Fig. 11.1
Global overweight and obesity prevalence by WHO region
After decades of increase, reports from several western countries suggest that the prevalence of obesity in adult and pediatric populations may be leveling off or beginning to decline. In contrast, in developing countries, particularly those undergoing modernization and urbanization, trend data suggest that obesity prevalence is on the rise. A 2012 publication from the global Organization for Economic and Cooperative Development (OECD) reports that obesity rates have stabilized in Korea at 3–4 %, in Switzerland at 7–8 %, in Italy at 8–9 %, in Hungary at 17–18 %, and in England at 22–23 % [25]. Analyses that compared US NHANES data from 1999–2000 to data from 2009–2010 yielded similar findings: obesity prevalence in American men and women did not change significantly between 2003 and 2010 [20], whereas data from the China Health and Nutrition Survey indicate that among Chinese adults, the prevalence of obesity is still increasing, with the prevalence of general obesity (BMI > 27.5) increasing from 2.9 to 11.4 % among men and from 5.0 to 10.1 % among women [26]. A similar pattern of obesity trends in developed and developing countries is also observed in children. US Pediatric Nutrition Surveillance System data, derived from surveys that track health and health-related outcomes in 2- to 5-year-old low-income children, show that the prevalence of obesity in children decreased in 18 states and in the US Virgin Islands from 2008 to 2011 [27]. In China, by contrast, the standardized prevalence of obesity increased in children of all ages between 2005 and 2008 [28]. Similarly, a study conducted in 24,000 school-age children in India, showed that the prevalence of overweight children increased from under 4.9 % in 2003 to 6.6 % in 2005 [29]. Whereas data from the West may be encouraging, obesity remains a significant problem worldwide. In Western countries, the current obesity prevalence is high, and obesity is increasing rapidly in both adults and children in developing nations.
The Demographics of Obesity
Like the differences in obesity trends between the developed and developing world, the demographics of overweight and obesity also vary globally. Differences in obesity prevalence by sex, socioeconomic status (SES), and rural versus urban living reflect the complexities of the descriptive epidemiology of obesity. Male–female differences in obesity prevalence are not a recent development: in all WHO regions women are more likely to be obese than men. Estimates by WHO region show that in 23 % of European women, 24 % of Eastern Mediterranean women and 29 % of women in the Americas regions are obese [2]. In Africa, prevalence studies conducted across the continent during the 2000s also show that obesity rates were higher in women than in men [30]. Recent evidence, however, suggests that in many African and Middle Eastern nations these gender differences in obesity prevalence may reflect a societal preference for female overweight rather than normal weight [31].
Additional new evidence supports the inter-dependence of sociodemographic factors on obesity risk. SES is typically operationalized by educational attainment or income in population studies. The effect of SES on obesity appears to differ by gender, and the effect of living in a rural versus urban environment on obesity differs by SES. A 2013 review of the role of education on obesity prevalence concluded that an inverse association between education and obesity is more common in higher-income, developed countries, whereas a positive association is more common in lower-income and developing countries, with observed differences in these relationships by gender [32]. Further support for these complex inter-relationships comes from a comparison of data from the UK and China. The Health Survey for England 2006–2010 found obesity rates over 30 % in both men and women with no qualification (secondary education) and lowest in men and women with graduate degrees [33]. In China, the relationship between educational attainment and obesity differs between women and men. Specifically, in Chinese women, lower SES is associated with a greater likelihood of obesity [34], while in Chinese men, income and obesity are positively linked [35]. Effects of rural versus urban settings on obesity risk are further influenced by socioeconomic indicators. Obesity rates are highest in rural and lower-income regions in Western countries. In the United States, for example, 2012 estimates from the CDC documented five states with obesity rates greater than 30 %: Louisiana, Mississippi, West Virginia, Alabama, and Michigan [36]; these states are rural and/or relatively poor. Furthermore, 26 of the 30 states in the United States with the highest obesity prevalence are in the South and the Midwest, which are traditionally more rural and have lower average incomes than Northeastern and Western states [37]. In developing countries, opposite patterns are evident. The higher obesity rates in urban areas compared to rural ones are generally attributed to dietary and lifestyle changes that accompany urbanization [38]. A recent review of studies conducted in Africa found that obesity prevalence was higher in urban versus rural populations [30]. The observed pattern appears to be attributable to the effects of modernization, including sedentary behavior and shift to a Western diet, behaviors manifested by more affluent Africans [30]. Populations worldwide are vulnerable to obesity regardless of age, gender, educational attainment, income level, or urbanicity. Clearly, globalization will continue to influence existing patterns, with region- and country-specific responses.
Determinants of Obesity
Obesity is a unique condition in that it can be viewed as an exposure or as a risk factor for comorbidities, as well as a health outcome itself. When considered as an outcome, understanding the exposures or behaviors that contribute to its development is imperative, particularly in the context of treating the obese patient. In previous and subsequent chapters of this book, many of these exposures, which were identified through epidemiologic investigation, are reviewed in detail. Specifically, Chap. 4 reviews the relationship between the perinatal period and obesity, Chap. 5 explores the emerging body of work on the role of the gut microbiome, and finally in Chap. 8, the authors examine the relationship between sleep and obesity. To avoid duplication, this chapter reviews recent developments in the genetics of obesity and more proximal factors and behaviors including, diet, physical activity, sedentary time, and stress.
Genetics
Understanding the genetics of metabolism and excess weight gain is important, particularly for the continued development of pharmacological and behavioral interventions and the movement toward personalized medicine. Over the last decade, large scale genome-wide association studies (GWAS) have identified a number of genes related to excess weight gain [37] and that number continues to grow. In 2013, a mutation in the MRAP2 gene, a gene involved in the signaling of melanocortin-4 receptors, was identified as being associated with severe, early onset of obesity in mice and humans [38]. The melanocortin-4 receptor plays a key role in hypothalamic control of appetite—individuals with this mutation are not able to identify satiety effectively. A second recently identified obesity-related gene is FTO, which affects ghrelin levels, a hunger hormone. In a study of 359 healthy, normal weight men, those with a mutation in their FTO gene (TT) had higher postprandial levels of ghrelin than those with the low-risk FTO genotype (AA) [39]. Accordingly, those with the homozygous recessive mutation continued to feel hungry after eating a meal. GWAS studies contribute importantly to obesity research in their identification of genetic variants that may explain the some of the heritability of obesity.
Given that the human genome has not changed substantially over the last three decades, the causes of the obesity epidemic cannot be purely genetic. Instead, it is more likely that the obesity epidemic reflects the intrinsic interplay of an individual’s genetics with environmental exposures. The field of epigenetics examines how developmental and environmental cues affect the expression of genes into various phenotypes [39]. The specifics of the mechanisms behind the alteration of gene expression are beyond the scope of this chapter; however, briefly, gene expression is affected via DNA methylation, histone-tail acetylation, poly-ADP-ribosylation, and ATP dependent chromatin remodeling processes, all of which can be attributed to specific environmental exposures. Because maternal or perinatal lifestyle choices may alter developmental programming of the fetus [40], epigenetic investigation in obesity focuses on in utero and early life exposures. The marked environmental shifts that accompany global modernization influence behavior at the individual, and in some instances, the epigenetic level. For example, the adoption of a Western diet and more a sedentary lifestyle appears to elicit changes in gene expression. The new wave of epigenetic studies informs our initial understanding of the interactions among genetics, biology, and environment in excess weight gain and obesity development.
Stress
Recent epidemiologic investigations indicate that stress may affect long-term obesity risk. Evidence supports that both physiological and psychological “chronic stress” contribute to the development of adiposity. Specifically, long-term exposure to physiologic stress mediators, such as cortisol, can induce chronic low-grade inflammation, which is associated with obesity and has been implicated in the pathogenesis of many health conditions (including type 2 diabetes, fatty liver disease, heart disease, metabolic syndrome). Psychological stress can lead to under- or over-eating, and several studies have shown that cortisol can stimulate appetite and disregulate the balance between hunger and satiety [41, 42]. In a recent population-based study of Canadians, self-perceived lifetime stress was related to obesity. As compared to individuals reporting they were not at all stressed, those who reported being extremely stressed had an increased risk of obesity (adjusted OR = 1.23, 95 % CI 1.13, 1.35). When stratified by gender, this effect was significant only among women [43]. In a study of 822 adults 18–83 years of age, those with the highest level of emotion- and stress-related eating were 13 times more likely to be overweight or obese, compared with those in the highest quartiles [44]. Perceived stress has been shown to modify the association between sleep quality and obesity in women [45].
Work-related stress, parenting stress, and posttraumatic stress disorder have also been associated with weight status in population-based studies. In a longitudinal study of perceived stress and weight gain in adolescence, Van Jaarsveld et al. found that although persistent stress was associated with higher waist circumference and BMI in adolescence, higher stress over the 5-year period was not prospectively associated with greater weight gain [46]. A meta-analysis of the relationship between stress and adiposity in longitudinal studies found that stress was associated with increasing adiposity overall. Moreover, stronger associations were observed between stress and adiposity in men compared with women, and in better quality studies with longer rather than shorter follow-up. The authors concluded that psychosocial stress is a risk factor for weight gain, but that the magnitude of the observed effects is very small [47].
Diet
At the most basic level, excess weight gain results from an imbalance between energy consumed and energy expended; therefore, dietary intake continues to be a focus of epidemiologic studies of obesity. Despite the simplicity of the energy balance equation, the role of diet in obesity is complex due to the unique characteristics of obesity as an outcome and diet as an exposure. First, obesity often develops through small weight gains over several years or decades. The fact that obesity has a variable latency period and affects people of all ages makes the entire life-course relevant to study. Further, diet is not a single exposure, but a complex set of correlated nutrients and foods which together comprise food groups and, with intake behaviors, manifest as overall dietary patterns [48]. Finally, diet is notoriously difficult to measure in population studies because dietary factors occur together, exposures are continuous, dietary behaviors shift over time, and dietary assessment methods are imperfect [49]. In light of these challenges, this review prioritizes recent evidence from prospective studies and those that consider dietary exposures using complementary approaches.
Recent research supports the notion that, independent of overall dietary patterns and behaviors, the consumption of single foods contributes to excess weight gain in adults. Specifically, longitudinal analyses combining data from three large prospective cohort studies of US health professionals found that 4-year weight gain was positively associated with consumption of potato chips (1.69 lb), potatoes (1.28 lb), sugar-sweetened beverages (SSB) (1.0 lb), unprocessed red meat (0.95 lb), and processed meats (0.93 lb). Conversely, intake of nutrient-rich, less energy-dense foods, including vegetables (−0.22 lb), whole grains (−0.37 lb), fruit (−0.49 lb), nuts (−0.57 lb), and yogurt (−0.82 lb) are associated with lower 4-year weight gain [50]. Of these single foods, the positive association between sugar-sweetened beverages and excess weight gain in adults has been replicated in prospective studies in Europe and the United States. Specifically, studies suggest that consuming as few as one SSB per day is associated with increased risk for obesity [51–53]. Researchers attribute this association to the inability of liquid foods to affect hunger and satiety cues in the same way that solid foods do, despite their high calorie and sugar content. The focus on the role of SSB in excess weight gain has fueled subsequent investigations into the role of sugar in obesity and other chronic diseases. Whereas some researchers attribute the entire obesity epidemic to sugar intake [54], a recent meta-analysis using data from trials of individuals eating ad libitum diets suggests that sugar intake is responsible for a gain of only 0.75 kg (95 % CI 0.30, 1.19) over the intervention period [55]. The evidence for the role of individual foods in excess weight gain and obesity development is compelling, particularly for SSB and added sugar.
Given the complexity of diet, epidemiological investigations into the role of diet in obesity development often utilize integrated measures of dietary exposures, such as dietary patterns and dietary quality indices. Dietary patterns analysis examines nutrient, food, and food group intake comprehensively to better understand how overall intake patterns affect disease risk. The Mediterranean diet pattern, which is characterized by high consumption of olive oil, whole grain cereals, legumes, fruits, vegetables and fish, moderate consumption of dairy and wine, and low consumption of meat and meat products, has been associated with lower odds of obesity development in prospective studies. In the European Prospective Investigation into Cancer and Nutrition-Physical Activity, Nutrition, Alcohol Consumption, Cessation from Smoking, Eating out of the Home, and Obesity (EPIC-PANACEA) project, men and women who closely followed the Mediterranean diet pattern were 10 % (95 % CI 4 %,18 %) less likely to become overweight or obese over 5 years of follow-up compared to those with low adherence to the dietary pattern [56]. In a 16-year follow-up of normal weight women participating in the Framingham Offspring and Spouse Study, Wolongevicz et al. found that women in the lowest tertile of diet quality, those with the lowest intakes of fiber and micronutrients and the highest intakes of alcohol and total, saturated and monounsaturated fats, were nearly two times more likely to become overweight or obese during follow up as compared to those in the highest tertile of diet quality (OR = 1.75,95 % CI 1.16, 2.39) [57]. Together with the findings on consumption of single food items and obesity risk, these results confirm that excess weight gain and obesity risk are associated with consistent consumption of nutrient-poor, energy-dense foods.
With growing appreciation of the central role that changing dietary behaviors play in obesity development, new dietary exposures, such as eating outside of the home, portion sizes, meal and snack patterns, and the timing of energy intake have received increasing attention [58]. Trend data from nationally representative surveys in the United States suggest that since the late 1970s, the average portion size per meal or snack has increased by more than 65 g. Whereas energy density has remained fairly constant, eating frequency, or the number of eating occasions consumed per day, has increased from 3.8 eating occasions per day in 1977–1978 to 4.9 eating occasions per day in 2005–2006 [59]. Although ecologic, the concurrent trends between these dietary patterns and the obesity epidemic are suggestive. In a prospective study designed to evaluate the association between eating away from home and the risk of weight gain in a cohort of young Mediterranean adults, those individuals who ate outside the home two or more times per week had higher adjusted weight gain as compared to those who ate out less frequently [51]. Given that restaurant portion sizes are typically larger than those served at home, it is not surprising that portion size is also positively associated with energy intake and weight gain. For example, in a randomized controlled trial, researchers found that mean energy intake over 4 days was significantly higher when participants were given “larger” compared to “standard” portion sizes (59.1 (SD 6.6) versus 52.2 (SD 14.3) MJ) [60]. Finally, meal and snack patterns along with the timing of energy intake during the day have both been related to weight gain. Although considerably more research has been conducted in children than adults, breakfast consumption is independently associated with lower body weight in adults [61]. Prospective dietary analyses from a 10-year follow up of the Health Professionals Follow-Up Study suggest that breakfast consumption is associated with reduced risk of 5-kg weight gain, independent of lifestyle factors and baseline BMI (hazards ratio = 0.87, 95 % CI 0.82, 0.93) [62]. In a second study conducted in the UK, consuming breakfast was associated with lower 3-year weight gain (adjusted β-coefficient = −0.021 kg, 95 % CI −0.035, −0.007) [63]. The timing of energy intake appears to drive these associations; in a study of American men and women, Wang et al. found those who consumed more than one-third of their daily energy intake in the morning were less likely to be overweight or obese compared to those who did not (OR = 0.34, 95 % CI 0.12, 0.95) [64].
The dietary risk factors for childhood obesity are very similar to those for adult obesity. A recent evaluation of 2007–2008 NHANES dietary data found that the top energy sources for American children ages 2–18 years include desserts, pizza, and soda, with almost 40 % of total energy consumed by 2- to 18-year olds as empty calories [65]. Further, increased consumption of sugar-sweetened beverages (SSB) [66–70], and eating more meals away from home [67, 71], and increased portion sizes [72, 73] are all established risk factors for childhood obesity.
Physical Activity
Physical activity represents the modifiable aspect of the energy expenditure side of the energy balance equation, and can be conceptually divided into occupational and leisure-time physical activity. For most adults, time spent in leisure-time activity accounts for a fairly small portion of any given day (especially work-days). Therefore, occupational activity may be a key factor in total caloric expenditure among adults. Church et al. examined energy expenditure for various occupations in the United States (private industry) using data from the US Bureau of Labor Statistics, in relation to body weight taken from NHANES [74]. They found that in the early 1960s, moderate intensity physical activity was required for nearly half of the jobs in private industry; in the 2010s, that number is approximately 20 %. Decreases in manufacturing, agriculture, mining, and logging occupations, and increases in professional services and leisure/hospitality jobs (which require more sitting) account for a large part of this decrease. Church et al. concluded that the reduction in occupational energy expenditure, which they estimate is in excess 100 cal, accounts for a significant portion of the increase in mean US body weight for women and men over the last 5 decades [74]. Similar decreases in occupational sitting have been observed in other developed countries [75].
In experimental studies in adults, prolonged sitting reduces insulin sensitivity and increases plasma glucose levels [76, 77]. Similar results showing the negative impact of prolonged sitting on cardio metabolic risk factors have been observed in cross-sectional studies [78]. In prospective cohort studies, all-cause mortality is higher among adults who do more sitting, after adjustment for physical activity. Statistical modeling of NHANES data projects that American adults could add an extra two years to their lifespan by reducing their daily sitting time to less than 3 h [79]. A recent review by Bauman et al. noted that more longitudinal data in diverse populations are needed to support a causal assertion that “not sitting” prevents weight gain [80]. Longitudinal data from the Helsinki Health Study showed that working conditions were largely unrelated to weight gain over a 5-to 7-year follow-up period [81].
Breaks in prolonged sitting have been shown to attenuate its negative metabolic, and work place interventions are effective at reducing sitting time when special devices are installed at employee work stations [82, 83]. The “Take-A-Stand” project reduced sitting time by 66 min per day [83]. Whether such reductions in sitting time translate into decreases in energy expenditure is unknown [84].
The Built Environment
Built environment is a newer construct that refers to the physical characteristics of places designed and built by humans, including the availability and safety of sidewalks, parks, trails, and public transportation in cities and neighborhoods. In addition to the individual-level factors that influence weight status, there is growing appreciation that the characteristics of this built environment may impact levels of obesity in a community by promoting or inhibiting physical activity, and by increasing energy intake via proximity to different types of food purveyors and eating establishments.
Several recent reviews have synthesized the extant research on the relationship between the built environment and physical activity and obesity [85–88]. The emergent built environment characteristics include street connectivity and density, land-use mix, and walkability. Availability and proximity to recreation facilities have been correlated with greater physical activity levels in several studies of children and adults [89]. For example, a recent study of over 300 children residing in East Harlem, New York found that the presence of at least one playground on a child’s block increased the odds of unscheduled outdoor physical activity about twofold (OR = 1.95, 95 % CI 1.1–3.4) and that the presence of an after-school program on a child’s block was strongly associated with increased hours of scheduled physical activity (OR = 3.25, 95 % CI 1.3–8.1) [90]. Safety from crime represents another key factor that has been positively associated with physical activity, especially in minority populations [89].
Less conclusive information is available about the link between built environment characteristics and weight status, particularly from longitudinal studies. Epstein et al. assessed whether neighborhood characteristics moderated the relationship between participation in one of four RCT’s for obesity treatment and weight loss in children 8–12 years old [91]. Greater reductions in BMI z-score were associated with more parkland and fewer convenience stores and supermarkets in all of the treatment programs. In one recent longitudinal cohort study of children, Wolch et al. found that the proximity of park acres and recreation programs was significantly and inversely related to attained BMI at age 18 [92]. Some longitudinal studies of adults have found associations between built environment characteristics [93, 94] and weight status, while others have observed no significant association [95]. In the review by Ferdinand et al., studies in which PA was measured objectively were less likely to find a beneficial relationship and the use of a direct measure of body weight was associated with a reduced likelihood of finding a beneficial relationship [87].
The spatial layout, density, and types of food establishments present in a community represent additional components of the built environment that may affect weight status on the energy intake side of the equation. Disparities in the prevalence of obesity among persons of lower SES and black race or Hispanic ethnicity might reflect exposure to different, and potentially obesogenic, environments. National zip-code level data have shown that poorer neighborhoods have less access to large chain supermarkets, but more access to small grocery and convenience stores [96], where the quality of produce is typically be more expensive and of lower quality. Living near a convenience store has been associated with a slightly higher prevalence of overweight and obesity (obesity prevalence ratios [PR] = 1.16, 95 % CI 1.05–1.27; overweight PR = 1.06, 95 % CI 1.02–1.10), whereas proximity to a supermarket has been associated with less overweight and obesity (obesity prevalence ratio [PR] = 0.83, 95 % CI 0.75–0.92; overweight PR = 0.94, 95 % CI 0.90–0.98) [97].
Although fast-food availability has been linked to fast-food consumption, and consumption in turn has been linked to weight status, the question of whether availability is related to weight status is methodologically difficult to assess. In predominantly white, rural samples no association between fast-food availability and weight status has been observed [98]. However, in a sample of non-white rural residents greater availability of fast-food was associated with the number of meals consumed and overall weight status [99].
Many new techniques for assessing characteristics of the built environment, such as omnidirectional imagery and the Microscale Audit of Pedestrian Streetscapes (MAPS) tool, have become available in the past 5 years [100] and show promise for improving the accuracy, reliability, and consistency of built environment measures. A growing interest in this area and appreciation of related methodological issues in spatial epidemiology, such as inconsistencies in the definition of place and how it is measured, and objective versus perceived measures of the built environment, are likely to spawn improvements in our approaches to linking built environment to chronic disease health outcomes like obesity.
Screen Time
An increase in time spent using electronic screen media is regarded as a key factor in the decline of physical activity levels, especially the amount of free and outdoor play among children. Video games are played in a large percentage of American households and are a popular leisure-time activity choice across all age groups [101]. Screen-time activities are increasingly done simultaneously with other sedentary and non-sedentary activities, with implications for public health messaging as well as measurement challenges.
Relatively new to the range of options are traditional video game systems that incorporate partial or whole-body physical activity (“active video gaming” or “exergaming”). According to a recent report on the use of media among children ages 8–18 years, over an hour per day is spent playing video games; 64 % of the respondents reported having ever played active video games such as WiiPlay/WiiSports [102]. Working within this new reality of an increasingly electronic world, the potential for these active video game systems to reduce the amount of time spent in sedentary behavior and increase energy expenditure is of particular interest.
In a recent systematic review which included 52 articles on active video gaming, LeBlanc et al. summarized the current state of knowledge about the potential for active video games (AVG’s) to impact the physical activity levels and overall health of children and youth [103]. Recent studies of AVG use have focused on several major outcomes: appeal, adherence, energy expenditure, body composition, energy intake, and use in special populations. There is evidence from cross-sectional and intervention studies that children using AVG’s increase energy expenditure both above rest and above levels that would be observed during passive video game use; however, they do not consistently result in physical activity levels that meet the current recommendation of 60 min of moderate-to-vigorous physical activity (MVPA), and they may not increase energy expenditure to the levels observed for traditional physical activities [103–105]. Of the 28 laboratory studies included in a systematic review, 12 found that the AVG’s assessed were equivalent to light-to-moderate physical activity, for both children and adults [104]. A study that compared traditional and two different AVG’s found that energy expenditure could be increased up to 2.9 kcal/min, or 172 kcal/h [106], roughly equivalent to an hour of heavy housework, doubles tennis, or brisk walking [107].
In overweight children, the use of AVG’s may attenuate weight gain. In a recent randomized controlled trial conducted in overweight/obese children, Maddison et al. found that compared to the control group, the AVG condition resulted in a small but statistically significant difference in BMI at the end of the 6-month trial [108]; however, no differences in levels of physical activity measured by accelerometry or by VO2max were observed. When asked to rate their perceived exertion while playing AVG’s, both children and adults rate it as similar to activities with lower intensities, suggesting that their engagement with the game may distract them from the physical exertion involved, especially at moderate intensities [104, 105]. Warburton argues that even this light-to-moderate intensity activity attained during AVG use has health benefits, and that comparison to the 60-min MVPA guideline is too narrow. In their review, LeBlanc et al. note that many studies of AVG use have small samples and are underpowered, and that future research should be designed with longer follow-up periods and should include both direct (accelerometry, heart rate) and indirect (self- and parent-report) outcomes [103]. Additionally, the heterogeneity in the type of AVG platforms, which focus on different types of body movements, limits comparison across studies, and the long-term effectiveness of AVG’s in non-structured and self-regulated settings remains largely unknown [104].
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
