Body weight in the United States has increased dramatically since 1980 (Figure 1). In epidemiological studies the body mass index (BMI) calculated as weight in kilograms divided by height in meters squared is used to classify body weight. Body weight is considered to be healthy when the BMI is 18.5-24.9 kg/m2. An individual is classified overweight with a BMI of 25-29.9 kg/m2 and obese with a BMI >30 kg/m2. The National Health and Nutrition Examination Survey (NHANES) has been collating data for BMI in the United States since 1960. The first report (NHANES I) found that 44.9% of adults aged 20-74 years were overweight or obese (1). Data from the last available report (NHANES 2003-2004) indicate a dramatic increase in the prevalence rates of overweight and obesity in adults (Figure 1). A staggering 66.3% of the population studied (n=4431) was overweight or obese (2). While the prevalence of overweight increased by only ~2% in forty years, the prevalence of obesity almost tripled from 13.3% in1960-62 to 32.9% in 2003-04 (1). Interestingly the prevalence of obesity continued to rise in men while the prevalence rates for women had leveled off (2). Certainly the most alarming finding surrounds body weight gain in children and adolescents. Among children aged 2-19 years, the prevalence of overweight (defined by weight for height above the 95th percentile for age) significantly increased from 5% in 1971-74 to ~17% in 2003-04 (1).
The secular rise in overweight and obesity can be explained by physiological and behavioral factors as well as changes in social and environmental cues (Figure 2). In the coming sections we will describe the physiological regulators of energy balance and hence body weight, as well as discussing the growing list of putative environmental and behavioral changes over the last 40 years that will most likely play a role in the epidemic of obesity.
The balance between energy intake (calories consumed) and energy expenditure (calories burned) determines body energy stores (Figure 3). Majority of the energy is stored in the body as fat, it is therefore, the balance between energy intake and energy expenditure that primarily determines whether body fat and hence body weight is gained or lost.
Given that all living organisms comply with the first law of thermodynamics, the energy balance equation has been used to predict changes in body weight when energy intake or expenditure is changed. However, the classic equation of energy balance, which states that the body energy store is equal to energy intake minus energy expenditure, has provided both insight and confusion in the understanding of energy balance in humans. Much of the confusion comes from inappropriate energy calculations using a static equation of energy balance. This equation is self-evident and reasonably accurate when body weight is maintained. Therefore if body weight and hence energy stores are stable it is likely that the energy flux is balanced between what is consumed and what is expended (Energy Intake = Energy Expenditure).
The most common equation used in discussions and calculations of energy balance is referred to as the static energy balance equation (below).
Change in Energy Stores = Energy Intake - Energy Expenditure
Intuitively, the equation appears valid, but Alpert (3) elegantly demonstrated that it is inadequate for calculations on living organisms, because it does not take into account the increasing energy expenditure that ensues as a result of the increasing weight (4-6). A small initial increase in energy intake sustained over a number of years therefore, cannot lead to a large weight gain, as is often claimed.
The more appropriate equation (below) would incorporate the use of the term "rates" because it introduces time dependency, thereby allowing the effect of changing energy stores (especially fat-free mass and weight) on energy expenditure to enter into the calculation (3).
Rate of Change of Energy Stores = Rate of Energy Intake - Rate of Energy Expenditure
This equation explains why a small initial positive energy balance (for example, from an increased energy intake or a lower thermic effect of food) will not lead to large weight increases over a number of years. After a short period of positive energy balance, the energy stores (fat mass and fat-free mass) will increase and cause an increase in energy expenditure which will balance the increased energy intake. The individual will then once again be in energy balance, but now with a higher energy intake, greater energy expenditure, and larger energy stores. Weight gain can therefore be viewed not only as the consequence of an initial positive energy balance but also as the mechanism by which energy balance is eventually re-established. How is intake balanced against expenditure, and how might a chronic mismatch between the two occur? A fruitful approach to these questions has been to dissect the energy balance equation into its various nutrient balance equations.
If the origins of a positive energy balance lie in the chronic imbalance of energy intake and expenditure, an appropriate question is: "What conditions allow a long-lasting imbalance between intake and expenditure”? An examination of each nutrient balance equation to determine if a chronic imbalance between nutrient intake and oxidation exists is only valid if each nutrient has a separate balance equation, implying separate regulation. In practical terms: Is each nutrient either oxidized or stored in its own compartment (separate regulation), or does it get converted into another compartment for storage?
Protein intake is usually about 15% of calories and the protein stores in the body represent about one-third of the total stored calories in a 70 kg man. The daily protein intake amounts to a little over 1% of the total protein stores (7, 8) (Figure 3). The protein stores increase in size in response to such growth stimuli as growth hormone, androgens, physical training, and weight gain, but do not increase simply from increased dietary protein. Protein stores are, therefore, tightly controlled and on a day-to-day basis protein balance is achieved (9). Therefore, protein imbalance is not implicated as a direct cause of obesity although, as with the other non-fat nutrients, protein intake may affect the fat balance equation (8).
Carbohydrate is usually the main source of dietary calories, yet the body stores of glycogen are very limited: 500-1000 g on average (10). Daily intake of carbohydrate corresponds to about 50-100% of the carbohydrate stores compared to about 1% for protein and fat (11) (Figure 3), so that over a period of hours and days, the carbohydrate stores fluctuate markedly compared with those of protein and fat. However, as with protein stores, carbohydrate sores are tightly regulated (12). Even if all the details of this regulation remain to be established, the control is based on humoral and/or nervous signals exchanged between the muscle, the liver and the brain. Dietary carbohydrate stimulates both glycogen storage and glucose oxidation and suppresses fat oxidation (13). That which is not stored as glycogen, is oxidized (not converted to fat), and carbohydrate balance is achieved (11, 13). Therefore, as with the other non-fat nutrients, a chronic imbalance between carbohydrate intake and oxidation cannot be the basis of weight gain because storage capacity is limited and controlled, conversion to fat is an option which occurs only under extreme conditions in humans, and oxidation is increased to match intake.
In marked contrast to the other nutrients, body fat stores are large and fat intake has no influence on fat oxidation (11, 13). As with protein, the daily fat intake represents less than 1% of the total energy stored as fat, but the fat stores contain about six times the energy of the protein stores (7) (Figure 3). These fat stores are the energy buffer for the body and the slope of the relationship between energy balance and fat balance is one in conditions of day-to-day small positive or negative energy balances (9). A deficit of 200 kcal of energy over 24 hours means 200 kcal comes from the fat stores, and the same holds true for an excess of 200 kcal of energy which ends up in the fat stores. Even in conditions of spontaneous overfeeding the entire excess fat intake is stored as body fat (12).
Ingestion of a mixed meal is followed by an increase in carbohydrate oxidation and a decrease in fat oxidation and the addition of extra fat does not alter that mix of nutrient oxidation (11, 13). What promotes fat oxidation if it is not dietary fat intake? The amount of total body fat exerts a small, but significant, effect on fat oxidation and this promotion of fat oxidation at higher body fat levels may represent a mechanism for attenuating the rate of weight gain (14). Energy balance is the driving force for fat oxidation (9, 14): when it is negative (i.e., energy expenditure exceeding intake), fat oxidation increases. After a few days, some individuals can however increase their fat oxidation in response to an increase in dietary fat under eucaloric conditions (15).
There is an inconsistent relationship between reported alcohol intake and body mass index, with many studies showing a negative relationship (16, 17). However, It has been shown that in healthy individuals the fate of ingested alcohol is oxidation and not storage (as fat), and therefore perfect alcohol balance is achieved (18). In the same manner as dietary carbohydrate and protein, alcohol diverts dietary fat away from oxidation and towards storage and inhibits lipolysis. Therefore, a chronic imbalance between alcohol intake and oxidation cannot be a direct cause of obesity, although by contributing to overall energy balance, it may indirectly influence fat balance (19).
Energy imbalance is buffered by fat stores
Dr JP Flatt introduced the concept that imbalance between energy intake and energy expenditure is buffered by changes in fat stores (20). As a consequence of the fact that amino acids, glucose and alcohol oxidation rates adjust themselves to the amount consumed, fat oxidation ends up to be determined primarily by the ‘gap’ between total energy expenditure and energy ingested in the form of carbohydrates, protein and alcohol rather than just the amount of fat consumed on a given day.
In summary, under physiological conditions, fat is the only nutrient capable to maintain a chronic imbalance between intake and oxidation thus directly contributing to the increase in adipose tissue. The other macronutrients will therefore indirectly influence adiposity by their contribution to overall energy balance and thus fat balance as emphasized by Frayn (21). The use of the fat balance equation instead of the energy balance equation offers a new framework for understanding the pathogenesis of obesity (Figure 3).
Determinants of weight gain at the turn of millenium
Over the past several decades, humans inhabiting industrialized countries have been challenged with relatively sudden and momentous changes in the environment. Agricultural and technological revolutions have coincided to allow significant changes to occur in both energy intake and energy expenditure (22). Specifically, the current environment supports sedentary behavior by allowing for profound reductions in the amount of physical activity required in order to exist and function successfully. Exacerbating the decreased need to expend energy is a surplus of low-cost, readily available, energy-dense, highly refined fats and carbohydrates that promote caloric intake in excess of energy needs (23). One of the most remarkable occurrences of the past 50 years is that humans have devised methods for producing mass quantities of food with very minimal input of physical labor. Combined, the excess food availability and physical inactivity have created an “obesogenic” environment that promotes energy imbalance and weight gain in everyone, but more in those genetically susceptible (Figure 2). The following section will outline trends in energy intake and expenditure over the past 20 – 50 years that have coincided with the significant increase in weight gain and the prevalence of obesity.
According to food supply data (which provides an estimate of food availability) and results from large-scale dietary surveys, American adults are consuming more calories today than they were in the 1970s (24). In particular, a large increase in caloric intake occurred between 1985 and 2000. In 2000 the average daily calorie consumption was 12 percent, or approximately 300 kcal, more than that consumed in 1985 (25). Of that 300 kcal increase, grains (mostly refined) accounted for 46%; added fats, 24%; added sugars, 23%; fruits and vegetables, 8%; and meat and dairy consumption declined 1%. The increase in fat intake over these years was not linear; in fact, total dietary fat intake remained steady during the 1980s and early 1990s and decreased slightly between 1993 and 1997. During the time between 1993 and 1997, consumers began to push for lower fat versions of popular products, and total calorie consumption in the form of mostly carbohydrates increased thus decreasing the overall percent of energy from fat. However, between 1997 and 2000, consumption of added fats rose dramatically, jumping 16% from 56 to 61 g per day on average, which was in addition to the fat that occurs naturally in foods (25). The increase in fat consumption is suggested to be a major player in the obesity epidemic, since consuming a high fat diet is associated with dysregulation of normal satiety mechanisms meant to regulate energy balance (i.e., humans overcompensate energy requirements when eating foods high in fat) (26). Lending support to this theory, there are numerous studies showing a relationship between body fatness and consuming a high fat diet (27).
Another major change in the American diet over the past 30 years is the consumption of refined sugars. The per capita consumption of all refined sugars in the United States in 1970 was 55.5 kg, and by 2000 this number rose to 69.1 kg (28). Whereas the Food Guide Pyramid recommends that consumers limit added sugars to 12 tsp a day for a 2,200 kcal diet, the food supply in 2000 provided approximately 31 tsp of total added sugars per day (25). The quality of added sugars has changed in recent years in addition to the increase in total quantity (28). The production of high-fructose corn syrup (HFCS) in mass quantity became economically feasible with the advent of chromatographic fructose enrichment technology in the late 1970s (29). This invention stimulated the 22% increase in average consumption of added sugars between 1980 and 2000, since the intake of HFCS rose from practically zero in the 1970s to 13.2 tsp per capita per day in 2000 (25). In addition to the increased availability, the existence of abundant carbohydrates and fats in a highly refined form has increased the energy density of the U.S. diet and has also decreased the satiety index and reduced the amount of physiologic work necessary for digestion and absorption (7, 22).
A change in the portion size of foods appears to have contributed to the increase in obesity rates. Compared with similar older data, current portions of restaurant foods, grocery products, and cookbook recipes have increased in size in every food category except bread (30). Portions started to expand in the 1970s, increased sharply in the 1980s, and continue to rise today. A startling example is that of “huge foods” (31), where muffins weigh half a pound, restaurant-sized pasta bowls hold over 2 pounds of food, and a medium-sized popcorn at a movie theater contains 16 cups and delivers approximately 1,000 kcals. Recent analyses of nationally representative survey data show that self-reported portion sizes of foods consumed has increased between 1977 and 1998 (32), (33) and this has occurred both inside and outside of the home, with the largest increase in portion size occurring at fast-food establishments (32). Indeed, the increase in obesity prevalence has coincided with an increase in the number of meals eaten outside of the home (24), and the frequency of eating out is associated with increased energy and fat intake (34) and higher BMI (35).
Changes in Energy Expenditure
A popular theory holds that the escalating rate of obesity can be attributed to the change in the American diet and the rise in overall energy intake without a compensatory increase in energy expenditure. Historical observations suggest that regulatory systems for maintaining body weight evolved against a backdrop of high physical activity and energy expenditure (36). Technological advancements within the past century, including mechanized equipment, have shifted the majority of the population away from work that demands physical labor to occupations that encourage mostly sedentary behavior. Furthermore affluence is associated with significant reductions in work-related energy expenditure. Compounding the sedentary nature of work is the development of power-assisted domestic appliances and tools that have allowed for significant reductions in the amount of energy expended in daily living and those conducted during leisure time. Further reductions in the propensity to expend energy during leisure time can be owed to devices such as television viewing, computer games and the Internet, all of which generally require minimal movement and low energy expenditure. In some opinions, modern obesity can be viewed as a normal and predictable physiological response to a new, pathogenic and “obesogenic” environment (Figure 2).
Despite the changing environment, nationally-representative survey data indicate that adults achieving the recommended physical activity levels actually increased by 9.7% for men and 5.8% for women between 1990 and 2000 (37). Vigorous physical activity during leisure time (LTPA), reported as metabolic equivalents (METs), increased by ~60 MET min/day in men between the 1960s and 1990s, which represented an average increase of approximately 10 min of high intensity LTPA per day. No change in vigorous LTPA during this time was found in women, and there were no changes in moderate LTPA in either men or women during this time period (38). Accordingly, the percentage of adults who were sedentary in their leisure time decreased from the 1960s to 1990s in both men (24% to 14%) and women (26% to 17%) (38). In adolescents, the percentage of youth who participated in vigorous activity (making them sweat or breathe hard for at least 20 min 3 times per week) remained stable from 1993 (65.8%) to 2001 (64.6%); however, the percentage of students who attended physical education classes decreased during that same time period, dropping from 41.6% to 32.2% (37).
The major change in activity patterns appears to have occurred in physical activity related to occupation. Over the last 50 years, the surge of women entering the workplace has transformed the structure and characteristics of the U.S. workforce, with the total number of adults employed more than doubling from 62.2 million in 1950 to 142.5 million in 2000. In 1992, it was estimated that only 20% of men and 10% of women were employed in active occupations ((39), U.K. data). Another significant change has been in the percentage of adults employed in agricultural work, typically associated with high physical activity levels, which declined from 12.2% in 1950 to less than 2% in 2000 (37). Overall, the percentage of adults who worked in low physical activity occupations increased by almost 83% from 1950 to 2000 and those employed in high activity jobs decreased by 25.2% (37).
Despite the suggestion that LTPA has increased over the past 50 years, there is reason to believe that changes in our physical environment have influenced leisure-time energy expenditure. For example, car ownership has increased significantly over the past 30 years and as of 2001, U.S. households on average had more vehicles than available drivers (40). The likelihood of walking or biking is inversely related to the number of automobiles per household, and this is independent of income level (37). Another influential change has been the television. In 1950, only about 10% of U.S. households owned a TV, whereas now, approximately 98% of households have at least one television set (37). This increase is associated with a doubling of average viewing hours per day; daily television viewing increased by 61.4% from 1950 to 2000 (37). There is growing evidence for a strong association between hours per day spent watching TV and obesity, in adults (41) and children (42). Overall, proxy measures of inactivity, such as automobile ownership and TV viewing, may be more closely related to overweight and obesity than purposeful physical activity (43).
Other Recent Developments
Recently, evidence has been collated to support ten other potential contributors to the obesity epidemic (44). That is besides the ‘Big Two’ (diet and physical activity) these 10 factors can increase adiposity and the frequency distribution has changed in the direction that would support an obesogenic response. For example;
a) Intrauterine and Intergenerational Effects: A recent hypothesis that is gaining momentum in the scientific community is that obesity (and others disorders) may be the result of genetic and epigenetic programming that occurs in utero up to two generations earlier (Figure 4). Thereby environmental changes that occurred during the lifetime of our grandparents could be implicated in the obesity epidemic today. In support of this hypothesis studies in rodents have shown that overfeeding results in increased body weight and adiposity in the offspring which continues for more that 3 generations (45). In humans, an infant born small for gestational age (SGA) is considered to be at risk for adult onset obesity (46). A low birth weight and the subsequent catch-up growth is associated with adult BMI and interestingly the incidence of low birth weight babies has increased in the US reaching ~8% in 2002 (47).
b) Reduced smoking: Nicotine suppresses appetite and increases thermogenesis. It is frequently reported that smokers have a lower body weight compared to non-smokers and that weight gain coincides with smoking cessation. Cigarette smoking has declined steadily over the past several decades and together with the change in overweight during the 12 years to 1990, smoking was estimated to account for approximately 25% of the increase in men and approximately 10% of the increase in women (48).
c) Sleep debt: Average hours of sleep per night has decreased from over 9 to around 7 hours per night (49) and this has been proposed to stimulate hyperphagia via endocrine changes including increased ghrelin and decreased leptin concentrations (50). Hours of sleep per night is inversely associated with BMI and obesity in cross-sectional studies of children (51) and adults (52).
d) Reduction in variability in ambient temperature: Greater use of central air conditioning and heating has increased the amount of time that U.S. adults spend in the thermoneutral zone, which is defined as the range of ambient temperature at which energy expenditure is not required for maintaining homeothermy (44). For example the number of homes in the US without air conditioning has decreased by almost 20% to a low 28% and is estimate to be as low as 7% in the Southern States where the obesity rates are at the highest levels in the country. Exposure to temperatures above or below the thermoneutral zone increases energy expenditure (53) and may significantly decrease food intake under conditions of elevated temperature (54).
e) Increased obesity in predisposed genotypes: The number of individuals from racial minorities at high susceptibility for the development of obesity has increased in the Western world. BMI and hence obesity is clearly heritable as shown in twins in whom up to 65% of BMI has been shown to be heritable (55). The second component of this hypothesis is that individuals with a higher BMI have an increased potential to reproduce. Supporting evidence suggests that the number of offspring is positively associated with BMI (56) in women and one explanation could be that obesity can lead to increased fertility. Certainly a low BMI in women is associated with infertility (57) and for men leanness too leads to a greater reduced in sperm count (58).
