) provide further robust evidence of the link between obesity and diabetes. Intensive lifestyle interventions in those with impaired glucose tolerance and obesity have focused on dietary change, increased physical activity and weight loss (82,83). With weight loss targets of 5-7%, and achieved weight loss of approx 5%, both the Finnish (82) and the American (83) studies showed a 58% reduction over 4-6 yr in the incidence of diabetes among those in the intensive lifestyle study arms compared to those in the control arms. Furthermore, a placebo

BMI group (kg/m2)

Fig. 4. The age-adjusted risk of developing diabetes over 8 years, according to baseline BMI. The nurse Health Study (80).

BMI group (kg/m2)

Fig. 4. The age-adjusted risk of developing diabetes over 8 years, according to baseline BMI. The nurse Health Study (80).

controlled RCT of the weight loss drug orlistat showed a 37% reduction in the incidence of diabetes and a 45% reduction within the subgroup with IGT (84).

Type and Measurement of Obesity

In the last 10-15 yr, it has become apparent that different fat depots have different properties. In particular, visceral fat has been found to be more metabolically active than subcutaneous fat. Circulating free fatty acids (FFAs) (as well as inflammatory cytokines) encourage insulin resistance in liver and muscle and are released at a greater rate by intra-abdominal compared to subcutaneous adipocytes. Furthermore, central fat deposits release FFAs into the portal circulation, which drains directly into the liver, further promoting hepatic insulin resistance and hyperglycemia. A study of second generation Japanese Americans showed that visceral fat, as measured by intra-abdominal fat area on CT scanning, predicted the development of diabetes, although other measures of total adiposity, including BMI, did not (85). For those of third generation Japanese descent, all the measured indicators of obesity were predictive of diabetes incidence. A number of large observational studies have relied on anthropometric measurements of adiposity to compare the impact of overall adiposity (as determined by BMI) with that of visceral adiposity (as measured by the waist circumference or the waist:hip ratio (WHR)) on the development of type 2 diabetes. Cross-sectional data from a study in Mauritius (86) showed that both BMI and WHR were independently associated with the presence of diabetes, with WHR being more important in women, and BMI more important in men. In the Health Professionals Follow-Up Study of over 27,000 men, waist circumference (WC), WHR, and BMI predicted the development of diabetes over 13 yr, with the risk being 7-12 times higher in those in the top quintile of each measurement, compared to those in the bottom quintile (87). Among those who were obese according to the BMI (BMI > 30 kg/m2), the risk of developing diabetes varied considerably according to the WC. However, the opposite was also true, in that for those with a WC ^102 cm (i.e., within the obese range), the risk also varied according to the BMI. Overall, BMI and WC were better predictors than was WHR, but it appears that each of the measures provides information about the risk of diabetes that is not captured in the other (i.e., they are statistically independent of each other). This is consistent with the hypothesis that both subcutaneous and visceral fat depots play a role in the development of diabetes. However, if one were to accept that, pathophysiologically, visceral fat is the key fat depot, an alternative explanation is that the inherent difficulties in accurately measuring WC mean that it is a relatively poor measure of an important physiological parameter (visceral fat), although BMI is a good measure of a less important physiological parameter (total fat), which itself is correlated with visceral fat.

Physical Activity and Exercise

Contracting skeletal muscle takes up more glucose from the circulation than it does at rest. This effect is partly mediated by adrenaline, and is responsible for the state of improved insulin sensitivity that is produced by exercise. The increased glucose uptake continues after exercise has been stopped, to replenish glycogen stores, and so regular exercise has the potential to improve carbohydrate metabolism in both diabetic and nondiabetic subjects. In addition, it has beneficial effects on lipid metabolism and its contribution to weight loss provides another mechanism whereby exercise may influence the development of type 2 diabetes.

Cross-sectional population based comparisons of diabetic with normoglycemic subjects have shown associations of diabetes with various different assessments of physical activity in populations as diverse as Asian Indians, Alaskan natives, and Chinese subjects. Prospective studies identifying risk factors for the development of type 2 diabetes in normoglycemic subjects also find physical activity to be correlated. In a large study of US male physicians, vigorous activity undertaken at least once a week led to a relative risk of developing type 2 diabetes of 0.71 (after adjusting for age and BMI), in comparison to those exercising less frequently (88). The effect was strongest in the most obese. A very similar result was found among a cohort of over 85,000 women (89), but the effect was significantly weakened after controlling for BMI. Moderate physical activity in British men also reduced the relative risk to 0.4 (90). Direct measurements of physical fitness have also been shown to be predictors of type 2 diabetes, and although less practical for screening programs, seem to provide more information than do physical activity scores.

Further evidence of the important role that exercise plays has recently been presented in RCTs targeting the prevention of diabetes. In studies from the US and Finland, lifestyle interventions that included both dietary change and increases in exercise levels led to a reduction in the incidence of diabetes of 58% among obese people with IGT (82,83).

Recently, an additional component to the role of physical activity in the development of diabetes and obesity has been identified. Measurements of sedentary behavior have been found to be independent predictors of obesity and of abnormal glucose tolerance. Cross-sectional studies have related the amount of time spent watching television to the risk of obesity and of having IGT or diabetes and have found a significant relationship (91,92). Indeed, the relationships appear to be stronger for television viewing time than they are for time spent undertaking physical activity. This suggests an additional health message focusing on avoiding sedentary behaviors in addition to the promotion of exercise sessions.

Dietary Factors. There seems to be little doubt that diet plays a significant role in the development of type 2 diabetes. However, it has been remarkably difficult to pin down the precise dietary constituents that are the key players. There are several reasons for this. Observational studies relate measurements of potential risk factors to outcomes, and rely on accurate measurements of both. Precise measurement of dietary intake has been particularly challenging, and although a variety of validated questionnaires have been developed to assess food intake, their accuracy is always limited by the ability of individuals to recall their intake and is also influenced by the patient's perceived rather than actual diet. Furthermore, observational studies can be confounded by associations with other factors. This is neatly demonstrated by the case of hormone replacement therapy (HRT). A number of large, well-conducted observational studies reported that women who were on HRT had lower rates of CVD and some cancers than women not using HRT, and concluded that HRT was protective against these diseases. However, clinical trials showed the opposite —women randomized to HRT actually had slightly higher rates of CVD and cancer than those on placebo (93). Thus, it is clear that although the reports from observational studies had attempted to adjust, statistically, for the fact that women who chose to go onto HRT might also have made a variety of other healthy lifestyle choices, which in themselves might reduce disease risk, this was never fully achieved, and led to an erroneous conclusion. RCTs provide an opportunity to assess causality, not just correlation. However, they also have some pitfalls. Although different groups within an RCT will be "instructed" to follow different diets, final results ultimately reflect the "achieved" diet (which, as described above, is difficult to measure), rather than the prescribed diet. Additionally, a number of studies of diet within the diabetes field have used diet as one component of a lifestyle program, making it difficult to tease out the precise roles of specific dietary components. With these limitations in mind, it is reasonable to draw some conclusions from the literature.

Observational Studies. The increased risk of diabetes with increasing intake of total fat has been reported in several studies using prospective data (94,95). However, this has not been a consistent finding, with other studies failing to find the link (96,97). A higher intake of saturated fat has also been associated with type 2 diabetes (98), whereas higher intakes of unsaturated fats appear to be protective, with those people in the top quintile of polyunsaturated fat intake having a 25% lower risk of developing diabetes compared to those in the bottom quintile (97). There may also be a role for trans fatty acids, which may also increase the risk of developing diabetes, with those in the top quintile of trans fatty acid intake having a 31% higher risk of developing diabetes compared to those in the bottom quintile (97).

The relationship of carbohydrate intake to diabetes is less clear than for fat intake, with a recent review concluding that there was no association between total carbohydrate intake and diabetes risk (99). However, there seems to be a fairly consistent finding in terms of the importance of dietary fiber. Three large longitudinal studies showed that a low intake of dietary fiber increased the risks of developing diabetes (96,100,101), such that those who were in the lowest quintile of dietary fiber intake had a 39-56% increased risk of developing diabetes, compared to those in the highest quintile of fiber intake.

Randomized controlled trials. The most robust data on lifestyle factors in the development of diabetes come from the diabetes prevention trials. There are now several such studies, each of which has convincingly shown that lifestyle changes focusing on weight loss, dietary change, and increasing physical activity, significantly reduce the risk of progressing to diabetes among people with IGT (82,83,102,103). The dietary targets of the Finnish DPS (82) were similar to those used in other studies, and included total fat intake <30% of energy intake, saturated fat intake <10% of energy intake, and fiber intake >15g/1,000 kcal. These targets, in combination with a weight loss and a physical activity target, led to a 58% reduction in the rate of developing diabetes. Furthermore, the risk of

Number of targets achieved

Fig. 5. The incidence of diabetes according to the number of lifestyle targets achieved. The Diabetes Prevention Study (82). Data shown for the intervention and control arms of the clinical trial.

Number of targets achieved

Fig. 5. The incidence of diabetes according to the number of lifestyle targets achieved. The Diabetes Prevention Study (82). Data shown for the intervention and control arms of the clinical trial.

developing diabetes fell progressively with increasing numbers of targets achieved (Fig. 5). Thus, it appears that each of the targets was contributing to the prevention of diabetes, and it is therefore reasonable to conclude that increased dietary fat and reduced levels of dietary fiber are important etiological factors in the development of type 2 diabetes.

36 Fat Burning Potent Foods

36 Fat Burning Potent Foods

If you’re overweight, you are not a bad person. You’re simply overweight. But it’s important to lose the extra pounds so you’ll look good, feel healthier and develop a sense of pride and self-esteem. Once you’ve lost the fat, you’ll need to maintain your weight.

Get My Free Ebook

Post a comment