The workshop’s first session was dedicated to framing the problem of environmental exposures and obesity. To do that, two speakers described the issue from two different perspectives: the public health perspective and the environmental health perspective.
The first speaker was William Dietz, the director of the Sumner Redstone Global Center for Prevention and Wellness at the Milken Institute of Public Health at George Washington University. He presented his observations via the telephone.
Background on Obesity
Dietz began by describing the standard criteria for obesity, which is based on body mass index (BMI). BMI is defined as weight in kilograms divided by height in meters squared. For example, a 6-foot man (or woman) who weighs 184 pounds has a BMI of 25, as does a woman (or man) who is 5 feet 4 inches and 145 pounds. A BMI of 25 is at the lower end of being overweight, which is defined as having a BMI from 25 to less than 30. A person with a BMI of 30 or above is said to have obesity. The classifications are not perfect, and a number of men are classified as overweight when in reality they simply have more muscle mass than normal. Thus, there is a lot of misclassification, particularly for men with BMIs between 25 and 30. However, “For both men and women,” Dietz said, “a BMI greater than 30 is invariably associated with increased body fat unless you play linebacker for the New England Patriots.”
The criteria for children are different because they are growing. A child who is above the 85th percentile in BMI for his or her age is said to be overweight, while those whose BMI is above the 95th percentile are said to have obesity. However, Dietz emphasized, these standards are based
on national surveys and historical data that were collected before the rapid increase in obesity. Thus, it is possible for 17 percent of children to have obesity, because 17 percent of a national sample of children from several decades ago have a BMI at or above the 95th percentile.
The percentage of children in the United States who have obesity has been increasing steadily since the late 1970s (see Figure 2-1). Currently, nearly 20 percent of children between 6 and 19 years old have obesity.
There have been similar increases in obesity among children in countries around the world. The worst problems have been in the developed world, but even in developing countries the prevalence of obesity has been increasing, although the prevalence still remains much lower than that in the developed world.
Obesity has also increased in the adult population. For instance, the prevalence of obesity among U.S. men doubled between 1976–1980 and 2003–2004. The increase was approximately uniform across Caucasians, African Americans, and Mexican Americans. There was a similar approximate doubling of the obesity rates among U.S. women during that same time, although the overall prevalence of obesity was significantly higher among women than among men (Flegal et al., 2010).
One of the prevalent misconceptions about obesity concerns its relationship with poverty, Dietz said. Among men, the prevalence of obesity differs very little among the various socioeconomic classes or among non-Hispanic whites, non-Hispanic blacks, and Mexican Americans. Interestingly, the only statistically significant relationship between socioeconomic class and obesity appears in African-American men and Mexican-American men, where upper-income men are significantly more likely to have obesity than those in other socioeconomic classes (Ogden et al., 2010).
The story among U.S. women is different. According to data from the 2005 to 2008 National Health and Nutrition Examination Survey, women in lower socioeconomic brackets are significantly more likely to have obesity than those in the middle or upper socioeconomic brackets. However, when broken into racial/ethnic categories, the relationship is significant only among non-Hispanic whites. There are no significant socioeconomic gradients for obesity among African-American women or Mexican-American women (Ogden et al., 2010).
Factors Leading to Obesity
The rapid increases in obesity seen over the past decade cannot be explained by genetics, Dietz said. More than 100 genes have been shown to be related to obesity, but those genes have always been present in the population in approximately the same proportions that exist today. Dietz explained that genes may affect susceptibility to obesity through their impact on the individual’s energy balance either by calorie intake or by calorie expenditure.
It does not take much of a shift in energy balance to produce obesity, Dietz said. For example, the shift in mean body weight of 2- to 5-year-olds since the 1970s can be accounted for by an excess of approximately 30 calories per day. However, the change in the energy balance necessary to reduce obesity is much greater than that necessary to produce it. Physical activity can play a significant role in changing body composition, but it is a poor way to lose weight because it is hard to achieve the major caloric deficits necessary for weight loss through physical activity.
Although on one level obesity is simply the product of an energy imbalance, it is actually extremely complex. Dietz illustrated this point with a slide showing various pathways inside and outside the body related to obesity. At the center of the illustration were the critical interactions between energy intake and energy expenditure, the
imbalance of which accounts for obesity. But many factors influence energy intake and expenditure, and these are loosely grouped into seven categories: food production, food consumption, societal influences, individual psychology, individual activity, the activity environment, and biology. Dietz said that he expected that much of the focus of the workshop would be on biological factors, which are various mechanisms that affect the pathways that regulate energy intake and expenditure. The nonbiological factors, in contrast, affect the susceptibility of individuals to an energy imbalance. They do not cause obesity in the traditional sense of biological agents, but they make it more or less likely.
Social and Behavioral Influences
For the rest of his presentation, Dietz focused on how the macro-environment and behavior influence susceptibility to obesity.
In the 1950s, he said, the typical diet consisted of milk and other dairy products, meat and eggs, potatoes, fruits, and vegetables—in short, mostly unprocessed foods that were prepared at home. Today, a much larger portion of the average American diet is highly processed: pizza, sodas, canned foods, and so on. The shifts in food practices from then until now have been enormous. There are a variety of reasons for this shift, Dietz said, including the increased availability and lower cost of highly processed foods as well as increased portion sizes.
“All of these factors promote increased food intake,” Dietz said. “The more variety an individual is exposed to, the more likely [he or she is] to overconsume foods. The greater the portion size, the more likely we are to overconsume,” Dietz said. Because of the reduced consumption of unprocessed foods like fruits and vegetables, higher-calorie foods account for a greater part of the diet.
There have been comparable changes in physical activity, Dietz said. They may be less quantifiable, but they are nonetheless important. Physical education and recess have been eliminated or reduced in schools. The time spent in front of television and computer screens has increased, particularly in children. The use of appliances has displaced what used to be household activities, like washing the dishes and hanging up the clothes to dry. The movement of large numbers of people to suburbs means that children are less likely to walk to school because many suburbs lack sidewalks, and even if they have sidewalks, they do not connect people with places where people want to go. As a result, people are increasingly reliant on cars. Research has shown that the more
time that someone spends in a car, the more likely he or she is to have obesity. Finally, because society is now in a postindustrial era, there has been a shift from manufacturing to services, which has led to a decrease in the amount of energy that people spend on performing physical activities.
Maternal behavioral factors associated with obesity in a child include a higher prepregnancy weight, excessive weight gain during pregnancy, gestational diabetes, and tobacco use during pregnancy.
Finally, early exposure to various adverse experiences—including physical and verbal abuse, family incarcerations, divorce, poverty, drug use, and alcohol use—is associated with an increased prevalence of severe obesity in adulthood. The connection between these early experiences and obesity in adulthood is in part due to the effects of the experiences on brain development.
There have recently been some encouraging data about obesity, Dietz said. For example, over the past decade or so there has been a plateau in the prevalence of obesity in both boys and girls 2 to 19 years old. Similarly, over the past decade there have been no significant increases in obesity among adults, either men or women. What is particularly encouraging, he said, is that recent data indicate that the obesity rate among 2- to 5-year-old U.S. children has actually started to drop, after nearly three decades of increases.
Local data have shown a similar trend. Although the quality of the local data is much more variable, six states and 16 communities have reported that the rates of childhood obesity have dropped. “There is still work that needs to be done to validate the samples to assure that the samples themselves are comparable,” Dietz said, “but I think we can say in some of the states and communities which have been examined pretty intensively … there are significant decreases in the prevalence of childhood obesity.”
Why have the rates started to decrease? Researchers are beginning to look into that question, Dietz said. One factor seems to be that nationally there have been substantial changes in food consumption. Between 1999–2000 and 2009–2010, the average consumption of sugar drinks in the United States dropped by 68 calories per day among 2- to 19-year-olds and by 45 calories per day among adults (Kit et al., 2013). Between 2003–2004 and 2007–2008, fast food consumption in the United States
dropped by 64 calories per day among 2- to 11-year-olds, by 14 calories per day among 12- to 19-year-olds, and by 33 calories per day among adults (Powell et al., 2012). An agreement between the companies that supply 25 percent of the calories in the United States and the Healthy Weight Commitment Foundation pledged in 2010 to reduce the number of calories in the U.S. food supply by 1.5 trillion calories; the actual reduction achieved in 2014 was 6.4 trillion calories, or 78 calories per person per day.
These changes in consumption could clearly account for flattening of the obesity prevalence curve, Dietz said. However, he said, “that is no cause for complacency because we still have a prevalence of about 20 percent obesity in 6- to 11-year-olds and about the same, maybe a little more, in 12- to 19-year-olds, and about 34 percent of the adult population is obese.” Thus, there is still work to do.
The next speaker was Jerry Heindel, a health science administrator in the Division of Extramural Research and Training at the National Institute of Environmental Health Sciences (NIEHS). He provided an environmental health perspective on the current obesity epidemic.
A Brief History of the Field
Heindel began by offering a brief history of the field of environmental exposures and obesity. The field got its start in 2002, he said, with a review article by Paula Baillie-Hamilton, “Chemical toxins: A hypothesis to explain the global obesity epidemic” (Baillie-Hamilton, 2002). In that article, Baillie-Hamilton offered a compelling chart showing how closely the rise in the rates of obesity correlated with the increase in chemical production, with a certain lag time (see Figure 2-2). “Of course, it was just a correlation,” Heindel said, “but it pulled together a lot of data and got people thinking.”
Even more compelling than the chart, however, was the literature search that Baillie-Hamilton had carried out. In that search she identified a number of toxicological studies going back to the 1970s and 1980s that had shown that various chemicals increased weight in experimental subjects. The types of chemicals on her list included pesticides, such as organophosphates and carbamates; polychlorinated biphenyls (PCBs); polybrominated biphenyls and fire retardants; heavy metals; solvents; and plastics, such as phthalates and bisphenol A (BPA). At the time that the studies had been done, Heindel said, no one was paying attention to increased weight in the subjects because the focus was on the decreases in weight and the general toxicity caused by high doses of the substances.
Then, over the next few years there were several commentaries on the subject published, and NIEHS funded an initiative to understand the fetal basis of disease, which included the fetal origins of obesity in its
purview. In 2004, Heindel and Ed Levin of Duke University held the first symposium on the fetal origins of and environmental influences on obesity. The biggest change occurred in 2006, when Bruce Blumberg of the University of California, Irvine, wrote a review article and coined the term “obesogen,”1 Heindel said. “I think that really stimulated the field because it caught on in the press.”
In just the past few years, NIEHS has funded another initiative on the role of environmental chemicals in the development of obesity, type 2 diabetes, and the metabolic syndrome with the goal of stimulating new research in the field. That initiative is ongoing.
Background Information on Obesity: Setting the Stage
Next, Heindel offered some background on obesity and its causes. A number of factors are involved in the development of obesity, he said, including one’s genetic background, congenital illness, drug use, viruses, antibiotics, and various environmental factors, including a lack of exercise, stress, a lack of sleep, and nutrition. The focus of his talk, he said, was on one particular environmental factor that leads to obesity: exposure to environmental chemicals.
Body weight is controlled by the endocrine system, Heindel explained. The endocrine system is highly complex and interrelated. There are hormones that dictate appetite and satiety as well as the development of adipose tissue. Because it is a finely tuned system, endocrine-disrupting chemicals can throw off its operation and lead to weight gain.
An endocrine disrupter, Heindel explained, is defined as an exogenous chemical or mixture of chemicals that interferes with any aspect of hormone action. More than 800 chemicals are now known to have some endocrine-disrupting activity. These chemicals fall into more than a dozen different classes, according to their intended uses, including pesticides, herbicides, flame retardants, plastics, plasticizers, surfactants, solvents, heavy metals, personal care products, sunscreens, and cosmetics. The point, Heindel said, is that these chemicals were designed for a specific purpose, but they also have the side effect that they can interfere with some aspect of the endocrine system.
1 “Obesogen” refers to chemical compounds that may have an impact on metabolic processes or may increase individuals’ susceptibility to obesity, or both. The term is used in this Proceedings of a Workshop in the manner in which the researchers used it at the workshop.
Do these chemicals in the environment make it into the human body, and is there enough exposure to these chemicals that some effects could be expected? The data say yes, Heindel said. He mentioned in particular a study from the Centers for Disease Control and Prevention (CDC) that found measurable amounts of nearly 300 different chemicals in cord blood from babies. In addition, a small study of 50 pregnant women found 47 chemicals in every one of the women tested. Further, some chemicals were found in every one of thousands of people tested by CDC. “Certainly,” he said, “there is significant exposure to these endocrine-disrupting chemicals.”
Heindel cautioned that the presence of these chemicals in the womb does not mean that they are causing any harm. However, he added, “it does mean we have accepted a strategy whereby every pregnant woman is contaminated with a variety of chemicals without her knowledge with the potential for harm to either her or the baby.”
The Obesogen Hypothesis
Data collected over the past 10 or 12 years clearly show that developmental exposures to environmental chemicals can lead to a variety of diseases and dysfunctions later in life, Heindel said. In particular, the period of development that takes place in utero and early in childhood is the period when the human body is most sensitive to exposure to environmental chemicals, and such exposures can disrupt development in ways that cause problems long after the chemicals are gone.
A variety of diseases are caused by such developmental exposures, Heindel said, and he believes that obesity is clearly one of them. He pointed out that although there are chemicals that will cause weight gain in adults, it is believed that the developmental stage is much more sensitive to metabolic disruptions and the development of obesity, and as a result, the field has been focused on developmental exposures to chemicals linked to an increased likelihood of obesity later in life.
This is the obesogen hypothesis: that the obesity epidemic is due, in part, to environmental exposures during development. In particular, the hypothesis is that a subset of endocrine-disrupting chemicals, which are called obesogens, act during development and disrupt adipose tissue development in such a way that the disruption alters the number of fat cells. The chemicals can also alter subsequent food intake and metabolism by having effects on the pancreas, adipose tissue, liver,
gastrointestinal tract, brain, or muscle. The ultimate result is that these environmental endocrine-disrupting chemicals alter the programming of the body’s set point or its sensitivity to the development of obesity later in life.
This is a very important point, Heindel said. The chemicals are not causing obesity per se, but rather they play a role increasing the body’s sensitivity to the development of obesity. “It is very important that you all realize that we who are working in this field understand that food intake and exercise are very important and that they are certainly key to the obesity epidemic,” Heindel said. “But we believe that environmental chemicals are altering the set point or sensitivity for gaining weight—that is, how much food does it take to put on weight and how much exercise does it take to reduce weight. Those effects are occurring via alterations in this developmental programming of this endocrine system that controls weight gain.”
Examples of Obesogens
There are a number of examples of such obesogens, Heindel said, with clear data showing a connection between environmental exposures and obesity. For example, more than 20 different epidemiological studies have shown that cigarette smoking by a mother during pregnancy results in her child having an increased likelihood of being obese. The obesity generally shows up at about the time that the child starts school, he said.
There is some interesting evidence related to prenatal exposure to diethylstilbestrol (DES). This is a drug that was given to millions of women to prevent miscarriage. It did not actually help with that, but it did cause a number of different diseases and dysfunctions, including some very rare cancers, in the children of the mothers who took it.
Although not demonstrated in humans, animal models have demonstrated that one of the possible effects of prenatal exposure to DES is obesity. In one experiment, newborn mice were given DES for 5 days beginning at birth. Once the exposed mice hit puberty, they began gaining weight significantly faster than control animals that did not get the drug. Then, by the time the exposed mice were 9 months old, they were morbidly obese. Interestingly, they got fat without eating any more or exercising any less than the control mice.
Other animal studies have contributed to a growing body of evidence indicating that environmental exposures can increase susceptibility to obesity. BPA, a chemical used to make various plastics such as the ones
used in water bottles, has shown a slightly different effect in mice. In one series of experiments, BPA exposure did not lead to weight gain but, rather, led to an increase in the percentage of body fat and a decrease in the percentage of lean body mass.
Another study of BPA looked for the mechanisms behind the chemical’s effect. In this case, the researchers used a different animal model, and they did see increased weight in the animals exposed to BPA as well as increased food intake. When the researchers examined the brains of the animals, they found an increased number of appetite neurons and a decreased number of satiety neurons, indicating that changes to the numbers of appetite and satiety neurons may have been how BPA exerts its influence.
Yet another study looked at exposure to diethylhexyl phthalate (DEHP), a plasticizer found in various plastics, including plastic toys. Developmental exposure to the chemical increased visceral fat tissue and also the number of fat cells in an animal model. The increase in weight in the exposed animals was relatively minor, but they had huge amounts of fat filling up their abdomens.
That experiment illustrated an important point about environmental exposures. The development of increased fat actually occurred at the lowest dose tested. At the highest dose—500 milligrams per kilogram, which is the usual dose that toxicologists use to look for effects of these chemicals on different systems—there was no increase in fat. The lesson, Heindel said, is that the experimenter must pay attention to the effects of very low doses because in many cases the dose–response curves are not linear.
In addition to animal models, Heindel said that about 33 human epidemiology studies have now linked developmental exposure to environmental chemicals to weight gains in children later in life. The chemicals that have exhibited such effects include PCBs, BPA, hexachlorobenzene, polycyclic aromatic hydrocarbons, and the chemicals produced by maternal smoking.
Recently, some troubling data in lab animals indicate that it is possible to have transgenerational inheritance of obesity—that is, that the chemically produced obesity can be passed along to subsequent generations. In these experiments, a pregnant female is exposed to an environmental chemical, and the offspring are examined for effects. Then, the male offspring are mated with females that have had no such exposure to get a third generation, and the process is repeated. Obesity from the second generation shows up again in the third and fourth
generations in animals that were never exposed to the chemical at all, Heindel explained. Some studies have shown such an effect with tributyltin, the pesticide dichlorodiphenyltrichloroethane (DDT), jet fuel, and a mixture of BPA and two phthalates.
This is very troubling, Heindel said, because it indicates that if a pregnant mother is exposed to a chemical, it may affect not only her children but also her grandchildren and her great grandchildren as well.
At this point there is a large list of chemicals for which data for either humans or animals suggest a metabolic disruption or an obesogenic effect, and it seems to be just the tip of the iceberg, Heindel said, because it seems that a new chemical is being added to the list every month or two.
Data Gaps and Needs
NIEHS is now funding 57 grants in the area of obesity and diabetes, Heindel said. Of those, 32 are in humans, 20 are in animals, and 5 are basic cellular and molecular studies. In the 32 studies with human birth cohorts, developmental exposures to various chemicals are assessed, and the children are followed later in life to see if they become overweight or obese. Thus, in the next 4 or 5 years, he said, there will be a huge increase in the amount of data available on this issue both from the human studies and from the animal studies, plus all of the other studies being funded around the globe.
Still, he added, the field is still young—only about 10 years old—and there are many data gaps and needs and many questions to be answered: screens need to be developed to determine which chemicals have the ability to cause weight gain, for instance; dose–responses need to be determined; the sites at which the chemicals act and their mechanisms need to be discovered; the animal studies need to be coordinated with the human studies so that their insights can be compared and combined; and so on.
Heindel closed by noting that because the field is so new, there are many opportunities to help direct the research. That is why meetings like this workshop are so valuable, he said, so that it is sooner, rather than later, that the field is able to understand the importance of environmental chemicals in the obesity epidemic.
Lynn Goldman of the Milken Institute School of Public Health at George Washington University opened the discussion session with a
comment about the relatively modest changes in caloric consumption or energy use that are necessary to either go from a normal weight to obese or move from obese to a normal weight. This would seem to have profound implications for policy, she said, and she asked both speakers to comment.
Heindel answered that the fact that it takes only a small increase in calories to result in weight gain over time helps people to accept the idea that environmental chemicals can play a role. For the most part, he said, the effects of environmental chemicals are not large; they are just increasing the susceptibility or altering the set point. But that has the ability to play a large role if only a small change in calorie intake is necessary to lead to obesity.
Dietz said that it is important to distinguish between levels of obesity. A person with a BMI of 32 is obese and could have gotten there with a daily excess of only 200 calories or so, but for a person with a BMI of, say, 44, the caloric gap would have to have been much greater—perhaps an additional 700 or 800 calories a day. This is also what they will need to cut from their diets to eventually return to a normal weight. It is a mistake to assume that all obesity is the same, for example, that a person with a BMI of 32 is the same as one with a BMI of 44, in terms of either the factors that got them there or the metabolic consequences. “I think we need to be much more sophisticated about understanding the different phenotypes of obesity and what the contributing factors are,” he said.
Bernie Goldstein of the University of Pittsburgh asked about the effects of DES on the children of mothers who took it. Is there any evidence that they were more likely to become obese?
Heindel answered that several researchers have been trying for many years to link DES exposure in women to weight gain in the offspring, but they have not been able to pin that down. The problem, he suggested, is that there is little information about the doses that these women received, and there was a huge variation in both the dosing and the timing of the dosing from woman to woman. Some received DES in the first trimester, some in their second, and some in their third, so no one has been able to prove a link. However, he noted that there are many anecdotal, but interesting, data about people who are morbidly obese, have been for their whole life, and did not know why. It turns out that they were the daughters of mothers who had taken DES.
Linda Birnbaum of NIEHS commented that, historically, one of the problems with developmental toxicology studies is that the animals were
not held long enough to see obesity develop. They were sacrificed just before birth, soon after birth, at weaning, or maybe even at puberty, but they were not kept until the time that a big difference in weight gain would become apparent. But, she said, with some of the new paradigms that are being used, researchers are doing long-term studies, starting with developmental in utero exposure and holding the animals until they are 2 years of age. Given that, researchers should start seeing some things that have been difficult to see before.
Birnbaum then asked Heindel whether preconceptional exposures might play a role in obesity. She also asked if exposures at puberty might have an effect, because that seems to be another time when there is increased susceptibility to endocrine-disrupting chemicals.
Heindel said that although the field began with a focus on developmental exposure in utero or neonatally, in the past 2 or 3 years researchers have begun to realize that there are probably other sensitive windows, such as paternal exposure and maternal exposure before pregnancy or prepuberty. “Any time there is a huge change in hormone levels,” he said, “there are going to be major changes in epigenetic regulation. If environmental chemicals can perturb that process, then the end result will be some problem later on. As we move forward, we are certainly going to look at other windows of exposure.”
Barbara Corkey of Boston University commented that the epidemiological studies that had been discussed during the session could form the basis for some interesting hypotheses, but they did not actually prove causation. Heindel responded that she was correct, that the field is very descriptive at this point. The hope is that researchers will work from the correlations in the epidemiological studies and test those in animal models in an attempt to show causality between the particular chemical exposure and weight gain. Once that happens, there will a much better understanding of exactly what is going on.
Birnbaum added that one of the most important directions in this area is the move toward looking at multiple chemical exposures. “None of us are exposed to one potential obesogen by itself,” she noted. “With some of the Tox21 [the Toxicology Testing in the 21st Century project] approaches that are being used by EPA [the U.S. Environmental Protection Agency] and NIEHS—and FDA [the U.S. Food and Drug Administration] is partnering with us as well—we are not testing 1 or 10 or 100 chemicals, but we are testing thousands and thousands of chemicals through large numbers of assays. And in fact, within the past couple of years, we have added assays that are involved in the integrated
pathways that are associated with obesity. I think that will at least provide us a great deal of screening prioritization, but eventually actual understanding.”
Baillie-Hamilton, P. F. 2002. Chemical toxins: A hypothesis to explain the global obesity epidemic. Journal of Alternative and Complementary Medicine 8(2):185–192.
CDC (Centers for Disease Control and Prevention). 2000. CDC Growth Charts: United States. Atlanta, GA: Centers for Disease Control and Prevention. http://www.cdc.gov/growthcharts/background.htm (accessed March 23, 2016).
CDC. 2014. National Health and Nutrition Examination Survey: Questionnaires, Datasets, and Related Documentation. Atlanta, GA: Centers for Disease Control and Prevention. http://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm (accessed March 23, 2016).
Flegal, K. M., M. D. Carroll, C. L. Ogden, and L. R. Curtin. 2010. Prevalence and trends in obesity among U.S. adults, 1999–2008. JAMA 303(3):235–241.
Kit, B. K., T. H. I. Fakhouri, S. Park, S. J. Nielsen, and C. L. Ogden. 2013. Trends in sugar-sweetened beverage consumption among youth and adults in the United States: 1999–2010. The American Journal of Clinical Nutrition 98(1):180–188.
Ogden, C. L., M. M. Lamb, M. D. Carroll, and K. M. Flegal. 2010. Obesity and socioeconomic status in adults: United States, 2005–2008. NCHS Data Brief No. 50. Hyattsville, MD: National Center for Health Statistics. http://www.cdc.gov/nchs/data/databriefs/db50.pdf (accessed March 23, 2016).
Powell, L. M., B. T. Nguyen, and E. Han. 2012. Energy intake from restaurants. American Journal of Preventive Medicine 43(5):498–504.
This page intentionally left blank.