Type II Diabetes Disparities in Diverse Women: The Potential Roles of Body Composition,Diet and Physical Activity

Abstract

The rates of diabetes in the USA are rapidly increasing, and vary widely across different racial/ethnic groups. This paper explores the potential contribution of body composition, diet and physical activity in explaining diabetes disparities across women of different racial and ethnic backgrounds. For body composition, racial/ethnic groups differ widely by BMI, distribution of body mass and quantity and type of adipose tissue. Dietary patterns that vary across race/ethnicity include consumption of meat, added sugars, high-glycemic carbohydrates and fast food. Additionally, physical activity patterns of interest include aerobic versus muscle-strengthening exercises, and the purpose of physical activity (leisure, occupation, or transportation). Overall, these variables provide a partial picture of the source of these widening disparities, and could help guide future research in addressing and reducing diabetes disparities.

Keywords

diabetes diet ethnicity obesity physical activity race women's health

Type II diabetes (diabetes) is a largely preventable disease, yet the prevalence of diabetes in the USA has reached epidemic proportions. Recent studies estimate that 40% of the US adults have diabetes or prediabetes [1]. This constitutes an approximate 50% increase in diabetes in the past decade [1], and numbers are expected to rise. Based on current trends, it is projected that the prevalence of diagnosed diabetes will more than double by 2050 [2]. Those with diabetes experience increased risk of a number of complications, including cardiovascular disease, chronic kidney disease and reduced lifespan. People who are diagnosed before age 30 lose an estimated 15 years of life, and 24 years of quality-adjusted life [3]. Additionally, treatment for diabetes is extremely costly. Recent studies estimate that the cost of diabetes in the USA has increased by 41% in just 5 years, totaling US$245 billion in 2012, and is more than 20% of total US healthcare costs [4].

While diabetes is prevalent across racial/ethnic groups, it shows marked health disparities between individuals of different races and ethnicities. Age-adjusted rates of diabetes for non-Hispanic blacks (blacks) and Hispanics are nearly double those of non-Hispanic whites (whites) [1]. While these differences are large, disparities with other racial groups are even more extreme: prevalence for white adults is approximately 7.6% [5], yet rates as high as 18.3% for US-dwelling Asian–Indians [6] and 38% in US-dwelling Pima Indians [7] have been reported. Nationally collected data from the CDC show significantly higher rates of diabetes among Hispanics, blacks, Asian–Americans (Asians) and American Indians compared with whites, with only Chinese–Americans and Alaska Natives showing lower diabetes prevalence [5].

Additionally, these disparities are expected to increase. While the overall number of individuals with diagnosed diabetes is projected to increase by approximately 99% among whites by 2050, the increase is estimated to be over 200% for blacks and nearly 500% for Hispanics [2]. Rates of diabetes are expected to increase more in women (220%) than in men (174%) by 2050, particularly for women of racial and ethnic minorities. Given current trends, it is expected that 49% of black women and 52.5% of Hispanic women will develop diabetes in their lifetime, compared with 31% of white women [3].

A key component of addressing rising rates of diabetes in women, then, will be addressing disparities in this disease. However, the causes of these wide disparities at this point are not well understood. The causes of diabetes are varied, and include genetics, physiology (such as body composition), behaviors (such as diet and physical activity) and healthcare access. Differences in any or all of these factors could contribute to different rates of diabetes. In the current paper, we will explore potential sources of diabetes disparities by examining past research on differences in health behavior and physiology between women of different racial/ethnic backgrounds. Specifically, we will examine patterns of body composition, diet and physical activity across women of varying racial/ethnic groups in order to explore the potential role each of these may play in differential rates of diabetes between women of different racial/ethnic groups.

Literature search

We searched PubMed for papers comparing body composition, dietary patterns and/or physical activity variables between two or more racial/ethnic groups. In the case of body composition, this included BMI, percent body fat, body fat distribution and specific fat depots (such as visceral and subcutaneous fat). For dietary patterns, we searched for papers on specific dietary patterns that have been linked to diabetes such as sugar and sugar-sweetened beverage consumption; consumption of red and processed meats; fast food consumption; and low intakes of fiber, fruits and vegetables. Physical activity was considered any sort of measured bodily movement, and included both aerobic and muscle-strengthening activities. Whenever possible, we isolated data that were specific to women. We restricted our search to papers published since 1990.

We focused primarily on data from white, black and Hispanic women, simply because the greatest amount of data were available for these groups. The majority of large, nationally representative datasets generally have not provided more detailed racial/ethnic categories beyond these. Data were seldom available, particularly for behavioral variables, for Asian, Pacific Islander, or American Indian women. Additionally, data on different subgroups within racial/ethnic categories were scarce, though this data was included when available.

Racial/ethnic differences in women's body composition

BMI & obesity

The vast majority of research exploring the association between body composition and diabetes has centered on excess adiposity, marked differences in which have been noted across racial/ethnic groups. Obesity, classified as a BMI greater than 30, has been strongly tied to risk of diabetes in women. In the Nurses’ Health Study, women in the 90th BMI percentile had 11 times greater risk of developing diabetes over the 8-year follow-up than women in the 10th percentile [8]. Higher BMI even in women within a normal weight range has been linked to greater incident diabetes, as has weight gain in normal weight women [9].

Measures of BMI show that black and Hispanic women typically have higher overall BMI and higher rates of overweight and obesity than white women. Recent estimates show that approximately 82% of black women and 74% of Hispanic women are classified as overweight (BMI >25), compared with 61% of white women, while obesity is seen in 58 and 41% of black and Hispanic women, respectively, compared with 33% of white women [10]. Few nationally representative studies have included racial/ethnic breakdowns beyond these, but those that have consistently show that Asian–Americans, particularly women, have a markedly lower prevalence of overweight and obesity, while obesity in Native American women is similar to rates in blacks and Hispanics [11].

However, there are drawbacks to using BMI to compare body composition across racial/ethnic groups. Actual body composition, such as percent body fat, appears to differ across different racial/ethnic backgrounds such that two women with identical BMIs may have very different physiology. Articles examining BMI and body fat percentage consistently show higher body fat percentage in South East Asians than whites at the same BMI [12]. For example, at a given percent body fat, BMI in whites is approximately three BMI points higher than in Singaporeans [13]. More extreme differences were shown when comparing body fat percentages in Asian Indian women and Pacific Island women, who had the same percentage of fat, but BMIs of 26 and 35, respectively [14]. One meta-analysis examined the association between percent body fat and BMI in American black, white, Chinese, Ethiopian, Indonesian, Polynesian and Thai women, and found that predictions of body fat percentage using standard BMI, age and gender-centered equations erred by a range of −3.9% (Polynesians) to +10% (Ethiopians) [15]. These racial/ethnic differences in body fat–BMI relationships also appear to be more pronounced in women than men [12,16].

Body fat distribution

In this respect, exploring overall rates of overweight and obesity between racial/ethnic groups is likely less useful to the discussion of disparities in diabetes than examining differences in more specific measures of body composition. While more exact measures of percent body fat are useful, a large body of research suggests that the distribution of body fat may be more important than the quantity. Fat in the abdominal region in particular has been shown to be strongly associated with diabetes. Among the most common measures of abdominal distribution of fat are waist circumference, which offers a crude measure of abdominal adiposity, and the ratio of the waist and hip circumference (waist–hip ratio), which compares distribution of fat in the abdomen and hip region. Waist circumference (WC) has been shown to be a markedly better predictor of prevalent and incident diabetes than overall weight or BMI [17]. Longitudinal data in the Nurses’ Health Study showed that, even after adjusting for BMI, risk of incident diabetes over an 8-year period was more than five times higher in the 90th percentile for WC than those in the 10th percentile, and more than three times higher in the 90th versus 10th percentile for waist–hip ratio (WHR) [8]. Similarly, in the Normative Aging Study, WHR in men was more strongly associated with diabetes than overall BMI and was positively associated with blood glucose independent of overall adiposity [18].

There is some evidence that distribution of fat differs between racial/ethnic groups. There are somewhat consistent findings showing that Hispanics have greater abdominal adiposity than white women [19]. For example, one study found relatively modest differences in average BMI (25.2 vs 23.9) and body fat percentage (38.4 vs 34.9) in Hispanic versus white women, yet found that abdominal adiposity was 30–40% greater in Hispanics [19]. In fact, waist circumference was identified as the best predictor of incident diabetes in a cohort of Mexican–Americans, with those in the highest quartile showing 11 times greater risk than those in the lowest quartile [20]. There is also evidence that Asian women (Filipino), who generally have lower BMIs than white women, have higher WHR [21], and black women generally have higher WC measurements than white women [22].

While mass and distribution of body fat clearly vary across women of different racial/ethnic backgrounds, it appears the story is even more complicated than that. Several studies have shown that these differences appear to change at different levels of BMI/WC [16,23], and also appear to change with age [24]. Thus, while BMI may be an equally good predictor of fat mass for both Chinese women and white women, this may only be true for middle aged, normal weight women. One study showed that accounting for both age and skinfold thickness improved correlations between BMI and percent body fat for white and Asian (primarily Chinese) women [12].

Abdominal fat deposits

While abdominal obesity is a superior predictor of diabetes than overall body fat or BMI, this measure is limited in that it cannot distinguish between types of fat in the abdomen. Fat in the abdominal region is generally deposited either outside of the abdominal wall (subcutaneous adipose tissue [SAT]) or within the abdominal wall (visceral adipose tissue [VAT]). Both of these fat depots are captured simultaneously when relying on WC or WHR, yet they differ markedly in their physiology and association with metabolic dysfunction. VAT in particular is robustly predictive of metabolic risk factors such as fasting glucose, metabolic syndrome, insulin sensitivity and diabetes [25]. SAT, on the other hand, often shows no association with metabolic markers, and has even been shown to be protective against diabetes [25]. Distribution of fat in these locations in the abdomen may vary across race/ethnicity, which could contribute to differential rates of diabetes beyond simple measures of overall abdominal obesity.

Research examining racial/ethnic differences in VAT and SAT distribution have shown remarkably consistent results (Table 1). Despite generally having higher weight, BMI, body fat percentage and waist circumference, black women consistently show lower areas of VAT and higher areas of SAT compared with women of other races/ethnicities [22,26 –30]. Measures of VAT in Southeast Asians tend to be significantly higher than in whites, despite smaller overall body size. One study with black, white and Filipino women showed that, consistent with previous research, black women had the highest average waist circumference and the highest amount of SAT, but the lowest amount of VAT [22]. Filipino women, on the other hand, had the highest amount of VAT despite having the smallest waist circumference, and also had the highest rates of diabetes. This is consistent with other studies showing significantly higher rates of VAT in Asian women than white women [23,31 –32], despite smaller overall body size.

Table 1.

Measures of visceral adipose tissue and subcutaneous adipose tissue in women of different racial/ethnic backgrounds.

Study (year)	Sample (sample size, race/ethnicity)	Measure	Findings	Ref.
Araneta and Barrett-Connor (2005)	196 white women; 181 Filipino women; 193 black women	CT between L4 and L5 vertebrae	Highest VAT mass in Filipino women despite lowest BMI; lowest VAT mass in black women despite the highest BMI and WC; most SAT in black women	[22]
Lear et al. (2007)	201 Europeans; 219 Chinese; 207 South Asians	CT between L4 and L5 vertebrae	Percent body fat was higher in South Asians than Chinese or Europeans; South Asians had more VAT than Europeans at all but the most extreme WC; Chinese had more VAT than Europeans at WC >71 cm	[23]
Despres et al. (2000)	240 white women; 143 black women	CT between L4 and L5 vertebrae	Higher total body fat in black women, but similar levels of VAT in black and white women	[26]
Lovejoy et al. (1996)	37 black women; 22 white women matched for obesity status	CT between L4 and L5 vertebrae	Lower VAT in black women despite similar BMI and WC	[27]
Carroll et al. (2008)	45 black women; 52 Latina women; 32 white women	CT between L4 and L5 vertebrae	Highest percent body fat and highest SAT in black women, but lowest total VAT in black women; Highest VAT in Latinas	[28]
Perry et al. (2000)	30 black women; 36 white women	MRI of total abdomen	Less VAT in black women despite similar total body fat percentage	[29]
Bray et al. (2008)	346 white women; 157 black women; 104 Latinas; 21 Asian/Pl women	CT at L2/L3 and L4/L5 vertebrae	Highest BMI, WC and SAT in black women, but lowest VAT in black women; highest VAT in white women; highest VAT/SAT ratio and lowest BMI in Asian/PI women	[30]
Raji et al. (2001)	12 Asian–Indians (7 women); 12 Caucasians (7 women) matched for BMI	CT at L2/L3 vertebrae and L3/L4 vertebrae	Higher fat mass and VAT mass in Asian–Indians despite similar BMI	[31]
Park et al. (2001)	36 white women; 18 Asian women	Whole body MRI	Greater VAT mass in Asian women despite lower BMI and less total body fat	[32]

CT: Computed tomography; SAT: Subcutaneous adipose tissue; VAT: Visceral adipose tissue; WC: Waist circumference.

The vast differences in body composition between racial/ethnic groups suggest that racial/ethnic-specific anthropometric targets, like BMI or WC, be established [33,34]. This may be particularly appropriate for Asians, given the large discrepancy between BMI/overall body size and body fat, especially VAT in the abdomen. Accordingly, the American Diabetes Association has recently released new BMI cut points specifically for Asian–Americans, identifying 23 as an appropriate BMI cut point for diabetes screening rather than 25, as is used for other racial/ethnic groups [35,36].

Muscle mass across racial/ethnic groups

Despite the fact that muscle plays a crucial role in glucose metabolism, very little data exist on how muscle quantity or quality relates to the risk of metabolic disorders. There is some evidence suggesting that lower ratios of muscle-to-total body weight is associated with greater insulin resistance [37], and lower abdominal muscle mass in ethnically diverse women has been associated with higher diabetes prevalence [38].

Few studies have examined differences in muscle mass between racial and ethnic groups. One examination of abdominal muscle in an ethnically diverse sample showed that Asian–Indians had less lean mass and skeletal muscle than European or Pacific Island individuals [14]. Similarly, one study examining fat and muscle mass in white and Hispanic women found that Hispanics not only had more abdominal fat mass, they also had less fat free mass in the trunk region [19].

Body composition & disease risk across diverse populations

Clearly there are differences in body composition in women of different racial/ethnic backgrounds; whether these differences translate to differences in disease risk is an important question that has received relatively little attention. Black women, for example, have relatively high rates of diabetes, but consistently had the lowest levels of VAT, thus the influence of body composition may differ by racial/ethnic group. In the Nurses’ Health Study, each 5-unit increment in BMI at baseline was associated with the greatest increase in risk of incident diabetes in Asian (RR: 2.36) and Hispanic women (RR: 2.21), and the lowest increase in risk in black women (RR: 1.55) [39]. Similarly, weight gain over the 20-year period increased the risk of diabetes most strongly for Asian women: each 5-kg gain increased risk by 84% in Asian women, compared with just 38% for blacks and 37% for whites. In a study of South Asians and Europeans, increasing WHR was associated with greater diabetes prevalence in both racial groups. However, this association was stronger for South Asians, with each increase in WHR being associated with more than double the prevalence of diabetes [40]. Additionally, higher rates of VAT appear to be only part of the explanation for the higher rates of diabetes in South Asians. While Filipino women had the highest rates of diabetes and the highest levels of VAT in the Rancho Bernardo cohort, the odds of diabetes in Filipinos were still more than twice that of black women and more than seven-times that of white women even after adjusting for VAT area [22]. Similarly, an examination of anthropometrics and metabolic markers in white and black women showed that, despite similar BMI and WHR between the two groups, insulin sensitivity was 36% lower in black women [27].

Finally, one study did directly assess the extent to which differences in body composition contributed to disparities in diabetes across women of different racial/ethnic groups. The authors found that differences in abdominal obesity specifically accounted for 12.1% of the difference in diabetes between white and black women and 9.8% of the difference between white and Hispanic women [41]. Differences in abdominal obesity, then, do play a role in diabetes disparities, yet clearly there are other factors responsible as well.

Racial/ethnic differences in diet

Dietary patterns have been strongly related to diabetes risk. While consumption of sugars [42 –46], foods with high glycemic indexes [47,48], red and processed meats [49 –53] and saturated fat [54] are associated with increased risk of diabetes, other foods such as green leafy vegetables [55,56] and whole grains [57,58] constitute protective factors. Thus, differences in food consumption among women of various ethnicities may partially account for disparities in their risks of diabetes. According to national data, for example, blacks are less likely to meet nutrition guidelines, compared with whites and Mexican–Americans [59]. A closer inspection of dietary practices among different racial/ethnic groups, detailed below and summarized in Table 2 , reveals disparities in eating behaviors associated with diabetes.

Table 2.

Dietary behaviors among women of different ethnic/racial backgrounds.

Study (year)	Sample (sample size, race/ethnicity)	Measure or survey used	Relevant findings	Ref.
O'Brien et al. (2013)	25,433 women: 4321 Hispanic and 21,112 non-Hispanic women	California Health Interview Survey	Hispanic women were more likely than non-Hispanic women to consume more fried potatoes, fast food and SSBs, but less likely to consume more desserts	[60]
Kong et al. (2013)	352 WIC mothers (170 Hispanic women, 182 black women) and their children	24-h recall with multiple-pass technique	Percentage of calories from fat, saturated fat and added sugars was lower among Hispanic compared with black mothers. Hispanic mothers had higher consumption of fiber, whole grains and fruits	[61]
Storey and Anderson (2014)	51.8% female, 68.7% white, 11.4% black, 13.6% Hispanic, 6.3% other	NHANES, 2009–2010	Men consumed more dietary fiber than women. Black women consumed less dietary fiber than women of other racial groups	[62]
Drewnowski and Rehm (2015)	16,628 individuals (11,182 adults, 8346 female, 7102 white, 3414 black, 3436 MA and 1843 other Hispanic)	NHANES, 2007–2010	Among adults, men consumed more fruit than women. Blacks were less likely to meet the guidelines for fruit consumption	[63]
Wang et al. (2010)	44,788 individuals (23,249 women, ethnic breakdown not specified)	1988–1994 and 1999–2004 NHANES, the 1994–1996 CSFII and DHKS	Between 1994–1996 and 1999–2004 white women saw a decrease in consumption of all meat, black women's consumption of all meat groups (except for other meat products) increased. MA women's decrease in consumption of other meat products was larger than that of women of other races	[64]
Chandran et al. (2013)	3148 women (1692 white, 1456 black)	GSEL-Food Frequency Questionnaire	White women showed higher consumption of red meat, compared with black women, while consumption of processed meat and poultry was higher among black women	[65]
Pereira et al. (2015)	3031 individuals (1444 black, 1587 white, 1580 women)	Structured interview with questions about dietary practices	At all time points, black individuals consumed fast food more frequently than whites, and men consumed fast food more frequently than women. White women had the lowest frequency of fast food consumption)	[66]

CSFII: Continuing Survey of Food Intake by Individuals; DHKS: Dietary Health and Knowledge Survey; MA: Mexican–American; NHANES: National Health and Nutrition Examination Survey; WIC: Women Infant Children.

Added sugars & sugar-sweetened beverages

Intake of added sugar is positively associated with incidence of diabetes, not only through weight gain but also through glycemic and metabolic processes [42 –46,67]. Additionally, increased consumption of sugar-sweetened beverages, which comprise the largest (42%) source of added sugar intake [67], has been associated with increased incidence of diabetes in various cohort studies [43 –46]. Multiple studies analyzing racial/ethnic differences in added sugar and sugar-sweetened beverage consumption yield that blacks consume significantly more added sugars than whites, and whites consume the least amount of added sugars of racial/ethnic groups [68 –70]. A study that used 2009 California Health Interview Survey data found Hispanic women to be more likely to consume larger quantities of sugar-sweetened beverages, compared with non-Hispanic women [60]. On the other hand, a comparison of black and Hispanic mothers enrolled in the Women, Infants and Children (WIC) program in Chicago found Hispanics to consume significantly less added sugars and sweetened beverages than blacks [61]. While there is less research regarding the consumption of sugar and sugar-sweetened beverages among Asian populations compared with other ethnicities, there is some evidence that acculturation plays a role in sugar consumption among Asians. In a sample of Korean Americans in New York [71], acculturation was positively associated with consumption of sweet foods, such as candies or chocolates. Less acculturated individuals ate fewer sweet foods, although they were more likely to add sugar to coffee or tea.

Carbohydrates & fiber

Carbohydrates are classified according to their glycemic index, which measures the rate at which a carbohydrate raises blood glucose after ingestion [48]. Highfiber, low-glycemic foods, such as whole grains, leafy greens and legumes, are associated with reduced risk of diabetes [47,57,72]. Conversely, consumption of low-fiber, high-glycemic foods, such as refined grains and white bread, is associated with an increased risk of diabetes [47,48].

Storey and Anderson [62] revealed that dietary fiber consumption is below adequate for women of all ethnicities in their analysis of NHANES data for white, black, Hispanic and ‘other race’ women. Women of ‘other races’ were found to have the highest intake of dietary fiber, followed by Hispanic women and white women, while black women had the lowest. Studies have consistently found blacks to have lower intake of dietary fiber compared with whites and Mexican–Americans, while Mexican–Americans have been found to have the highest intake of dietary fiber [73,74]. In their study of WIC mothers in Chicago, Kong and colleagues [61] found significantly higher consumption of fiber among Hispanic women, compared with black women; whole grain and fruit consumption was also significantly higher among Hispanic women, compared with black women. Nationwide, blacks have lower intakes of fruits and vegetables, compared with whites and Mexican–Americans [59]. Drewnowski and Rehm [63] also suggested that black women are less likely to consume an adequate amount of fruit than white, Mexican–American and other Hispanic women.

Red & processed meat

Positive associations have been found between meat consumption and diabetes, particularly processed and red meat [49 –53]. Nevertheless, more research is necessary to understand why consumption of red and processed meat is associated with an increased risk of diabetes. Saturated fat, found in meat products, and trans fatty acids from hydrogenated oils have also been associated with increased risk of diabetes [54]. These types of fat are believed to increase insulin resistance [54]. Conversely, certain types of unsaturated fats, like linoleic acid, decrease risk of diabetes through improved insulin sensitivity [54].

NHANES data from 1999–2004 show that black women consumed more red meat and other meat products, and Mexican–American women consumed more red meat, compared with white and ‘other race’ women [64]. However, a study comparing white and black women in New York and New Jersey found that red meat consumption was higher for white women, while processed meat consumption was higher for black women [65]. Additionally, a comparison of black and Hispanic mothers enrolled in WIC programs in Chicago revealed that Hispanic women consumed significantly less saturated fat than black women [61].

Fast food consumption

Various studies have documented periodic increases in fast food consumption throughout the years, which is correlated with increased consumption of saturated fat, carbohydrates, added sugars, soft drinks and meats, and with decreased intake of fruits, nonstarchy vegetables and fiber [75 –77]. Nationwide, fast food consumption has been found to be higher among blacks compared with all other ethnicities [75]. A comparison of blacks and whites in a multicenter cohort study revealed fast-food consumption to be higher among blacks at all time points [66]. White women had the lowest consumption of fast food compared with black women, white men and black men; additionally, white women's consumption of fast food decreased over time, while for black women it increased [66]. Hispanic women appear to be at increased risk for fast-food consumption compared with other ethnicities, as revealed by O'Brien and colleagues [60] in their study comparing Hispanic women with non-Hispanics in California.

Overall, a recommended diet for the prevention of diabetes includes a low intake of sugars, starches and saturated fats, and favors consumption of fiber-rich fruits, vegetables and whole grains. National data reveal disparities in diabetes-related eating behaviors among different ethnicities. Black women particularly are more likely to engage in unhealthy eating practices compared with whites and Hispanics. Nevertheless, certain regional studies obtained different results, as demonstrated by Chandran and colleagues’ study of women in New York and New Jersey [65], where red meat consumption was higher for white women. Newby and colleagues [78] suggested that research should focus not only on ethnicity but also on region when analyzing dietary habits; in their study of white and black men, they found an interaction between race and region for consumption of trans fat. In our review of literature, we found no epidemiological research that compared women's eating behaviors across regions, in addition to race/ethnicity.

Racial/ethnic differences in physical activity Benefits of physical activity

Physical activity is associated with a decreased likelihood of developing Type 2 diabetes. In the Women's Health Study, women who walked 5 h per week were 33% less likely to develop diabetes than women who did not walk regularly [79,80]. One longitudinal study of women at risk for developing diabetes found physical activity to be associated with risk for diabetes independently from BMI – with each 100-minincrement of moderate physical activity per week being associated with a 9% decreased risk for diabetes [81].

Both aerobic and muscle-strengthening exercises are protective against diabetes, and can improve glycemic control in people with diabetes. Combined exercise regimens include both aerobic and muscle-strengthening exercise, and are effective in improving insulin sensitivity and decreasing insulin requirement [82]. The CDC recommends that adults get at least 150 min of aerobic activity per week and that they engage in muscle-strengthening exercise at least 2 days per week [83]. Aerobic exercise is recommended to be spread between multiple days throughout the week, and to be at least moderate intensity [83].

Engagement in physical activity

Less than half of all women in the USA meet physical activity (PA) guidelines [84]. Engagement in PA varies markedly by racial/ethnic group. National level surveys have primarily focused on measuring aerobic PA, some through self-report and some through objective measures such as accelerometers. Those relying on self-report measures have typically found higher rates of PA and meeting guidelines in white women compared with black and Hispanic women. A comparison of the self-report measures included in NHANES, the National Health Interview Survey (NHIS), and the Behavioral Risk Factor Surveillance System (BRFSS) all showed a higher percentage of whites meeting PA recommendations compared with blacks and Hispanics, and whites being less likely to be classified as ‘inactive’ [85]. According to the 2011 version of the NHIS, 50% of white women reported meeting aerobic exercise guidelines, compared with just 34.6% of black women and 36.2% of Hispanic women. Black and Hispanic women were also much more likely to be classified as inactive (41.8 and 41.6%, respectively) compared with white women (26.7%) [86]. Similarly, the 2001 version of the BRFSS showed that, compared with white women, odds of meeting PA guidelines were 37 and 27% lower in black and Hispanic women, respectively [87]. Similarly, the National Physical Activity and Weight Loss Survey, conducted in 2002, found that only 12% of white women reported doing no PA, compared with 25.2% of black women and 27.3% of Hispanic women [88].

Data derived from accelerometers has generally shown much lower rates of PA, as well as different pat terns between racial/ethnic groups, than self-reported PA. Accelerometer data from NHANES compiling total PA per day (not in 10-min bouts) show that Hispanic women engaged in 21.7 min per day of moderate PA, compared with 19.4 min in black women and 18.6 min in white women [89]. When looking at the same data for 10-min bouts of exercise, Troiano et al. found that black women engaged in the most PA, followed by Hispanic women, and then white women [89]. Accelerometer data from the CDC show that white women are more likely to engage in vigorous activity, with 10% of white women engaging in five or more 10-min bouts of vigorous activity per week compared with 6% of black women, and 7% of Hispanic women [90].

The discrepancy between self-report and accelerometer-derived data could reflect less overreporting bias on the part of black and Hispanic women, and/or an emphasis in self-report measures on leisure time physical activity (LTPA), rather than occupational (OPA) or transportation activity (TPA; see Table 3 ). Studies distinguishing between types of activity have found that they often differ by race/ethnicity, and that engagement in OPA is independent of engagement in LTPA [88]. One study found that blacks and Hispanics engaged in less LTPA than whites, but more OPA [91]. When adjusting for age, sex and social class, Hispanics were 25% more likely to engage in OPA than whites and blacks [91]. Marshall et al. observed that blacks were the least likely of any ethnic group to engage in a high volume of OPA [88].

Table 3.

Measures of leisure-time physical activity, occupational physical activity and transportation-related physical activity in women of different ethnic/racial backgrounds.

Study (year)	Sample (sample size, race/ethnicity of women)	Measures	Instrument	Findings	Ref.
He and Baker (2005)	3640 white women; 974 black women; 266 Latina women (English-speaking); 236 Latina women (Spanish-speaking)	OPA, LTPA	Self-report questionnaire, created by authors	Spanish-speaking Latinas had the highest amount of total PA and OPA, but the lowest LTPA; white women engaged in the highest amount of LTPA	[91]
Neighbors et al. (2008)	(Men and women) 83,813 whites; 2191 Puerto Rican, 8027 Mexican; 5455 Mexican–American; 1253 Cuban, 641 Dominican; 3129 Central/South American	LTPA	Self-report data from the 2000 and 2003 National Health Interview Surveys	Compared to white women, Latina women's odds of engaging in LTPA: Puerto Rican (67%), Mexican (72%), Mexican–American (78%), Cuban (43%), Dominican (48%), Central/ South American (78%)	[92]
Kandula and Lauderdale (2005)	(Men and women) Asian–American subgroups: 962 Chinese; 667 Filipinos; 766 South Asians; 623 Koreans, 658 Vietnamese; 550 Japanese	LTPA, TPA	Self-report physical activity data from CHIS, collected November 2000–October 2001	Compared to Chinese women, all other Asian-American subgroups are equally or up to 50% more likely to meet LTPA guidelines. Transportation PA varies greatly across Asian-American subgroups: Chinese are most likely to engage in TPA; Koreans are least likely to engage in TPA	[92]
Kandula and Lauderdale (2005)	1865 Asian–American women; 16,951 white/black/Latina women	LTPA, OPA, TPA	Self-report physical activity data from CHIS, collected November 2000–October 2001	Across age groups, white/black/Latina women engaged in much more LTPA than Asian–American women, and only slightly more OPA. Asian–American women consistently engaged in more TPA than white/black/Latina women	[93]
Marshall et al. (2006)	3952 white women; 805 black women; 798 Latina women	LTPA (inactivity)	National Physical Activity and Weight Loss Survey data, collected September–December (2002) Self-report physical activity questions taken from the (2001) BRFSS	Latina women had most leisure-time inactivity (least LTPA) of all groups. LTPA did not vary by marital status. All ethnicities reported more inactivity (less LTPA) with lower education and socioeconomic level. OPA is independent of LTPA	[88]

BRFSS: Behavioral Risk Factor Surveillance System; CHIS: California Health Interview Survey; LTPA: Leisure time physical activity; OPA: Occupational physical activity

PA: Physical activity; TPA: Transportation activity.

Engagement in different types of activity has also changed over time. From 2000–2012, whites and Chinese significantly increased their amount of LTPA, while Hispanics significantly increased their engagement in TPA [94]. In the 2001 California Health Interview Survey, Asian–American women reported engaging in 11.6 min of TPA per day, compared with whites, blacks and Hispanics, who reported an average of 8.1 min per day [93]. Although Asian–American women were more likely to engage in TPA, Asian–American women were 52% less likely than other women to meet LTPA guidelines and were twice as likely to be physically inactive [93].

Similar to PA differences between racial groups, significant PA differences exist between racial subgroups. For example, Japanese are 4% less likely than Chinese to meet LTPA recommendations, while Filipinos are 51% more likely [93]. South Asians are 38% more likely, Koreans 42% more likely and Vietnamese 23% more likely to meet LTPA recommendations than are Chinese [93]. Among Hispanic women, Puerto Ricans are 33% less likely to meet LTPA recommendations than whites, Mexicans are 28% less likely, Mexican–Americans are 22% less likely, Cubans are 67% less likely, Dominicans are 62% less likely and Central or South Americans are 34% less likely [92]. Similarly, a study comparing PA between racial groups found a significant difference in OPA among English-speaking and Spanish-speaking Hispanic women. Only 11.6% of English-speaking Hispanic women reported engaging in OPA three or more times per week, compared with 27.7% of Spanish-speaking Hispanic women [88]. It is likely that women within the same ethnicity vary in PA engagement by other factors, such as age, education level and socioeconomic status.

Muscle-strengthening physical activity

It is also important to understand how adherence to muscle-strengthening guidelines varies by ethnicity. The 2012 NHIS reported that 30.9% of white women met aerobic guidelines only, compared with 24.5% of black women and 26% of Hispanic women [86]. Fewer women met both muscle-strengthening and aerobic guidelines: 19.9% of white women, 10.8% of black women and 12.2% of Hispanic women [86]. Additionally, women were less likely to engage in both aerobic and muscle-strengthening exercises than men: 19.9 versus 26% in whites, 10.8 versus 23.7% in blacks and 12.2 versus 19.3% in Hispanics, respectively [86]. NHIS data from 1998 through 2004 indicate that muscle-strengthening exercises are becoming more common in women: among white women, 16.2 compared with 20.4% (1998, 2004, respectively); among black women, 9.4 compared with 11.3%; and among Hispanic women, 8.9 compared with 9.1% [95]. Women's adherence to muscle-strengthening exercises also varies by age. In 2004, 20.1% of women aged 18–24 years engaged in muscle-strengthening exercises at least twice a week, compared with only 10.7% of women aged 65 and older [95]. Information regarding muscle-strengthening adherence in Asians, American Indians/Alaskan Natives (AI/AN) and Native Hawaiian/Pacific Islanders (NH/PI) are only available for men and women combined: 18.7% of AI/ANs, 17.1% of Asians and 31% of NH/PIs, compared with 21.4% of whites and 16.8% of blacks, meet both aerobic and muscle-strengthening guidelines [86].

Overall, women of different racial groups vary significantly in their engagement in physical activity. Further differences exist between engagement in LTPA, OPA and TPA. When examining percentages of people engaged in muscle-strengthening exercise compared with aerobic exercise, it is clear that few people engage in muscle-strengthening exercise. There was very little data on women-specific muscle-strengthening exercise, despite the importance of resistance training to metabolic health.

Discussion

Overall, data from body composition and behavioral studies showed patterns consistent with patterns of diabetes prevalence. Women from racial/ethnic groups with higher prevalence of diabetes – namely, black, Hispanic, Pacific Islander and Native American – tended to have higher rates of obesity, with the most notable exception being Asian women, who consistently had lower BMI and WC despite a higher prevalence of diabetes. More specific measures of body composition were less consistent with trends in diabetes prevalence, and there was marked variation in body composition measures between racial/ethnic groups. While black women generally had higher BMIs, WCs and overall body fat, they consistently had the lowest amount of visceral fat. Asian women, conversely, despite small body size, had the highest amounts of visceral fat, followed by Hispanics.

Data on diet were somewhat more consistent, though largely limited to whites, blacks and Hispanics. Though there were some regional differences, generally black women in particular showed more metabolically unhealthy dietary practices compared with white women. Data comparing dietary practices in other racial/ethnic groups was extremely limited. Findings from physical activity studies varied greatly depending on the measures used. Virtually all self-report studies showed greater activity and lower inactivity in white women compared with black and Hispanic women, while accelerometer data showed the opposite. Leisure time physical activity was highest among white women, while Hispanic women reported the most occupational activity. Given the variations in findings for physical activity, it is difficult to comment on how these may contribute to differential rates of diabetes. Overall, however, physical activity was quite low, which could contribute to rising rates of diabetes across racial/ethnic groups.

Of course, commenting on the contribution these variables may make to diabetes disparities in complicated by the fact that they are not independent of one another. Body composition can be directly influenced by both diet and physical activity, beyond simple associations with overall adiposity. Physical fitness has been associated with less VAT [96], and increasing physical activity appears to selectively reduce VAT deposits rather than SAT [97]. Weight loss strategies relying on dietary change, however, do not show this selective reduction [98], though dietary patterns, such as fiber intake, fast food intake, a ‘southern’ diet, sugar-sweetened beverages and overall calorie consumption have all been associated with specific fat depots [99 –101]. These relationships are further complicated by the fact that they may differ by racial/ethnic group. For example, a weight loss study with black and white women showed that, with similar overall weight loss, black women selectively reduced SAT while white women selectively reduced VAT [102].

The physical environment may play a key role in differences in health behavior. A review found evidence (although not always consistent) of positive associations between engagement in physical activity and different characteristics of the built environment, including aesthetics, safety and access to trails, parks and walkable destinations [103]. Moreover, studies have found associations between various traits of the built environment and neighborhoods’ racial composition. Moore and colleagues analyzed North Carolina, New York and Maryland census data and found an increased likelihood of availability of recreational facilities and parks in predominantly white neighborhoods, compared with ethnic minority neighborhoods (i.e., predominantly black and Hispanic neighborhoods) 104].

Likewise, there is scientific evidence regarding racial/ethnic disparities in access to healthy and unhealthy food. Neighborhoods consisting of ethnic minority populations generally have increased concentrations of fast-food outlets and decreased access to healthy food outlets, although these results have not been completely consistent [105 –108]. There is also evidence regarding associations between increased availability of healthy food through neighborhood outlets and increased consumption of healthy food [106,107]. For example, in one study higher availability of dark green and orange vegetables in the neighborhood led to increased intake of these vegetables [109]. Understanding determinants of behavior, such as physical environment, will better equip researchers to address health disparities in diabetes among different racial/ethnic groups.

Conclusion & future perspective

This review revealed important gaps in the literature regarding availability of data beyond the major racial/ethnic groups (i.e., whites, blacks and Hispanics, and in lesser degree Asians and Mexican–Americans). For example, not until 2011–2012 did NHANES include a category for Asians (previous racial/ethnic categories included whites, blacks, Mexican–Americans, other Hispanic and other races). The need for data specific to subgroups is emphasized by the fact that diabetes prevalence itself varies remarkably across subgroups, often showing a larger range within than between racial/ethnic groups. Rates in Alaska Natives/American Indians, for example, range from 6% in Alaska Natives to 24.1% in American Indians in Southern Arizona, and within Asian subgroups prevalence ranges from 4.4% in Chinese to 13% in Asian–Indians [110]. Given these vast differences in diabetes prevalence between subgroups, it is likely that data on risk factors is similarly misleading when generalized across racial/ethnic groups.

Continued investigation of ethnic differences by subgroups and by region is necessary; this type of data reveals important distinctions that are not captured by more general national data, which is limited in its treatment of race/ethnicity. In the nutrition realm, for example, national data identifies black women as having higher odds of most risk factors (e.g., higher consumption of added sugars, red meat, processed meat, etc); however, smaller regional studies might contradict these findings. For example, Chandran and colleagues studied differences between white and black women in New York and New Jersey, and found white women to have higher consumption of red meat [64].

Another large gap revealed in the literature was in muscle and strength training measures. Although muscle is highly active in glucose metabolism, and muscle-strengthening exercise increases insulin sensitivity, both the body composition literature and the physical activity literature lack information on the association between muscle and diabetes. Future research should focus on racial/ ethnic differences in muscle quantity and type, in conjunction with insulin sensitivity and diabetes risk. In the physical activity literature, there is scant data regarding adherence to muscle-strengthening guidelines between women across ethnic/ racial groups. Exercise physiology studies illustrate improvements in insulin sensitivity from muscle-strengthening exercise and combined aerobic/ muscle-strengthening exercise regimens [111]. However, NHIS data report that very few women participate in muscle-strengthening exercises [112]. Promoting strength training, therefore, may be a key future area of health behavior research, particularly for women from racial/ ethnic groups at high risk for diabetes.

Executive summary

Body composition

In addition to BMI, distribution of adipose tissue is associated with risk of diabetes.

Correlations between BMI and more specific body composition measures, such as percent body fat and waist circumference, vary widely across racial/ethnic groups.

Women of different races/ethnicities vary greatly in accumulations of visceral versus subcutaneous adipose tissue. Asian women are most likely to have a high amount of visceral adipose tissue at lower BMI cut points, and therefore have the highest risk of diabetes at the lowest BMIs when comparing across ethnicities.

Diet

Added sugar consumption is correlated with increased risk of diabetes. Black women consumed the highest amount of added sugars, and Hispanic women consumed the highest quantity of sugar-sweetened beverages, when comparing blacks, Hispanics and whites.

Red and processed meat consumption is correlated with increased risk of diabetes, while fruit and vegetable intake is correlated with reduced risk of diabetes. Measures of meat, fruit and vegetable intakes across ethnicity vary by study.

Physical activity

Self-report physical activity data yield that white women engage in the most physical activity, followed by Hispanics and blacks.

Accelerometer data yield that black and Hispanic women engage in the most physical activity, followed by whites.

Less than 50% of women in the USA meet government-recommended physical activity guidelines; even fewer women engage in muscle-strengthening exercises.

Hispanics are most likely to engage in occupational physical activity, while whites are most likely to engage in leisure time physical activity, and Asians are the most likely to engage in transportation-related physical activity.

Future directions

Future research should include more racially/ethnically-specific data beyond typical broad categories (white, black and Hispanic), including data specific to subgroups within racial/ethnic groups and regionally-specific data.

Future interventions should consider environmental determinants of behavior, and should shift a greater focus to muscle and strengthening exercises in order to prevent and manage diabetes in diverse women.

Footnotes

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

References

Papers of special note have been highlighted as: • of interest; •• of considerable interest.

Cowie

Rust

Ford

Full accounting of diabetes and pre-%diabetes in the U.S. population in 1988–1994 and 2005–2006. Diabetes Care 32(2), 287–294 (2009).

Narayan

Boyle

Geiss

Saaddine

Thompson

. Impact of recent increase in incidence on future diabetes burden: U.S., 2005–2050. Diabetes Care 29(9), 2114–2116 (2006).

Narayan

Boyle

Thompson

Sorensen

Williamson

. Lifetime risk for diabetes mellitus in the United States. J. Am. Med. Assoc. 290(14), 1884–1890 (2003).

Economic costs of diabetes in the U.S. In 2012. Diabetes Care 36(4), 1033–1046 (2013).

National Center for Chronic Disease Prevention and Health Promotion. National Diabetes Statstics Report. Promotion Centers for disease control (2014). www.cdc.gov/diabetes.

Venkataraman

Nanda

Baweja

Parikh

Bhatia

. Prevalence of diabetes mellitus and related conditions in Asian–Indians living in the United States. Am. J. Cardiol. 94(7), 977–980 (2004).

Schulz

Bennett

Ravussin

Effects of traditional and western environments on prevalence of Type 2 diabetes in Pima Indians in Mexico and the US. Diabetes Care 29(8), 1866–1871 (2006).

Carey

Walters

Colditz

Body fat distribution and risk of non-%insulin-dependent diabetes mellitus in women. The Nurses' Health Study. Am. J. Epidemiol. 145(7), 614–619 (1997).

10.

Colditz

Willett

Stampfer

Weight as a risk factor for clinical diabetes in women. Am. J. Epidemiol. 132(3), 501–513 (1990).

11.

Flegal

Carroll

Kit

Ogden

. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999–2010. J. Am. Med. Assoc. 307(5), 491–497 (2012).

12.

Wang

Beydoun

. The obesity epidemic in the United States – gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and metaregression analysis. Epidemiol. Rev. 29, 6–28 (2007).

13.

Wang

Thornton

Russell

Burastero

Heymsfield

Pierson

Jr.

Asians have lower body mass index (BMI) but higher percent body fat than do whites: comparisons of anthropometric measurements. Am. J. Clin. Nutr. 60(1), 23–28 (1994).

14.

Deurenberg-Yap

Chew

Deurenberg

. Elevated body fat percentage and cardiovascular risks at low body mass index levels among Singaporean Chinese, Malays and Indians. Obes. Rev. 3(3), 209–215 (2002).

15.

Rush

Freitas

Plank

. Body size body composition and fat distribution: comparative analysis of European, Maori, Pacific Island and Asian Indian adults. Br. J. Nutr. 102(4), 632–641 (2009).

16.

Deurenberg

Yap

van Staveren

. Body mass index and percent body fat: a meta analysis among different ethnic groups. Int. J. Obes. Relat. Metab. Disord. 22(12), 1164–1171 (1998).

17.

This meta-analysis is unique in that it examines data from a wide range of subgroups which are often excluded from research or collapsed into larger categories.

18.

Fernandez

Heo

Heymsfield

Is percentage body fat differentially related to body mass index in Hispanic Americans, African Americans, and European Americans?

Am. J. Clin. Nutr. 77(1), 71–75 (2003).

19.

Kissebah

Peiris

. Biology of regional body fat distribution: relationship to non-%insulin-dependent diabetes mellitus. Diabetes Metab. Rev. 5(2), 83 – 109 (1989).

20.

Cassano

Rosner

Vokonas

Weiss

. Obesity and body fat distribution in relation to the incidence of non-%insulin-dependent diabetes mellitus. A prospective cohort study of men in the normative aging study. Am. J. Epidemiol. 136(12), 1474–1486 (1992).

21.

Casas

Schiller

DeSouza

Seals

. Total and regional body composition across age in healthy Hispanic and white women of similar socioeconomic status. Am. J. Clin. Nutr. 73(1), 13–18 (2001).

22.

Wei

Gaskill

Haffner

Stern

. Waist circumference as the best predictor of noninsulin dependent diabetes mellitus (NIDDM) compared with body mass index, waist/hip ratio and other anthropometric measurements in Mexican–Americans – a 7-year prospective study. Obes. Res. 5(1), 16–23 (1997).

23.

Wardle

Wrightson

Gibson

. Body fat distribution in South Asian women and children. Int. J. Obes. Relat. Metab. Disord. 20(3), 267–271 (1996).

24.

Araneta

Barrett-Connor

. Ethnic differences in visceral adipose tissue and Type 2 diabetes: Filipino, African-American, and white women. Obes. Res. 13(8), 1458–1465 (2005).

25.

Lear

Humphries

Kohli

Birmingham

. The use of BMI and waist circumference as surrogates of body fat differs by ethnicity. Obesity (Silver Spring) 15(11), 2817–2824 (2007).

26.

Camhi

Bray

Bouchard

The relationship of waist circumference and BMI to visceral, subcutaneous, and total body fat: sex and race differences. Obesity (Silver Spring) 19(2), 402–408 (2011).

27.

Hughes-Austin

Larsen

Allison

. Visceral adipose tissue and cardiovascular disease risk. Curr. Cardiovasc. Risk Rep. 2 (7) 95–101 (2013).

28.

Despres

Couillard

Gagnon

Race, visceral adipose tissue, plasma lipids, and lipoprotein lipase activity in men and women: the Health, Risk Factors, Exercise Training, and Genetics (HERITAGE) family study. Arterioscl. Throm. Vas. 20(8), 1932–1938 (2000).

29.

Lovejoy

de la Bretonne

Klemperer

Tulley

. Abdominal fat distribution and metabolic risk factors: effects of race. Metabolism 45(9), 1119–1124 (1996).

30.

Carroll

Chiapa

Rodriquez

Visceral fat, waist circumference, and BMI: impact of race/ethnicity. Obesity (Silver Spring) 16(3), 600–607 (2008).

31.

Perry

Applegate

Jackson

Racial differences in visceral adipose tissue but not anthropometric markers of health-related variables. J. Appl. Physiol. (1985) 89(2), 636–643 (2000).

32.

Bray

Jablonski

Fujimoto

Relation of central adiposity and body mass index to the development of diabetes in the Diabetes Prevention Program. Am. J. Clin. Nutr. 87(5), 1212–1218 (2008).

33.

Raji

Seely

Arky

Simonson

. Body fat distribution and insulin resistance in healthy Asian–Indians and Caucasians. J. Clin. Endocrinol. Metab. 86(11), 5366–5371 (2001).

34.

Park

Allison

Heymsfield

Gallagher

. Larger amounts of visceral adipose tissue in Asian–Americans. Obes. Res. 9(7), 381–387 (2001).

35.

Misra

Wasir

Vikram

. Action and research are needed for evaluation of optimal definitions of anthropometric parameters and metabolic syndrome for Asians. Diabetes Res. Clin. Pract. 68(2), 178–179 (2005).

36.

Shiwaku

Anuurad

Enkhmaa

Kitajima

Yamane

. Appropriate BMI for Asian populations. Lancet 363(9414), 1077 (2004).

37.

Araneta

Kanaya

Hsu

Optimum BMI cut points to screen Asian–Americans for Type 2 diabetes. Diabetes Care 38(5), 814–820 (2015).

38.

This is of considerable interest as it outlines new standards for clinical practice.

39.

Hsu

Araneta

Kanaya

Chiang

Fujimoto

. BMI cut points to identify at-risk Asian–Americans for 925. Type 2 diabetes screening. Diabetes Care 38(1), 150–158 (2015).

40.

Srikanthan

Karlamangla

. Relative muscle mass is inversely associated with insulin resistance and prediabetes. findings from the Third National Health and Nutrition Examination Survey. J. Clin. Endocrinol. Metab. 96(9), 2898–2903 (2011).

41.

Larsen

Allison

Laughlin

The association between abdominal muscle and Type II diabetes across weight categories in diverse post-menopausal women. J. Clin. Endocrinol. Metab. 100(1), E105–E109 (2015).

42.

Shai

Jiang

Manson

Ethnicity, obesity, and risk of Type 2 diabetes in women: a 20-year follow-up study. Diabetes Care 29(7), 1585–1590 (2006).

43.

McKeigue

Shah

Marmot

. Relation of central obesity and insulin resistance with high diabetes prevalence and cardiovascular risk in South Asians. Lancet 337(8738), 382–386 (1991).

44.

Okosun

. Ethnic differences in the risk of Type 2 diabetes attributable to differences in abdominal adiposity in American women. J. Cardiovasc. Risk 7(6), 425–430 (2000).

45.

Malik

. Sweeteners and risk of obesity and Type 2 diabetes: the role of sugar-sweetened beverages. Curr. Diabetes Rep. 12(2), 195–203 (2012).

46.

de Koning

Malik

Rimm

Willett

. Sugar-sweetened and artificially sweetened beverage consumption and risk of Type 2 diabetes in men. Am. J. Clin. Nutr. 93(6), 1321–1327 (2011).

47.

Malik

Popkin

Bray

Despres

J-P

Willett

. Sugar-sweetened beverages and risk of metabolic syndrome and Type 2 diabetes. Diabetes Care 33(11), 2477–2483 (2010).

48.

Palmer

Boggs

Krishnan

Singer

Rosenberg

. Sugar-sweetened beverages and incidence of Type 2 diabetes mellitus in African American women. Arch. Intern. Med. 168(14), 1487–1492 (2008).

49.

Schulze

Manson

Ludwig

Sugar-sweetened beverages, weight gain, and incidence of Type 2 diabetes in young and middle-aged women. J. Am. Med. Assoc. 292(8), 927–934 (2004).

50.

Krishnan

Rosenberg

Singer

Glycemie index, glycemie load, and cereal fiber intake and risk of Type 2 diabetes in US black women. Arch. Intern. Med. 167(21), 2304–2309 (2007).

51.

Ludwig

DDS

. The glycemic index – physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. J. Am. Med. Assoc. 287(18), 2414–2423 (2002).

52.

Palli

InterAct

. Association between dietary meat consumption and incident Type 2 diabetes: the EPIC-InterAct study. Diabetologia 56(1), 47–59 (2013).

53.

Micha

Wallace

Mozaffarian

. Red and processed meat consumption and risk of incident coronary heart disease, stroke, and diabetes mellitus: a systematic review and meta-analysis. Circulation 121(21), 2271–U2252 (2010).

54.

Pan

Sun

Bernstein

Red meat consumption and risk of Type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis. Am. J. Clin. Nutr. 94(4), 1088–1096 (2011).

55.

Steinbrecher

Erber

Grandinetti

Kolonel

Maskarinec

. Meat consumption and risk of Type 2 diabetes: the multiethnic cohort. Public Health Nutr. 14(4), 568–574 (2011).

56.

van Woudenbergh

Kuijsten

Tigcheler

Meat consumption and its association with C-reactive protein and incident Type 2 diabetes: the Rotterdam Study. Diabetes Care 35(7), 1499–1505 (2012).

57.

Riserus

Willett

. Dietary fats and prevention of Type 2 diabetes. Prog. Lipid Res. 48(1), 44–51 (2009).

58.

Carter

Gray

Troughton

Khunti

Davies

. Fruit and vegetable intake and incidence of Type 2 diabetes mellitus: systematic review and meta-analysis. BMJ 341, c4229 (2010).

59.

Fan

Zhang

Hou

Tang

. Fruit and vegetable intake and risk of Type 2 diabetes mellitus: meta-analysis of prospective cohort studies. BMJ Open 4(11), e005497 (2014).

60.

Meyer

Kushi

Jacobs

Slavin

Sellers

Folsom

. Carbohydrates dietary fiber, and incident Type 2 diabetes in older women. Am. J. Clin. Nutr. 71(4), 921–930 (2000).

61.

de Munter

JSL

Spiegelman

Franz

van Dam

. Whole grain bran and germ intake and risk of Type 2 diabetes: a prospective cohort study and systematic review. PLoS Med. 4(8), 1385–1395 (2007).

62.

Kirkpatrick

Dodd

Reedy

Krebs-Smith

. Income and race/ethnicity are associated with adherence to food-based dietary guidance among US adults and children. J. Acad. Nutr. Diet. 112(5), 624–635 (2012).

63.

This article provides important information regarding ethnic differences for various diabetes-related dietary behaviors, including consumption of fruits, vegetables, grains and added sugar.

64.

O'Brien

Davey

Alos

Whitaker

. Diabetes-related behaviors in Latinas and non-Latinas in California. Diabetes Care 36(2), 355–361 (2013).

65.

Kong

Odoms-Young

Schiffer

Racial/ethnic differences in dietary intake among WIC families prior to food package revisions. J. Nutr. Educ. Behav. 45(1), 39–46 (2013).

66.

Storey

Anderson

. Income and race/ethnicity influence dietary fiber intake and vegetable consumption. Nutr. Res. 34(10), 844–850 (2014).

67.

While there is substantial evidence regarding the benefits of dietary fiber intake, this article is one of the few to thoroughly analyze age, gender, income and racial/ethnic differences in the consumption of dietary fiber.

68.

Drewnowski

Rehm

. Socioeconomic gradient in consumption of whole fruit and 100% fruit juice among US children and adults. Nutr. J. 14, 3 (2015).

69.

Wang

Beydoun

Caballero

Gary

Lawrence

. Trends and correlates in meat consumption patterns in the US adult population. Public Health Nutr. 13(9), 1333–1345 (2010).

70.

Chandran

Zirpoli

Ciupak

Racial disparities in red meat and poultry intake and breast cancer risk. Cancer Causes Control 24(12), 2217–2229 (2013).

71.

Pereira

Kartashov

Ebbeling

Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet 365(9453), 36–42 (2005).

72.

Drewnowski

Rehm

. Consumption of added sugars among US children and adults by food purchase location and food source. Am. J. Clin. Nutr. 100(3), 901–907 (2014).

73.

Ogden

Kit

Carroll

Park

. Consumption of sugar drinks in the United States, 2005–2008. NCHS Data Brief (71), 1–8 (2011).

74.

Park

Onufrak

Blanck

Sherry

. Characteristics associated with consumption of sports and energy drinks among US Adults: National Health Interview Survey, 2010. Am. J. Clin. Nutr. 113(1), 112–119 (2013).

75.

Park

Pan

Sherry

Blanck

. Consumption of sugar-sweetened beverages among US adults in 6 states: Behavioral Risk Factor Surveillance System, 2011. Prev. Chronic Dis. 11, E65–E65 (2014).

76.

Kim

Chan

. Acculturation and dietary habits of Korean Americans. Brit. J. Nutr. 91(3), 469–478 (2004).

77.

Wannamethee

Thomas

Whincup

Sattar

. Associations between dietary fiber and inflammation hepatic function and risk of Type 2 diabetes in older men: potential mechanisms for the benefits of fiber on diabetes risk. Diabetes Care 32(10), 1823–1825 (2009).

78.

Grooms

Ommerborn

Pham

Djousse

Clark

. Dietary fiber intake and cardiometabolic risks among US adults, NHANES 1999–2010. Am. J. Med. 126(12), 1059–1067 (2013).

79.

King

Mainous

III Lambourne

. Trends in dietary fiber intake in the United States, 1999–2008. J. Acad. Nutr. Diet. 112(5), 642–648 (2012).

80.

Bowman

Vinyard

. Fast food consumption of US adults: impact on energy and nutrient intakes and overweight status. J. Am. Coll. Nutr. 23(2), 163–168 (2004).

81.

Guthrie

Lin

Frazao

. Role of food prepared away from home in the American diet, 1977–78 versus 1994–96: changes and consequences. J. Nutr. Educ. Behav. 34(3), 140–150 (2002).

82.

Paeratakul

Ferdinand

Champagne

Ryan

Bray

. Fast-food consumption among US adults and children: dietary and nutrient intake profile. J. Am. Diet. Assoc. 103(10), 1332–1338 (2003).

83.

Newby

Noel

Grant

Judd

Shikany

Ard

. Race and region are associated with nutrient intakes among black and white men in the United States. J. Nutr. 141(2), 296–303 (2011).

84.

Krishnan

Rosenberg

Palmer

. Physical activity and television watching in relation to risk of Type 2 diabetes: the black women's health study. Am. J. Epidemiol. 169(4), 428–434 (2009).

85.

Paynter

LaMonte

Manson

Comparison of lifestyle-based and traditional cardiovascular disease prediction in a multiethnic cohort of nonsmoking women. Circulation 130, 1466–1473 (2014).

86.

Bao

Tobias

Bowers

. Physical activity and sedentary behaviors associated with risk of progression from gestational diabetes mellitus to Type 2 diabetes mellitus: a prospective cohort study. JAMA Intern. Med. 174(7), 1047–1055 (2014).

87.

Kang

Woo

Shin

Circuit resistance exercise improves glycemic control and adipokines in females with Type 2 diabetes mellitus. J. Sport. Sci. Med. 8, 682–688 (2009).

88.

2008 Physical Activity Guidelines for Americans Summary. Office of Disease Prevention and Health Promotion. www.health.gov.

89.

Mozaffarian

Roger

Heart disease and stroke statistics–2013 update: a report from the American Heart Association. Circulation 127, 42–43 (2013).

90.

Carlson

Densmore

Fulton

Your

Kore

. Differences in physical activity prevalence and trends from 3 US surveillance systems: NHIS, NHANES, and BRFSS. J. Phys. Act. Health 6(S1), S18–S27 (2009).

91.

Blackwell

Lucas

Clarke

. Summary health statistics for U.S. adults: National Health Interview Survey, 2012. Vital Health Stat. 10. (260), 1–161 (2014).

92.

Macera

Ham

Yore

Prevalence of physical activity in the United States: Behavioral Risk Factor Surveillance System, 2001. Prev. Chronic Dis. 2(2), A17 (2005).

93.

Marshall

Jones

Ainsworth

Reis

Levy

Macera

. Race/ethnicity social class, and leisure-time physical inactivity. Med. Sci. Sports Exerc. 39(1), 44–51 (2007).

94.

Troiano

Berrigan

Dodd

Masse

Tilert

Mcdowell

. Physical activity in the United States measured by accelerometer. Med. Sci. Sports Exerc. 40(1), 181–188 (2008).

95.

This article explains important discrepancies in analysis techniques of accelerometer data, which influence the number of people meeting physical activity guidelines.

96.

Pleis

Lethbridge-Cejku

. Summary health statistics for U.S. adults: National Health Interview Survey, 2006. Vital Health Stat. (235), 1–153 (2007).

97.

Baker

. Differences in leisure-time, household, and work-related physical activity by race, ethnicity, and education. J. Gen. Intern. Med. 20, 259–266 (2005).

98.

Neighbors

Marquez

Marcus

. Leisure-time physical activity disparities among Hispanic subgroups in the United States. Am. J. Public Health 98 (8), 1460–1464 (2008).

99.

Kandula

Lauderdale

. Leisure time non-leisure time and occupational physical activity in Asian–Americans. Ann. Epidemiol. 15, 257–265 (2005).

100.

This article highlights physical activity differences between Asian ethnic subgroups, and also compares Asian women to non-Asian women. By describing the unique habits of specific subgroups, this article illustrates the limitation of generalizing physical activity habits by ethnicity.

101.

Hirsch

Moore

Clarke

Changes in the built environment and changes in the amount of walking over time: longitudinal results from the multi-%ethnic study of atherosclerosis. Am. J. Epidemiol. 180(8), 799–809 (2014).

102.

This article explains ethnic differences in leisure-time versus transportation-related physical activity in response to the built environment.

103.

Centers for Disease Control and Prevention (CDC). Trends in strength training – United States, 1998–2004. MMWR Morb. Mortal. Wkly Rep. 769–772 (2006).

104.

Brock

Irving

Gower

Hunter

. Differences emerge in visceral adipose tissue accumulation after selection for innate cardiovascular fitness. Int. J. Obes. (Lond.), 35(2), 309–312 (2011).

105.

Kay

Singh

Fiatarone MA

. The influence of physical activity on abdominal fat: a systematic review of the literature. Obes. Rev. 7(2), 183–200 (2006).

106.

Murphy

McDaniel

Mora

Villareal

Fontana

Weiss

. Preferential reductions in intermuscular and visceral adipose tissue with exercise-induced weight loss compared with calorie restriction. J. Appl. Physiol. 112(1), 79–85 (2012).

107.

Kazlauskaite

Karavolos

Janssen

The association between self-reported energy intake and intra-%abdominal adipose tissue in perimenopausal women. J. Obes. 2012, 567320 (2012).

108.

Odegaard

Choh

Czerwinski

Towne

Demerath

. Sugar-sweetened and diet beverages in relation to visceral adipose tissue. Obesity (Silver Spring) 20(3), 689–691 (2012).

109.

Liu

Hickson

Musani

Dietary patterns, abdominal visceral adipose tissue and cardiometabolic risk factors in African Americans: the Jackson Heart Study. Obesity (Silver Spring) doi: 10.1002/oby.20265 (2012).

110.

Weinsier

Hunter

Gower

Schutz

Darnell

Zuckerman

. Body fat distribution in white and black women: different patterns of intraabdominal and subcutaneous abdominal adipose tissue utilization with weight loss. Am. J. Clin. Nutr. 74(5), 631–636 (2001).

111.

Frost

Goins

Hunter

Effects of the built environment on physical activity of adults living in rural settings. Am. J. Health Promot. 24(4), 267–283 (2010).

112.

Moore

Roux

AVD

Evenson

McGinn

Brines

. Availability of recreational resources in minority and low socioeconomic status areas. Am. J. Prev. Med. 34(1), 16–22 (2008).

113.

Block

Scribner

DeSalvo

. Fast food race/ethnicity and income – a geographic analysis. Am. J. Prev. Med. 27(3), 211–217 (2004).

114.

Ford

Dzewaltowski

. Disparities in obesity prevalence due to variation in the retail food environment: three testable hypotheses. Nutr. Rev. 66(4), 216–228 (2008).

115.

Larson

Eisenberg

Berge

Arcan

Neumark-Sztainer

. Ethnic/racial disparities in adolescents' home food environments and linkages to dietary intake and weight status. Eat. Behav. 16, 43–46 (2015).

116.

Walker

Keane

Burke

. Disparities and access to healthy food in the United States: a review of food deserts literature. Health Place 16(5), 876–884 (2010).

117.

Izumi

Zenk

Schulz

Mentz

Wilson

. Associations between neighborhood availability and individual consumption of dark-green and orange vegetables among ethnically diverse adults in Detroit. J. Am. Diet. Assoc. 111(2), 274–279 (2011).

118.

National Center for Chronic Disease Prevention and Health Promotion. National Diabetes Statistics Report, 2014. Centers for Disease Control and Prevention (2014). www.cdc.gov.

119.

Fenkci

Sarsan

Rota

Ardic

. Effects of resistance or aerobic exercises on metabolic parameters in obese women who are not on a diet. Adv. Ther. 23(3), 404–413 (2006).

120.

Report

. Centers for Disease Control & Prevention (CDC) Weekly in Strength in – United States. 55(28), 769–772 (2015). www.cdc.gov.