Population Health: Behavioral and Social Science Insights

By Wendy Baldwin

Abstract

Developing countries are undergoing an epidemiological transition, with a shift from mortality predominately driven by infectious diseases to mortality driven largely by non-communicable diseases (NCDs). NCDs—primarily cardiovascular disease, diabetes, chronic respiratory disease, and most cancers—are strongly influenced by social and behavioral factors. These behaviors, which have multiple drivers, are tobacco use, excessive alcohol use, poor nutrition, and sedentary lifestyle. The increasing impact of these behaviors on mortality highlights the importance of a lifecycle approach to lowering risk of disease. While there is much to do at different life cycle stages, there is a window of opportunity to focus research, policy, and programmatic attention on adolescence. Adolescence is the time when both tobacco and alcohol use typically begin, and it is a time of increasing independence. It is also a time to ensure that healthy eating and regular exercise are solidified as habits for a lifetime. At a time of rising concern about the current and future burden of NCDs in the developing world, there are a number of opportunities involving data, research, programmatic, and policy initiatives centered on the role of adolescents that should be pursued.

Introduction

The predominance of death from non-communicable diseases (NCDs), as opposed to infectious diseases, is the hallmark of the epidemiological transition.¹ In high-income countries, this transition occurred generations ago, and now NCDs generally represent diseases and causes of death of the elderly. However, NCDs are not just diseases of older, wealthier populations; they are now poised to be the dominant cause of death in developing countries. The earlier onset of NCDs in low-income countries means that deaths and disability occur during economically productive ages. Since the behavioral risk factors that drive most of these diseases can be addressed during adolescence, there is an opportunity for primary prevention. While primary prevention does not eliminate the need for secondary prevention and treatment of disease, it does offer an opportunity to lower the burden on individuals, families, health systems, communities, and nations.

The burden of NCDs on the developing world has been recognized by the World Health Organization. In 2011, the United Nations General Assembly held a special meeting on the topic, only the second such meeting focused on a health issue.² Notably, the first was 10 years earlier and addressed HIV and AIDS. There has been continuing attention from WHO on the challenges that NCDs present for low- and middle-income countries, as well as for more wealthy nations, as can be seen in the NCD action plan which includes specific targets for improved use of medications as well as reductions in risk factors.³ The WHO action plan provides a wide-ranging blueprint of actions, targets, and potential actors that, together, could be a formidable global response to the crisis NCDs represent. Clearly, much needs to be done to strengthen screening, diagnosis, and treatment. Also, for a comprehensive response, many sectors outside of the health care arena need to be involved. One way to heighten attention to primary prevention and articulate what other sectors could do is through attention to adolescence as a critical life transition point and an opportunity for intervention to lower lifelong risks for NCDs.

While the focus in this chapter is on the potential for intervention during adolescence, other life stages also represent important intervention points. A broad life-cycle approach has many potential benefits, since there is a growing literature linking nutrition in early years—even prenatally— to diet and the ability to maintain a healthy weight in later life. However, two of the main risk factors—tobacco and alcohol use—typically begin during adolescence, and as young people take more responsibility for their lifestyle choices, adolescence may be an ideal time to reinforce healthy habits. Establishing and/or reinforcing healthy habits during adolescence is not without challenges, but changing well-ingrained habits later in life is likely more difficult. Adolescence may not be the best time to instill positive behaviors, nor is it the last time, but it presents a valuable window of opportunity at a critical life cycle stage.

Non-Communicable Diseases in Low- and Middle-Income Countries

Developing countries—that is, the low- and middle-income countries (LMICs)— now face an increase in the proportion of deaths that result from NCDs (Figure 1). Of course, this would be expected with the decline in infectious diseases. But, the proportion of deaths attributable to NCDs is only part of the story. The mortality rate—deaths per 100,000 population of all ages—is 705 for males and 520 for females for the world as a whole. However, in Sub-Saharan Africa the comparable figures are 869 and 746.⁴ It might be tempting to think that these are diseases of the elderly. Death is still a lamentable (but unavoidable) outcome for all of us, but age of death does vary and can be influenced. In LMICs, the age of death from NCDs is far earlier and is more likely to occur during the economically productive years than in higher income countries. Premature death is an additional trauma for families that may rely on that individual for their economic well-being, happiness, and even survival. In high-income countries, 13 percent of deaths due to NCDs were premature, i.e. before the age of 60. However, in middle- income countries, the proportion of premature NCD mortality is 28 percent, and in low-income countries it is 41 percent.⁵ Furthermore, the morbidity that often precedes a death caused by an NCD can include serious, life- altering consequences, such as amputations from complications of diabetes or disability from chronic cardiac or respiratory diseases.

Figure 1. Projected burden of deaths from chronic diseases

Source: Baldwin W, Kaneda T, Amato L, et al. Noncommunicable diseases and youth: a critical window of opportunity for Latin America and the Caribbean. Washington, DC: Population Reference Bureau; 2013. Used with permission.

The economic impact of NCDs has been assessed by the World Economic Forum, which projects the cumulative impact by 2030 of $47 trillion or 5 percent of the global GDP.⁶ In fact, the costs of addressing NCDs could actually undermine the economic growth that has been shown to date. The World Bank has stated that "the developing countries cannot afford to treat their way out of this,"⁷ and the United Nations Development Program has declared that the rise in NCDs threatens those economic gains made in the developing world.⁸

It is appropriate to ask about the ability of health systems to meet this challenge, and clearly, they will have to be strengthened. NCDs are typically, but not always, chronic diseases that may have an onset years before a death occurs. In addition to the impact on the individual and the family, this also means that there is a prolonged impact on the health care delivery system. Much of health care around the world is provided by the family, but the advances in drugs to help manage diabetes, hypertension, cardiovascular disease, chronic respiratory diseases, and many cancers bring expectations for the health care delivery system as well. Discovery of elevated blood glucose, blood pressure, or cholesterol usually comes with access to a robust primary care, including specific screening programs. The potential for management of risk (secondary prevention) through medications is quite impressive, but there needs to be health education, availability of essential medicines, and access to a health system that is able to sustain potentially decades of need for those affected.

Lessons learned from dealing with diseases such as HIV infection and AIDS can be helpful, but the potential costs of strengthening health systems to meet the kinds of demands NCDs will present is daunting. Also, the low- and middle-income countries have not conquered the challenges of infectious diseases, and thus, many will face a "double burden" of disease. The outbreak of Ebola in West Africa in 2014 is a reminder of the potential of infectious diseases and the limitations of health systems in many countries.⁹ A growing awareness of the burden imposed by NCDs is leading to increased surveillance of mortality and risk factors. The Institute for Health Metrics and Evaluation at the University of Washington has been a leader in evaluating the current and future burden of disease. In their report "Global Burden of Disease 2010," they concluded there were several big challenges, all of which are relevant for the present analysis.¹⁰ First, there is the demographic transition and projected longer lifespans that make healthy aging critically important. Second, there is an epidemiologic transition that will result in the move toward NCDs as the overwhelming cause of death and a shift to concern for disability and not just mortality. The growth in disability will be influenced by the extent to which healthy lifestyles are established and maintained. Also, there is a change in the risk factors from those that reflected environmental factors (e.g., clean water) to those that are behavioral (e.g., healthy lifestyle). It is not surprising that the WHO has focused on the four behavioral risk factors that form the basis for the rise in NCDs. Finally, while health systems are facing enormous challenges, attention to primary prevention of NCD risks involves many other sectors besides the health system that can lower future burdens on the health system itself.¹⁰

The World Bank has projected that there will be a slight rise in deaths attributed to NCDs in high- income countries.¹¹ They expect the rise to be slight because the proportion is already so high (Figure 1); in other parts of the world, however, the increase will be substantial. In Latin America and the Caribbean, the proportion of deaths due to NCDs is fairly high now at 72 percent, but this figure is projected to rise to 81 percent by 2030. In South Asia the current level is about half (51 percent), but it is projected to rise to 72 percent by 2030. Africa is still heavily influenced by infectious diseases, so the proportion of deaths due to NCDs is only 28 percent, but the increase will be stunning over the next 15 years or so, to 46 percent by 2030.

The continued rise in deaths attributable to NCDs is foreshadowed by the rise in risk behaviors, such as tobacco use, obesity, and sedentary lifestyle. These risk behaviors are typically higher in urban areas than in rural areas, so the trend toward greater urbanization portends rising risk factors. Also, unlike in industrialized countries where these risks are often higher in lower income groups, in developing countries the levels are often higher in higher income groups.^12,13 The National Institute on Aging has supported considerable research on non-communicable diseases in the elderly in low- and middle-income countries and observed the pattern that as economic conditions improve in a country, more people may have discretionary income to spend on tobacco, alcohol, and higher calorie food.¹⁴ They may be able to alter the conditions that kept them physically active (car ownership rises with rising income) and so their risks rise. Urban areas that are magnets for growth may provide fewer opportunities for physical activity and healthy diet. Many developing countries have growing economies, and as people are able to move into a better standard of living (frequently in an urban area), we can expect to see the risk factors for NCDs rise as well.

The World Health Organization focused on the four NCDs that are driving the epidemiological transition—diabetes, cardiovascular disease, chronic respiratory disease, and most cancers.¹⁵ These four NCDs are responsible for 80 percent of the NCD mortality rate, and they also tie to four shared underlying risk factors: tobacco use, excessive alcohol consumption, unhealthy diet/obesity, and sedentary lifestyle.

Tobacco use is the single most important modifiable risk factor; cigarette smoking is responsible for 71 percent of lung cancer deaths, 42 percent of chronic respiratory disease, and 10 percent of cardiovascular disease (mortality). Excessive use of alcohol has many unhealthy sequelae, including heart disease and some cancers. Unhealthy diet—i.e., a diet that is high is sugars, saturated fats, trans-fatty acids, and salt and low in fruits and vegetables—and resulting obesity are linked to high blood pressure, type 2 diabetes, and heart disease. Poor diet and low physical activity often result in obesity, which is now viewed as a global epidemic.¹⁵

Obesity is not just a health risk, it is also a reflection of reduced ability to function productively. Finally, sedentary lifestyle is estimated to result in a 20-30 percent increase in all-cause mortality.⁵ Two of these risk factors—tobacco and alcohol use—typically begin during adolescence, and adolescence provides an opportunity to establish and solidify good eating and activity levels. This is not without challenges, but perhaps it is not as daunting as waiting and trying to change deeply entrenched behaviors later in life.

In addition to their role in the four NCDs described above, these risk behaviors affect other health issues as well. Tobacco use is a risk factor for poor pregnancy outcome, for example. Excessive alcohol consumption is often a trigger for interpersonal violence and for accidents and injuries, especially those that are traffic-related. Road traffic accidents and interpersonal violence are the leading causes of death among young people worldwide.¹⁶ The link to productivity can perhaps best be seen in the fact that in the United States, the major health reason for young men and women being refused entry into the military is overweight or obesity.^a,17

Risk Behaviors Among Adolescents

A first step is to understand the risk behaviors of youth. There are several key data sources that provide insight into the level and nature of compromising health behaviors among adolescents in the developing world. The Global Youth Tobacco Survey (GYTS) is a school-based survey that collects data on students aged 13-15 using standardized methods across the multiple countries implementing the survey.¹⁸ That ensures that the method for establishing the sampling frame, selecting schools and classes, and processing of data are comparable. The structure of the sample allows for national estimates with regional level stratification possible. While there is a common protocol, countries may add questions. The questionnaire is self-administered and address multiple measures of consumption along with other knowledge and attitude measures. To date, 145 countries have participated. Funding is provided by the Canadian Public Health Association, the National Cancer Institute (NCI), the United Nations Children's Fund (UNICEF), and the World Health Organization (WHO).¹⁹

The Global School-Based Student Health Survey (GSHS) was developed by WHO in collaboration with UNICEF, UNESCO (United Nations Educational, Scientific, and Cultural Organization), and UNAIDS (the Joint United Nations Programme on HIV/AIDS) and with technical assistance from the Centers for Disease Control and Prevention (CDC).²⁰ This, too, is a school-based survey conducted primarily among students aged 13-17. The purpose is to collect data on health behaviors and protective factors that can be used for international comparisons and the establishment of trends and serve as a basis for countries to develop policies, establish programs, and address youth health needs. This survey also uses a standardized sampling process and core questionnaire modules. In some locations, countries have added specific questions to supplement the core questions. Data are collected with a self-administered questionnaire that can be completed in one regular class period. The core modules are alcohol use, dietary behaviors, drug use, hygiene, mental health, physical activity, protective factors, sexual behaviors, tobacco use, and violence/unintentional injury. More than 100 countries have had people trained in the methodology, and 73 have completed at least one GSHS. More than 420,000 students have participated in the surveys.²⁰

A third survey provides international comparisons with data gathered from a household survey instead of a school- based survey. The Demographic and Health Surveys (DHS) have been conducted in more than 90 countries and, for many, in multiple waves.²¹ Funded by USAID (the U.S. Agency for International Development), the DHS are a valued resource for international comparisons of health and reproductive behavior. The core DHS survey is of women in the reproductive years, with data collection generally beginning at age 15. There are some country variations, and some countries also include men in their surveys. Sharing a common core protocol, there are country-country variations within the standardized format and training for implementing the DHS. Since the DHS typically includes adolescent women, it provides an additional data point for measures of obesity, smoking, and alcohol consumption.

Each of these surveys provides a view into the health behavior of youth. The GSHS and GYTS are school-based surveys, which could of course limit their coverage. However, school enrollment/attendance is increasing around the globe, helped by the focus given with the United Nations' millennium development goals and activities.²² But, there could be some gaps in coverage among population groups that have low school involvement. Also, the risk of gaps in coverage is greater when older teens are being considered, but even at young adolescent ages, this can be problematic if specific groups (girls or rural or ethnic groups) have low school enrollment. The DHS is primarily a survey of women in their reproductive years, although some sites include men in the survey. All three of these surveys generate data that are publicly available and built on a shared platform so that the survey design and questionnaires are comparable.

Individual countries, or groups of countries, may conduct other surveys of youth risk behaviors covering an entire country or specific regions, cities, or schools. These surveys provide an excellent complement to the broad national data and, depending on the local circumstances, may go into more details on specific risk factors. The STEPS (STEPwise approach to surveillance) provides a tool to gather data about NCD risks. Supported by WHO, it enables countries to gather data about youth risks, such as tobacco, alcohol, physical activity, obesity, and diet. STEPS is a valuable resource, but individual country surveys are not necessarily focused on youth.^b,23

We might conclude that basic surveillance of youth risk factors does exist for many countries, although not for all countries and not always for identical time periods and ages. There is also a need for more surveys and repeat surveys to assess changes over time in the levels of behaviors that place youth at risk for later disease. The desire for more and better data is universal, but it should not detract from the basic picture of risks that is before us.

Extent of Risk to Youth in Developing Countries

The Population Reference Bureau (PRB) has arrayed data for youth risk behaviors in LAC (Latin America and the Caribbean) to provide an overview of risks in that region (Figure 2). The measures were mostly from school-based surveys and involved 13-15 year olds.²⁴ There were 27 countries with recent (2008 or later) data on at least three risk categories. For each risk behavior, cut offs were determined to identify countries as being at high, medium, or low risk (technical notes to the PRB data sheet for a full discussion).

Tobacco

For tobacco use, countries were considered at high risk if the proportion of youth aged 13-15 smoking in the past 30 days was 16 percent or more and at medium risk if the proportion was 7-15.9 percent. For the region, eight countries were at high risk for tobacco use for male youth and five for females. In no case was the risk level for females greater than the risk level for males. Fourteen countries were at medium risk for tobacco for males and 13 for females. Canada, Puerto Rico, Antigua and Barbuda, St. Kitts-Nevis, and Brazil were at low risk for both males and females. Discussion of the definitions and cut offs for each risk factor can be found on the PRB Web site.²⁴

Research generally shows that early onset of tobacco use is associated with high levels of use in later life and greater difficulty in quitting. Also, it also signals a likely longer period of exposure to tobacco.^25,26 The ongoing accumulation of evidence about the risks of smoking and early onset of smoking has led to policy and programmatic strategies to impede initiation and facilitate cessation of smoking. There has been concern that the tobacco companies specifically target youth with advertising and marketing approaches.²⁷ One might conclude that there is no health benefit of tobacco, youth are smoking at significant and often increasing rates, and prevention efforts need to accommodate to the potential for directed marketing toward young people.

While there is less information about changes over time in risk factors, there are indications that tobacco use may be declining in some of the most developed countries. In the United States, for example, youth smoking declined between 2000 and 2011, but the decline has not continued.²⁸ In some countries where overall levels of smoking have declined, the gender differential has narrowed. Generally, males are more likely to smoke than females, but the rates for females seem to be rising. In some countries, such as Argentina and Uruguay, the rates of smoking for girls exceeds that for boys, with 27 percent of girls and 21 percent of boys in Argentina having smoked in the past 30 days; in Uruguay, the rates of smoking are 23 percent for girls and 16 percent for boys.²⁹

Overall, of those surveyed in the GYTS who were aged 13-15 at the time, the median percentage for having ever smoked cigarettes was 33 percent. One benefit of these data is that they also ask about age of initiation, which for most is before age 18 and for a significant number of respondents, is before age 10. For all countries surveyed, the median percentage who had their first cigarette before age 10 was 23.9 percent, although it is above 80 percent in parts of India.³⁰

Figure 2. Youth risk data sheet

Source: Baldwin W, Kaneda T, Amato L, et al. Noncommunicable diseases and youth: a critical window of opportunity for Latin America and the Caribbean. Washington, DC: Population Reference Bureau; 2013. Used with permission.

Alcohol

For alcohol use among adolescents, countries were identified as at high risk if the proportion of youth aged 13-15 who had consumed alcohol in the past 30 days was 40 percent or more and at medium risk if the proportion was 20- 39.9. For alcohol, as for tobacco, the measure chosen was use in the past 30 days.²⁴

Alcohol consumption is a more difficult risk behavior to assess than tobacco use, since there can be positive or at least neutral effects of moderate drinking. The "risk factor" is usually described as excessive or harmful alcohol consumption. While it might be good to have more nuanced measures of risk, we presumed that reporting alcohol use at ages 13-15 constituted some level of risk behavior.

In the Latin American/Caribbean countries included in the study,²⁴ 10 countries were at high risk for youth alcohol use for both males and females and an additional three countries for males only. Eleven countries were at medium risk for both males and females. Only El Salvador and Guatemala were at low risk for both males and females in the country as a whole.

Excessive alcohol use is clearly a risk factor for later disease as well as a trigger for interpersonal violence, including gender-based violence and road traffic accidents (Figure 3).³¹ Earlier onset of alcohol use, with or without a family history of alcoholism, is a risk for alcohol abuse or alcoholism.^32,33 This provides further reason for concern for drinking among adolescents, especially when alcohol use among 13-15 year olds is over 50 percent in a number of countries.²⁴

Figure 3. Onset of alcohol use and alcohol dependence in the United States

Source: Baldwin W, Kaneda T, Amato L, et al. Noncommunicable diseases and youth: a critical window of opportunity for Latin America and the Caribbean. Washington, DC: Population Reference Bureau; 2013. Used with permission.

Reports from the Global School-Based Student Health Survey²⁰ show the proportion of school age youth who report drinking in the previous 30 days. Some countries with largely Muslim populations—where alcohol use rates are very low—do not collect data about adolescent experiences with alcohol.34 Although some countries have similar rates for males and females, when they differ it is where males are more likely to drink than females. Perhaps more alarming is the percentage reporting binge drinking, i.e. five or more drinks at one time. For example, in Guatemala City, over 11 percent of school-going youth aged 13-15 reported drinking so much that they were "really drunk" one or more times in their life, a figure that was about the same for boys as for girls.³⁵

Diet, Physical Activity, and Body Mass Index

While the many aspects of diet are more difficult to measure and track across countries, it is possible to look at the levels of obesity or overweight, which is the typical outcome of a diet that is low in fruits and vegetables and high in sugar and processed carbohydrates combined with a sedentary lifestyle. Data from the Demographic and Health Surveys (DHS) show worrying levels and rises in body weight among women across many countries.³⁶ This may seem out of place for countries still battling malnutrition and childhood stunting, but India has as many obese in their population as they do stunted, a situation mirrored in many countries in the Middle East.³⁷

In the 22 countries in Latin America and the Caribbean where there were data for female obesity/overweight, none were below 10 percent of the population being overweight or obese; five were at moderate risk (10-19.9 percent), and 17 were at high risk, with 20 percent or more of the adolescent female population being overweight or obese.²⁴ Recent press attention has focused on Mexico, which now exceeds the United States in obesity,³⁸ but Mexico is taking innovative steps toward prevention, as well as steps to ban television ads promoting high calorie foods and soft drinks.³⁹

Data for physical activity among males and females in Latin America and the Caribbean show that 14 countries are at high risk for insufficient physical activity, with 70 percent or more youth reporting that they did not meet the WHO guidelines of 60 minutes or more a day for 5 out of the last 7 days.^24,40 Where there is a gender difference, it favors males—with males being more physically active. Levels of physical activity appear to fall with urban living and with many youth failing to meet the minimum levels of activity of 1 hour a day, at least 5 days a week. NCDs are a burden in the developing world now, and yet the underlying risk behaviors are increasing. This would lead us to expect an even greater burden of NCDs in the future if there is no improvement in the risk behaviors discussed here.

Searching for Successful Interventions

Although a comprehensive review of all interventions on these four risk factors—tobacco use, excessive alcohol consumption, unhealthy diet/obesity, and sedentary lifestyle—is beyond the scope of this paper, there are indications that where there is appreciation of the risks, there have been attempts to address them. These actions may be at the level of individual behavior change up to broad policy changes. There have been few rigorous evaluations of such interventions in the developing world, but there are some promising approaches. Interventions may be aimed at individual behavior change or focused on changing the social and cultural environment. Some broad initiatives may not have a specific focus on youth and, if aimed at the total population, may have a differential impact on youth. For example, analyses in different countries show that an increase in taxes on tobacco is associated with declines in use (Figure 4).⁴¹ Also, the impact on use appears greater for younger smokers (or potential smokers). In the United States, as well as in other countries, a growing awareness of the risks of tobacco have restricted where one can smoke, making it both more inconvenient and less acceptable to smoke. While there may be concern that adolescents would still seek to smoke as a sign of adult behavior or defiance, even in most high-prevalence countries, the majority of youth do not smoke, and rates of smoking have fallen in many countries that have launched broad anti-smoking campaigns.^42-44

Figure 4. U.S. cigarette prices vs. consumption, 1970-2012

Source: Chaloupka F, Hugan J. Cigarette prices and cigarette sales, United States, 1970-2012. Chicago: University of Illinois at Chicago, Health Policy 2014. Used with permission.

In Mexico, an increase in the tax on tobacco is estimated to have reduced consumption by 5 percent, and the relationship of taxes to consumption looks very similar for the United States, Mexico, and South Africa. Several countries have taken at least some of the proceeds from increased taxes to support public health programs or smoking cessation programs. These are not specifically focused on youth, but youth are more price-sensitive and so the impact is greater for them. Approaches that make it less acceptable or convenient to smoke can have the same overall effect. There are some useful approaches to explore, and this is an ideal time to do so when there has been a shift in marketing by tobacco companies away from the wealthier countries and to the developing world.⁴⁵

Primary prevention is often located in the broader social network, rather than just focused on individual behaviors. In the case of levels of physical activity, communities (which includes schools and religious organizations, as well as families) can ensure support for activities such as sports, dance, safe transportation, bike paths, and other opportunities for youth to be active. Mexico and other countries periodically close off major roads to car traffic to encourage recreational biking.^46,47 Several African countries, such as Ethiopia, have supported running clubs for youth. The success of Kenyan and Ethiopian runners in the New York and Boston marathons from 2000 to the present speaks to the feasibility and acceptability of running as an activity for youth.

While it is important to build positive food choices at the individual level, there are also activities at the community level that can be helpful. Clearly, schools have an opportunity to increase exposure to and support for healthy food choices. Broader campaigns, such as the "Less Salt, More Life" campaign in Argentina,⁴⁸ is an agreement between the government and 25 companies to reduce the salt content of packaged foods.49 One study that looked at the impact of taxing sugar content had some encouraging findings.⁵⁰ John Jemmot and colleagues have tested a rigorous study of a cognitive-behavioral approach to influencing diet and physical activity in South Africa showing that such an intervention was effective.⁵¹

What We Know and What We Need to Know

There are useful data already available that allow us to anchor our understanding of the issues of youth and risks for later non-communicable diseases. These are a product of far-sighted surveys that include a variety of measures of youth health, health risks, and related social factors. Agencies with a global view of these issues, such as WHO, CDC, and USAID, are providing assistance for youth- oriented data collection activities that provide generally comparable data for a number of countries and where the resulting data are made available for researchers and policymakers in local, national, and global settings.

A first step to advancing our global understanding of these issues would be for more countries to participate in the surveys and, for those who do participate, to repeat the measures at regular intervals. This would facilitate temporal study as well as enhance the ability to do cross-national studies. Of course, the drawbacks of such global studies is that they may miss issues of local relevance. However, it is possible to supplement core data collection with measures that are especially meaningful in specific settings.

In this chapter, I have used a single indicator for each risk factor, each one representing judgment about the value of that single measure. However, it is clear that there are different ways to measure each risk factor. Researchers could develop and test other measures. For example, would a measure of binge drinking be more useful in understanding the risks for NCDs? Overall, one would hope that future research would add to our understanding of the benefits of specific measures, the development of new measures, and greater understanding of how to get the best information from questions about risk factors. A recent study pointed out that people overestimate how much they are exercising. Self-report measures could probably be enhanced by taking advantage of such research. In this case, providing a definition of vigorous exercise (unable to carry on a conversation) could improve reporting.⁵² Youth surveys could ask about the intensity of activity, not just the time spent doing the activity.

The issues around measuring physical activity go further into just what counts as physical activity. Most surveys ask about a simple, overall measure, such as "over the past 7 days, in how many were you physically active for at least 60 minutes a day"?⁵³ However, some surveys break up such questions, most often into three parts. In the STEPS survey in Colombia, the question has three parts and asks about physical activity related to work, transportation (e.g. walking or biking to school or work), and planned physical activity (e.g. exercise).⁵⁴ Results reveal gender differences that are not so surprising, since boys and men typically are involved in more vigorous work. But, are all forms of physical activity the same? One view is that only activity that is being structured as "exercise" should be considered. However, other research in the United States found that people who defined their activity as "exercise" were prone to reward themselves for doing it, and that rewards could undermine the benefits of the exercise if the rewards include foods high in sugar or fat.⁵⁵

Measures of smoking and alcohol consumption are already quite varied and may include age of first experience, frequency, intensity, and consistency. While clearly one would like to avoid labeling as a "user" someone who had a single experience with alcohol or tobacco, but what would be the most meaningful measure that is plausible to include in broad-based youth surveys? Clearly there are different types of tobacco use (smoked, chewed, etc) that can be distinguished in data collection, but how much detail adds to our understanding of whether youth are beginning a potentially dangerous habit? Do youth distinguish between alcohol that is home-brewed from that commercially produced? In terms of alcohol use, can we extend research on how the circumstances of use—at home or with friends—affect future use?

Body mass index (BMI) seems to be a straightforward measure. However, measuring BMI among youth means that the measures are taken at different stages of pubertal development. Does that affect the meaning of those measures? Higher fat in early adolescence may fuel puberty among girls at least,⁵⁶ and so perhaps in thinking about the linkages of youth risks for NCDs and sexual and reproductive health, the assessments should start earlier than usual for adolescents. Surveys typically include a direct measure of weight and height from which to calculate BMI, are there other simpler techniques that would still be robust?

Improvements in data collection activities need to be based on knowledge about a behavior to obtain the most meaningful measures and important correlates of the behavior. Also, there is a need to develop measures that are valid and useful in survey settings. Much can be done in highly controlled or laboratory settings, but there is a value to measures that can be widely adopted in data collection efforts to allow for studies of differentials within and among countries and over time. In addition to data collected specifically on NCD risk behaviors, there is also a benefit from broader data initiatives that help set the context for young lives. Surveys of education, labor force participation, and family and household functioning could also be useful.

At the most basic level, vital statistics systems are a valuable adjunct to the work of understanding youth and NCDs. Birth registration is a key step to ensuring that young people know how old they are and can document it if needed for participation in school, work, or programs. Death registration, including cause of death reporting, is important for documenting the ultimate toll of NCDs, and for youth, it can also help to document the impact of causes of death, such as accidents, injuries, and suicide among youth.

Research Opportunities

Closely linked to the data issues is the research potential of using those data. Currently available data allow for many analyses within countries and comparisons across countries. A valuable addition to this dialogue would be research to model the impact of the present risk behaviors, the apparent trends, and the impact of changes in those anticipated trends. This modeling could then be used to estimate the likely burden of either not taking any action or the benefits of different interventions to lower risks. Since the cost of treating resulting diseases may vary country-by-country, it would be valuable to have this research conducted at the national level. National level estimates would likely be more useful to policymakers and those who are developing interventions.

The National Institutes of Health (NIH) has already supported many individual projects that deal with elements of youth risks for NCDs and called attention to the need to address the underlying factors that are involved in establishing and maintaining positive health behaviors in children and adolescents. See Table 1 for some examples of NIH program announcements calling for research on youth risks for NCDs. Initiatives by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) address not only individual behaviors but the role of advertising in influencing drinking by young adolescents. Research on obesity pursues the ties to specific diseases and assesses the impact of broad social initiatives in influencing behavior. An innovative announcement specifically sought to improve the understanding of policy changes by facilitating research on the natural experiments that emerge through policy or program changes. There is a wealth of research on tobacco use and growing interest in how the built environment influences activity levels. The landmark Add Health Study⁵⁷ has shown the power of a large, representative sample of youth surveyed over time, as it has added to our understanding of how adolescent behaviors influence later health.

Table 1. Examples of behavioral health research funded by the National Institutes of Health

Key: NCI = National Cancer Institute; NHLBI = National Heart, Lung, and Blood Institute; NIAAA = National Institute on Alcohol Abuse and Alcoholism; NIBIB = National Institute of Biomedical Imaging and Bioengineering; NIDCR = National Institute of Dental and Craniofacial Research; NICHD = National Institute of Child Health and Human Development; NIDDK = National Institute of Diabetes and Digestive and Kidney Diseases; National Institute on Drug Abuse; NIEHS = National Institute of Environmental Health Sciences; NIGMS = National Institute of General Medical Sciences; NIMH = National Institute of Mental Health; National Institute of Nursing Research; OBSSR = Office of Behavior and Social Science Research; ODS = Office of Dietary Supplements.

There is a body of work on youth risks for NCDs that can be a platform for relevant research and program or policy development in low- and middle-income countries, but such work also needs to be developed and adapted to fit the circumstances faced by youth in those settings. In combination, such work could provide a guide for how a society could foster the development of a broad set of positive health behaviors among youth and lower the future risk and burden of NCDs. Youth in developing countries are often facing rapid urbanization that may be disruptive to family and social supports. Urban areas may make an active lifestyle and healthy eating more difficult, but the density found there may present opportunities for interventions. Urban living may simply be a current example of how researchers could take context into account in studying these health behaviors.

There is an opportunity for research to consider the four risk behaviors together to further our understanding of the collective impact that they have on future NCDs. Modeling the impact of changes in risk behavior in relation to later expected disease burden could advance our understanding of the likely impact of interventions or the cost of neglect.

In the 50 years since the U.S. Surgeon General's report²⁶ on the health risks of tobacco, there have been many interventions to address tobacco use and many tools to help spread awareness of those tools (e.g. Campaign for Tobacco Free Kids).⁵⁸ There is an opportunity now to ensure that those tools are widely available and culturally appropriate for the millions of young people in developing countries. One intervention that has spread is the increase in the cost of tobacco products through taxation.⁴¹ Increased costs of tobacco, lower social tolerance of smoking, and a broader awareness of the health risks associated with smoking are all programmatic approaches that can be both broadly applied (tax increases apply to all) but have greater impact on youth. The large and growing portfolio of interventions could allow the developing world to move more quickly on addressing youth smoking than was the case in higher income countries.

Interventions to address youth alcohol consumption can include taxation, since increasing the price of alcohol tends to lower youth consumption and also provides a funding stream for other health interventions. Laws to prohibit the provision of alcohol to youth can help address underage drinking. Of course, the enforcement of such laws is critical. Developing parental, family, and community understanding of the risks of underage and excessive drinking may be a way to support enforcement. There are growing examples of campaigns to address problematic drinking, such as those focused on drinking and driving, that provide concrete tools (e.g. designated drivers, taxi services after partying) to help mitigate the risks of excessive alcohol consumption and provide a reminder of the inherent risks.

A great deal of attention has been given to the rise in overweight and obesity, and there have been a plethora of approaches to deal with it after the fact. This is not the place to review the myriad tools to help people lose weight; instead we need to ask how we can support youth to establish healthy eating patterns and maintain them into adulthood. Some social interventions, such as taxes on sodas or required labeling of foods and restaurant meals, have been tried with mixed results in terms of support for them and evidence of their effectiveness. A cross-national study⁵⁹ found a significant relationship between soft drink consumption and overweight and obesity, as well as a relationship to the prevalence of diabetes. Further research could sharpen our understanding of how soft drinks come to replace healthier options in different settings. Such research would have to confront the role of commercial interests in the transition in food choices in developing countries. Schools offer one setting to provide both nutritious food and education about the importance of healthy eating.

Low income urban areas, typically slums or informal settlements, may become food deserts where there is limited access to a wide range of nutritious, fresh foods. Policies that limit unhealthy fats, salt, and sugar in prepared foods or in restaurant foods are gaining traction and can be tested in a variety of settings with different forms of implementation.

It is unusual to see small children who are not physically active, but levels of activity seem to decline during adolescence. As young people begin to live lives filled with education, work, and family responsibilities, physical activity may suffer. That follows with certain types of employment and typically with urban living. Recent research has shed light on the linkage of physical activity and diet and shown that when physical activity is labeled as being "exercise" it is less effective than physical activity that just appears to be enjoyable, even when it is the same activity.⁵⁵ Individuals who saw the activity as "exercise" were more likely to reward themselves with treats, which were often ones that undermined healthy eating. Findings such as these could provide guidance in the ways interventions are developed and marketed for adolescents.

Policy Implications

In addition to numerous data and research opportunities, there are also opportunities at the policy level to address youth risks for later NCDs. Countries facing this looming shift in burden of disease could make an "adolescent health report card" to guide their actions. This could ask basic questions about the steps that were being taken to address each of the risk factors. For example, What steps are being taken to reduce youth smoking—taxation, health education, enforcement of laws about selling tobacco to minors? Are there laws about selling alcohol to minors, and are they enforced? Are there health education programs about the risks of early and extreme drinking? Do they include tools for youth to use to offset peer pressure? Are parents provided tools to help them identify problems and manage alcohol and tobacco use among their children? What steps are schools taking to support healthy eating and physical activity? Is there an infrastructure to support physical activity? In urban areas, are there ways to access healthy foods? Does the tax structure support unhealthy foods? Are economic development tools used to advance healthy eating and physical activity?

While the specific approaches will vary with different settings, a policy approach can be one that is implemented at different levels. Action at the national level would be needed for taxation approaches, but the enforcement of laws typically happens at the community level. Where there is widespread understanding of the risks of underage drinking and smoking, enforcement may be more rigorous. Community, cultural, and religious institutions can sponsor activities that support physical activity, such as sports. In most developing countries the health sector is already burdened, but there are NCD primary prevention activities that lie outside the health sector. That creates an opportunity to have more players who are challenged to contribute to primary prevention. The World Health Organization has articulated an essential package of "Best Buys" for population-based interventions.⁶⁰

NIH has long supported research on health behaviors and is well- positioned to contribute to understanding the threats faced by the developing world in terms of NCDs. A review of the initiatives NIH is promoting shows that many could enlighten us about how youth form, establish, and maintain heathy behaviors. There is an opportunity to specifically focus on the policy issues that have been raised about tobacco, alcohol, nutrition, and physical activity, and there are structural interventions that may address alcohol use and HIV/AIDS and, at the same time, enlighten us about structural interventions more generally. There is even an expressed interest in improving the measures used for diet and physical activity. Other initiatives call up the role of schools, social, and other media and highlight gender/ethnic issues, setting the stage for this work to focus on adolescents and include the considerations of these behaviors in developing countries. In other words, NIH has already established a framework of interest in important key issues related to the risk behaviors among youth in developing countries. Signaling to the investigator and reviewer communities that work in such settings is desired would help mobilize current initiatives to advance work in this area (Table 1 for examples of these NIH-supported initiatives). This existing background work and interest could be directed to include relevant studies that address adolescent risk behaviors in developing countries. Research in non-U.S. settings often provides an opportunity to not only advance understanding of important health issues among our global partners, but it also may allow for the testing of hypotheses, developing novel measures, and testing of different types of interventions.

Conclusion

In sum, the developing world is facing an epidemiologic transition that makes non-communicable diseases the most prominent cause of death. Despite ongoing needs to address infectious disease and malnutrition, we also must pay attention now to the morbidity and mortality associated with heart disease, diabetes, cancer, and chronic respiratory disease. In addition to the need to screen, diagnose, and treat those with disease, we also need to articulate a primary prevention approach that can lower the future burden of NCDs on individuals, families, communities, and nations. Two of the primary behavioral risk factors—use of tobacco and alcohol—begin during adolescence. To promote a healthy diet and active lifestyle in later years, adolescence is an important time to solidify positive health behaviors that will, among other things, lower the risk from overweight and obesity.

Basic data are available about youth risk behaviors in the developing world, but vigilance is needed to enhance those data systems. Research on these risk factors is more heavily concentrated in settings where the burden of NCDs has been felt for longer; there is a need to extend this research attention to youth in developing countries. The policy community has an opportunity to take the available data, research findings, and experiences with interventions to make youth prevention of NCD risks a key, measurable part of their overall planning. Benefits would be felt not only in the health sector, but also more broadly, since rising levels of NCDs threaten economic productivity.

Acknowledgments

Special thanks go to Toshiko Kaneda of the Population Reference Bureau who has participated in the development of data on youth in Latin America and the Caribbean. The opinions presented herein are those of the author and do not necessarily represent the position of the Agency for Healthcare Research and Quality, the National Institutes of Health, or the U.S. Department of Health and Human Services.

Author's Affiliation

Wendy Baldwin, PhD, Former President and CEO, Population Reference Bureau, Washington, DC.

Address correspondence to: Wendy Baldwin, PhD, 1104 Mourning Dove Drive, Blacksburg, VA 24060; email wendybburg@gmail.com.

References

1. Omram AR. The epidemiologic transition: a theory of the epidemiology of population change. Milbank Q 1971;40:509-38.
2. United Nations high-level meeting on noncommunicable disease prevention and control: NCD summit to shape the international agenda. September 19-20, 2011, New York City. Available at www.who.int/nmh/events/un_ncd_summit2011/en/. Accessed January 12, 2015.
3. Global action plan for the prevention and control of NCDs, 2013-2020. Geneva: World Health Organization; 2013. Available at http://www.who.int/nmh/publications/ncd-action-plan/en/. Accessed January 12, 2015.
4. 2012 World population data sheet. Washington, DC: Population Reference Bureau; 2012. Available at http://www.prb.org/Publications/Datasheets/2012/2012-world-population-data-sheet.aspx. Accessed January 12, 2015.
5. Noncommunicable diseases country profiles 2011: WHO global report. Geneva: World Health Organization; 2011. Available at http://www.who.int/nmh/publications/ncd_profiles2011/en/. Accessed January 12, 2015.
6. Bloom DE, Cafiero E, Jané-Llopis E, et al. The global economic burden of noncommunicable diseases. PGDA working paper No. 87. Boston, MA: Harvard Institute for Global Health, Program on the Global Demography of Aging; 2012. Available at http://www.hsph.harvard.edu/program-on-the-global-demography-of-aging/WorkingPapers/2012/PGDA_WP_87.pdf. Accessed January 12, 2015.
7. Scaling up action against NCDs: How much will it cost? Geneva: World Health Organization; 2011. Available at http://www.who.int/nmh/publications/cost_of_inaction/en/. Accessed January 20, 2015.
8. Malik K. Sustaining human progress: reducing vulnerabilities and building resilience. UNDP human development report, 2014. New York: United Nations Development Program; 2014. Available at www.undp.org/content/undp/en/home/presscenter/events/2014/july/HDR2014.html. Accessed January 12, 2015.
9. Chan M. Report by the Director-General to the special session of the Executive Board on Ebola. Geneva: World Health Organization; 2015. Available at www.who.int/dg/speeches/2015/executive-board-ebola/en/. Accessed March 12, 2015.
10. Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2095- 128.
11. Nikolic IA, Stanciole AE, Zaydman M. Chronic emergency: why NCDs matter. Washington, DC: World Bank; 2011. Available at http://tinyurl.com/knvnbph. Accessed March 12, 2015.
12. Gupta DK, Shah P, Misra A, et al. Secular trends in prevalence of overweight and obesity from 2006 to 2009 in urban Asian Indian adolescents aged 14-17 years. PLoS One 2011;6(2):e17221.
13. Reddy SP, Resnicow K, James S,. Rapid increases in overweight and obesity among South African adolescents: comparison of data from the South African National Youth Risk Behaviour Survey in 2002 and 2008. Am J Public Health 2012;102 (2):262-8.
14. Noncommunicable diseases among older adults in low- and middle-income countries. Population Reference Bureau, Today's Research on Aging, Issue 26; 2012, August. Available at www.prb.org/pdf12/TodaysResearchAging26.pdf. Accessed January 13, 2015.
15. Obesity: preventing and managing the global epidemic. Report of a WHO consultation, WHO technical report series 894. Geneva: World Health Organization; 2000.
16. Viner RM, Coffey C, Mathers C, et al. 50-year mortality trends in children and young people: a study of 50 low-income, middle-income, and high-income countries. Lancet 2011;377 (9772):1162-74.
17. Cawley J, MacLean JC. Unfit for service: the implications of rising obesity for the U.S. military. Health Econ 2012;21(11):1348- 66.
18. Global Youth Tobacco Survey. Atlanta, GA: Centers for Disease Control and Prevention. Available at http://www.cdc.gov/tobacco/global/gtss/tobacco_atlas/pdfs/part3.pdf. Accessed January 13, 2015.
19. Global tobacco control. Atlanta, GA: Centers for Disease Prevention and Control, Office on Smoking and Health, National Center for Chronic Disease Prevention and Health Promotion; 2013. Available at http://www.cdc.gov/tobacco/global/index.htm. Accessed January 13, 2015.
20. Global School-Based Student Health Survey. World Health Organization and Centers for Disease Control and Prevention. Available at http://www.cdc.gov/gshs/. Accessed January 13, 2015.
21. The DHS Program: Demographic and Health Surveys. Rockville, MD: USAID. Available at www. dhsprogram.com. Accessed January 13, 2015.
22. Millennium Development Goals and Beyond 2015. Goal 2: Achieve universal primary education. United Nations. Available at http://www.un.org/millenniumgoals/education.shtml. Accessed January 13, 2015.
23. STEPwise approach to chronic disease risk factor surveillance. 2010 STEPS survey report. Geneva: World Health Organization; 2010. Available (Spanish only) at http://www.who.int/chp/steps/2010_STEPS_Survey_Colombia.pdf. Accessed March 12, 2015.
24. Baldwin W, Kaneda T, Amato L. Noncommunicable diseases in Latin America and the Caribbean: Youth are key to prevention. Washington, DC: Population Reference Bureau; June 2013. Available at http://www.prb.org/pdf13/TechnicalNotes.pdf. Accessed January 13, 2015.
25. Taioli E, Wynder EL. Effect of the age at which smoking begins on frequency of smoking in adulthood. N Engl J Med 1991;325(13):968-9.
26. Reducing the health consequences of smoking: 25 years of progress. A report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services; 1989. Available at http://profiles.nlm.nih.gov/ps/access/NNBBXS.pdf. Accessed January 14, 2015.
27. You're the target: global Marlboro campaign found to target teens. Washington, DC: Campaign for Tobacco-Free Kids; 2014. Available at http://global/tobaccofreekids.org/en/industry_watch/marketing/youre_the_target. Accessed January 14, 2015.
28. Tobacco product use among middle and high school students-United States, 2011 and 2012. MMWR 2013 Nov 15;62(45):893-7.
29. Global Tobacco Surveillance System Data (GTSSData). Atlanta GA: Centers for Disease Control and Prevention; 2010. Available at http://nccd.cdc.gov/GTSSData/default/default.aspx. Accessed January 14, 2015.
30. Global Youth Tobacco Survey Collaborative Group. Tobacco use among youth: a cross country comparison. Tob Control 2002;11(3):252- 70.
31. Alcohol, violence, and aggression. Alcohol Alert from NIAAA, No. 38, October 1997. Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. Available at http://pubs.niaaa.nih.gov/publications/aa38.htm. Accessed January 14, 2015.
32. Grant BF. Impact of a family history of alcoholism on the relationship between age at onset of alcohol use and DSM-IV alcohol dependence. NIAAA Epidemiological Bulletin 1998;22(2):144- 8.
33. Grant BF, Dawson DA. Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: results from the National Longitudinal Alcohol Epidemiologic Survey. J Subst Abuse 1997;9:103-10.
34. Kaneda T, Naik R, Baldwin W. Noncommunicable diseases—risk factors among young people in Africa. Washington, DC: Population Reference Bureau; 2014. Available at http://www.prb.org/Publications/Reports/2014/ncd-risk-youth-africa.aspx. Accessed January 14, 2015.
35. Guatemala (Capital City Oficial): 2009 Fact Sheet. Global School-based Student Health Survey. Available at http://www.who.int/chp/gshs/Guatemala_capital_city_oficial_2009_FS.pdf. Accessed January 14, 2015.
36. Global Health Observatory data: obesity among women. Geneva: World Health Organization; 2011. Available at http://www.who.int/gho/urban_health/risk_factors/women_obesity/en/. Accessed March 12, 2015.
37. Kolcic I. Double burden of malnutrition: a silent driver of double burden of disease in low- and middle-income countries. J Glob Health 2012;2 (2):020303.
38. The state of food and agriculture. Rome: Food and Agriculture Organization of the United Nations; 2013. Available at http://www.fao.org/docrep/018/i3300e/i3300e.pdf. Accessed March 12, 2015.
39. OECD obesity update, 2014. Victoria, BC, Canada: Jesse Ferreras; 2014.
40. Global recommendations on physical activity for health. Geneva: World Health Organization; 2010. Available at http://whqlibdoc.who.int/publications/2010/9789241599979_eng.pdf?ua=1. Accessed January 14, 2015.
41. Raising cigarette taxes reduces smoking, especially among kids (and the cigarette companies know it). Fact sheet. Washington, DC: Campaign for Tobacco-Free Kids; 2012.
42. Holford TR, Mexa R, Warner KE, et al. Tobacco control and the reduction in smoking-related premature deaths in the United States, 1964-2012. JAMA 2014;311(2):164-71.
43. Ng M, Freeman MK, Fleming TD, et al. Smoking prevalence and cigarette consumption in 187 countries, 1980-2012. JAMA 2014;311(2):183-92.
44. Youth and tobacco use. Fact sheets. Atlanta, GA: Centers for Disease Control and Prevention; 2014. Available at http://www.cdc.gov/tobacco/data_statistics/fact_sheets/youth_data/tobacco_use/index.htm. Accessed March 12, 2015.
45. Bornhaeuser A, Bloom J. Smokefree air law enforcement: lessons from the field. Washington, DC: Global Smokefree Partnership; 2009.
46. Ciclovias—the 20 longest cycling streets in the world. City Clock Magazine (online); 2014. Available at http://www.cityclock.org/ciclovias-20-longest-cycling-streets-world/#.VM2YO9LF90Y. Accessed March 12, 2015.
47. Ciclovis recreativisde las Americas (Web site). Available at http://www.cicloviasrecreativas.org/en/map. Accessed March 12, 2015.
48. Argentine initiative to reduce salt consumption. "Less salt, more life." Washington, DC: Pan American Health Organization. Available at http://www.paho.org/panamericanforum/wp- content/uploads/2012/08/less-salt-more-life_PAHO-consortium_ARG.pdf. Accessed January 15, 2015.
49. Legetic B. America's initiative for cardiovascular disease prevention through dietary salt reduction. Washington, DC: Pan American Health Organization; 2012. Available at http://www.iom.edu/~/media/Files/Activity%20Files/Nutrition/ ConsequencesSodiumReduction/2012-DEC-04/Presentations/06_Branka%20Legetic.pdf. Accessed January 15, 2015.
50. Cabrera Escobar M, Veerman JL, Tollman SM, et al. idence that a tax on sugar sweetened beverages reduces the obesity rate: a meta-analysis. BMC Public Health 2013;13(1):1072.
51. Jemmott JB, Jemmott LS, O'Leary A, et al. Cognitive-behavioural health-promotion intervention increases fruit and vegetable consumption and physical activity among South African adolescents: a cluster randomized controlled trial. Psychol Health 2011;26(2):167- 85.
52. Canning KL, Brown RE, Jamnik VK, et al. Individuals underestimate moderate and vigorous intensity physical activity. PLoS One 2014;9 (5):e97927.
53. Global School-based Student Health Survey: 2009 Guatemala GSHS questionnaire. Atlanta, GA: Centers for Disease Control and Prevention; 2009. Available at http://www.who.int/chp/gshs/2009_Guatemala_GSHS_Questionnaire.pdf. Accessed January 15, 2015.
54. STEPwise approach to chronic disease risk factor surveillance, Colombia. Geneva: World Health Organization; 2010. Available at http://www.who.int/chp/steps/colombia/en/. Accessed January 15, 2015.
55. Werle CO, Wansink B, Payne CR. (2014). Is it fun or exercise? The framing of physical activity biases subsequent snacking. Marketing Letters 2014;1-12. Available at http://tinyurl.com/njdqw2e. Accessed January 19, 2015.
56. Kaplowitz PB. Link between body fat and the timing of puberty. Pediatrics 2008;121(Suppl 3):S208-17.
57. Add Health. The National Longitudinal Study of Adolescent to Adult Health. University of North Carolina Population Center. Available at http://www.cpc.unc.edu/projects/addhealth. Accessed January 19, 2015.
58. Campaign for Tobacco-Free Kids; Web site. Washington, DC; 2014. Available at http://www.tobaccofreekids.org/. Accessed January 19, 2015.
59. Basu S, McKee M, Galea G, et al. Relationship of soft drink consumption to global overweight, obesity, and diabetes: a cross-national analysis of 75 countries. Am J Public Health 2013;103(11):2071- 7.
60. From burden to "best buys": reducing the economic impact of non-communicable diseases in low- and middle-income countries. Geneva: World Health Organization; 2011. Available at http://www.who.int/nmh/publications/best_buys_summary.pdf?ua=1. Accessed January 19, 2015.

^a. Jack Dilbeck, Research Analyst, United States Accessions Command, Fort Knox, KY, as cited in Ready, Willing, Unable to Serve, report by Mission Readiness: Military Leaders for Kids; cdn.missionreadiness.org/MR-Ready-Willing-Unable.pdf. Accessed January 13, 2015.

^b. Note that in Colombia, the age range for the STEPwise Survey was 15-64, but in some countries it was 25 or older. The survey from Colombia (Spanish language only) can be found at http://www.who.int/chp/steps/2010_STEPS_Survey_Colombia.pdf.

Wendy Baldwin, PhD, is a social demographer with a background that spans government, academic, and private-sector organizations. At the National Institute of Child Health and Human Development, she led a program to advance research on adolescent sexual and reproductive health, the role of child care in fertility and employment decisions, behavioral aspects of HIV/AIDS, and fertility in the United States. As the Deputy Director for Extramural Research, she advanced policies to support the sharing of research data and streamlined peer review and the electronic submission and processing of grants. At the Population Council in New York City, she led a program of research on poverty, gender, and youth in international settings that emphasized girls' schooling, contraception and childbearing, and access to demographic data and its use by policymakers. At the Population Reference Bureau, Dr. Baldwin focused on developing data about youth risks for non-communicable diseases in developing countries and communicating with policymakers. Now retired, she maintains her professional interests and activities from her home in Blacksburg, VA.