Global Trade Liberalization

Between the 1970s and 1990s, many countries underwent economic structural adjustments, including implementation of more market-oriented or liberal agricultural trade policies (Hawkes 2006). These policies have altered the food supply and are thought to have had direct effects on the obesity epidemic, contributing to nutritional transition and changing food choice and availability. At the same time, however, some of these changes to the food system have led to improvements in quality of life and food security and to reductions in poverty. Trade liberalization can affect the availability of certain foods by enabling the trade of greater amounts and varieties of food, by removing barriers to foreign investment in food distribution, and by expanding multinational food companies and fast-food chains (Kearney 2010). For example, in 1998, transnational food companies based in the United States invested US$5.7 billion in establishing outlets globally (Harris and others 2002). Analysis shows that LMICs that enter free trade agreements with the United States have a 63.4 percent higher consumption of sugar-sweetened beverages per capita than those that do not, after adjusting for a given country’s level of gross domestic product (GDP) per capita and urbanization (Stuckler and others 2012). Some scholars have suggested that implementation of the North American Free Trade Agreement has coincided with the burgeoning obesity epidemic in Mexico through increased sales of low-quality processed and fast foods (Clark and others 2012). The relationship between economic policies, global trade agreements, and obesity should be investigated as points of intervention, given the potential scale of their impact.

Income and Socioeconomic Status

During the next three to four decades, global income per capita is projected to rise at a rate of more than 2 percent per year, with more rapid rates expected in LMICs (Kearney 2010). The prevalence of obesity correlates positively with the initial stages of economic growth and development, as populations of rapidly developing LMICs undergo nutritional and lifestyle transitions while having little access to health services and education. An analysis of global patterns of nutritional risks in relation to economic development in 100 countries showed that BMI increased rapidly in relation to national income (Ezzati and others 2005). This association declined as countries achieved upper-middle-income and high-income status, primarily as a result of improved access to health services and education as well as behavioral changes. As average income rises, habits associated with obesity are adopted, such as watching television, purchasing convenience foods at supermarkets, and consuming highly processed fast food. However, access to health care, education, healthy food, and recreational activities that support weight maintenance remain limited.

Within countries, obesity is related to socioeconomic status. In many LMICs, body weight is positively associated with socioeconomic status, which contrasts with general patterns observed in the United States and other HICs, where body weight tends to be inversely associated with socioeconomic status. The association between socioeconomic status and body weight is thought to depend on the level of economic development in the country. In a review of nationally representative surveys of women in 37 LMICs, the burden of obesity was shown to shift toward low-socioeconomic groups as the country’s gross national product per capita increased to about US$2,500 (Monteiro and others 2004). Countries with gross national product per capita greater than US$2,500 include Brazil, Mexico, South Africa, and Turkey.

Urbanization

The proportion of the world’s population living in urban areas has increased markedly, from 13 percent in 1900 to almost 50 percent in 2005 (Kearney 2010). This trend is expected to continue, primarily in countries where the vast majority of the population is currently rural. Low- to medium-density residential areas around urban centers have been associated with obesity in the United States (Hu 2008) and are starting to appear in some LMICs. Globally, 93 percent of urban growth is estimated to occur in LMICs, with 80 percent of that urban growth occurring in Africa and Asia (UN Population Fund 2007). In China, for example, more than 1 billion people are projected to be living in urban centers by 2050, which is nearly twice the size of China’s current urban population (UN Department of Economic and Social Affairs 2009).

The consequences of urban living on the development of obesity are numerous and occur largely as a result of changes in the built living environment, range of available food options, and lifestyles related to technological advancement and mechanization. Collectively, these changes have played a role in lowering the quality of urban diets and the expenditure of energy. These trends of positive energy balance are expected to continue as urbanization continues. At the same time, urbanization facilitates greater access to health care and education, which have beneficial effects on obesity. However, many LMICs undergo urbanization at such a rapid pace that the development of essential infrastructure lags behind. It is also important to note that some inhabitants of rural areas have worse diet quality than those of urban areas, especially diet diversity, but it is unclear how this lower quality may affect obesity risk. One review noted that the relative annual change in weighted obesity prevalence is higher for rural (3.9 percent) than for urban (2.5 percent) women, suggesting that women in rural areas are catching up to their urban counterparts (Popkin, Adair, and Ng 2012). Overall, the trends likely differ between HICs and LMICs and are influenced by socioeconomic status (Darmon and Drewnowski 2008).

Physical Activity

Densely populated areas with little outdoor recreational space provide limited opportunities for walking and leisure-time physical activity. Physical activity, including walking and cycling, has modest benefits for weight, with more appreciable effects seen when lifestyle changes occur in combination with dietary intervention. The combination of increased caloric intake and reduced energy expenditure can have a significant impact on the development of obesity. Lower energy expenditure decreases energy requirements, allowing excess calories to accumulate faster. Current physical activity guidelines for preventing chronic diseases recommend 150 minutes per week of moderate-intensity aerobic activity or 75 minutes per week of vigorous-intensity activity or a combination thereof (U.S. DHHS 2008). A recent paper published as part of the Lancet’s second series on physical activity estimated that the prevalence of inactivity (defined as not meeting the physical activity guidelines above) among adult populations worldwide was 23.3 percent in 2016 (Sallis, Bull, and Guthold 2016). The trend data from the United States show that leisure-time physical activity has been relatively stable or increased slightly, while activities related to work, transportation, and household chores have declined dramatically and sedentary behaviors, such as television viewing and other activities involving digital devices, have increased substantially. These changes have led to an overall reduction in total physical activity (Brownson, Boehmer, and Luke 2005).

In many LMICs, a pronounced movement has occurred away from jobs with high energy expenditure, such as farming, mining, and forestry and toward jobs in the more sedentary sectors of manufacturing, services, and office-based work. This trend is typical of countries experiencing economic growth and urbanization, and these processes will determine the timeframe over which such movement takes place. In addition, computer technology, factories, and mechanization have become widespread in jobs that previously required high energy expenditure. In China, for example, the proportion of the population working in very light-activity jobs increased to 66 percent from 44 percent between 1989 and 2004 (Popkin 2009). Household chores also have become increasingly mechanized with the emergence of appliances, such as washing machines and vacuum cleaners. It was recently found that in contrast to evidence from HICs, urban (versus rural) residence was an inverse correlate of physical activity in LMICs; this finding is a concern, given global trends toward urbanization (Sallis, Bull, and Guthold 2016).

Leisure-time activities also have become more sedentary, shifting from outdoor play to indoor entertainment, such as television viewing, Internet use, and computer gaming. In China, a study published in 2005 showed that fewer than 25 percent of adults reported participating in at least 30 minutes of moderate physical activity per day, and television ownership increased from 38 sets per 1,000 persons in 1985 to 155 in 1990 and 270 in 1997 (Wang and others 2007). Data on television viewing and screen time in LMICs are sparse, but many epidemiological studies have shown a link between time spent watching television and weight gain in both children and adults, and reducing sedentary behavior has been shown to have beneficial effects on weight independent of exercise (Muntner and others 2005).

Changes to the built living environment include the construction of roads and highways and the implementation of motorized transportation systems that have created sprawl that limits the opportunities for walking or bicycling and provides less healthy alternatives. The proportion of journeys made on foot or by bicycle has declined from 37 percent and 26 percent, respectively, in cities with 100,000–250,000 inhabitants to 28 percent and 9 percent, respectively, in cities with more than 5 million inhabitants (Singh 2005). Another major shift affecting energy expenditure in LMICs is the displacement of human-propelled modes of transportation, such as bicycles, in favor of motorized transportation, including cars and mopeds. In India, the annual rate of motor vehicle ownership increased 11 percent between 1997 and 2008; in China, new car sales are increasing an estimated 30 percent per year (Kjellstrom, Hakansta, and Hogstedt 2007; Siegel, Narayan, and Kinra 2008).

Behavioral Change and Sociocultural Norms

Certain behavioral changes brought about by urban living could also contribute to the development of obesity in LMICs. Although not formally documented, urbanization is associated with a decrease in sleep duration, because noise pollution, street and domestic lighting, access to television and the Internet, shift work, and nighttime social activities are more common in urban than in rural areas. In epidemiological studies, short sleep duration has been associated with weight gain in both children and adults (Berkey, Rockett, and Colditz 2008; Patel and Hu 2008). Similar evidence is starting to emerge from various LMICs, including Brazil, Senegal, and Tunisia, as well as from Taiwan, China (Hu 2008). Stress, which has been associated with obesity (Hu 2008), may also be a risk factor in rapidly urbanizing LMICs because of occupational demands and lack of the kind of social support that is available in traditional villages.

Obesity can also be affected by cultural and social norms, with wide variation across LMICs. These norms include cultural food preferences, societal norms for body shape, cultural practices surrounding the use of leisure time and physical activity, gender norms, and academic expectations, all of which can interact with other risk factors to contribute to rising rates of obesity. For example, Polynesians, who have one of the highest rates of obesity in the world, equate large body size with power, beauty, and affluence (Brewis and others 1998), and in India and many LMICs, there is a general misconception that an obese child is a healthy child (Bhardwaj and others 2008). In contrast, in the United States and some other cultures, some demographics consider being underweight to be the epitome of beauty (Low and others 2003). Although cultural and social norms are embedded in society, they are not static. Globalization of the food supply, including through multinational supermarkets, fast-food chains, mass media, and marketing, has altered consumer preferences and behaviors in many LMICs in a relatively short period.

Diet

As many countries experience the rapid economic growth and urbanization that have changed the choice and availability of food, concomitant shifts in dietary structure or nutritional transitions occur that promote a positive energy balance. These transitions are also fueled by reductions in the price of low-quality foods that are high in energy and by increases in GDP, which leads to higher family income and enhanced purchasing power. Weight gain and obesity in free-living populations result from the cumulative effects of small changes in daily energy balance. A typical diet is composed of energy-bearing macronutrients, including carbohydrates, protein, and fat, as well as micronutrients, including vitamins and minerals. These dietary factors can directly or indirectly tip the balance in energy intake and expenditure and thus effect changes in body weight. Similar to noncommunicable diseases, weight gain and obesity are complex processes that are caused by perturbations of multiple dietary habits and behaviors rather than by changes in any single dietary factor. This course contrasts with deficiency syndromes, which can usually be ascribed to one or a few factors. While dietary patterns vary enormously between and within countries and limited data are available to document dietary shifts, some broad themes are apparent, such as the global decrease in diet quality—excess consumption of unhealthful and highly processed foods and inadequate consumption of healthful whole foods. Many of the dietary risk factors for weight gain and obesity are covered in chapter 6.

Another major shift contributing to poor diets has been an increase in the consumption of highly refined carbohydrates, such as polished white rice and refined wheat flour. Milling and processing of whole grains to produce refined grains remove the fiber and numerous micronutrients, which have a variety of health benefits. Fiber is largely responsible for the beneficial effects of whole grains on body weight, promoting satiety and leading to decreased energy intake. Given the number and variety of nutrients, whole grains can also be useful for addressing undernutrition. Temporal data from the United States show an association between intake of refined carbohydrates and risk of obesity and type 2 diabetes mellitus (Gross and others 2004). In urban South India, nearly half of an average individual’s daily energy intake comes from refined grains, with polished white rice constituting more than 75 percent of refined grain intake (Mattei and others 2015). In China, white rice accounts for more than 30 percent of an average individual’s daily caloric intake (Mattei and others 2015). A meta-analysis of prospective cohort studies in Asian and Western populations found that, for each increment of one serving per day of white rice, the risk of type 2 diabetes increased 11 percent (Hu and others 2012). Associations were stronger in Asian populations than in Western populations because Asian populations consume larger quantities of white rice. The adverse effects of refined grains tend to be more evident in individuals who are overweight or obese and generally more insulin resistant than individuals who are lean. This finding is of great concern because as obesity becomes more prevalent globally, more people will be susceptible to the adverse effects of diets high in refined carbohydrates. Although some refined carbohydrates, such as white rice, have been staples in countries such as China for hundreds of years, the negative health effects of refined grains were likely offset by high levels of physical activity.

In addition to the intake of specific foods, certain dietary habits have been associated with body weight. Regular consumption of breakfast has been widely recommended for preventing obesity. Skipping breakfast increases the production of appetite-stimulating hormones, which may lead to overeating throughout the day. In a randomized controlled trial (RCT), omitting breakfast led to increased energy intake and adverse effects on blood lipids and glycemic control (Farshchi, Taylor, and Macdonald 2005). Data on breakfast consumption and body weight are limited, but some prospective cohort studies have reported inverse associations between breakfast consumption and weight gain in European and U.S. populations (Hu 2008). Increased consumption of fast food is thought to be a major factor contributing to rising rates of obesity, and the intake of fast food is on the rise in many LMICs, as multinational chains seek large new markets. Epidemiological studies from Europe and the United States have shown positive associations between the consumption of fast food and weight gain and adverse metabolic outcomes (Rosenheck 2008). These associations are likely due to a combination of large portion sizes and high calorie content; high amounts of processed meat, refined carbohydrates, sugary beverages, and unhealthful fats; enhanced palatability; and low cost, resulting in overeating and positive energy balance.

From a public health point of view, identifying the dietary determinants of weight gain is critical for reducing the prevalence of obesity because once an individual develops obesity, it is metabolically difficult to lose weight and maintain weight loss. However, numerous short- and long-term studies across a variety of general and clinical populations have attempted to identify the optimal ratio of macronutrients for weight loss. Lowering the proportion of daily calories consumed from fat has been targeted for many reasons, including the high energy density of fat (that is, a single gram of fat contains more than twice the calories of a gram of carbohydrates or protein) and enhanced palatability of high-fat foods. Thus, prevailing guidelines for weight loss have recommended reducing the intake of fat, but studies have produced inconsistent findings regarding the benefits of restricting fat for weight loss. One reason is the reciprocal relationship between energy from fats and carbohydrates in most diets; in addition, as previously discussed, intake of refined carbohydrates has been positively associated with weight gain. Although the relative influence of macronutrients on body weight remains unclear, accumulating evidence suggests that low-carbohydrate and Mediterranean-style diets, which are rich in plant-based foods and have a moderate fat content, have benefits for weight loss (Tobias and others 2015). These regimens may be more sustainable than other strategies because of the greater diversity and palatability of foods.

Genetic Risk of Obesity

The search for human obesity genes began several decades ago with findings from studies of adopted twins suggesting that obesity and obesity-related traits have a substantial heritable component, although the exact degree of genetic heritability is still debatable. Because the prevalence of obesity in many countries has tripled during the past three decades, it is unlikely that genetics is the primary cause of obesity. Compelling evidence indicates that obesity is driven largely by changes in diet and lifestyle. However, ethnic differences in obesity rates cannot be explained by these factors alone. A more probable hypothesis is that obesity is the outcome of an adverse obesogenic environment interacting with a susceptibility genotype. Genetic factors that underlie susceptibility to obesity may be amplified in the presence of certain environmental factors or, given the same diet and lifestyle factors, some individuals may be genetically more prone to obesity than others. Epigenetic mechanisms, in which environmental factors cause changes in the expression of genes but do not involve changes in DNA (deoxyribonucleic acid) sequence, could also help explain the global increase in obesity prevalence.

Efforts to identify obesity genes have intensified in recent years with advances in genotyping technology and genetic epidemiologic methods. Several genetic factors responsible for rare monogenic forms of obesity have been identified; however, genes for common forms of obesity remain an active area of research. Unlike monogenic obesity, which results from an alteration in a single gene, the genetic profile of common or polygenic obesity is complex and likely results from the effects of several altered genes. Genome-wide association studies, which determine whether an association between a genetic variation and an obesity-related trait exists by surveying the entire genome for causal genetic variants in a comprehensive and unbiased manner, have identified more than 52 loci associated with obesity traits (Albuquerque and others 2015). Of all the currently identified loci, the fat mass and obesity-associated (FTO) gene has the largest effect on susceptibility to obesity (Albuquerque and others 2015). However, FTO polymorphism has a modest effect on BMI, explaining only 0.31 percent of the variation in BMI among individuals (Albuquerque and others 2015). These findings have been independently replicated and confirmed in several African, Asian, and European populations in both children and adults, although the functional mechanism underlying the role of FTO in obesity remains unknown.

Identification of gene-environment interactions related to obesity has been challenging because many genetic association studies lack detailed information on diet or other types of exposure. However, some recent observational studies and trials have evaluated gene-diet interactions. For example, the association between sugar-sweetened beverages and obesity was strengthened by an analysis examining whether consumption of sugar-sweetened beverages can modify the genetic risk of obesity, using a genetic predisposition score based on 32 obesity genes identified from genome-wide association studies (Qi and others 2012). Based on data from three large cohorts, greater consumption of sugar-sweetened beverages was associated with a more pronounced genetic effect on elevated BMI and an increased risk of obesity. Individuals who consumed one or more servings of sugar-sweetened beverages per day had genetic effects on BMI and obesity risk that were approximately twice as large as those who consumed less than one serving per month. These data suggest that regular consumers of sugar-sweetened beverages may be more susceptible to the genetic effects on obesity, implying that a genetic predisposition to obesity can be partly offset by healthier beverage choices. Alternatively, persons with a greater genetic predisposition to obesity may be more susceptible to the deleterious effects of sugar-sweetened drinks on BMI. These findings may partly explain individual differences in the metabolic response to sugar-sweetened beverages. A better understanding of gene-environment interactions and epigenetics could lead to more accurate estimates of the impact of environmental factors on genetically susceptible individuals and identify high-risk populations for targeted prevention and intervention.

Nutritional and Agricultural Policies

Combining incentives and deterrents can be an effective strategy for encouraging production and consumption of nutritionally beneficial foods and discouraging production and consumption of unhealthful foods. Some governments are considering taxing select foods and beverages, particularly sugar-sweetened beverages, as a means to improve consumer choice and generate revenue. Whether these programs will have the desired effect is yet to be determined. Some studies have suggested that, for such interventions to have an appreciable impact, tax increases of at least 10 percent are needed (Gortmaker and others 2011). The most relevant data currently available come from Mexico, where a peso-per-liter tax on sugar-sweetened beverages (roughly US$0.08 per liter, equivalent to a 10 percent tax increase) resulted in a 10 percent decline in purchases of taxed beverages in the first quarter of 2014, compared with the first quarter of 2013. These preliminary results also show roughly a 7 percent increase in purchases of untaxed beverages (such as diet sodas, sparkling and still plain water, 100 percent fruit juices, flavored water with noncaloric sweeteners, and milk without added sugar), including approximately a 13 percent increase in purchases of plain water.¹ A more recent evaluation found that purchases of taxed beverages decreased by an average of 6 percent 1 year after implementation of the tax. Purchases of untaxed beverages were 4 percent higher over this period, mainly driven by an increase in purchases of bottled plain water (Colchero and others 2016). Removing subsidies on animal-based foods, unhealthful oils, and sugar in exchange for subsidies on fruits, vegetables, legumes, nuts, and whole grains could be an effective strategy for improving diet quality. In 2000, India produced 26.6 million tons of fruits and 96.5 million tons of vegetables (Siegel, Narayan, and Kinra 2008). However, inaccessibility and high costs prohibited consumption of these foods in all but affluent, urban populations. In China, subsidies on fruits, vegetables, and soybeans have increased the production and consumption of these products (Zhai and others 2002). Voluntary actions and regulations made by industry to reduce calories in the food supply are underway in the United Kingdom and the United States (Malik, Willett, and Hu 2013). Translating similar initiatives to other countries represents a promising long-term goal for preventing obesity.

Food Labeling

Nutritional labeling is emerging as a major global initiative, and some LMICs, including Brazil, Chile, China, India, Mexico, and South Africa, are also considering developing systems to identify nutritionally beneficial foods and beverages (Popkin, Adair, and Ng 2012). In comparison with nutritional facts panels, which consumers use to draw their own conclusions about how healthy a product is on the basis of the nutrient content of foods, these systems would identify foods that benefit health, such as whole grains, to help consumers make healthy choices. An increasingly popular strategy being considered by various jurisdictions is calorie labeling on restaurant menus. The U.S. Food and Drug Administration recently finalized two rules requiring that calorie information be listed on menus and menu boards in chain restaurants as well as in similar retail food establishments and vending machines. A systematic review and meta-analysis of 19 studies in the United States found that menu labeling was associated with a reduction of 18.13 kilocalories ordered per meal, but significant heterogeneity between studies limits interpretation of the results (Long and others 2015). Another small meta-analysis found that labeling menus with calories alone significantly reduced the amount of calories ordered by 31 kilocalories and the amount of calories consumed by 13 kilocalories (Sinclair, Cooper, and Mansfield 2014). In addition, it found that the use of interpretative nutritional information on menus assisted consumers in selecting and consuming foods with fewer calories, saving 67 kilocalories and 81 kilocalories, respectively. To provide some context, one small apple provides approximately 55 kilocalories.

These initiatives hold promise for LMICs, where the increasing availability of processed and packaged foods and fast-food chains is accompanying increasing rates of urbanization. Educational campaigns must precede or accompany both food package and point-of-purchase nutrition labeling to raise awareness about these initiatives and provide context, so that individuals know to look for the labels and understand why they are important and how to interpret them. Other important caveats include the need to have consensus on the definition of specific healthy foods such as whole grains and, in the context of LMICs, a system in place to ensure accurate labeling.

Food Advertising

A growing body of evidence indicates that food marketing can influence food preferences and consumption habits, especially of children (Hawkes 2007). However, evidence from systematic reviews is lacking, and few studies have evaluated the impact of advertising on energy intake or body weight. A systematic review of seven RCTs aiming to assess the effect of television advertising on food intake of children ages 4–12 years concluded that there is a positive association between television and energy intake but that this association is based on a very limited number of trials lacking a solid ground of first-level evidence (Gregori and others 2014). Despite the lack of systematic evidence, regulation of advertising targeted toward children through television, the Internet, or other media is thought to be a potentially effective strategy that should be adopted globally to reduce the harmful effects of marketing unhealthful high-energy foods. In 2010, the WHO released a set of recommendations on the marketing of foods and nonalcoholic beverages high in fat, sugar, and salt to children in an effort to encourage healthy dietary choices and promote the maintenance of healthy weight (WHO 2010). To date, several countries, including Brazil, Chile, the Islamic Republic of Iran, Ireland, the Republic of Korea, Mexico, Peru, and the United Kingdom, as well as economies such as Taiwan, China, have taken steps to reduce such marketing (World Cancer Research Fund International 2016). Since 2007, France has banned the marketing of foods high in fat, sugar, and salt unless they are taxed and labeled with a health warning. In other countries, including Switzerland, Thailand, and the United States, industry has made voluntary pledges to restrict marketing to children. At the same time, governments can institute zoning laws, if available, that limit the number of fast-food restaurants in a given area.

Mass Media Campaigns

The mass media, including national-level social marketing and public service campaigns, have the potential to be very useful tools in delivering public health messages about healthy diets and lifestyles, either independently or as part of multicomponent interventions. Concurrently, these strategies should be carefully monitored and evaluated to gauge their effectiveness. However, very few studies have examined the impact of mass media campaigns on diets and lifestyle behaviors or on body weight. This is an important knowledge gap.

School and Workplace Interventions

School-based programs and policies to increase physical activity by requiring physical education classes or breaks and to improve diet by providing healthy school meals and heathier snack options in vending machines and cafeterias are effective strategies to address childhood obesity and should be part of the global obesity prevention agenda. These strategies are likely to be more effective if reinforced through curriculum-based education about healthy diets and active lifestyles and efforts to engage parents and families. A recent systematic review including 115 school-based interventions (mostly in the United States) concluded that moderately strong evidence supports the effectiveness of school-based interventions for preventing childhood obesity (Wang and others 2015). However, more evidence is needed to evaluate programs in other settings or with other types of design, especially interventions oriented toward providing environmental, policy, and consumer health information.

School meal programs, which provide low-cost or free meals to ensure nutritional adequacy among schoolchildren, provide a unique opportunity to encourage healthy eating habits while preventing undernutrition. WHO Europe has a food and nutrition policy, which can be adapted for use in schools in individual countries, that emphasizes the importance of breakfast clubs; intake of fruits, vegetables, and milk; access to water; and removal of vending machines containing snacks and beverages of poor nutritional quality (WHO Europe 2006).

Similar to the school setting, worksite-based interventions can overcome barriers to choosing a healthy lifestyle by providing resources and a socially supportive environment for change at a place where individuals spend much of their week and by offering programs at low or no cost. A meta-analysis of worksite-based physical activity programs in HICs (mostly the United States) showed significant positive improvements in body weight, cardiometabolic risk factors, physical activity and fitness, and diet quality as well as lower absenteeism and job stress (Conn and others 2009). A systematic review of 17 studies in Europe focusing on promoting a healthy diet in the workplace found limited to moderate evidence of effectiveness for prevention of obesity and obesity-related conditions (Maes and others 2012). Another systematic review of 16 studies, mostly in Europe and North America, found that diet-based worksite interventions of moderate methodological quality led to positive changes in fruits, vegetables, and total fat intake (Mhurchu, Aston, and Jebb 2010). To improve effectiveness, future programs should aim to intervene at multiple levels of the worksite environment and ensure stronger adherence to established quality criteria.

Urban Planning

Evidence from various countries supports the relationship between physically active modes of transport and obesity. An analysis of aggregate cross-sectional health and travel data from the United States, 14 comparison HICs, and 47 large U.S. cities found a significant inverse relationship between active travel and self-reported obesity at all three geographic levels (Pucher and others 2010). A systematic review of 43 incidental physical activity community interventions from high-income economies or regions (Australia; Canada; Europe; Japan; New Zealand; the United Kingdom; the United States; and Hong Kong SAR, China) found that, primarily, active transport (walking, bicycling) interventions and, secondarily, children’s play interventions, and to a lesser extent, use of stairs can be effective ways of increasing physical activity (Reynolds and others 2014). Comparisons were control groups that did not receive the intervention or baseline data of participants acting as their own control.

Given the potential for health and environmental benefits, governments should promote and facilitate the use of public transportation and bicycles by providing incentives, such as discounted transportation fares, bicycle-sharing programs, cycling safety classes, and secure bicycle parking. Lower health care premiums for active commuting in countries where health care is not state run would also be beneficial. Using public transportation encourages people to be more active generally, by walking and standing, than if they were using cars. Creating a central policy for urban transportation could be a first step for some countries. Governments should also mandate the construction of sidewalks and safe bicycle lanes and the construction of buildings with features that promote fitness, such as accessible staircases. Urban planning initiatives at the national or regional level should also encourage the development of safe, pedestrian-friendly communities with green spaces and access to public transportation. A systematic review found that physical changes to the built environment can increase urban green space and encourage physical activity (Hunter and others 2015). Careful evaluation of the effectiveness and cost-effectiveness of such policies is needed to support policy strategies.

Popular Weight-Loss Diets

Numerous branded weight-loss programs are broadly available to the general public in many countries, providing structured dietary and lifestyle recommendations via popular books and in-person or online behavioral support. Limited data are available for evaluating the clinical effectiveness of these strategies. However, a recent meta-analysis of trials examining the impact of popular self-administered brand-name diets on body weight compared with no diet found losses in body weight ranging from 4.10 to 6.55 kilograms after 12 months, with no appreciable difference between diets (Johnston and others 2014). Whether such programs are sustainable is not known.

Pharmacological Strategies

Lifestyle interventions for weight loss are often characterized by high rates of recidivism or weight regain, which may be due in part to complex metabolic processes and biological adaptations that defend against subsequent weight loss and promote weight regain. The primary aim of pharmacological treatment for obesity is to suppress the biological drivers of weight gain or dampen the counterregulatory response to weight loss and thereby enable patients to achieve and sustain clinically meaningful reductions in body weight.

Currently, five drugs have been approved by the U.S. Food and Drug Administration for chronic weight management in obese adults: orlistat, lorcaserin, phentermine/topiramate extended release, naltrexone/bupropion extended release (Kakkar and Dahiya 2015), and most recently, liraglutide. Very little data are available regarding the approval, availability, and use of these therapies in LMICs. For countries where these drugs are available, drug safety monitoring systems should be implemented—if not already in existence—to ensure safety of use. A recent report from India documented orlistat abuse in a case of bulimia nervosa and noted that the drug is available over the counter (Deb, Gupta, and Varshney 2014). Use of anti-obesity medications in conjunction with lifestyle modification can lead to weight loss in the range of 5 percent to 15 percent (Kakkar and Dahiya 2015). One long-term study found that, compared with lifestyle changes alone, orlistat plus lifestyle changes resulted in a greater reduction in the incidence of type 2 diabetes over four years and produced greater weight loss in a clinically representative obese population (Torgerson and others 2004). However, further studies are needed to assess the long-term benefits and cost-effectiveness of these new agents as well as potential pharmacological interactions and adverse effects. Several anti-obesity therapies have been taken off the market because of significant side effects (Kakkar and Dahiya 2015), underscoring the importance of close monitoring and evaluation.

Bariatric Surgery

Management of obesity usually begins in primary care, but surgical approaches for weight loss may be considered when initial measures have failed or the patient’s degree of obesity and presence of comorbid conditions are great. The most common types of bariatric surgical procedures include sleeve gastrectomy and gastric bypass along with adjustable gastric banding, which is technically a medical device. Evidence from trials in HICs has shown that these strategies are clinically effective for moderately to severely obese patients compared with nonsurgical interventions (Picot and others 2009). However, further research is needed to provide data on patient quality of life, impact of surgeon experience on outcome, late complications leading to reoperation, duration of comorbid condition remission, and resource use.

The strength of the evidence for these select interventions has been summarized based on data from the Australian Assessing Cost-Effectiveness (ACE) in Obesity and ACE Prevention studies (Gortmaker and others 2011), which assessed the effectiveness and cost-effectiveness of various obesity prevention interventions in Australia (table 7.1). A scale from 1 to 5 was used to rank evidence (with 1 = strongest; 5 = weakest), and the assessment was made by a stakeholders’ group. Strategies with the greatest strength of evidence included gastric banding in children and adults, pharmacologic therapy (orlistat), combined diet and exercise, weight-loss diets (low fat, Weight Watchers), and a family-based targeted intervention for obese children. The evidence for unhealthful food and beverage taxes and front-of-package nutrition labeling was considered weak, suggesting a need for additional data.

Table 7.1

Effectiveness and Cost-Effectiveness of Select Obesity Prevention Interventions from the Australian Assessing Cost-Effectiveness (ACE) in Obesity and ACE Prevention Studies.

ACE Obesity Study

In the ACE study, a decision threshold of $A 50,000 (US$49,500) per DALY averted was used to establish whether an intervention was cost-effective, which is in line with empirical evidence on what constitutes acceptable value for money in Australia. Use of standard methods improves the comparability of results, although lower strength of evidence for many interventions limits the generalizability of findings, and costs can vary.

Of the 20 interventions evaluated, 8 were found to improve health and save costs: taxes on unhealthful foods and beverages, front-of-package nutrition labeling, reduction of advertising of unhealthful foods and beverages to children, school-based education programs to reduce television viewing, multifaceted school-based programs including nutrition and physical activity, school-based education programs to reduce the consumption of sugar-sweetened beverages, family-based targeted programs for obese children, and multifaceted school-based targeted programs for overweight and obese children (table 7.1). Gastric banding in adults and adolescents as well as family-based, general practitioner–mediated programs for overweight and obese children were found to be very cost-effective in that they improved health at a cost of less than $A 10,000 per DALY averted, while multifacted school-based programs without a functioning physical activity component, diet and exercise, and low-fat diets were found to improve health at a cost of between $A 10,000 and $A 50,000 per DALY averted. The top three cost-saving interventions—food and beverage tax, nutritional labels, and reduction in advertising to children, all of which are environmental—showed modest effects at the individual level but were highly cost-effective because benefits accrue to the entire population and the cost of implementation is relatively low. However, these interventions vary in their effectiveness and in the likelihood of implementation. In Australia, regulation of advertising is not on the political agenda, which means that reducing advertising to children, one of the most cost-effective interventions, is unlikely. Also, policy makers in Australia considered the evidence for front-of-package nutrition labeling to be insufficient to warrant support, despite plausible outcomes.

Overall, based on assessments made in the ACE study, policy approaches generally showed greater cost-effectiveness than either health promotion or clinical interventions. To prevent obesity, policy makers should consider the strategies found to be cost saving or highly cost-effective. However, the decision about whether to implement a specific obesity prevention strategy in a given country will be based on a combination of factors aside from cost-effectiveness, including the strength and generalizability of the evidence base, feasibility of implementation, impacts on equity, and acceptability to stakeholders. Because the evidence base is constantly evolving, particular emphasis should be given to identifying updated studies in the field. A good example of updated studies pertains to trials examining the effectiveness of low-fat diets on weight loss. Although the ACE study found strong evidence that low-fat diets are effective among overweight adults, recent meta-analyses of trials found negligible effects, suggesting that low-fat diets may not be an effective strategy for weight control (Tobias and others 2015). Other than for this intervention, data from the ACE study are generally consistent with data from other HICs for the same interventions.

OECD and WHO Study

In the OECD and WHO study, seven interventions aimed at tackling rapidly escalating rates of obesity through healthy dietary habits and increased physical activity—school-based health promotion, worksite interventions, mass media campaigns, counseling of individuals at risk in primary care, fiscal measures affecting the price of fruits and vegetables and foods high in fat, regulation of food advertising to children, and food labeling—were evaluated for their cost-effectiveness for preventing related chronic diseases (stroke, ischemic heart disease, and cancer, including lung, colorectal, and female breast cancer) in two time periods—20 years and 50 years—in six LMICs. Additionally, a prevention strategy combining a mass media campaign, fiscal measures, regulation of food advertising, and food labeling was assessed on the basis of the assumption that the effects of the individual interventions, measured by the relative risk of risk factors or chronic diseases, would combine multiplicatively. The analysis was based on a microsimulation model (chronic disease prevention model) that implements a causal web of lifestyle risk factors for selected chronic diseases. Whereas individual-level effectiveness was based mostly on studies from high-income settings, country-specific information was used to establish potential population coverage and to adapt effectiveness to the local distribution of risk factors.

Based on U.S. dollars per DALY averted, relative to a comparator situation of treatment only and no prevention, fiscal measures were consistently cost saving in all LMICs considered and generated the largest (for example, in China) or second largest health effects in both 20 years and 50 years (table 7.2). The health effect of fiscal measures was substantially lower in India than in other countries because Indians consume fewer foods high in fat. Food labeling was also cost saving in many settings, but with smaller health effects than fiscal measures. Regulation of food advertising to children and mass media campaigns had very favorable cost-effectiveness ratios. In 50 years, regulation of food advertising was cost saving in several countries, although its health effect was still very small compared with other interventions. Worksite health promotion initiatives had favorable cost-effectiveness, with quicker health returns than regulation of advertising, although returns were lower in some countries over the entire simulation. Physician counseling of individuals at risk in primary care was one of the most effective interventions, but its health effect was greatest and cost-effectiveness was highest in countries where a larger proportion of the population had regular access to primary care physicians and facilities. School-based interventions consistently had unfavorable cost-effectiveness ratios up to 50 years from their initial implementation. However, the cost-effectiveness of interventions targeting young children tends to improve in a longer timeframe (greater than 50 years), as these interventions realize their full potential in improving health. A multiple-intervention strategy would achieve substantially larger health gains than would individual interventions, often with an even more favorable cost-effectiveness profile. Such a strategy would be cost saving in about half the countries examined.

Table 7.2

Cost-Effectiveness Ratios of Select Obesity Prevention Interventions after 20 Years and 50 Years in Seven Countries: Organisation for Economic Co-operation (OECD) and WHO Analysis.

Taken together, findings from these two studies suggest that the most cost-effective approaches to preventing obesity and downstream chronic disease are price interventions and regulation, such as taxes on unhealthful foods and beverages and subsidies on healthy foods, food labeling, and regulation of advertising of unhealthful foods and beverages. As the OECD and WHO study found, fiscal measures are the only interventions likely to pay for themselves since they can generate larger savings in health expenditures than the costs of delivery. What sets these interventions apart from other more targeted strategies is their greater coverage in the population and relatively low cost of implementation. These strategies could feasibly be added to existing measures for preventing chronic disease, such as demand-reduction strategies for tobacco and alcohol (that is, higher excise taxes, advertising bans, and improved labeling) and salt-reduction strategies (via mass media campaigns or regulation of the salt content in manufactured foods). A strategy of several interventions would generate larger health gains and have a more cost-effective profile than would individual interventions. School-based interventions can be cost-effective strategies for preventing obesity, but their impact on future chronic disease risk may not be realized until many decades later. Regulation of food advertising to children would be a more effective and efficient strategy for targeting children.