Chronic diseases, such as heart disease, stroke, and cancer, often arise as a by-product of how people live and behave. The epidemiologic transition to chronic diseases—in which conditions like atherosclerosis and diabetes have replaced infectious diseases as the leading causes of death—has focused increasing attention on the central role of personal health behaviors in health (Olshansky and Ault, 1986). Two decades ago, in an article titled “Actual Causes of Death,” tobacco use, diet, physical inactivity, and other personal behaviors were identified as the leading killers in the modern age (McGinnis and Foege, 1993). That work has since been refined in the United States and replicated throughout the world. It is now widely recognized that health behaviors are the leading contributors to the global burden of disease, especially in high-income countries where the epidemiologic transition has been longstanding (Lopez et al., 2006; Olshansky and Ault, 1986; World Health Organization, 2008a).
Can the U.S. health disadvantage be explained by a prevalence of unhealthy behaviors that is higher among Americans than among people in other high-income countries? Some analyses have questioned whether health behaviors are at the root of the U.S. health disadvantage (see, e.g., Avendano et al., 2009). This chapter examines these behaviors, and it also considers sexual practices, drug use, and injurious behaviors that might also contribute to the U.S. health disadvantage documented in Part I, particularly among younger adults.
A life-course perspective is important in studying health behaviors, many of which are adopted at young ages (especially in adolescence)
before becoming life-long habits (see Chapter 3). Smoking is a good example. According to Viner (2012, p. 8):
[T]he tracking history of smoking is of initiation and increasing use in adolescence, followed by relative stability from the late teens onwards. Compared with other forms of drug use, cigarette smoking shows the least decline in young adulthood. Those who start early in adolescence, e.g., less than 14 years, are more likely to become lifelong smokers.
Comparing health behaviors across high-income countries is difficult due to a scarcity of data and extensive challenges with measurement. In contrast with such health indicators as blood pressure or serum lipid concentrations, which are easy to measure in standardized units across countries, health behaviors are rarely examined in a uniform fashion, even within the same country, and in some countries they are not measured at all. For example, a survey about tobacco use can ask about cigarette smoking only or can also include cigar and pipe smoking or the use of smokeless tobacco. Respondents can be asked how many cigarettes they smoke each day or whether they smoke “currently,” have “ever smoked,” have ever smoked 100 cigarettes, or have smoked in the past 30 days.
Similarly, questions about physical activity can differentiate between leisure-time “exercise” and physical activity, between regular and episodic physical activity, between levels of exertion, and between duration and frequency of activity per day, per week, or per month. They can measure sedentary behavior, moderate activity, and intense physical activity. They can document types of activity, such as sitting, walking, cycling, running, or gardening. No single measure is the established standard and, unlike smoking, no single measure is the strongest predictor of premature death or morbidity or has the strongest link to obesity.
Questions about diet are perhaps the most varied. They can address food groups (e.g., fruits and vegetables, grains, meats), specific nutrients (e.g., saturated and unsaturated fats, trans fats, sodium, calcium), and caloric intake. A proper dietary history involves numerous questions (Paxton et al., 2011) and can produce different results depending on how the data are collected (Erinosho et al., 2011). Nutrition science has yet to identify the most important dietary predictors of longevity or of specific disease outcomes, such as cancer or myocardial infarctions. Few countries include sufficient questions on population surveys to evaluate the national diet, and very few questions are asked in common across countries, confounding efforts to make meaningful or accurate comparisons. Often, the only common data across countries are governmental statistics on national caloric expenditures and food consumption divided by the population size to derive per capita estimates.
With these caveats in mind, this chapter examines six behaviors: (1) tobacco use, (2) diet, (3) physical inactivity, (4) alcohol and other drug use, (5) sexual practices, and (6) injurious behaviors.1 As in other chapters, the panel poses three questions in reference to each behavior:
• Does the health behavior matter to health?
• Is the behavior more prevalent in the United States than in other high-income countries?
• Does the difference in behavior explain the U.S. health disadvantage?
Question 1: Does Tobacco Use Matter to Health?
The enormous health consequences associated with tobacco use were first recognized in the 1950s. Beginning with the landmark 1964 report by U.S. Surgeon General Luther Terry, the health consequences of tobacco use have since become one of the most extensively documented topics in public health history (U.S. Department of Health and Human Services, 1964, 2000). The tobacco literature is too extensive to recapitulate here, nor is it necessary, other than to list the tobacco-related diseases that contribute to the U.S. health disadvantage (see Part I): noncommunicable diseases, such as coronary artery disease, stroke, cancer, and chronic lung disease, as well as adverse birth outcomes, such as low birth weight and infant mortality.
Is Tobacco Use More Prevalent in the United States
Than in Other High-Income Countries?
A generation ago, Americans led the Western world in tobacco use (Forey et al., 2002)—reaching peak rates in the 1950s—but decades of tobacco control efforts following the 1964 Luther Terry report led to the marked reductions in smoking rates shown in Figure 5-1, from 42 percent in 1965 to 21 percent in 2009 (Garrett et al., 2011). Cigarette consumption rose and fell earlier and more dramatically in the United States than in many other countries (Cutler and Glaeser, 2006; Pampel, 2010). Figure 5-2 shows that the United States now has the lowest adult smoking rates of all peer countries but Sweden and one of the lowest among OECD countries. In many countries, the mass adoption of smoking occurred earlier among males than among
1The list of behaviors examined in this chapter is not exhaustive. Other behaviors or health practices, such as getting adequate sleep or reducing stress, are also important to health promotion and injury prevention (Smolensky et al., 2011) but are not examined here.
FIGURE 5-1 Percentage of U.S. adults age 18 and older who were current smokers, by sex and race/ethnicity, 1965-2008.
NOTES: Data are based on the National Health Interview Survey (NHIS). For NHIS survey years 1965-1991, current smokers included adults who reported that they had smoked 100 or more cigarettes in their lifetimes and currently smoked. Since 1992, current smokers included adults who reported smoking 100 or more cigarettes during their lifetimes and specified that they currently smoked every day or on some days. Figure depicts trend over time. Data were not available for 1967-1969, 1971-1973, 1975, 1981, 1982, 1984, 1986, 1989, and 1996, because the NHIS did not include questions about smoking.
SOURCE: Garrett et al. (2011, p. 110).
females (Pampel, 2010). In the United States, the highest rates of smoking among males occurred in cohorts born around 1915-1920 and among women in cohorts born around 1940-1944 (Preston and Wang, 2006).
Question 3: Does Tobacco Use Explain the U.S. Health Disadvantage?
Various studies suggest that a large proportion of differences in mortality (notably sex differences) among high-income countries are attributable to tobacco use (Bongaarts, 2006; Janssen et al., 2007; Pampel, 2002; Peto et al., 1992; Retherford, 1975; Valkonen and Van Poppel, 1997; Waldron, 1986). There is a 20-30 year lag between changes in smoking rates and the resulting effects on smoking-related mortality (Lopez et al., 1994): thus, these effects are manifest mainly among older adults. According to Staetsky (2009, p. 892): “The impact of smoking related mortality on old-age mortality in the end of the 20th century is a function of the exposure of middle-aged adults to smoking approximately during the 1970s.” Using coefficients derived from lung cancer, she concluded that smoking accounted for the slower decline in mortality among women in Denmark,
FIGURE 5-2 Prevalence of daily smoking in 17 peer countries.
NOTES: Prevalence rates are for 2005-2009. Prevalence rates for the most current year available are reported.
SOURCE: Adapted from OECD (2011c).
the United States, and the Netherlands when compared with France and Japan (Staetsky, 2009).
Using an innovative macrostatistical method, Preston and colleagues (2010a, 2010b) estimated the attributable fraction of deaths after age 50 from smoking2 and its effect on life expectancy at age 50 among 10 high-income countries in 1955, 1980, and 2003. The authors calculated that by 2003 smoking accounted for 41 percent of the difference in male life expectancy at age 50 between the United States and 9 comparison countries and for 78 percent of the difference in female life expectancy at age 50 (Preston et al., 2010b).3 Smoking appeared to have a larger impact
2The results were modeled on lung cancer mortality, see Preston et al. (2010b).
3Attributable fractions produce estimates that are subject to some errors, as reported by the authors. Slightly different results are reported in Preston et al. (2010a).
on women because of the later uptake of smoking by U.S. women (U.S. Department of Health and Human Services, 2000, 2002) (see Figure 5-3). The smoking-attributable fraction of U.S. deaths among males age 50 and older was 23 and 22 percent in 1980 and 2003, respectively, but during the same years increased from 8 to 20 percent among females of the same age (Preston et al., 2010b). Based on the researchers’ assumptions, smoking accounted for 67 percent of the shortfall in life expectancy gains that U.S. women experienced relative to 20 other countries between 1950 and 2003.
These findings implicate smoking as a potential cause of the shorter life expectancy of adults age 50 and older, but they do not explain the lower life expectancy observed in younger people. The U.S. health disadvantage before age 50 has worsened over the same time that smoking prevalence rates in this population have decreased. The reduction in smoking rates will produce benefits in years to come. Wang and Preston (2009) predicted that the future will bring a decline in deaths attributable to smoking among men but that improvements for women will occur later.
FIGURE 5-3 Four stages of the U.S. tobacco epidemic.
SOURCE: Thun et al. (2012, Figure 1).
Question 1: Does Diet Matter to Health?
As with tobacco, the evidence linking diet to health is extensive and unnecessary to review here.4 Caloric intake relative to energy expenditure is a major contributor to obesity and obesity-related diseases such as diabetes. Levels of consumption of specific nutrients (e.g., saturated fats, trans fats, sodium, calcium, iron, vitamins) are linked to a variety of diseases (e.g., hypertension, cardiovascular disease, osteoporosis, anemia) (U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2010; Willett, 1998). Consumption of food groups such as fibers and grains are associated with lower risks of hyperlipidemia and some cancers. Low fruit and vegetable intake is associated with increased risk of obesity, coronary heart disease, stroke, diabetes, hypertension, and colorectal cancer (World Health Organization, 2004b). Worldwide in 2000, 5 percent of deaths were attributable to low fruit and vegetable intake (Lock et al., 2004). The World Cancer Research Fund (1997) estimated that up to 30-40 percent of global cancers could be prevented through a change in diet. However, considerable controversy and scientific uncertainty still surround the strength of the evidence linking specific nutrients to disease risks and longevity.
Is an Unhealthy Diet More Prevalent in the United States
Than in Other High-Income Countries?
Comparable cross-national data on dietary practices are limited for the reasons noted above, including the challenges that countries face in evaluating the diets of their populations and inconsistencies across countries in food culture, defining indicators, sampling respondents, and administering surveys. Data collected within the United States suggest that the American diet has become less nutritious over time. Between 1971 and 2000, average daily caloric consumption increased from 2,450 kcals to 2,618 kcals among men and from 1,542 kcals to 1,877 kcals among women;5 similarly, carbohydrate intake increased by 67.7 grams and 62.4 grams, respectively for men and women, and total fat intake increased by 6.5 grams and 5.3
4We do not review the extensive evidence regarding the dietary benefits of breastfeeding, but we do discuss below the relatively low prevalence of breastfeeding in the United States.
5The physical unit “calorie” is the energy required to increase the temperature of 1 gram of water by 1 degree Celsius. The commonly used dietary term “calorie” is shorthand for the scientific term kilocalorie (kcal), which equals 1,000 calories.
grams, respectively, for men and women (Centers for Disease Control and Prevention, 2004).
Between 1950 and 2000, annual per capita food consumption in the United States increased by 20 percent for fruits and vegetables but also for grains (by 44.5 pounds, a 29 percent increase), meats (by 57 pounds, a 41 percent increase), cheese (by 22.1 pounds, a 287 percent increase), and caloric sweeteners (by 42.8 pounds, a 39 percent increase). High-fructose corn syrup consumption per capita rose from zero in 1950 to 85.3 pounds by 2000. Some of these increases may be associated with an increase in dining out, which increased from 18 percent of total food energy consumption in 1977-1978 to 32 percent in 1994-1996 (U.S. Department of Agriculture, 2012).
How do these trends compare with other rich nations? Americans consumed 3,770 kcals per person per day in 2005-2007,6 more than any other country in the world (see Figure 5-4). This trend is not new: the United States also had the highest caloric consumption in 2003-2005 and ranked fourth in the world in 1999-2001 (behind Austria, Belgium, and Italy). Between 1999-2001 and 2005-2007, the U.S. ranking on fat intake rose from seventh to fourth in the world, with Americans consuming an average of 161 grams per person per day. By comparison, in 2005-2007 the average Swede consumed 17 percent fewer calories and 24 percent less fat (Food and Agriculture Organization, 2010).
Whether other eating habits in the United States differ from those in other countries is less clear due to inadequate data. Only 23 percent of Americans consume fruits or vegetables five times a day (Centers for Disease Control and Prevention, 2012a), as recommended by dietary guidelines (U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2010), but the European Union also reports inadequate intake of fruits and vegetables in its member countries (European Food Information Council, 2012). Some studies have reported that Americans consume fewer fruits and vegetables than people in other countries, such as Canada (Richards and Patterson, 2005) and France (Tamers et al., 2009), but the evidence is inconclusive. Another study reported that Americans consumed more calories per minute than those in Austria, France, Germany, Italy, and the Netherlands; they also spent the least time cooking or eating at home and the most time eating at restaurants (after the French) (Brunello et al., 2008; see also Michaud et al., 2007).7 Among the 17 peer countries
6Per capita consumption statistics do not account for wastage and therefore overestimate actual nutrient intake.
7A two-country comparison showed a rate of consumption of 53.8 calories per minute for the United States and 28.4 minutes for France (Brunello et al., 2008).
FIGURE 5-4 Global map of per capita caloric intake.
SOURCES: United Nations Development Programme, “Human Development Report 2000,“ Table 23, http://www.undp.org/hdro/HDR2000.htmlFAO, “The Developing World’s New Burden: Obesity,” http://www.fao.org/FOCUS/E/Obesity/obes1.htm; World Health Organization, “Global Strategy on Diet, Physical Activity, and Health: Obesity and Overweight,” http://www.who.int/dietphysicalactivity/publications/facts/obesity/en/; World Health Organization, “Nutrition for Health and Development,” http://www.who.int/nut/publications.htm.
SOURCE: Global Education Project (2004).
examined in Part I, only two countries (Spain and France) report lower rates of breastfeeding of infants (Palloni and Yonker, 2011).
Question 3: Does Diet Explain the U.S. Health Disadvantage?
Being the global leader in caloric intake (see Figure 5-4, above), combined with inadequate physical activity, helps explain escalating obesity rates in the United States. It could plausibly contribute to higher rates of diseases attributable to obesity, such as diabetes and heart disease. High intake of saturated fats and inadequate intake of fresh produce and other healthy foods could explain a variety of diet-related noncommunicable diseases that are more prevalent in the United States than in comparable countries (see Part I). Using the macrostatistical method discussed above for tobacco but with a different set of comparison countries, Preston and Stokes (2011) estimated that obesity accounted for 42 percent of the shortfall in female life expectancy at age 50 years in the United States relative to countries with higher life expectancies, and 67 percent of the shortfall among males.
Question 1: Does Physical Activity Matter to Health?
Although the health consequences of physical inactivity8 are often difficult to disentangle from the morbidity and mortality associated with obesity and unhealthy diet, decades of research suggest that avoiding sedentary behavior and engaging in regular physical activity (independent of body weight, body mass index, and dietary habits) exerts its own protective effect on the risk of heart disease and stroke and possibly other conditions, including cancer, depression, and dementia (Lee et al., 2012; U.S. Department of Health and Human Services, 2008a). Physical activity among children offers unique age-specific benefits (Janssen and Leblanc, 2010), as does exercise among elderly people. Weight-bearing exercise plays a role in maintaining bone density and preventing osteoporosis and related morbidity (e.g., from hip fractures). The risk of injuries (e.g., falls among the elderly) is related to muscle strength and flexibility and obesity.
Is Physical Inactivity More Prevalent in the United States
Than in Other High-Income Countries?
As noted above, comparisons of the activity levels of Americans and people in other countries are difficult because of differences in definitions of activity and intensity levels and in survey and sampling methods. There is also inherent imprecision in self-reported activity levels (Sallis and Saelens, 2000). Furthermore, the same questions are rarely asked in each country (Hallal et al., 2012). One study administered the International Physical Activity Questionnaire to adults ages 18-65 in 20 countries. In the United States, 84.1 percent of respondents reported engaging in moderate or vigorous activity in a typical week, higher than all but five countries, only one of which was a high-income country (New Zealand) (Bauman et al., 2009).
In a study of physical activity in adults age 50 and older in the United States and Europe, Steptoe and Wikman (2010) found that the proportion of respondents who were moderately or vigorously active at least once a week ranged from 56 percent in Poland to 83 percent in Sweden, with the U.S. rate (69 percent) in the middle.9 However, the United States
8Physical activity is any body movement that activates muscles and requires more energy than resting; exercise (including leisure-time exercise) is a type of physical activity that is planned and structured.
9The data were from the Health and Retirement Study (HRS) in the United States, the Survey of Health, Ageing and Retirement in Europe (SHARE), and the English Longitudinal Study of Ageing (ELSA) in England, all of which use similar measures of physical activity.
and Poland had the highest proportion (22 percent) of adults age 50 and older who reported no moderate or vigorous physical activity (Steptoe and Wikman, 2010). Michaud and colleagues (2007, p. 1) concluded that “sedentary lifestyle or a lack of vigorous physical activity may also explain a substantial share of the cross-country differences” between the United States and Europe. However, a recent cross-national comparison of data from 155 surveys in 122 countries reported that the prevalence of physical inactivity in the United States among those age 15 and older (40.5 percent) was lower than in 7 of the 16 peer countries that provided data (Hallal et al., 2012).
In a comparison with Austria, France, Germany, Italy, and the Netherlands, American adults age 20 and older spent 16.8 minutes per day doing sports, more than people in any other country but the Netherlands; however, Americans also spent 157.9 minutes per day watching television or listening to the radio, more than people in any other country (Brunello et al., 2008).
The international Health Behaviour in School-Aged Children (HBSC) survey of children ages 11, 13, and 15 ranked the United States in the bottom third of countries based on the percentage of children who exercised at least two times a week (World Health Organization, 2000a). However, these data are from the 1990s, before the popularity of video games and electronic devices. When the same survey was administered in 2001-2002, the United States had the largest proportion among 34 countries of children ages 10-16 who reported regular physical activity, but 46.7 percent watched 3 or more hours of television per day, more than all high-income countries but Norway (48 percent), Scotland (50.1 percent), and England (51.9 percent) (Janssen et al., 2005b). Other cross-national data on regular physical activity (e.g., walking), rather than leisure-time activity, are limited.
Question 3: Does Physical Inactivity Explain
the U.S. Health Disadvantage?
Current evidence is inadequate to compare the physical activity of Americans with people in other countries, let alone to explore its role in explaining the U.S. health disadvantage. The limited data available paint a conflicting picture of whether Americans are more active than others, and many of the data predate the mass popularity of electronic devices that may encourage more sedentary life-styles. Even with ideal physical activity data, strong causal relationships between activity and health outcomes remain inadequately understood. Steptoe and Wikman (2010, p. 208) found that “countries with a higher proportion of individuals who are physically active have a lower prevalence of fair or poor self-rated health” but emphasized
that cross-sectional associations do not prove causality.10 In one study, countries’ child obesity rates correlated with reported physical activity levels but not with time spent on computers (Janssen et al., 2005b). Studies based on current life-styles will clarify these relationships and may elucidate whether physical activity levels can ultimately explain some of the U.S. health disadvantages reported in Part I, such as chronic diseases or accidents (e.g., falls among the elderly) that could be attributed to lack of exercise or obesity. The explanation for other conditions lies elsewhere.
Question 1: Does Alcohol or Other Drug Use Matter to Health?
Although there is evidence that moderate consumption of alcohol can offer some health benefits, excessive drinking increases the risk of liver disease, various forms of cancer, hypertensive heart disease, stroke, pregnancy complications, and many other conditions (see, e.g., Rehm et al., 2009). Worldwide, alcohol is implicated in 30 percent of violent injuries, 21 percent of road traffic accidents, 19 percent of poisonings, 16 percent of drownings, and 11 percent of falls and self-inflicted injuries (World Health Organization, 2011a). In the United States, alcohol use during adolescence has an important linkage to violence and homicide (Green et al., 2011; Parker et al., 2011b).
Drug use also affects the risk of disease and injury. Among fatally injured drivers in the United States in 2009, 18 percent tested positive for at least one drug (e.g., illicit, prescription, or over-the-counter) (National Highway Traffic Safety Administration, 2010). Use of illicit drugs (e.g., marijuana, cocaine, methamphetamines, heroin), misuse of prescription medication, including medical errors in the clinical setting (Chen and Lin, 2009), and the growing problem of prescription opiate abuse (Centers for Disease Control and Prevention, 2011g; Kuehn, 2012; U.S. Food and Drug Administration, 2011) can affect health in ways that range from mild side effects to permanent complications, including accidental or intentional overdoses. Cocaine was involved in almost 40 percent of U.S. drug deaths in 2008 (United Nations Office on Drugs and Crime, 2010). Injection drug users are at risk of infection with the hepatitis B and C viruses and HIV.
Both alcohol and other drug dependency can increase vulnerability to crime and related violent injuries, such as drug-related homicides. Intoxicated individuals are prone to other risky behaviors, such as unprotected sexual activity. Forensic testing of U.S. victims of violent deaths in 2008
10It is also important to note that self-rated health does not necessarily correlate with objective indicators of health.
commonly revealed toxicological evidence of alcohol intoxication11 (62 percent) and ingestion of opiates (26 percent), antidepressants (20 percent), cocaine (11 percent), marijuana (11 percent), amphetamines (3 percent), and other drugs (42 percent) (Karch et al., 2011).
Is Alcohol or Other Drug Use More Prevalent in the United States
Than in Other High-Income Countries?
Alcohol consumption has decreased in recent years in most OECD countries, and U.S. alcohol consumption in 2009 (8.8 liters per capita) was below the consumption levels of 11 peer countries and below the OECD average (OECD, 2011b). The World Health Organization (WHO) also reported that people in the United States consume less alcohol per capita than those in France, Germany, Italy, Sweden, and the United Kingdom, although they consume slightly more than the Japanese (World Health Organization, 2011). Whether patterns of alcohol consumption (e.g., binge drinking, alcohol dependence) differ in the United States is less clear. According to the WHO Global Status Report on Alcohol (World Health Organization, 2004a),12 the prevalence of both heavy drinking among U.S. adults (6.4 percent for males, 5.0 percent for females)13 and of heavy episodic drinking among U.S. youth was lower than in other high-income countries and the prevalence of lifetime abstinence was among the highest. However, another WHO study reported that the prevalence of heavy alcohol use14 among U.S. and Canadian males ages 15-29 was 23.8 percent, a higher prevalence than in any other region of the world (Chisholm et al., 2004). This could bear on the high rate of adolescent deaths from car crashes in the United States (see Chapter 1). In contrast, the 2007 European School Survey Project on Alcohol and Other Drugs reported higher rates of drinking among young people in Europe than in the United States, including a greater percentage of youth ages 15-16 reporting drinking in the past 30 days, higher intoxication rates, and a greater percentage reporting intoxication before age 13 (Friese and Grube, 2001).
11In most jurisdictions in the United States, intoxication is defined as a blood alcohol content of 0.08 milligrams per deciliter (mg/dL) or greater.
13Higher rates of binge drinking are reported in U.S. surveys. According to the 2010 Behavioral Risk Factor Surveillance System (BRFSS) survey, the prevalence of binge drinking in the United States was 17.1 percent (Centers for Disease Control and Prevention, 2012d).
14Heavy alcohol use was defined as an average rate of consumption of more than 20 grams of pure alcohol daily for women and more than 40 grams daily for men (Chisholm et al., 2004).
United Nations data suggest that drug use may be more prevalent in the United States than in many comparable high-income countries. Fully 12.5 percent of Americans ages 15-64 report using cannabis (marijuana) in the past year, a larger proportion than in all of Europe (except Italy and the Czech Republic) and Australia, although less than in Canada and New Zealand. Although cocaine use has declined significantly in the United States, 2.6 percent of people ages 15-64 report using cocaine each year, a higher rate than in Canada, all of Europe (except Spain and the United Kingdom), Australia, and New Zealand. The United States accounts for 36 percent of the global consumption of cocaine.
The reported use of amphetamine-type stimulants was 1.3 percent in the United States, higher than in Europe (except for Scotland and the Czech Republic) but lower than in Canada, Australia, and New Zealand. Use of opiates and ecstasy is not exceptionally higher in the United States than in other countries, but use of prescription opioids (pain killers) is markedly higher (United Nations Office on Drugs and Crime, 2010). Substance use is common among U.S. adolescents—37 percent of those ages 12-17 report using alcohol, marijuana, analgesic opioids, or other drugs in the past year, and 8 percent meet criteria for a substance-related disorder (Wu et al., 2011)—but comparable data from other countries are lacking.
Question 3: Does Alcohol or Other Drug Use
Explain the U.S. Health Disadvantage?
As noted in Chapter 1, alcohol- and other drug-related deaths claim extra years of life in the United States relative to other high-income countries. However, there is little definitive evidence that heavy drinking is more prevalent in the United States than in other high-income countries, except perhaps for young adults. Use of marijuana, cocaine, and amphetamine-type stimulants appears to be more prevalent in the United States than in many high-income countries. In theory, higher rates of substance abuse could explain higher rates of drug-related outcomes—from deaths due to medication errors and poisonings15 to intoxication-related injuries (e.g., motor vehicle crashes), violence (e.g., homicides), unsafe sex (e.g., unintended pregnancies, sexually transmitted infections), and injection drug use (e.g., hepatitis B virus and HIV infections). However, empirical data are lacking to assert a causal link between substance abuse and the U.S. health disadvantage. Conditions that give rise to substance abuse, such as stressful living conditions (see Chapters 6-7) and mental illness (see Chapter 2), may have a more direct causal link to the health outcomes.
15In 2008, 9 of 10 poisoning deaths in the United States were caused by medications or illicit drugs (Kuehn, 2012).
Question 1: Do Sexual Practices Matter to Health?
Failure to use contraception and sexual activity with multiple partners can lead to unintended pregnancies and its consequences and can expose the participants to sexually transmitted infections, such as chlamydia, gonorrhea, syphilis, herpes zoster, HIV, and human papilloma virus (which can cause cervical cancer). Exposure to these risks is highest among sexually active adolescents and young adults and even higher when sexual activity begins at an early age or occurs under conditions (e.g., while intoxicated) when protective measures are often neglected. The factors that influence adolescent pregnancy rates include the timing of first intercourse, the frequency of intercourse, condom and contraceptive use, and the efficacy of the chosen contraceptive method(s) (Hatcher et al., 2007).
Are High-Risk Sexual Practices More Prevalent in the United States
Than in Other High-Income Countries?
Oral contraceptives are generally used less regularly in the United States than in other countries (Mosher and Jones, 2010). International data on the prevalence of unsafe sex among adults, such as a history of multiple sexual partners or the prevalence of unprotected intercourse among men engaging in sex with other men, are lacking. Data comparing adolescent sexual behavior in the United States and other countries are somewhat outdated, but they suggest that unsafe sex is more common among U.S. adolescents than their peers in other countries.
A comparison of survey data in the United States and four other countries—Canada, France, Sweden, and the United Kingdom—found few cross-country differences in the proportion of adolescents who were sexually active (Darroch et al., 2001). However, the study found that U.S. adolescents were more likely to report having sex before age 15 (14 percent) than were those in Canada, France, and the United Kingdom (4-9 percent) and more likely to report having multiple sexual partners (two or more in the past year). Female U.S. teenagers were less likely to report using contraceptives at either their first or most recent intercourse (25 and 20 percent, respectively) than were those in France (11 and 12 percent, respectively), the United Kingdom (21 and 4 percent, respectively), and Sweden (22 and 7 percent, respectively). Condom use at the first sexual encounter was lower in the United States than in France, similar to the United Kingdom, and higher than in Sweden, but sole use of condoms at last intercourse (an unreliable contraceptive method) was higher in the United States than in the other study countries. Overall condom use (including dual use with
hormonal contraceptives, a more effective method) was lower in the United States than in the United Kingdom and similar to levels in France (Darroch et al., 2001).
In an examination of sexual practices among 15-year-olds, Santelli and colleagues (2008) compared responses to the 2001 and 2003 Youth Risk Behavioral Survey (YRBS) in the United States to responses to similar questions about sexual behavior that adolescents in 24 countries completed on the HBSC survey of 2001-2002, as reported by Godeau and colleagues (2008). In the United States, reported use of condoms after last intercourse was 66 percent among adolescent females and 75 percent among adolescent males compared with 74 percent for both genders in the HBSC survey.16 Reported use of oral contraceptives was 11 and 6 percent, respectively, among U.S. adolescent females and males but 24 percent for adolescents in other countries. Dual use (condom and oral contraceptive) at last intercourse was 4 and 3 percent, respectively, in adolescent females and males, compared with 16 percent for adolescents in other countries. Use of either the pill or condoms was 72 and 77 percent, respectively, for U.S. adolescent girls and boys and 82 percent in the HBSC survey.
Emergency contraception was not examined in the U.S. survey, but Santelli et al. (2008) noted that the rates in other countries for emergency contraception after last sexual intercourse (9 percent) matched the lifetime rate (8 percent) reported for youths aged 15-19 in the U.S. National Survey of Family Growth (Mosher et al., 2004). Santelli and colleagues (2009) examined subsequent data, which showed a similar pattern: in the 2005-2006 HBSC survey (Currie et al., 2008), 72 and 81 percent, respectively, of female and male 15-year-olds in the surveyed countries reported using condoms at last intercourse, compared with 62 and 75 percent for females and males, respectively, in the 2005 and 2007 YRBS. Contraceptive use was also much higher in Europe (Eaton et al., 2006, 2008; Santelli et al., 2009).
Do High-Risk Sexual Practices Explain the U.S. Health Disadvantage?
The apparent tendency of U.S. adolescents to have multiple sexual partners, to not use oral contraceptives as often as their peers in other countries, to rely on less effective barrier methods, such as condoms, and to use them less often than their counterparts in some other countries could explain the higher rates of pregnancies and sexually transmitted infections among U.S. adolescents, the high burden of HIV/AIDS in the United States, and perhaps the excess deaths from some congenital anomalies. However, the
16In the high-income countries included in the HBSC survey (Austria, Canada, England, Finland, France, the Netherlands, Portugal, Spain, Sweden, Switzerland), condom use ranged from 53 to 86 percent (Godeau et al., 2008).
data available to compare countries are not equivalent due to differences in instrument design and sampling methods.
Question 1: Do Injurious Behaviors Matter to Health?
Injuries are the leading cause of death among U.S. children and adults from ages 1-45 (National Center for Health Statistics, 2012) and, in the case of nonfatal injuries, are responsible for a heavy burden of lifelong neurologic and other disabilities. As detailed in Chapter 1, transportation-related injuries and violence account for many of the extra years of life lost in the United States relative to other high-income countries. Unintentional injuries include poisonings, motor vehicle crashes, falls, drowning, fires, asphyxiation, and burns (National Center for Health Statistics, 2012). Although such injuries are sometimes true “accidents” that could not have been prevented—and others result from unsafe product designs or weak safety provisions (see Chapter 8)—a large fraction of injuries result from the behavior of individuals. Examples include operating a motor vehicle while intoxicated, not using occupant restraints such as seatbelts and child safety seats, riding motorcycles or bicycles without a helmet, not installing smoke detectors, setting water heaters at scalding temperatures, unsafe boating practices, and failure to secure firearms and medications from children. Intentional injuries (including assault, murder, rape, child abuse and neglect, and intimate partner violence) and self-inflicted injuries (including suicidal behaviors) claim lives but also inflict both physical disabilities and emotional scars (e.g., posttraumatic stress disorder) (Felitti et al., 1998).
Are Injurious Behaviors More Prevalent in the United States
Than in Other High-Income Countries?
Data are lacking to determine whether injurious behaviors are more common in the United States than elsewhere. For example, although Chapter 1 reported that poisoning accounts for two-thirds of U.S. nontransportation-related injury deaths before age 50 (and has recently replaced motor vehicles crashes as the leading cause of U.S. injury deaths) (Warner et al., 2011), there are no data to assess whether, for example, U.S. children have easier access to unsecured medications or toxic chemicals than children in peer countries. Nor are there data to know whether Americans are more susceptible to risks from falls, drowning, asphyxiation, or burns.
There is some evidence to compare driving practices. For example, data
|Country||Drivers Seatbelt Front||Wearing
|Road Traffic Deaths Attributable to Alcohol (%)|
aUse of bicycle safety helmets in Europe is thought to be less common.
SOURCE: Data from World Health Organization (2009, Table A.3, Table A.4, Table A.6).
in Table 5-1, taken from the WHO Global Status Report on Road Safety, suggest that Americans are less likely to fasten front17 seatbelts than those in most high-income countries (World Health Organization, 2009). And only 58 percent of motorcyclists in the United States wear helmets, far less than the rate reported in most other high-income countries, where more than 95 percent of motorcyclists reportedly wear helmets (World Health Organization, 2009). However, other reports suggest that helmet use in Europe is probably lower, especially among bicyclists. For example, the International Transport Forum (2011) reports that helmets are worn by 49 percent of cyclists in Norway and 31 percent of cyclists in Finland, and these figures may be overestimates of actual use.
Almost one-third (32 percent) of U.S. road traffic deaths are attributable
17Use of rear seatbelts in the United States is 76 percent, a higher rate than in many (but not all) high-income countries.
to alcohol, a higher proportion than in other high-income countries (see Table 5-1), including countries with greater per capita alcohol consumption.18 (See Chapter 8 for a more detailed discussion of cross-national differences in traffic fatalities.)
There is little evidence that violent acts occur more frequently in the United States than elsewhere.19 Crime statistics cannot be compared accurately across countries because offences have different legal definitions, but the rate of criminal (police-recorded) nonviolent assaults in the United States is in the middle of the range reported by high-income countries (United Nations Office on Drugs and Crime, 2012). Among the study’s 17 peer countries (see Chapter 1), the United States had the sixth highest rate of physical or sexual assaults on partners (intimate partner violence) (OECD, 2012k). Studies that have examined the incidence of violent behavior (e.g., engaging in fights) among U.S. adolescents relative to their peers in other countries have reported mixed results (Smith-Khuri et al., 2004; U.S. Department of Health and Human Services, 2001b).
One behavior that probably explains the excess lethality of violence and unintentional injuries in the United States is the widespread possession
18Data from the 2010 BRFSS indicate that 1.8 percent of Americans report at least one alcohol-impaired driving episode in the past 30 days (Centers for Disease Control and Prevention, 2011f). Drunk driving appears to have decreased in the United States in recent decades. For example, the percentage of weekend nighttime drivers and underage drivers with a blood alcohol concentration (BAC) of 0.8 g/L or greater decreased from 1973 to 2007 (Transportation Research Board, 2011). The percentage of U.S. crash fatalities involving a driver with a BAC greater than zero decreased from 55 to 38 percent between 1982 and 1995, and it was 37 percent in 2008. However, such statistics are of limited value because BAC tests are performed on only 40 percent of U.S. drivers involved in fatal crashes (Transportation Research Board, 2011). Comparisons with other countries are complicated by differences in measurement methods and legal BAC limits. Nonetheless, despite having higher per capita rates of alcohol consumption, Australia, Germany, Sweden, and the United Kingdom appear to have achieved lower rates of alcohol-related traffic fatalities than the United States, both as a percentage of fatalities and as measured per vehicle kilometer of travel (Transportation Research Board, 2011). This may also reflect differences in management, planning, and enforcement policies (see Chapter 8).
19Some surveys hint at greater acceptability of violence in the United States. For example, a survey of students in cities in Estonia, Finland, Romania, Russia, and the United States found that American students were the most likely to justify killing to defend property and were more supportive of war (McAlister et al., 2001).
of firearms and the common practice of storing them (often unlocked) at home. The statistics are dramatic:
• The United States is home to approximately 35-50 percent of the world’s civilian-owned firearms.
• Fully 48 percent of all violent deaths (66 percent of homicides) in the United States involve firearms (Karch et al., 2011).
• As of 2004, 38 percent of U.S. households and 26 percent of individuals reported owning at least one firearm, and almost one-half of individual owners reported owning four or more firearms (Hepburn et al., 2007).
• Although U.S. youth may be no more violent than those in other countries, they are more likely to carry a firearm (Pickett et al., 2005). In a survey of high school students in Boston, 5 percent reported carrying a firearm (Hemenway et al., 2011).
• U.S. civilians own four times the number of automatic and semi-automatic rifles owned by the U.S. Army (Small Arms Survey, 2007).
International comparisons draw an association between firearm ownership rates and the rate of deaths from homicide and suicide (Hemenway and Miller, 2000; Killias, 1993).21 Among 23 OECD countries, 80 percent of all firearm deaths occurred in the United States (Richardson and Hemenway, 2011). Consistent with the findings reported in Chapter 1, Richardson and Hemenway (2011) reported that the excess homicide rate in the United States relative to 22 high-income OECD countries was driven by firearm homicide rates that were 20 times higher (43 times higher for those ages 15-24). Firearms are also associated with deaths from causes other than homicide. The presence of a firearm in the home is a risk factor for suicide (Johnson et al., 2010; Miller and Hemenway, 2008): fully 52 percent of all U.S. suicides involve a firearm (Karch et al., 2011). Suicide rates are not higher in the United States than in peer countries, but firearm suicide rates are 5.8 times higher, and unintentional firearm deaths are 5.2 times higher (Richardson and Hemenway, 2011).
20Because many owners have more than one firearm, the actual proportion of Americans who own firearms is far less than 89 percent.
21A notable exception is Switzerland, which has one of the highest rates of gun ownership but relatively low crime rates.
FIGURE 5-5 Civilian firearm ownership in 16 peer countries.
NOTES: The data reflect the number of firearms owned per 100 persons. Because many people own multiple firearms, the proportion of people who own firearms is lower.
SOURCE: Data from Small Arms Survey (2007).
Do Injurious Behaviors Explain the U.S. Health Disadvantage?
There are inadequate data to know whether higher death rates from unintentional injuries in the United States are the result of more injurious behaviors or environmental factors. Countries do not collect similar data on behaviors that affect the risk of falls, poisoning, drowning, or other behaviors, making valid international comparisons of these behaviors impossible. It may be the severity, and not the incidence, of injuries that differs in the United States, a factor influenced not only by personal actions—such as the above evidence that Americans may be less apt to use seatbelts or helmets and are more involved in accidents involving alcohol—but also by deficiencies in product and roadway designs (e.g., crash protection) and resources that protect public safety (e.g., law enforcement). However, the prevalence of firearms in the United States looms large as an explanation for higher death rates from violence, suicidal impulses, and accidental shootings.22
22The validity of the correlation between firearm ownership and homicide is strongly debated by opponents of stricter gun control laws in the United States.
Individual behaviors contribute to each of the nine domains in which the United States demonstrates a health disadvantage relative to other countries. Smoking contributes to adverse birth outcomes, heart disease, and chronic pulmonary disease, although smoking rates are now lower in the United States than in other countries and would explain little of the U.S. health disadvantage among adults younger than age 50. Unhealthy diet and low physical activity contribute to higher rates of obesity and diabetes. Alcohol consumption, other drug use, and unsafe sexual practices contribute to drug-related mortality, HIV/AIDS, sexually transmitted infections, and adolescent pregnancies. Substance abuse also contributes to injuries (unintentional and intentional), as do injurious practices and the prevalence of firearms in the United States. These conditions are causally interconnected. For example, obesity in early life can give rise to diabetes and, over time, the macrovascular complications of heart disease. Thus, health behaviors may play a pivotal role over the life course in promoting the conditions responsible for the U.S. health disadvantage.
Whether health behaviors in the United States differ significantly from those in other countries and the extent to which they explain the U.S. health disadvantage cannot be determined without better cross-national data. Further research is needed to define the specific behaviors that are predictive of adverse health outcomes, based on rigorous science, and to validate appropriate metrics and sampling methods for measuring those behaviors. Modes of administration, such as using accelerometers and other sensors instead of relying on self-report of physical activity, may also need to evolve. Countries would need to adopt a consistent battery of questions about health behaviors to enable meaningful international comparisons. Crime statistics would have to be more consistent across countries to understand international differences in violence. Historical cohort data on behavior patterns in prior years, or in prior decades, may be important in order to understand current disparities in the prevalence of diseases that result from a lifetime of sedentary behavior, unhealthy diet, or substance abuse.
Although no single behavior can explain the U.S. health disadvantage, the high prevalence of multiple unhealthy behaviors in the United States (Fine et al., 2004) may play a large role. Advocates of “personal responsibility” would note that people choose to engage in all of the behaviors discussed in this chapter, from eating sweets to carrying handguns, and they should be free to make those choices and bear the consequences. But such choices may not always be made “freely”: they are made in a societal and environmental context (Brownell et al., 2010). Parents may want to serve healthy meals but may not be able to do so without nearby retailers that sell fresh produce (Larson et al., 2009). They may want their children to
play outside but the neighborhood may be unsafe. As Chapter 7 details, Americans do not make product choices in a vacuum, and they face both environmental supports and barriers to healthy behaviors. Advertising and marketing of tobacco, alcohol, and unhealthy foods; sexualized content in television, film, and musical entertainment; and exposure to violence and stress have known consequences to health (Gordon et al., 2010; Harris et al., 2009; Lovato et al., 2003; Mosher, 2011; Nestle, 2007; Robinson et al., 2007; Spano et al., 2012; Vermeiren et al., 2003).
Socioeconomic conditions also limit choices and play an important role in the prevalence of unhealthy behaviors and their effect on mortality (Lantz et al., 2010). Tobacco use and other unhealthy behaviors are significantly more prevalent among adults with limited education and limited incomes (National Center for Health Statistics, 2012; Pampel et al., 2010) (see Chapters 6 and 7). Barriers in access to medical care generally, which is discussed in Chapter 4, and to specific health care services (e.g., pharmacotherapy for smoking cessation, substance abuse counseling, oral contraceptives) also limit the ability of individuals to adopt and sustain behavioral changes. Finally, policy and culture are important factors in understanding why health behaviors might differ between high-income countries: these are discussed in Chapter 8.