Heart Disease and Stroke Statistics—2018 Update: A Report From the American Heart Association

Each chapter listed in the Table of Contents (see next page) is a hyperlink to that chapter. The reader clicks the chapter name to access that chapter.

Table of Contents

Each chapter listed here is a hyperlink. Click on the chapter name to be taken to that chapter.

Summary

Each year, the American Heart Association (AHA), in conjunction with the Centers for Disease Control and Prevention, the National Institutes of Health, and other government agencies, brings together in a single document the most up-to-date statistics related to heart disease, stroke, and the cardiovascular risk factors listed in the AHA’s My Life Check - Life’s Simple 7 (Figure¹), which include core health behaviors (smoking, physical activity, diet, and weight) and health factors (cholesterol, blood pressure [BP], and glucose control) that contribute to cardiovascular health. The Statistical Update represents a critical resource for the lay public, policy makers, media professionals, clinicians, healthcare administrators, researchers, health advocates, and others seeking the best available data on these factors and conditions. Cardiovascular disease (CVD) and stroke produce immense health and economic burdens in the United States and globally. The Update also presents the latest data on a range of major clinical heart and circulatory disease conditions (including stroke, congenital heart disease, rhythm disorders, subclinical atherosclerosis, coronary heart disease [CHD], heart failure [HF], valvular disease, venous disease, and peripheral artery disease) and the associated outcomes (including quality of care, procedures, and economic costs). Since 2007, the annual versions of the Statistical Update have been cited >20 000 times in the literature. From January to July 2017 alone, the 2017 Statistical Update was accessed >106 500 times.

• Download figure
• Download PowerPoint

Each annual version of the Statistical Update undergoes revisions to include the newest nationally representative data, add additional relevant published scientific findings, remove older information, add new sections or chapters, and increase the number of ways to access and use the assembled information. This year-long process, which begins as soon as the previous Statistical Update is published, is performed by the AHA Statistics Committee faculty volunteers and staff and government agency partners. This year’s edition includes new data on the monitoring and benefits of cardiovascular health in the population, new metrics to assess and monitor healthy diets, new information on stroke in young adults, an enhanced focus on underserved and minority populations, a substantively expanded focus on the global burden of CVD, and further evidence-based approaches to changing behaviors, implementation strategies, and implications of the AHA’s 2020 Impact Goals. Below are a few highlights from this year’s Update.

Current State of Cardiovascular Health in the United States: What’s New? (Chapter 2)

• Findings continue to accumulate from US and international studies showing a strong protective association between ideal cardiovascular health metrics (My Life Check - Life’s Simple 7¹) and many clinical and preclinical conditions, including premature all-cause mortality, CVD mortality, ischemic heart disease mortality, HF, carotid arterial wall stiffness, coronary artery calcium progression, impaired physical function, cognitive decline, stroke, depression, end-stage renal disease, chronic obstructive pulmonary disease, deep venous thromboembolism, and pulmonary embolism.

• New data reported this year expand this list to include the positive association of cardiovascular health metrics with better leukocyte telomere length and reduced mean healthcare expenditures.

• A recent major study in blacks found that the risk of incident HF was 61% lower among those with ≥4 ideal cardiovascular health metrics than among those with 0 to 2 ideal metrics.

Smoking and Tobacco Use (Chapter 3)

• The prevalence of current smoking in the United States in 2015 was 15.1% for adults and 4.2% for adolescents. Although there has been a consistent decline in tobacco use in the United States, significant disparities persist. Substantially higher tobacco use prevalence rates are observed in American Indian/Alaska Natives and lesbian, gay, bisexual, and transgender populations, as well as among individuals with low socioeconomic status, those with mental illness, individuals with HIV who are receiving medical care, and those who are active-duty military.

• Tobacco use remains the leading cause of preventable death in the United States and globally. It was estimated to account for 7.2 million deaths worldwide in 2015.

• Over the past 5 years, there has been a sharp increase in e-cigarette use among adolescents.

• Policy-level interventions such as Tobacco 21 Laws and MPOWER are being adopted and have been associated with reductions in tobacco use incidence and prevalence.

Physical Inactivity (Chapter 4)

• According to the 2015 National Health Interview Survey, only 21.5% of American adults reported achieving adequate leisure-time aerobic and muscle-strengthening activities to meet the physical activity guidelines; however, the national survey did not record activity accumulated in occupational, transport, and domestic domains. These other domains could provide substantial contributions to total physical activity, especially in minority populations.

• Community physical activity interventions and strategies at worksites, schools, and the built environment have proven cost-effective in reducing medical spending. For example, nearly $3 in medical cost savings is realized for every $1 invested in building bike and walking trails.

• The current physical activity guidelines focus almost entirely on achieving moderate- and vigorous-intensity physical activity, but recent evidence suggests that substituting even 10 minutes of sedentary time with 10 minutes of light-intensity activity is associated with a 9% lower mortality risk.

• The prevalence of adolescents meeting the aerobic physical activity guidelines is 27.1%, almost as low as their adult counterparts. Moreover, the amount of time children spend sitting may be on the rise. From 2009 to 2015, the proportion of youth who reported spending at least 3 hours per day on computers for activities other than school work (eg, videogames or other computer games) increased from 24.9% to 41.7%.

Nutrition (Chapter 5)

• A recent study using a comparative risk assessment model estimated that 45.4% of US deaths caused by heart disease, stroke, and type 2 diabetes mellitus (DM) (cardiometabolic mortality) in 2012 were attributable to poor dietary habits. The top contributing poor dietary factors included high sodium intake, low levels of nuts and seeds, high intake of processed meats, low consumption of seafood omega-3 fats, low intake of vegetables, low intake of fruits, and high consumption of sugar-sweetened beverages. Between 2002 and 2012, diet-associated cardiometabolic death rates decreased for polyunsaturated fats (−20.8%), nuts and seeds (−18.0%), and sugar-sweetened beverages (−14.5%) but increased for sodium (5.8%) and unprocessed red meats (14.4%).

• A report from the Global Burden of Disease Study stated that an estimated 22.4% of all male deaths and 20.7% of all female deaths in 2015 were attributable to poor dietary factors.

• Data from the US Department of Agriculture’s Economic Research Service showed that the proportion of food expenditures away from home increased from 47.2% in 2004 to 50.1% in 2014.

Overweight and Obesity (Chapter 6)

• Obesity prevalence increased in the National Health and Nutrition Examination Survey (NHANES) 2013 to 2014 survey years in both adults and children compared with NHANES 2011 to 2012 (Chart 6-11). In adults, the prevalence increased from 34.9% in 2011 to 2012 to 37.7% in 2013 to 2014. In children (ages 2–19 years), the prevalence of obesity increased from 16.9% in 2011 to 2012 to 17.2% in 2013 to 2014.

• Significant race and sex differences in obesity rates were still observed in the most recent survey years, with female and minority populations more likely to be obese than their male and white counterparts.

• A meta-analysis from 2016 suggested that CVD risk was higher (relative risk, 1.45) in obese individuals without metabolic syndrome than in metabolically healthy normal-weight participants, which suggests that obesity is a risk factor even in the absence of high blood pressure, high cholesterol, and DM.

High Blood Cholesterol and Other Lipids (Chapter 7)

• 21% of youths 6 to 19 years of age have at least 1 abnormal cholesterol measure.

• The Healthy People 2010 target of an age-adjusted mean total cholesterol level of ≤200 mg/dL has been achieved in the overall adult population, in males, in females, and in all racial/ethnic and sex subgroups. The Healthy People 2020 target is a mean total blood cholesterol of 178 mg/dL for adults, which has not yet been achieved for any subgroup.

• An estimated 28.5 million adults ≥20 years of age have serum total cholesterol levels ≥240 mg/dL, with a prevalence of 11.9%.

• 56.0 million (48.6%) US adults ≥40 years of age are eligible for statin therapy based on the 2013 American College of Cardiology/AHA guidelines for the management of blood cholesterol; however, even after the guidelines were published, statin use and mean low-density lipoprotein cholesterol level remained unchanged, which suggests that increased efforts to understand and implement these guidelines are needed.

High Blood Pressure (Chapter 8)

• From 2005 to 2015, the death rate attributable to high BP increased by 10.5%, and the actual number of deaths attributable to high BP rose by 37.5%.

• On the basis of data from 135 population-based studies (N=968 419 adults from 90 countries), it was estimated that 31.1% (95% confidence interval, 30.0%–32.2%) of the world’s adult population had hypertension in 2010.

• There were 3.47 billion adults worldwide with systolic BP of 110 to 115 mm Hg or higher in 2015. In 2015, 874 million adults had systolic BP ≥140 mm Hg.

• Between 1999 to 2000 and 2011 to 2012, the prevalence of prehypertension decreased among US adults, from 31.2% to 28.2%.

Diabetes Mellitus (Chapter 9)

• There is substantial heterogeneity in the prevalence of DM across US counties. In 2012, the overall county-level prevalence of DM ranged from 8.8% to 26.4%; the prevalence of diagnosed DM was estimated to range from 5.6% to 20.4%, and the prevalence of undiagnosed DM ranged from 3.2% to 6.8%.

• In 2015, an estimated 5.2 million deaths globally were attributed to DM.

• The prevalence of concurrent DM, hypertension, and hypercholesterolemia among US adults increased over time from 3% in 1999 to 2000 to 6.3% in 2011 to 2012.

• The incidence rate of DM per 1000 person-years associated with having 0, 1, 2, 3, 4, and 5 or 6 ideal cardiovascular health factors was 21.8, 18.6, 13.0, 11.2, 4.7, and 3.6, respectively.

Metabolic Syndrome (Chapter 10)

• On the basis of NHANES data from 2001 to 2012, the prevalence of metabolic syndrome was higher among females (34.4%) than males (29%) and increased with advancing age. The prevalence of metabolic syndrome was 17% among people <40 years old, 29.7% for people 40 to 49 years old, 37.5% among those 50 to 59 years old, and >44% among people ≥60 years of age.

• Isolated metabolic syndrome, which could be considered an earlier form of metabolic syndrome, has been defined as the presence of ≥3 metabolic syndrome components without overt hypertension or DM. In a community-based random sample of residents of Olmsted County, MN, those with isolated metabolic syndrome were found to be at increased risk of incident hypertension, DM, diastolic dysfunction, and reduced renal function (glomerular filtration rate <60 mL/min) compared with healthy control subjects (P<0.05).

Kidney Disease (Chapter 11)

• According to the Global Burden of Disease study, the prevalence of chronic kidney disease is rising in almost every country of the world, primarily because of aging populations. In 2015, the total estimated prevalence was 323 million people (95% confidence interval, 313–330 million people), a 27% increase since 2005.

• In the United States, Medicare spent more than $50 billion in 2014 caring for people ≥65 years of age with chronic kidney disease. More than 70% of this spending was attributable to patients who had comorbid DM or congestive HF.

• CVD is the leading cause of death among people with kidney disease. Among those with end-stage renal disease, CVD accounts for more than half of the deaths with known causes, with arrhythmias and sudden cardiac death accounting for nearly 40%.

Total Cardiovascular Diseases (Chapter 12)

• The mortality for males and females in the United States declined from 1979 to 2015 (Chart 12-19). In 2015, 2 712 630 resident deaths were registered in the United States, and 10 leading causes accounted for 74.2% of all registered deaths. The 10 leading causes of death were heart disease (No. 1), cancer (No. 2), chronic lower respiratory diseases (No. 3), unintentional injuries (No. 4), stroke (No. 5), Alzheimer disease (No. 6), DM (No. 7), influenza and pneumonia (No. 8), kidney disease (No. 9), and suicide (No. 10).

• CHD (43.8%) is the leading cause of deaths attributable to CVD in the United States, followed by stroke (16.8%), high BP (9.4%), HF (9.0%), diseases of the arteries (3.1%), and other CVDs (17.9%).

• By 2035, >130 million adults in the US population (45.1%) are projected to have some form of CVD, and total costs of CVD are expected to reach $1.1 trillion in 2035, with direct medical costs projected to reach $748.7 billion and indirect costs estimated to reach $368 billion.

Stroke (Cerebrovascular Disease) (Chapter 13)

• Although there has been considerable reduction in stroke risks and stroke outcomes, the racial and geographic disparities remain significant, with African Americans and residents of the southeastern United States experiencing the greatest excess disease burden. According to data from the Centers for Disease Control and Prevention 2015 Behavioral Risk Factor Surveillance System, 2.6% of non-Hispanic whites, 4.1% of non-Hispanic blacks, 1.5% of Asian/Pacific Islanders, 2.3% of Hispanics (of any race), 5.2% of American Indian/Alaska Natives, and 4.7% of other races or multiracial people had a history of stroke. Stroke prevalence in adults is 2.7% in the United States, with the lowest prevalence in Minnesota (1.9%) and the highest prevalence in Alabama (4.3%).

• The impact of hypertension management on stroke risk is evident with the greater risk reduction among those with more intense treatment.

Congenital Cardiovascular Defects and Kawasaki Disease (Chapter 14)

• Between 0.4 and 1 per 100 US infants are born with congenital heart disease, ranging from mild asymptomatic lesions to severe life-threatening conditions; approximately one fourth of these lesions will require intervention during infancy.

• Kawasaki disease, an acquired acute inflammatory condition associated at its worst with coronary artery dilation, aneurysms, and coronary ischemia, is most common in Japanese, Taiwanese, and Korean populations and is increasing in incidence over time.

Disorders of Heart Rhythm (Chapter 15)

• There is increasing awareness that atrial fibrillation (AF) is frequently unrecognized. At present, the detection of AF, even in an asymptomatic stage, is the basis for risk stratification for stroke and appropriate decision making about the need for anticoagulant therapy. Ongoing trials are evaluating the risks and benefits of anticoagulation therapy among patients at high risk for stroke but without a prior history of AF. The findings from these studies will help to determine optimal strategies for subclinical AF screening and treatment.

• A prospective registry from 47 countries reported substantial variability in annual AF mortality by region. Annual AF mortality in South America (17%) and Africa (20%) was double the mortality rate in North America, Western Europe, and Australia (10%; P<0.001). In individuals with AF, HF deaths (30%) exceeded deaths caused by stroke (8%).

• Observational data have suggested that overweight and obese individuals with symptomatic AF who opted for weight loss and aggressive risk factor management had fewer hospitalizations, cardioversions, and ablation procedures than their counterparts who declined enrollment. The risk factor management group was associated with a predicted 10-year cost savings of $12 094 (in 2010 Australian dollars).

Sudden Cardiac Arrest (Chapter 16)

• Sudden cardiac death appears among the multiple causes of death on 13.5% of death certificates (366 807 of 2 712 630), which suggests that 1 of every 7.4 people in the United States will die of sudden cardiac death. Because some people survive sudden cardiac arrest, the lifetime risk of cardiac arrest is even higher.

• Survival after out-of-hospital cardiac arrest (2006–2015, Resuscitation Outcomes Consortium Epistry data) and in-hospital cardiac arrest (2000–2015, Get With The Guidelines data) has continued to improve over time.

• Large regional variations in survival to hospital discharge (range, 3.4%–22.0%) and survival with functional recovery (range, 0.8%–20.1%) have been observed between 132 counties in the United States. Differences in rates of layperson cardiopulmonary resuscitation (CPR) explained much of this variation.

• In a survey of 9022 people in the United States in 2015, 18% of participants reported current training in CPR and 65% reported having CPR training at some point. The prevalence of CPR training was lower among Hispanic/Latino people, older adults, individuals with less formal education, and lower-income groups.

Subclinical Atherosclerosis (Chapter 17)

• Although the presence and a higher burden of coronary artery calcification (CAC) are well established to be associated with an increased risk of CHD, there are few data highlighting the associations with risk of stroke. A recent meta-analysis reporting data from 13 262 asymptomatic patients (mean age 60 years, 50% males) reported a nearly 3-fold higher risk of stroke with presence versus absence of CAC.

• Absence of CAC was a powerful negative marker for clinical atherosclerotic CVD in a study of African American adults with up to 10 years of follow-up. A significant proportion of those considered statin eligible based on the current cholesterol guideline had no detectable CAC, and the observed atherosclerotic CVD risk was below the threshold for statin consideration.

• A recent study from MESA (Multi-Ethnic Study of Atherosclerosis) demonstrated that during a median 10-year follow-up, among 12 negative risk markers (CAC=0, carotid intima-media thickness, absence of carotid plaque, >5% change in brachial flow-mediated dilation, ankle-brachial index >0.9 and <1.3, high-sensitivity C-reactive protein <2 mg/L, homocysteine <10 µmol/L, N-terminal pro-B-type natriuretic peptide <100 pg/mL, no microalbuminuria, no family history of CHD [any/premature], absence of metabolic syndrome, and healthy lifestyle), CAC=0 had the lowest diagnostic likelihood ratio for all CHD (0.41) and CVD (0.54).

Coronary Heart Disease, Acute Coronary Syndrome, and Angina Pectoris (Chapter 18)

• On the basis of data from NHANES 2011 to 2014, an estimated 16.5 million Americans ≥20 years of age have CHD.

• This year, ≈720 000 Americans will have a new coronary event (defined as first hospitalized myocardial infarction [MI] or CHD death), and ≈335 000 will have a recurrent event.

• CHD mortality dropped by 34.4% from 2005 to 2015, with a predicted continued decline (27% reduction by 2030); however, race disparities are projected to persist.

• Whites had a higher rate of recognized MI than blacks (5.04 versus 3.24 per 1000 person-years) in the Atherosclerosis Risk in Communities Study.

• In individuals ≥45 years of age, median survival (in years) after a first MI is 8.4 for white males, 5.6 for white females, 7.0 for black males, and 5.5 for black females.

• Individuals self-reporting low income and low education have twice the incidence of CHD as those reporting high income and high education (10.1 per 1000 person-years versus 5.2 per 1000 person-years, respectively).

Cardiomyopathy and Heart Failure (Chapter 19)

• The prevalence of HF continues to rise over time with the aging of the population. An estimated 6.5 million American adults ≥20 years of age had HF between 2011 and 2014 compared with an estimated 5.7 million between 2009 and 2012.

• Although survival of adults who develop HF after an MI has improved over time, likely because of secondary prevention therapies, CHD remains a major risk factor for HF, along with hypertension and DM.

• Primary prevention of HF can be augmented by greater adherence to the AHA’s My Life Check - Life’s Simple 7 goals; optimal profiles in smoking, body mass index, physical activity, diet, cholesterol, blood pressure, and glucose are associated with a lower lifetime risk of HF and more favorable cardiac structure and functional parameters by echocardiography.

• Of incident hospitalized HF events, approximately half are characterized by reduced ejection fraction and the other half by preserved ejection fraction. Black males had the highest proportion of presentations with reduced ejection fraction (≈70%); white females had the highest proportion of HF hospitalizations with preserved ejection fraction (≈60%).

Valvular Diseases (Chapter 20)

• Although rheumatic heart disease is uncommon in high-income countries such as the United States, it remains an important cause of morbidity and mortality in low- and middle-income countries.

• Both administrative and community-based data report that the incidence of infective endocarditis did not change after the publication of the 2007 AHA guidelines for prevention of infective endocarditis,² which restricted the indications for antibiotic prophylaxis before dental procedures.

• From the time of initial US Food and Drug Administration approval in late 2011 through 2014, >26 000 transcatheter aortic valve replacements for calcific aortic stenosis were performed at 348 centers in 48 states.

• A recently published meta-analysis that included 50 studies that enrolled >44 000 patients, with a mean duration of follow-up of 21.4 months, compared transcatheter to surgical aortic valve replacement. Compared with surgical aortic valve replacement, balloon-expandable transcatheter aortic valve replacement in symptomatic high-risk patients had similar all-cause mortality; lower incidence of stroke, AF, major bleeding, and acute kidney injury; and higher incidence of vascular complications, aortic regurgitation, and pacemaker implantation.

• Percutaneous mitral valve repair techniques are becoming a common treatment option for high-risk patients with mitral regurgitation who are not deemed candidates for surgical repair. Data from the Society of Thoracic Surgeons/American College of Cardiology’s TVT Registry (transcatheter valve therapy registry) on patients commercially treated with the MitraClip percutaneous mitral valve repair device showed a reduction in the severity of mitral regurgitation and procedural success in >90% of cases, although mitral valve dysfunction at 12 months was more common with percutaneous mitral valve repair than with surgical repair.

Venous Thromboembolism (Deep Vein Thrombosis and Pulmonary Embolism), Chronic Venous Insufficiency, Pulmonary Hypertension (Chapter 21)

• Although there is no formal surveillance for deep vein thrombosis (DVT) or pulmonary embolism in the United States, national hospitalization data from 1996 to 2014 were compiled in this edition of the update. From 2005 to 2014, there was a 34% increase in hospitalizations for DVT and a 53% increase in hospitalizations for pulmonary embolism.

• Using 2014 data for DVT cases treated in the hospital (outlined above), if it is assumed that 30% of DVT cases were treated in the outpatient setting in 2014, then an estimated 676 000 cases of DVT occurred in 2014.

Peripheral Artery Disease and Aortic Diseases (Chapter 22)

• A 2016 study using data from the Nationwide Inpatient Sample demonstrated that admission rates related to critical limb ischemia remained constant from 2003 to 2011.

• A few recent studies have demonstrated that even individuals with low-normal ankle-brachial index (0.91–0.99) have reduced physical function compared with those with normal ankle-brachial index.

• US patients demonstrated higher rates of abdominal aortic aneurysm repair, a smaller abdominal aortic aneurysm diameter at the time of repair, and lower rates of abdominal aortic aneurysm rupture and abdominal aortic aneurysm–related death than patients in the United Kingdom.

Quality of Care (Chapter 23)

• Performance on inpatient measures of quality of care or measures at discharge in patients after MI or stroke exceeds 90% for most measures. Performance on outpatient measures was <80% for body mass index assessment, counseling for physical activity, and advising smokers and tobacco users to quit.

• Among patients who had out-of-hospital cardiac arrest, bystander CPR rates were 45.7% in adults and 61.4% in children.

Medical Procedures (Chapter 24)

• The mean hospital charges for cardiovascular procedures in 2014 ranged from $43 484 for carotid endarterectomy to $808 770 for heart transplantations.

• The total number of inpatient cardiovascular operations and procedures decreased 6%, from 8 461 000 in 2004 to 7 971 000 in 2014.

• Of the 10 leading diagnostic groups in the United States, the greatest number of surgical procedures were conducted in the cardiovascular system and obstetric procedures groups.

• In 2016, 3191 heart transplantations were performed in the United States, the most ever.

Economic Cost of Cardiovascular Disease (Chapter 25)

• In 2013 to 2014, the annual direct and indirect cost of CVD and stroke in the United States was an estimated $329.7 billion.

• CVD and stroke accounted for 14% of total health expenditures in 2013 to 2014, more than any major diagnostic group.

Conclusions

The AHA, through its Statistics Committee, continuously monitors and evaluates sources of data on heart disease and stroke in the United States to provide the most current information available in the Statistical Update. The 2018 annual Statistical Update is the product of a full year’s worth of effort by dedicated volunteer physicians and scientists, committed government professionals, and outstanding AHA staff members, without whom publication of this valuable resource would be impossible. Their contributions are gratefully acknowledged.

Emelia J. Benjamin, MD, ScM, FAHA, Chair

Paul Muntner, PhD, MHS, FAHA, Co-Vice Chair

Salim S. Virani, MD, PhD, FAHA, Co-Vice Chair

Sally S. Wong, PhD, RD, CDN, FAHA, AHA Science and Medicine Advisor

On behalf of the American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee

Note: Population data used in the compilation of NHANES prevalence estimates are for the latest year available. Extrapolations for NHANES prevalence estimates are based on the census resident population for 2014 because this is the most recent year of NHANES data used in the Statistical Update.

Acknowledgments

We wish to thank our colleagues: Dr David Goff, Dr Paul Sorlie, and Lucy Hsu, National Heart, Lung, and Blood Institute; Dr Elizabeth B. Pathak, Dr Mary G. George, Dr Stuart K. Shapira, and Dr Tiffany J. Colarusso, Centers for Disease Control and Prevention; Dr Fran F. Thorpe and Jim Beachy, American College of Cardiology; Dr Chris Shay, Ian Golnik, and Kathleen Smith, American Heart Association; and all the dedicated staff of the Centers for Disease Control and Prevention and the National Heart, Lung, and Blood Institute for their valuable comments and contributions.

Disclosures

Footnotes

1. About These Statistics

The AHA works with the CDC’s Division for Heart Disease and Stroke Prevention and the NCHS, the NHLBI, and other government agencies to derive the annual statistics in this Heart Disease and Stroke Statistics Update. This chapter describes the most important sources and the types of data used from them. For more details, see Chapter 27 of this document, the Glossary.

The surveys used are the following:

• ARIC—CHD and HF incidence rates

• BRFSS—ongoing telephone health survey system

• GCNKSS—stroke incidence rates and outcomes within a biracial population

• HCUP—hospital inpatient discharges and procedures (discharged alive, dead, or status unknown)

• MEPS—data on specific health services that Americans use, how frequently they use them, the cost of these services, and how the costs are paid

• NAMCS—physician office visits

• National Vital Statistics System—national and state mortality data

• NHAMCS—hospital outpatient and ED visits

• NHANES—disease and risk factor prevalence and nutrition statistics

• NHHCS—staff, services, and patients of home health and hospice agencies

• NHIS—disease and risk factor prevalence

• NIS of the AHRQ—hospital inpatient discharges, procedures, and charges

• NNHS—nursing home residents

• USRDS—kidney disease prevalence

• WHO—mortality rates by country

• YRBSS—health-risk behaviors in youth and young adults

Disease Prevalence

Prevalence is an estimate of how many people have a condition at a given point or period in time. The NCHS/CDC conducts health examination and health interview surveys that provide estimates of the prevalence of diseases and risk factors. In this Update, the health interview part of the NHANES is used for the prevalence of CVDs. NHANES is used more than the NHIS because in NHANES, AP is based on the Rose Questionnaire; estimates are made regularly for HF; hypertension is based on BP measurements and interviews; and an estimate can be made for total CVD, including MI, AP, HF, stroke, and hypertension.

A major emphasis of this Statistical Update is to present the latest estimates of the number of people in the United States who have specific conditions to provide a realistic estimate of burden. Most estimates based on NHANES prevalence rates are based on data collected from 2011 to 2014. These are applied to census population estimates for 2014. Differences in population estimates cannot be used to evaluate possible trends in prevalence because these estimates are based on extrapolations of rates beyond the data collection period by use of more recent census population estimates. Trends can only be evaluated by comparing prevalence rates estimated from surveys conducted in different years.

A major enhancement in the 2018 Statistical Update is a greater emphasis on the global burden of CVD. The Institute for Health Metrics and Evaluation contributed statistics and maps that characterize the distribution and prognosis of CVD and its risk factors.

Risk Factor Prevalence

The NHANES 2011 to 2014 data are used in this Update to present estimates of the percentage of people with high lipid values, DM, overweight, and obesity. The NHIS 2015 data are used for the prevalence of cigarette smoking and physical inactivity. Data for students in grades 9 through 12 are obtained from the YRBSS.

Incidence and Recurrent Attacks

An incidence rate refers to the number of new cases of a disease that develop in a population per unit of time. The unit of time for incidence is not necessarily 1 year, although incidence is often discussed in terms of 1 year. For some statistics, new and recurrent attacks or cases are combined. Our national incidence estimates for the various types of CVD are extrapolations to the US population from the FHS, the ARIC study, and the CHS, all conducted by the NHLBI, as well as the GCNKSS, which is funded by the NINDS. The rates change only when new data are available; they are not computed annually. Do not compare the incidence or the rates with those in past editions of the Heart Disease and Stroke Statistics Update (also known as the Heart and Stroke Statistical Update for editions before 2005). Doing so can lead to serious misinterpretation of time trends.

Mortality

Mortality data are generally presented according to the underlying cause of death. “Any-mention” mortality means that the condition was nominally selected as the underlying cause or was otherwise mentioned on the death certificate. For many deaths classified as attributable to CVD, selection of the single most likely underlying cause can be difficult when several major comorbidities are present, as is often the case in the elderly population. It is useful, therefore, to know the extent of mortality attributable to a given cause regardless of whether it is the underlying cause or a contributing cause (ie, the “any-mention” status). The number of deaths in 2015 with any mention of specific causes of death was tabulated by the NHLBI from the NCHS public-use electronic files on mortality.

The first set of statistics for each disease in this Update includes the number of deaths for which the disease is the underlying cause. Two exceptions are Chapter 8 (High Blood Pressure) and Chapter 19 (Cardiomyopathy and Heart Failure). HBP, or hypertension, increases the mortality risks of CVD and other diseases, and HF should be selected as an underlying cause only when the true underlying cause is not known. In this Update, hypertension and HF death rates are presented in 2 ways: (1) as nominally classified as the underlying cause and (2) as any-mention mortality.

National and state mortality data presented according to the underlying cause of death were computed from the mortality tables of the NCHS/CDC website or the CDC compressed mortality file. Any-mention numbers of deaths were tabulated from the electronic mortality files of the NCHS/CDC website.

Population Estimates

In this publication, we have used national population estimates from the US Census Bureau for 2015¹ in the computation of morbidity data. NCHS/CDC population estimates² for 2015 were used in the computation of death rate data. The Census Bureau website contains these data, as well as information on the file layout.

Hospital Discharges and Ambulatory Care Visits

Estimates of the numbers of hospital discharges and numbers of procedures performed are for inpatients discharged from short-stay hospitals. Discharges include those discharged alive, dead, or with unknown status. Unless otherwise specified, discharges are listed according to the first-listed (primary) diagnosis, and procedures are listed according to all listed procedures (primary plus secondary). These estimates are from HCUP 2014. Ambulatory care visit data include patient visits to physician offices and hospital outpatient departments and EDs. Ambulatory care visit data reflect the first-listed (primary) diagnosis. These estimates are from the NAMCS and NHAMCS of the NCHS/CDC. Data for community health centers, which were included in estimates in previous years, were not available for 2014 NAMCS estimates included in this Update.

International Classification of Diseases

Morbidity (illness) and mortality (death) data in the United States have a standard classification system: the ICD. Approximately every 10 to 20 years, the ICD codes are revised to reflect changes over time in medical technology, diagnosis, or terminology. If necessary for comparability of mortality trends across the 9th and 10th ICD revisions, comparability ratios computed by the NCHS/CDC are applied as noted.³ Effective with mortality data for 1999, we are using the 10th revision (ICD-10).⁴

Age Adjustment

Prevalence and mortality estimates for the United States or individual states comparing demographic groups or estimates over time are either age specific or age adjusted to the 2000 standard population by the direct method.⁶ International mortality data are age adjusted to the European standard.⁷ Unless otherwise stated, all death rates in this publication are age adjusted and are deaths per 100 000 population.

Data Years for National Estimates

In this Update, we estimate the annual number of new (incidence) and recurrent cases of a disease in the United States by extrapolating to the US population in 2014 from rates reported in a community- or hospital-based study or multiple studies. Age-adjusted incidence rates by sex and race are also given in this report as observed in the study or studies. For US mortality, most numbers and rates are for 2015. For disease and risk factor prevalence, most rates in this report are calculated from the 2011 to 2014 NHANES. Because NHANES is conducted only in the noninstitutionalized population, we extrapolated the rates to the total US population in 2014, recognizing that this probably underestimates the total prevalence, given the relatively high prevalence in the institutionalized population. The numbers and rates of hospital inpatient discharges for the United States are for 2014. Numbers of visits to physician offices and hospital EDs are for 2014, whereas hospital outpatient department visits are for 2011. Except as noted, economic cost estimates are for 2013 to 2014.

Cardiovascular Disease

For data on hospitalizations, physician office visits, and mortality, CVD is defined according to ICD codes given in Chapter 12. This definition includes all diseases of the circulatory system, as well as congenital CVD. Unless otherwise specified, an estimate for total CVD does not include congenital CVD. Prevalence of CVD includes people with hypertension, HD, stroke, PAD, and diseases of the veins.

Race/Ethnicity

Data published by governmental agencies for some racial groups are considered unreliable because of the small sample size in the studies. Because we try to provide data for as many racial and ethnic groups as possible, we show these data for informational and comparative purposes.

Contacts

If you have questions about statistics or any points made in this Update, please contact the AHA National Center, Office of Science & Medicine. Direct all media inquiries to News Media Relations at http://www.newsroom.heart.org/contacts or 214-706-1173.

The AHA works diligently to ensure that this Update is error free. If we discover errors after publication, we will provide corrections at http://www.heart.org/statistics and in Circulation.

2. Cardiovascular Health

See Tables 2-1 through 2-6 and Charts 2-1 through 2-13

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

In 2011, the AHA created a new set of central Strategic Impact Goals to drive organizational priorities for the current decade:

These goals introduced a new concept of cardiovascular health, characterized by 7 metrics (Life’s Simple 7²), including health behaviors (diet quality, PA, smoking, BMI) and health factors (blood cholesterol, BP, blood glucose). Ideal cardiovascular health is defined by the absence of clinically manifest CVD together with the simultaneous presence of optimal levels of all 7 metrics, including not smoking and having a healthy diet pattern, sufficient PA, normal body weight, and normal levels of TC, BP, and fasting blood glucose, in the absence of drug treatment (Table 2-1). Because a spectrum of cardiovascular health is possible and the ideal cardiovascular health profile is known to be rare in the US population, a broader spectrum of cardiovascular health can also be represented as being “ideal,” “intermediate,” or “poor” for each of the health behaviors and health factors.¹Table 2-1 provides the specific definitions for ideal, intermediate, and poor cardiovascular health for each of the 7 metrics, both for adults and for children.

This concept of cardiovascular health represented a new focus for the AHA, with 3 central and novel emphases:

• An expanded focus on CVD prevention and promotion of positive “cardiovascular health,” in addition to the treatment of established CVD.

• Efforts to promote both healthy behaviors (healthy diet pattern, appropriate energy intake, PA, and nonsmoking) and healthy biomarker levels (optimal blood lipids, BP, glucose levels) throughout the lifespan.

• Population-level health promotion strategies to shift the majority of the public toward greater cardiovascular health, in addition to targeting those individuals at greatest CVD risk, because healthy lifestyles in all domains are uncommon throughout the US population.

Beginning in 2011, and recognizing the time lag in the nationally representative US data sets, this chapter in the annual Statistical Update has evaluated and published metrics and information to provide insights into both progress toward meeting the 2020 AHA goals and areas that require greater attention to meet these goals. The AHA has advocated for raising the visibility of patient-reported cardiovascular health status, which includes symptom burden, functional status, and health-related quality of life, as an indicator of cardiovascular health in future organizational goal setting.³

Relevance of Ideal Cardiovascular Health

• Since the AHA announced its 2020 Impact Goals, multiple independent investigations (summaries below) have confirmed the importance of these metrics and the concept of cardiovascular health. Findings include strong inverse, stepwise associations in the United States of the metrics and cardiovascular health with all-cause mortality, CVD mortality, and HF; with preclinical measures of atherosclerosis such as carotid IMT arterial stiffness and coronary artery calcium prevalence and progression; with physical functional impairment and frailty⁴; and with cognitive decline and depression.^4,5 Similar relationships have also been seen in non-US populations.^4,6–9

• A recent study in a large Hispanic/Latino cohort study in the United States found that associations of CVD and cardiovascular health metrics compared favorably with existing national estimates; however, some of the associations varied by sex and heritage.¹⁰

• A recent study in blacks found that risk of incident HF was 61% lower among those with ≥4 ideal cardiovascular health metrics than among those with 0 to 2 ideal metrics.¹¹

• Ideal health behaviors and ideal health factors are each independently associated with lower CVD risk in a stepwise fashion (Chart 2-1). In other words, across any level of health behaviors, health factors are associated with incident CVD; conversely, across any level of health factors, health behaviors are still associated with incident CVD.¹²

• Analyses from the US Burden of Disease Collaborators demonstrated that poor levels of each of the 7 health factors and behaviors resulted in substantial mortality and morbidity in the United States in 2010. The top risk factor related to overall disease burden was suboptimal diet, followed by tobacco smoking, high BMI, raised BP, high fasting plasma glucose, and physical inactivity.¹³

• A stepwise association was present between the number of ideal cardiovascular health metrics and risk of death based on NHANES 1988 to 2006 data.¹⁴ The HRs for people with 6 or 7 ideal health metrics compared with 0 ideal health metrics were 0.49 (95% CI, 0.33–0.74) for all-cause mortality, 0.24 (95% CI, 0.13–0.47) for CVD mortality, and 0.30 (95% CI, 0.13–0.68) for IHD mortality.¹⁴

• A recent meta-analysis of 9 prospective cohort studies involving 12 878 participants reported that achieving the most ideal cardiovascular health metrics was associated with a lower risk of all-cause mortality (RR, 0.55; 95% CI, 0.37–0.80), cardiovascular mortality (RR, 0.25; 95% CI, 0.10–0.63), CVD (RR, 0.20; 95% CI, 0.11–0.37), and stroke (RR, 0.31; 95% CI, 0.25–0.38).¹⁵

• The adjusted population attributable fractions for CVD mortality were as follows¹⁴:

  	— 40.6% (95% CI, 24.5%–54.6%) for HBP
  	— 13.7% (95% CI, 4.8%–22.3%) for smoking
  	— 13.2% (95% CI, 3.5%–29.2%) for poor diet
  	— 11.9% (95% CI, 1.3%–22.3%) for insufficient PA
  	— 8.8% (95% CI, 2.1%–15.4%) for abnormal glucose levels

• The adjusted population attributable fractions for IHD mortality were as follows¹⁴:

  	— 34.7% (95% CI, 6.6%–57.7%) for HBP
  	— 16.7% (95% CI, 6.4%–26.6%) for smoking
  	— 20.6% (95% CI, 1.2%–38.6%) for poor diet
  	— 7.8% (95% CI, 0%–22.2%) for insufficient PA
  	— 7.5% (95% CI, 3.0%–14.7%) for abnormal glucose levels

• Data from the REGARDS cohort also demonstrated a stepwise association between cardiovascular health metrics and incident stroke. Using a cardiovascular health score scale ranging from 0 to 14, every unit increase in cardiovascular health was associated with an 8% lower risk of incident stroke (HR, 0.92; 95% CI, 0.88–0.95), with a similar effect size for white (HR, 0.91; 95% CI, 0.86–0.96) and black (HR, 0.93; 95% CI, 0.87–0.98) participants.¹⁶

• The Cardiovascular Lifetime Risk Pooling Project showed that adults with all-optimal risk factor levels (similar to having ideal cardiovascular health factor levels of cholesterol, blood sugar, and BP, as well as not smoking) have substantially longer overall and CVD-free survival than those who have poor levels of ≥1 of these cardiovascular health factor metrics. For example, at an index age of 45 years, men with optimal risk factor profiles lived on average 14 years longer free of all CVD events, and 12 years longer overall, than people with ≥2 risk factors.¹⁷

• Better cardiovascular health is associated with less incident HF,¹⁸ less subclinical vascular disease,^19,20 better global cognitive performance and cognitive function,^21,22 lower prevalence²³ and incidence²⁴ of depressive symptoms, lower loss of physical functional status,²⁵ longer leukocyte telomere length,²⁶ less ESRD,²⁷ and less pneumonia, chronic obstructive pulmonary disease,^27a and venous thromboembolism/PE.^27b

• The AHA’s 2020 Strategic Impact Goals are to improve cardiovascular health among all Americans. On the basis of NHANES 1999 to 2006 data, several social risk factors (low family income, low education level, minority race, and single-living status) were related to lower likelihood of attaining better cardiovascular health as measured by Life’s Simple 7 scores.²⁸

• Cardiovascular health metrics are also associated with lower healthcare costs. A recent report from a large, ethnically diverse insured population found that people with 6 or 7 and those with 3 to 5 of the cardiovascular health metrics in the ideal category had a $2021 and $940 lower annual mean healthcare expenditure, respectively, than those with 0 to 2 ideal health metrics.²⁹

Cardiovascular Health: Current Prevalence

(See Table 2-2 and Charts 2-2 through 2-10)

• The most up-to-date data on national prevalence of ideal, intermediate, and poor levels of each of the 7 cardiovascular health metrics are shown for adolescents and teens (Chart 2-2) and for adults (Chart 2-3).

• For most metrics, the prevalence of ideal levels of health behaviors and health factors is higher in US children than in US adults. The main exceptions are diet and PA, for which the prevalence of ideal levels in children is similar to (for PA) or worse (for diet) than in adults (unpublished AHA tabulation).

• Among US children aged 12 to 19 years (Chart 2-2), the prevalence (unadjusted) of ideal levels of cardiovascular health behaviors and factors currently varies from <1% for the healthy diet pattern (ie, <1 in 100 US children meets at least 4 of the 5 dietary components or a corresponding AHA diet score of at least 80) to >80% for the smoking, BP, and fasting glucose metrics (unpublished AHA tabulation).

• Among US adults (Chart 2-3), the age-standardized prevalence of ideal levels of cardiovascular health behaviors and factors currently varies from <1% for having a healthy diet pattern to up to 77% for never having smoked or being a former smoker who has quit for >12 months.

• Age-standardized and age-specific prevalence estimates for ideal cardiovascular health and for ideal levels of each of its components are shown for 2011 to 2012 and 2013 to 2014 in Table 2-2. NHANES 2011 to 2012 data were used for some of the statistics that required nutritional data. The prevalence of ideal levels across 7 health factors and health behaviors generally was lower with age, with much lower prevalence among older versus younger age groups. The exception was diet, for which prevalence of ideal levels was highest in older adults.

• Chart 2-4 displays the prevalence estimates for the population of US children (12–19 years of age) meeting different numbers of criteria for ideal cardiovascular health (of 7 possible) in 2011 to 2012.

  	— Few US children 12 to 19 years of age (≈5%) meet only 0, 1, or 2 criteria for ideal cardiovascular health.
  	— Approximately half of US children (54%) meet 3 or 4 criteria for ideal cardiovascular health, and ≈41% meet 5 or 6 criteria.
  	— <1% of children meet all 7 criteria for ideal cardiovascular health.

• Charts 2-5 and 2-6 display the age-standardized prevalence estimates of US adults meeting different numbers of criteria for ideal cardiovascular health (of 7 possible) in 2011 to 2012, overall and stratified by age, sex, and race.

  	— Approximately 3% of US adults have 0 of the 7 criteria at ideal levels, and another 15% meet only 1 of 7 criteria. This is much worse than among children (12–19 years of age).
  	— Most US adults (≈65%) have 2, 3, or 4 criteria at ideal cardiovascular health, with ≈20% adults within each of these categories.
  	— Approximately 13% of US adults have 5 criteria, 5% have 6 criteria, and virtually 0% have 7 criteria at ideal levels.
  	— Presence of ideal cardiovascular health is both age and sex related (Chart 2-5). Younger adults are more likely to meet greater numbers of ideal metrics than are older adults. More than 60% of Americans >60 years of age have ≤2 metrics at ideal levels. At any age, females tend to have more metrics at ideal levels than do males.
  	— Presence of ideal cardiovascular health also varies by race (Chart 2-6). Blacks and Hispanics tend to have fewer metrics at ideal levels than whites or other races. Approximately 6 in 10 white adults and 7 in 10 black or Hispanic adults have no more than 3 of 7 metrics at ideal levels.

• Chart 2-7 displays the age-standardized percentages of US adults and the percentages of children who have ≥5 of the metrics (of 7 possible) at ideal levels.

  	— Approximately 41% of US children 12 to 19 years of age have ≥5 metrics at ideal levels, with similar prevalence in boys (42%) as in girls (41%).
  	— In comparison, only 17% of US adults have ≥5 metrics at ideal levels, with lower prevalence in males (13%) than in females (21%).
  	— All populations have improved since baseline year 2007 to 2008.

• Chart 2-8 displays the age-standardized percentages of US adults and percentages of children by race/ethnicity who have ≥5 of the metrics (of 7 possible) at ideal levels.

  	— In both children and adults, NH Asians tend to have higher prevalence of having ≥5 metrics at ideal levels than other race/ethnic groups.
  	— Approximately 4.7 in 10 NH Asian children, 4.4 in 10 NH white children, 3.5 in 10 NH black children, and 3.7 in 10 Hispanic children have ≥5 metrics at ideal levels.
  	— By comparison, among adults, ≈2.6 in 10 NH Asians, 1.8 in 10 NH whites, 1.3 in 10 Hispanics, and 1 in 10 NH blacks have ≥5 metrics at ideal levels.

• Chart 2-9 displays the age-standardized percentages of US adults who meet different numbers of criteria for both poor and ideal cardiovascular health. Meeting the AHA 2020 Strategic Impact Goals is predicated on reducing the relative percentage of those with poor levels while increasing the relative percentage of those with ideal levels for each of the 7 metrics.

  	— Approximately 92% of US adults have ≥1 metric at poor levels.
  	— Approximately 34% of US adults have ≥3 metrics at poor levels.
  	— Few US adults (2.5%) have ≥5 metrics at poor levels.
  	— More US adults have 4 to 6 ideal metrics than 4 to 6 poor metrics.

• Using data from the BRFSS, Fang et al³⁰ estimated the prevalence of ideal cardiovascular health by state (all 7 metrics at ideal level), which ranged from 1.2% (Oklahoma) to 6.9% (District of Columbia). Southern states tended to have higher percentages of poor cardiovascular health, lower percentages of ideal cardiovascular health, and lower mean cardiovascular health scores than New England and Western states (Chart 2-10).

Cardiovascular Health: Trends Over Time

(See Charts 2-11 through 2-13)

• The trends over the past decade in each of the 7 cardiovascular health metrics (for diet, trends from 1999–2000 through 2013–2014) are shown in Chart 2-11 (for children 12–19 years of age) and Chart 2-12 (for adults ≥20 years of age).

  	— The prevalence of both children and adults meeting the dietary goals improved between 2003 to 2004 and 2011 to 2012. The prevalence of ideal levels of diet (AHA diet score >80) increased from 0.2% to 0.6% in children and from 0.7% to 1.5% in adults. The prevalence of intermediate levels of diet (AHA diet score 40–79) increased from 30.6% to 44.7% in children and from 49.0% to 57.5% in adults. These improvements were largely attributable to increased whole grain consumption and decreased SSB consumption in both children and adults, as well as small, nonsignificant trends in increased fruits and vegetables (Chapter 5). No major trends were evident in either children or adults meeting the target for consumption of fish or sodium.
  	— Fewer children over time are meeting the ideal BMI metric, whereas more are meeting the ideal smoking and TC metrics. Other metrics do not show consistent trends over time in children.
  	— More adults over time are meeting the smoking metric, whereas fewer are meeting the BMI and glucose metrics. Other metrics do not show consistent trends over time in adults.

• On the basis of NHANES data from 1988 to 2008, if current trends continue, estimated cardiovascular health is projected to improve by 6% between 2010 and 2020, short of the AHA’s goal of 20% improvement (Chart 2-13).³¹ On the basis of current trends among individual metrics, anticipated declines in prevalence of smoking, high cholesterol, and HBP (in males) would be offset by substantial increases in the prevalence of obesity and DM and smaller changes in ideal dietary patterns or PA.³¹

• On the basis of these projections in cardiovascular health factors and behaviors, CHD deaths are projected to decrease by 30% between 2010 and 2020 because of projected improvements in TC, SBP, smoking, and PA (≈167 000 fewer deaths), offset by increases in DM and BMI (≈24 000 more deaths).³²

Achieving the 2020 Impact Goals¹

(See Tables 2-3 through 2-6)

• To achieve the AHA’s 2020 Impact Goals of reducing deaths attributable to CVD and stroke by 20%, continued emphasis is needed on the treatment of acute CVD events and secondary prevention through treatment and control of health behaviors and risk factors.

• Taken together, these data continue to demonstrate both the tremendous relevance of the AHA 2020 Impact Goals for cardiovascular health and the progress that will be needed to achieve these goals by the year 2020.

• For each cardiovascular health metric, modest shifts in the population distribution toward improved health would produce appreciable increases in the proportion of Americans in both ideal and intermediate categories. For example, on the basis of NHANES 2013 to 2014, the current prevalence of ideal levels of BP among US adults is 45.4%. To achieve the 2020 goals, a 20% relative improvement would require an increase in this proportion to 54.4% by 2020 (45.4% × 1.20). On the basis of NHANES data, a reduction in population mean BP of just 5 mm Hg would result in 55.3% of US adults having ideal levels of BP, which represents a 21.8% relative improvement in this metric (Table 2-3). Larger population reductions in BP would lead to even greater numbers of people with ideal levels of BP. Such small reductions in population BP could result from small health behavior changes at a population level, such as increased PA, increased fruit and vegetable consumption, decreased sodium intake, decreased adiposity, or some combination of these and other lifestyle changes, with resulting substantial projected decreases in CVD rates in US adults.³³

• A range of complementary strategies and approaches can lead to improvements in cardiovascular health.³² These include the following:

  	— Individual-focused approaches that target lifestyle and treatments at the individual level (Table 2-4).
  	— Healthcare systems approaches that encourage, facilitate, and reward efforts by providers to improve health behaviors and health factors (Table 2-5).
  	— Population approaches that target lifestyle and treatments in schools or workplaces, local communities, and states, as well as throughout the nation (Table 2-6).

• Such approaches can focus on both (1) improving cardiovascular health among those who currently have less than optimal levels and (2) preserving cardiovascular health among those who currently have ideal levels (in particular, children, adolescents, and young adults) as they age.

• The metrics with the greatest potential for improvement in the United States are health behaviors, including diet quality, PA, and body weight. However, each of the 7 cardiovascular health metrics can be improved and deserves major focus.

3. Smoking/Tobacco Use

See Table 3-1 and Charts 3-1 through 3-8

Tobacco use is one of the leading preventable causes of death in the United States and globally.¹ Tobacco smoking, the most common form of tobacco use, is a major risk factor for CVD and stroke.² The AHA has identified never having tried smoking or never having smoked a whole cigarette (for children) and never having smoked or having quit >12 months ago (for adults) as 1 of the 7 components of ideal cardiovascular health in Life’s Simple 7.³ Life’s Simple 7 uses NHANES data. According to the most recently available NHANES data, which sampled from 2013 to 2014, 91.4% of adolescents and 77.1% of adults met these criteria. Unless otherwise stated, throughout the rest of the chapter we will report tobacco use and smoking estimates from the NSDUH¹ for adolescents (12–17 years of age) and from the NHIS for adults (≥18 years of age), because these data sources have more recent data.⁴

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Chart 3-8. Age-standardized global mortality rates attributable to tobacco per 100 000, both sexes, 2015. Please click here to view the chart and its legend.

Other forms of tobacco use are becoming increasingly common. Electronic cigarette (e-cigarette) use, which involves inhalation of a vaporized liquid that includes nicotine, solvents, and flavoring (“vaping”), has risen dramatically, particularly among young people. The variety of e-cigarette–related products has increased exponentially, giving rise to the more general term electronic nicotine delivery systems.⁵ Use of cigarillos and other mass market cigars, hookahs, and water pipes also has become increasingly common in recent years. Thus, each section below will address the most recent statistical estimates for combustible cigarettes, electronic nicotine delivery systems, and other forms of tobacco use if such estimates are available.

Prevalence

(See Charts 3-1 through 3-5)

Youth

• According to the NSDUH 2015 data, for adolescents aged 12 to 17 years¹:

  	— 6.0% (1 492 000) used tobacco products, and 4.2% (1 039 000) smoked cigarettes in the past month.
  	— 1.5% (367 000) used smokeless tobacco in the past month.
  	— Of adolescents who smoked, 20% (208 000) smoked cigarettes daily, and 5% (52 000) smoked ≥1 pack per day in the past month.
  	— 2.1% (517 000) were current cigar smokers, 0.3% (84 000) were current pipe tobacco smokers, and 1.5% (367 000) used smokeless tobacco.

• In 2015, tobacco use within the past month for youth 12 to 17 years of age varied modestly by region: 6.2% in the Northeast, 7.0% in the Midwest, 6.0% in the South, and 4.9% in the West.¹

• In 2015, 21.6% of adults aged 18 to 20 years were current cigarette smokers compared with 8.7% of adolescents aged 16 to 17 years.¹

• In most states, the minimum age for purchasing tobacco products is 18 years.

Adults

• According to the NHIS 2015 data, among adults ≥18 years of age⁶:

  	— 15.1% of adults (36.5 million) are current smokers.
  	— 16.7% of males and 13.6% of females are current smokers.
  	— 13% of those 18 to 24 years of age, 17.7% of those 25 to 44 years of age, 17.0% of those 45 to 64 years of age, and 8.4% of those ≥65 years old are current smokers.
  	— 21.9% of American Indians or Alaska Natives, 16.7% of blacks, 7% of Asians (with the highest rates in Vietnamese [16.3%] and Filipinos [12.6%]), 10.1% of Hispanics (highest rates in Puerto Ricans [28.5%] and Cubans [19.8%]), and 16.6% of whites are current smokers.
  	— 26.1% of people below the poverty level are current smokers.

• 20.6% of lesbian/gay/bisexual/transgender individuals were current smokers.⁷

• By region, the prevalence of current cigarette smokers was highest in the Midwest (18.7%) and lowest in the West (12.4%).⁷

• In 2009, 42.4% of adults with HIV receiving medical care were current smokers.⁸

• Using data from BRFSS 2015, the state with the highest age-adjusted percentage of current cigarette smokers was West Virginia (17.3%). The state with the lowest age-adjusted percentage of current cigarette smokers was Utah (9.0%) (Chart 3-5).⁹

• In 2015, smoking prevalence was higher among adults ≥18 years of age who reported having a disability or activity limitation (21.5%) than among those reporting no disability or limitation (13.8%).⁷

  — 40% of males and 34% of females with mental illness were current smokers.7

• In 2012 to 2013, among females 15 to 44 years of age, past-month cigarette use was lower among those who were pregnant (15.4%) than among those who were not pregnant (24.0%). Rates were higher among females 18 to 25 years of age (21.0% versus 26.2% for pregnant and nonpregnant females, respectively) than among women 26 to 44 years of age (11.8% versus 25.4%, respectively). Smoking declines by pregnancy trimester, from 19.9% of females 15 to 44 years of age in the first trimester of pregnancy to 12.8% in the third trimester (NSDUH).¹⁰

Incidence

• According to 2015 NSDUH¹:

  — Approximately 1.96 million people ≥12 years of age smoked cigarettes for the first time within the past 12 months, down from 2.3 million in 2012. The 2015 estimate averages out to ≈5390 new cigarette smokers every day. Of new smokers in 2015, 823 000 (42.1%) were 12 to 17 years old, 762 000 (39%) were 18 to 20 years old, and 287 000 (14.7%) were 21 to 25 years old, and only 84 000 (4.3%) were ≥26 years when they first smoked cigarettes.

  — The number of new smokers 12 to 17 years of age (823 000) was down from 2002 (1.3 million); new smokers 18 to 25 years of age increased from ≈600 000 in 2002 to 1.05 million in 2015.

• In the NHIS⁴:

  — In 2015, the average age for initiation of cigarette use was 17.9 years.

Lifetime Risk and Cumulative Incidence in Youth (12 to 17 Years Old) in 2015 (NSDUH Data)

• Per NSDUH data for individuals aged 12 to 17 years, overall the lifetime use of tobacco products declined from 18.5% to 17.3%, with lifetime cigarette use declining from 14.2% to 13.2% during the same time period (P<0.05 for both).¹

• The lifetime use of tobacco products among adolescents 12 to 17 years old varied by the following¹:

  	— Sex: Lifetime use was higher among boys (19.1%) than girls (15.3%).
  	— Race/ethnicity: Lifetime use was highest among whites (19.9%), followed by American Indians or Alaskan Natives (19.6%), Hispanics or Latinos (14.5%), African Americans (13.8%), and Asians (7.7%).
  	— Geographic division: The highest lifetime use was observed in the South (East South Central 22.8%), and the lowest was observed in the Pacific West (13.0%).

Adults

• According to NSDUH data, the lifetime use of tobacco products in individuals aged ≥18 years declined significantly (P<0.01) between 2014 and 2015, from 71.1% to 68.7%, with cigarette use declining in the same interval from 65.9% to 63.1% (both P<0.01).¹ Similar to the patterns in youth, lifetime risk of tobacco products varied by demographic factors:

  	— Sex: Lifetime use was higher in men (78.1%) than women (60.0%).
  	— Race/ethnicity: Lifetime use was highest in American Indians or Alaskan Natives (75.9%) and whites (75.9%), followed by blacks (58.4%), Native Hawaiian or Other Pacific Islander (56.8%), Hispanics or Latinos (56.7%), and Asians (37.9%).
  	— Geographic division: The highest lifetime use was observed in the South (East South Central 19.8%), and the lowest was observed in New England (10.6%).

• In 2015, the lifetime use of smokeless tobacco for adults ≥18 years of age was 17.4%.

• Lifetime tobacco use for people with psychiatric diagnoses was 56.1%, 55.6%, and 70.1% in patients with mood disorders, anxiety disorders, and schizophrenia, respectively.¹¹

Mortality

• Tobacco use is the largest preventable cause of death in the United States, killing >480 000 Americans per year; 41 000 of these deaths were attributed to secondhand smoke exposure.¹²

• Overall mortality among US smokers is 3 times higher than that for never-smokers.¹³

• On average, male smokers die 13.2 years earlier than male nonsmokers, and female smokers die 14.5 years earlier than female nonsmokers.²

• Increased CVD mortality risks persist for older (≥60 years old) smokers as well. A meta-analysis comparing CVD risks in 503 905 cohort participants ≥60 years of age reported an HR for cardiovascular mortality of 2.07 (95% CI, 1.82–2.36) compared with never-smokers and 1.37 (95% CI, 1.25–1.49) compared with former smokers.¹⁴

• In a sample of Native Americans (Strong Heart Study), among whom the prevalence of tobacco use is highest in the United States, the PAR for total mortality was 18.4% for males and 10.9% for females.¹⁵

• Since the first report on the dangers of smoking was issued by the US Surgeon General in 1964, tobacco control efforts have contributed to a reduction of 8 million premature smoking-attributable deaths.¹⁶

• If current smoking trends continue, 5.6 million US children will die of smoking prematurely during adulthood.¹⁷

Secular Trends

(See Charts 3-2 and 3-3)

Youth

The percentage of adolescents (12–17 years old) who reported smoking in the past month declined from 13% in 2003 to 4.2% in 2013.¹⁰

Adults

Since the US Surgeon General’s first report on the health dangers of smoking, age-adjusted rates of smoking among adults have declined, from 51% of males smoking in 1965 to 16.7% in 2015 and from 34% of females in 1965 to 13.6% in 2015, according to NHIS data.^4,7 The decline in smoking, along with other factors (including improved treatment and reductions in the prevalence of risk factors such as uncontrolled hypertension and high cholesterol), is a contributing factor in the sharp decline in the HD death rate during this period.¹⁷

• On the basis of weighted NHIS data, the current smoking status among 18- to 24-year-old men declined 46.5%, from 28.0% in 2005 to 15.0% in 2015, and for 18- to 24-year-old women, smoking declined 47.0%, from 20.7% to 11.0%, over the same time period.⁷ On the basis of age-adjusted estimates in 2015, among people ≥65 years of age, 9.7% of men and 8.3% of women were current smokers.

• From 2005 to 2015, adjusted prevalence rates for tobacco use in individuals with serious psychological distress (according to the Kessler Scale) went from 41.9% to 40.6%, which represents a nonsignificant 3.1% decline; however, rates for people without serious psychological stress declined significantly, from 20.3% to 14.0%.⁷

Cardiovascular Health Impact

• A 2010 report of the US Surgeon General on how tobacco causes disease summarized an extensive body of literature on smoking and CVD and the mechanisms through which smoking is thought to cause CVD.¹² There is a sharp increase in CVD risk with low levels of exposure to cigarette smoke, including secondhand smoke, and a less rapid further increase in risk as the number of cigarettes per day increases. Similar health risks for CHD events are reported with regular cigar smoking as well.¹⁸

• Smoking is an independent risk factor for CHD and appears to have a multiplicative effect with the other major risk factors for CHD: high serum levels of lipids, untreated hypertension, and DM.¹²

• Cigarette smoking and other traditional CHD risk factors may have a synergistic interaction in HIV-positive individuals.¹⁹

• A meta-analysis of 75 cohort studies (≈2.4 million individuals) demonstrated a 25% greater risk for CHD in female smokers than in male smokers (RR, 1.25; 95% CI, 1.12–1.39).²⁰

• Cigarette smoking is an independent risk factor for both ischemic stroke and SAH and has a synergistic effect on other stroke risk factors such as SBP²¹ and oral contraceptive use.^22,23

• A meta-analysis comparing pooled data of ≈3.8 million smokers and nonsmokers found a similar risk of stroke associated with current smoking in females and males.²¹

• Current smokers have a 2 to 4 times increased risk of stroke compared with nonsmokers or those who have quit for >10 years.^24,25

• Short-term exposure to water pipe smoking is associated with a significant increase in SBP and heart rate compared with nonsmoking control subjects,²⁶ but long-term effects remain unclear. Current use of smokeless tobacco is associated with an increased risk of CVD events in cigarette nonsmokers.²⁷

• The CVD risks associated with e-cigarette use are not known.

Healthcare Utilization: Hospital Discharges/Ambulatory Care Visits

Cost

• Each year from 2005 to 2009, US smoking-attributable economic costs were between $289 billion and $333 billion, including $133 billion to $176 billion for direct medical care of adults and $151 billion for lost productivity related to premature death.¹⁷

• In the United States, cigarette smoking was associated with 8.7% of annual aggregated healthcare spending from 2006 to 2010.²⁸

• In 2014, $9 billion was spent on marketing cigarettes and smokeless tobacco in the United States.²⁹

• 258 billion cigarettes were sold in the United States in 2016, which is a 2.5% decrease from the number sold in 2015.²⁹

• Cigarette prices in the United States have increased 283% between the early 1980s and 2011, in large part because of excise taxes on tobacco products. Higher taxes have decreased cigarette consumption, which fell from ≈30 million packs sold in 1982 to ≈14 million packs sold in 2011.²⁹

Smoking Prevention

Tobacco 21 Laws increase the minimum age of sale for tobacco products from 18 to 21 years old.

• Such legislation would likely reduce the rates of smoking during adolescence, a time during which the majority of smokers start smoking, by limiting access, because most people who buy cigarettes for adolescents are <21 years of age.³⁰

• In several towns where Tobacco 21 laws have been enacted, 47% reductions in smoking prevalence among high school students have been reported.³⁰

• Furthermore, the Institute of Medicine estimates that a nationwide Tobacco 21 law would result in 249 000 fewer premature deaths, 45 000 fewer lung cancer deaths, and 4.2 million fewer lost life-years among Americans born between 2010 and 2019.³¹

• As of May 5, 2017, the states of California and Hawaii and at least 230 communities in Arizona, Arkansas, Illinois, Kansas, Maine, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, New Jersey, New York, Ohio, Oregon, Rhode Island, and Washington, DC, have set the minimum age for the purchase of tobacco to 21 years.^32,33

• A federal Tobacco 21 law was introduced to the Congress in September 2015; review and a vote are pending.

Awareness, Treatment, and Control

Smoking Cessation

• According to NHIS 2015 data, 59.1% of adult ever-smokers had stopped smoking.³⁴

  	— The majority (68.0%) of adult smokers wanted to quit smoking, 55.4% had tried in the past year, 7.4% had stopped recently, and 57.2% had received healthcare provider advice to quit.
  	— Receiving advice to quit smoking was lower in uninsured smokers and varied by race, with a lower prevalence in Asian (34.2%), American Indian/Alaska Native (38.1%), and Hispanic (42.2%) smokers than in white smokers (60.2%).
  	— The period from 2000 to 2015 revealed significant increases in the prevalence of smokers who had tried to quit in the past year, had stopped recently, had a health professional recommend quitting, or had used cessation counseling or medication.
  	— In 2015, less than one third of smokers attempting to quit used evidence-based therapies: 4.7% used counseling and medication, 6.8% used counseling, and 29.0% used medication (16.6% nicotine patch, 12.5% gum/lozenges, 2.4% nicotine spray/inhaler, 2.7% bupropion, and 7.9% varenicline).

• Smoking cessation reduces the risk of cardiovascular morbidity and mortality for smokers with and without CHD.

  	— There is no convincing evidence to date that smoking fewer cigarettes per day reduces the risk of CVD, although in several studies, a dose-response relationship has been seen among current smokers between the number of cigarettes smoked per day and CVD incidence.12
  	— Quitting smoking at any age significantly lowers mortality from smoking-related diseases, and the risk declines more the longer the time since quitting smoking.35 Cessation appears to have both short-term (weeks to months) and long-term (years) benefits for lowering CVD risk. Overall, CVD risk appears to approach that of nonsmokers after ≈10 years of cessation.
  	— Smokers who quit smoking at 25 to 34 years of age gained 10 years of life compared with those who continued to smoke. Those aged 35 to 44 years gained 9 years and those aged 45 to 54 years gained 6 years of life, on average, compared with those who continued to smoke.13

• Cessation medications (including sustained-release bupropion, varenicline, and nicotine gum, lozenge, nasal spray, and patch) are effective for helping smokers quit.³⁶

• EVITA was an RCT that examined the efficacy of varenicline versus placebo for smoking cessation among smokers who were hospitalized for ACS. At 24 weeks, rates of smoking abstinence and reduction were significantly higher among patients randomized to varenicline. Point-prevalence abstinence rates were 47.3% in the varenicline group and 32.5% in the placebo group (P=0.012; NNT=6.8). Continuous abstinence rates and reduction rates (≥50% of daily cigarette consumption) were also higher in the varenicline group.³⁷

• The EAGLES trial demonstrated efficacy and safety of 12 weeks of varenicline, bupropion, or nicotine patch in motivated-to-quit smoking patients with major depressive disorder, bipolar disorder, anxiety disorders, posttraumatic stress disorder, obsessive-compulsive disorder, social phobia, psychotic disorders including schizophrenia and schizoaffective disorders, and borderline personality disorder. Of note, these participants were all clinically stable from a psychiatric perspective and were believed not to be at high risk for self-injury.³⁸

• Extended use of a nicotine patch (24 weeks compared with 8 weeks) has been demonstrated to be safe and efficacious in recent randomized clinical trials.³⁹

• An RCT demonstrated the effectiveness of individual- and group-oriented financial incentives for tobacco abstinence through at least 12 months of follow-up.⁴⁰

• In addition to medications, smoke-free policies, increases in tobacco prices, cessation advice from healthcare professionals, and quit-lines and other counseling have contributed to smoking cessation.⁴¹

• Mass media antismoking campaigns, such as the CDC’s Tips campaign (Tips From Former Smokers), have been shown to reduce smoking-attributable morbidity and mortality and are cost-effective.⁴²

• Despite states having collected $25.6 billion in 2012 from the 1998 Tobacco Master Settlement Agreement and tobacco taxes, <2% of those funds are spent on tobacco prevention and cessation programs.⁴³

Electronic Cigarettes

(See Charts 3-6 and 3-7)

• Electronic nicotine delivery systems, more commonly called electronic cigarettes or e-cigarettes, are battery-operated devices that deliver nicotine, flavors, and other chemicals to the user in an aerosol. Although e-cigarettes were introduced less than a decade ago, there are currently >450 e-cigarette brands on the market, and sales in the United States were projected to be $2 billion in 2014.⁴⁴

• Teenagers are increasingly trying e-cigarettes. According to the National Youth Tobacco Survey, in 2015, e-cigarettes were the most commonly used tobacco products in youth: in the prior 30 days, 5.3% of middle school and 16.0% of high school students endorsed use (Charts 3-3 and 3-4).⁴⁵ In 2015, ever use of e-cigarettes was reported by 13.5% of middle school students and 37.7% of high school students,⁴⁶ which was a substantial increase from 2011 (1.4% of middle and 4.7% of high school students)⁴⁷ (Charts 3-6 and 3-7).

• In 2014, 18.3 million US middle and high school students (68.9%) were exposed to e-cigarette advertising.⁴⁸

• Among US adults during 2010 to 2013, awareness and use of e-cigarettes increased considerably, and more than one third of current cigarette smokers reported ever having used e-cigarettes.⁴⁹

• According to NHIS 2014 data, among US working adults, 3.8% (≈5.5 million) currently used e-cigarettes. The use of e-cigarettes was significantly higher among current smokers (16.2%) than among past smokers (4.3%) or never-smokers (0.5%). Similarly, prevalence was significantly higher among males (4.5%), NH whites (4.5%), younger adults (18–24 years; 5.1%), individuals without health insurance (5.9%), individuals with incomes <$35 000, and those residing in the Midwest (4.5%).⁵⁰

• As of June 30, 2017, 46 states, the District of Columbia, Guam, Puerto Rico, and the US Virgin Islands have passed legislation prohibiting the sale of e-cigarettes to minors. Four states (Maine, Massachusetts, Michigan, and Pennsylvania) do not have any legislation requiring a minimum age restriction on the purchase of e-cigarettes.⁵¹

Secondhand Smoke

• Data from the US Surgeon General on the consequences of secondhand smoke indicate the following:

  	— Nonsmokers who are exposed to secondhand smoke at home or at work increase their risk of developing CHD by 25% to 30%.12
  	— Short exposures to secondhand smoke can cause blood platelets to become stickier, damage the lining of blood vessels, and decrease coronary flow velocity reserves, potentially increasing the risk of an AMI.12
  	— Exposure to secondhand smoke increases the risk of stroke by 20% to 30%, and it is associated with increased mortality (adjusted mortality rate ratio, 2.11) after a stroke.52 A meta-analysis of 23 prospective and 17 case-control studies of cardiovascular risks associated with secondhand smoke exposure demonstrated 18%, 23%, 23%, and 29% increased risks for total mortality, total CVD, CHD, and stroke, respectively, in those exposed to secondhand smoke.53 A meta-analysis of 24 studies demonstrates that secondhand smoke can increase risks for preterm birth by 20%.54

• As of June 30, 2017, 8 states (California, Delaware, Hawaii, New Jersey, North Dakota, Oregon, Utah, and Vermont), the District of Columbia, and Puerto Rico have passed comprehensive smoke-free indoor air laws that include e-cigarettes. These laws prohibit smoking and the use of e-cigarettes in indoor areas of private work sites, restaurants, and bars.⁵¹

• Pooled data from 17 studies in North America, Europe, and Australia suggest that smoke-free legislation can reduce the incidence of acute coronary events by 10%.⁵⁵

• The percentage of the US nonsmoking population with serum cotinine ≥0.05 ng/mL (which indicates exposure to secondhand smoke) declined from 52.5% in 1999 to 2000 to 25.3% in 2011 to 2012, with declines occurring for both children and adults. During 2011 to 2012, the percentage of nonsmokers with detectable serum cotinine was 40.6% for those 3 to 11 years of age, 33.8% for those 12 to 19 years of age, and 21.3% for those ≥20 years of age. The percentage was also higher for NH blacks (46.8%) than for NH whites (21.8%) and Mexican Americans (23.9%). People living below the poverty level (43.2%) and those living in rental housing (36.8%) had higher rates of secondhand smoke exposure than their counterparts (21.1% of those living above the poverty level and 19.0% of those who owned their homes; NHANES).⁵⁶

Global Burden of Smoking

(See Table 3-1 and Chart 3-8)

• Although tobacco use in the United States has been declining, the absolute number of tobacco users worldwide has climbed steeply.⁴³

• Based on the GBD synthesis of >2800 data sources, the daily smoking age-standardized global prevalence of smoking in 2015 was 25.0% (95% UI, 24.2%–25.7%) in males and 5.4% (95% UI, 5.1%–5.7%) in females. The investigators estimate that since 1990, smoking rates have declined globally by 28.4% in males and 34.4% in females.⁵⁷

• The GBD 2015 study used statistical models and data on incidence, prevalence, case fatality, excess mortality, and cause-specific mortality to estimate disease burden for 315 diseases and injuries in 195 countries and territories. Greenland had the highest mortality rates attributable to tobacco (Chart 3-8).⁵⁸

• The number of smokers was estimated to have grown from 721 million in 1980 to 967 million in 2012.⁵⁹

• Worldwide, ≈80% of smokers live in low- and middle-income countries.⁶⁰

• Tobacco smoking (including secondhand smoke) caused an estimated 7.2 million deaths in 2015. Among the leading risk factors for death, smoking ranked fifth in DALYs in 109 countries.⁵⁷

• The WHO estimated that the economic cost of smoking-attributable diseases accounted for $422 billion US dollars, which represented ≈5.7% of global health expenditures.⁶¹ The total economic costs, including both health expenditures and lost productivity, amounted to approximately US $1436 billion, which was roughly equal to 1.8% of the world’s annual gross domestic product. The WHO further estimated that 40% of the expenditures were in developing countries.

• To help combat the global problem of tobacco exposure, in 2003 the WHO adopted the Framework Convention on Tobacco Control treaty. From this emerged a set of evidence-based policies with the goal of reducing the demand for tobacco, entitled MPOWER. MPOWER policies outline the following strategies for nations to reduce tobacco use: (1) monitor tobacco use and prevention policies; (2) protect individuals from tobacco smoke; (3) offer to help with tobacco cessation; (4) warn about tobacco-related dangers; (5) enforce bans on tobacco advertising; (6) raise taxes on tobacco; and (7) reduce the sale of cigarettes. More than half of all nations have implemented at least 1 MPOWER policy.⁶²

4. Physical Inactivity

See Charts 4-1 through 4-11

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Chart 4-11. Age-standardized global mortality rates attributable to low physical activity per 100 000, both sexes, 2015. Please click here to view the chart and its legend.

Physical inactivity is a major risk factor for CVD and stroke.¹ Meeting the guidelines for PA is one of the AHA’s 7 components of ideal cardiovascular health for both children and adults.² The AHA and 2008 federal guidelines on PA recommend that children get at least 60 minutes of PA daily (including aerobic and muscle- and bone-strengthening activity). In 2015, on the basis of survey interviews,³ only 27.1% of high school students reported achieving at least 60 minutes of daily PA, which is likely an overestimation of those actually meeting the guidelines.^4,5 The guidelines recommend that adults get at least 150 minutes of moderate-intensity or 75 minutes of vigorous-intensity aerobic activity (or an equivalent combination) per week and perform muscle-strengthening activities at least 2 days per week.⁶ In a nationally representative sample of adults,⁷ only 21.5% reported participating in adequate leisure-time aerobic and muscle-strengthening activity to meet these criteria, but they were not asked to report activity accumulated during occupational, transportation, or domestic duties. Being physically active is an important aspect of overall health.⁶ PA not only reduces premature mortality but also improves risk factors for CVD (such as HBP and high cholesterol) and reduces the likelihood of diseases related to CVD, including CHD, stroke, type 2 DM, and sudden heart attacks.⁶ Benefits from PA are seen for all ages and groups, including older adults, pregnant females, and people with disabilities and chronic conditions. Therefore, the federal guidelines recommend being as physically active as abilities and conditions allow and, if not currently meeting the recommendations, increasing PA gradually.

Defining and Measuring PA

There are 4 dimensions of PA (mode or type, frequency, duration, and intensity) and 4 common domains for adults (occupational, domestic, transportation, and leisure time).⁴ For children, there are additional considerations of structured PA in schools and communities. The federal guidelines specify the suggested frequency, duration, and intensity of activity. Historically, recommendations on PA for health purposes have focused on leisure-time activity. However, because all domains of PA could have an impact on health, and because an increase in 1 domain may sometimes be compensated for by a decrease in another domain, ideally data will be collected on all dimensions and domains of PA.⁴

There are 2 broad categories of methods to assess PA: (1) subjective methods that use questionnaires and diaries/logs and (2) objective methods that use wearable monitors (pedometers, accelerometers, etc). Studies that compare the findings between subjective and objective methods have found that there is marked discordance between self-reported and measured PA, with respondents often overstating their PA, especially the intensity.^4,5

Another consideration in the measurement of PA is that surveys often ask only about leisure-time PA, which represents PA obtained from a single domain. People who get a lot of PA in other domains may be less likely to engage in leisure-time PA. Although they may meet the federal PA guidelines, people who spend considerable time and physical effort in occupational, domestic, or transportation activities/domains may be less likely to be identified as meeting the guidelines.

PA and cardiorespiratory fitness provide distinct metrics in assessment of CVD risk.⁸ Poor cardiorespiratory (or aerobic) fitness may be a stronger predictor of adverse cardiometabolic and cardiovascular outcomes than traditional risk factors.⁹ Although many studies have shown that increasing the amount and quality of PA can improve cardiorespiratory fitness, other factors can contribute, such as a genetic predisposition to perform aerobic exercise.¹⁰ Because cardiorespiratory fitness is directly measured and reflects both participation in PA and the state of physiological systems affecting performance, the relationship between cardiorespiratory fitness and clinical outcomes is stronger than the relationship of PA to a series of clinical outcomes.⁸ Unlike health behaviors such as PA and risk factors that are tracked by federally funded programs (NHIS, NHANES, etc),^5,7 there are no national data on adult cardiorespiratory fitness, although the development of a national cardiorespiratory fitness registry has been proposed.⁸ Such additional data on the cardiorespiratory fitness levels of Americans could give a fuller and more accurate picture of physical fitness levels.⁸

Prevalence

Youth

Meeting the Activity Recommendations

(See Charts 4-1 through 4-3)

• On the basis of self-reported PA (YRBSS, 2015)³:

  	— The prevalence of high school students who met aerobic activity recommendations of ≥60 minutes of PA on all 7 days of the week was 27.1% nationwide and declined from 9th (31.0%) to 12th (23.5%) grades. At each grade level, the prevalence was higher in boys than in girls.
  	— More than double the percentage of high school–aged boys (36.0%) than girls (17.7%) reported having been physically active ≥60 minutes per day on all 7 days.
  	— The prevalence of students meeting activity recommendations on ≥5 days per week was higher among NH white boys (62.0%), NH black boys (52.2%), and Hispanic boys (53.5%) than NH white girls (43.5%), NH black girls (33.4%), and Hispanic girls (33.1%) (Chart 4-1).
  	— 14.3% of students reported that they did not participate in ≥60 minutes of any kind of PA on any 1 of the previous 7 days (Chart 4-2). Girls were more likely than boys to report this level of inactivity (17.5% versus 11.1%).

• With regard to objectively measured moderate to vigorous PA (accelerometer counts per minute >2020; NHANES, 2003–2004)⁵:

  	— Only 8% of 12- to 19-year-olds accumulated ≥60 minutes of moderate to vigorous PA on ≥5 days per week, whereas 42% of 6- to 11-year-olds achieved similar activity levels.5
  	— More boys than girls met PA recommendations (≥60 minutes of moderate to vigorous activity) on ≥5 days per week.5

• With regard to objectively measured cardiorespiratory fitness (NHANES, 2012)¹¹:

  — For adolescents aged 12 to 15 years, boys in all age groups were more likely to have adequate levels of cardiorespiratory fitness than girls (Chart 4-3).11

• With regard to self-reported muscle strengthening activities (YRBSS, 2015):³

  	— The proportion of high school students who participated in muscle-strengthening activities on ≥3 days of the week was 53.4% nationwide and declined from 9th (males 64.9%, females 48.2%) to 12th (males 59.9%, females 39.9%) grades.
  	— More high school boys (63.7%) than girls (42.7%) reported having participated in muscle-strengthening activities on ≥3 days of the week.

Structured Activity Participation in Schools and Sport

• In 2015, only 29.8% of students attended physical education classes in school daily (33.8% of boys and 25.5% of girls) (YRBSS).³

• Daily physical education class participation declined from the 9th grade (44.6% for boys, 39.5% for girls) through the 12th grade (27.9% for boys, 16.0% for girls) (YRBSS).³

• Just over half (57.6%) of high school students played on at least 1 school or community sports team in the previous year: 53.0% of girls and 62.2% of boys (YRBSS).³

Television/Video/Computers

(See Chart 4-4)

Research suggests that screen time (watching television or using a computer) may lead to less PA among children.¹² In addition, television viewing time is associated with poor nutritional choices, overeating, and weight gain (Chapter 5, Nutrition).

• In 2015 (YRBSS)³:

  	— Nationwide, 41.7% of high school students used a computer for activities other than school work (eg, videogames or other computer games) for ≥3 hours per day on an average school day.
  	— The prevalence of using computers ≥3 hours per day (for activities other than school work) was highest among NH black girls (48.4%), followed by Hispanic girls (47.4%), Hispanic boys (45.1%), NH black boys (41.2%), NH white boys (38.9%), and NH white girls (38.3%) (Chart 4-4).
  	— The prevalence of watching television ≥3 hours per day was highest among NH black girls (41.5%) and boys (37.0%), followed by Hispanic girls (29.2%) and boys (27.4%) and NH white boys (21.4%) and girls (18.8%).

• A report from the Kaiser Family Foundation (using data from 2009) reported that 7th to 12th graders spent an average of 90 minutes per day text messaging, 33 minutes talking on the phone, and 49 minutes using their phone to access media (music, games, or videos).¹³ Surveys such as YRBSS have not historically asked about cell phone use specifically and are thus likely underestimates of total screen-time use.

Adults

Meeting the Activity Recommendations

(See Charts 4-5 through 4-9)

• With regard to self-reported leisure-time aerobic and muscle-strengthening PA (NHIS, 2015)⁷:

  	— 21.5% of adults met the 2008 federal PA guidelines for both aerobic and strengthening activity, an important component of overall physical fitness, based on only reporting leisure-time activity (Chart 4-5).
  	— 30.4% do not engage in any leisure-time PA (“no leisure time PA/inactivity” refers to no sessions of light/moderate or vigorous PA of ≥10 minutes’ duration).

• For self-reported leisure time aerobic PA (NHIS, 2015)⁷:

  	— The age-adjusted proportion who reported meeting the 2008 aerobic PA guidelines for Americans (≥150 minutes of moderate PA or 75 minutes of vigorous PA or an equivalent combination each week) through leisure-time activities was 49.7%, with 52.9% of males and 46.7% of females meeting the aerobic guidelines.
  	— Age-adjusted prevalence was 50.9% for NH whites, 41.7% for NH blacks, and 43.3% for Hispanics. Among both males and females, NH whites were more likely to meet the PA aerobic guidelines with leisure-time activity than NH blacks and Hispanics (Chart 4-6).
  	— Education was positively associated with meeting the federal guidelines. Among adults ≥25 years of age, 30.0% of participants with no high school diploma, 38.9% of those with a high school diploma or GED high school equivalency credential, 46.8% of those with some college, and 62.8% of those with a bachelor’s degree or higher met the federal guidelines for aerobic PA through leisure-time activities (Chart 4-7).

• Adults residing in urban areas (metropolitan statistical areas) are more likely to meet the federal aerobic PA guidelines through leisure-time activities than those residing in rural areas (51.4% versus 41.1%) (Chart 4-8).⁷

• Adults living below 200% of the poverty level are less likely to meet the federal PA guidelines through leisure-time activities than adults living at >200% above the poverty level (Chart 4-9).⁷

• 30.4% of adults report that they do not engage in leisure-time PA (no sessions of PA of ≥10 minutes’ duration).⁷

• With regard to objectively measured moderate to vigorous PA (accelerometer counts per minute >2020; NHANES, 2003–2004)⁵:

  	— Among those ≥60 years of age, adherence to PA recommendations was 2.5% in males and 2.3% in females.5
  	— In contrast to self-reported PA, which suggested that NH whites had the higher levels of PA,7 data from objectively measured PA revealed that Hispanic participants had higher total PA and moderate to vigorous PA compared with NH white or black participants (≥20 years old).5,14

• Among adults 20 to 59 years of age, 3.8% of males and 3.2% of females met recommendations to engage in moderate to vigorous PA for 30 minutes (in sessions of ≥10 minutes) on ≥5 of 7 days.⁵ It is also important to consider that using data accumulated only in ≥10-minute bouts could remove up to 75% of moderate activity.¹⁵

• Levels of activity declined sharply after the age of 50 years in all groups.¹⁵

• These data also revealed that rural US residents performed less moderate to vigorous PA than urban residents, but rural residents spent more time in lighter-intensity activities (accelerometer counts per minute >760) than their urban resident counterparts.¹⁶

• In a review examining self-reported versus actual measured PA (eg, accelerometers, pedometers, indirect calorimetry, doubly labeled water, heart rate monitor), 60% of respondents self-reported higher values of activity than what was measured by use of direct methods.¹⁷ Among males, self-reported PA was 44% greater than actual measured values; among females, self-reported activity was 138% greater than actual measured PA.¹⁷

Structured Activity Participation in Leisure-Time, Domestic, Occupational, and Transportation Activities

• Individuals from urban areas who participated in NHANES 2003 to 2006 reported participating in more transportation activity, but rural individuals reported spending more time in household PA and more total PA than urban individuals, possibly explaining the higher levels of light activity of rural individuals observed by objective methods.¹⁶

• NH whites and blacks were more likely to report meeting the PA guidelines through leisure-time PA but had lower objectively measured PA than Hispanic individuals.¹⁶ At this time, it is unclear which construct of PA (domestic, occupational, or transportation) contributes to the higher observed PA for Hispanic individuals or whether these differences are caused by overreporting or underreporting leisure-time PA.

• A 1-day assessment indicated that the mean prevalence of any active transportation was 10.3%. NH whites reported the lowest active transport, only 9.2%, of any racial/ethnic group. Roughly 11.0% of Hispanics, 13.4% of blacks, and 15.0% of other NH individuals reported participating in any active transportation on the previous day (Active Transportation Surveillance, 2012).¹⁸

Mortality

• Physical inactivity is the fourth-leading risk factor for global death, responsible for 1 to 2 million deaths annually.^19,20

• A study of US adults that linked a large, nationally representative sample of 10 535 participants (NHANES) to death records found that meeting the aerobic PA guidelines reduced all-cause mortality, with an HR of 0.64, after adjustment for potential confounding factors. Furthermore, for adults not meeting the aerobic PA guidelines, engaging in muscle-strengthening activity ≥2 times a week was associated with a 44% lower adjusted HR for all-cause mortality.²¹

• In smaller samples of NHANES (participants with objectively measured PA and between the ages of 50 and 79 years [n=3029]), models that replaced sedentary time with 10 minutes of moderate to vigorous PA were associated with lower all-cause mortality (HR, 0.70; 95% CI, 0.57–0.85) after 5 to 8 years of follow-up. Even substituting in 10 minutes of light activity was associated with lower all-cause mortality (HR, 0.91; 95% CI, 0.86–0.96).²²

• Adults ≥40 years of age who reported participating in moderate to vigorous PA once a week or more had a lower mortality risk than those who were inactive. Furthermore, older adults who engaged in moderate to vigorous PA ≥5 times per week had a significantly lower mortality risk than those who participated in a lower frequency of PA.²³

• A meta-analysis of 9 cohort studies, representing 122 417 patients, found that as little as 15 minutes of daily moderate to vigorous PA reduced all-cause mortality in adults ≥60 years of age. This protective effect of PA was dose dependent; the most rapid reduction in mortality per minute of added PA was for those at the lowest levels of PA. These findings suggest that older adults can benefit from PA time far below the amount recommended by the federal guidelines.²⁴

• A pooled study of >600 000 participants found that even low levels of leisure-time PA (up to 75 minutes of brisk walking per week) were associated with a reduced risk of mortality compared with participants with no PA. If this is a causal relationship, this low level of PA would confer a 1.8-year gain in life expectancy after age 40 years compared with no PA. For participants meeting the PA guidelines, the gain in life expectancy was 3.4 years over those with no PA.²⁵

• In further analysis of this pooled sample, an inverse dose-response relationship was observed between level of leisure-time PA (HR, 0.80 for less than the recommended minimum of the PA guidelines; HR, 0.69 for 1–2 times the recommended minimum; and HR, 0.63 for 2–3 times the minimum) and mortality, with the upper threshold for mortality benefit occurring at 3 to 5 times the PA recommendations (HR, 0.61). Furthermore, there was no evidence of harm associated with performing ≥10 times the recommended minimum (HR, 0.69).²⁶

• Similarly, a population-based cohort in New South Wales, Australia, of 204 542 adults followed up for an average of 6.5 years evaluated the relationship of PA to mortality risk. It found that compared with those who reported no moderate to vigorous PA, the adjusted HRs for all-cause mortality were 0.66 for those reporting 10 to 149 min/wk, 0.53 for those reporting 150 to 299 min/wk, and 0.46 for those reporting ≥300 min/wk of activity.²⁷

• A meta-analysis of 17 eligible studies on PA in patients with DM revealed that the highest PA category in each study was associated with a lower RR (0.61) for all-cause mortality and CVD (0.71) than the lowest PA category. Although more PA was associated with larger reductions in future all-cause mortality and CVD, in patients with DM, any amount of habitual PA was better than inactivity.²⁸

• Meta-analysis has also revealed an association between participating in more transportation-related PA and lower all-cause mortality risk.²⁹ In contrast, higher occupational PA has recently been associated with higher mortality in males (but not females).³⁰ It is unclear whether confounding factors such as fitness, SES, or other domains of PA might impact this relationship.

• In a study involving 55 137 adults followed up over an average of 15 years, running even 5 to 10 min/d and at slow speeds (<6 mph) was associated with a markedly reduced risk of CVD and of death attributable to all causes.³¹

• The Cooper Center Longitudinal Study, an analysis conducted on 16 533 participants, revealed that across all risk factor strata, the presence of low cardiorespiratory fitness was associated with a greater risk of CVD death over a mean follow-up of 28 years.³²

• In a retrospective cohort study of 57 085 individuals who were clinically referred for stress testing (but without established CAD or HF), cardiorespiratory fitness–associated “biological age” was a stronger predictor of mortality over 10 years of follow-up than chronological age.³³

• In the Southern Community Cohort Study of 63 308 individuals followed up for >6.4 years, more time spent being sedentary (>12 h/d versus <5.76 h/d) was associated with a 20% to 25% increased risk of all-cause mortality in both black and white adults. Both PA (beneficial) and sedentary time (detrimental) were associated with mortality risk.³⁴

• A study that prospectively assessed the association of continuous inactivity and of changes in sitting time for 2 years with subsequent long-term all-cause mortality found that compared with people who remained consistently sedentary, the HRs for mortality were 0.91 in those who were newly sedentary, 0.86 in formerly sedentary individuals, and 0.75 in those who remained consistently nonsedentary. Thus, participants who reduced their sitting time over 2 years experienced a reduction in mortality.³⁵

• A meta-analysis including >1 million participants across 16 studies compared the risk associated with sitting time and television viewing in physically active and inactive study participants. For inactive individuals (defined as the lowest quartile of PA), those sitting >8 h/d had a higher all-cause mortality risk than those sitting <4 h/d. For active individuals (top quartile for PA), sitting time was not associated with all-cause mortality, but active people who watched television ≥5 h/d did have higher mortality risk.³⁶

Secular Trends

Youth

In 2015 (YRBSS)³:

• Among students nationwide, there was a significant increase in the number of individuals reporting having participated in muscle-strengthening activities on ≥3 days per week, from 47.8% in 1991 to 53.4% in 2015; however, the prevalence did not change substantively from 2013 (51.7%) to 2015 (53.4%).

• A significant increase occurred in the number of youth reporting having used computers not for school work for ≥3 hours per day compared with 2003 (22.1% versus 41.7% in 2015). The prevalence increased from 2003 to 2009 (22.1% versus 24.9%) and then increased more rapidly from 2009 to 2015 (24.9% versus 41.7%).

  — From 2004 to 2009, the Kaiser Family Foundation reported that the proportion of 8- to 18-year-olds who own their own cell phone has increased from 39% to 66%,13 which could also contribute to higher exposure to screen time in children.

• Nationwide, the number of high school students who reported attending physical education classes at least once per week did not change substantively between 2013 (48.0%) and 2015 (51.6%).

  — The number of high school students reporting attending daily physical education classes changed in nonlinear ways over time. Attendance initially decreased from 1991 to 1995 (from 41.6% to 25.4%) and did not substantively change between 1995, 2013, and 2015 (25.4%, 29.4%, and 29.8%, respectively).

• The prevalence of high school students playing ≥1 team sport in the past year did not substantively change between 2013 (54.0%) and 2015 (57.6%).

  — In 2012, the prevalence of adolescents aged 12 to 15 years with adequate levels of cardiorespiratory fitness (based on age- and sex-specific standards) was 42.2% in 2012 (Chart 4-3), down from 52.4% in 1999 to 2000.11

Adults

(See Chart 4-10)

• The age-adjusted percentage of US adults who reported meeting both the muscle-strengthening and aerobic guidelines increased from 14.4% in 1998 to 21.4% in 2015 (Chart 4-10).⁷ The percentage of US adults who reported meeting the aerobic guideline increased from 40.1% in 1998 to 49.7% in 2015.^7,37

• Also, the number of adults reporting no leisure-time PA decreased from 36.2% in 2008 to 30.0% in 2014,³⁸ surpassing the target for Healthy People 2020, which was 32.6%.

• A 2.3% decline in physical inactivity between 1980 and 2000 was estimated to have prevented or postponed ≈17 445 deaths (≈5%) attributable to CHD in the United States.³⁹

Complications of Physical Inactivity: The Cardiovascular Health Impact

Youth

• Among children aged 4 to 18 years, both higher activity levels and lower sedentary time measured by accelerometry were associated with more favorable metabolic risk factor profiles.⁴⁰

• Among children 4 to 18 years of age, more time spent engaging in moderate to vigorous PA was associated with higher HDL-C and lower waist circumference, SBP, fasting triglycerides, and insulin. These findings were significant regardless of the amount of children’s sedentary time.⁴⁰

• For elementary school children, engagement in organized sports for ≈1 year was associated with lower clustered cardiovascular risk.⁴¹

Adults

Cardiovascular and Metabolic Risk

• A review of the US Preventive Services Task Force recommendations examined the evidence on whether relevant counseling interventions for a healthful diet and PA in primary care modify self-reported behaviors, intermediate physiological outcomes, DM incidence, and cardiovascular morbidity or mortality in adults with CVD risk factors. It was concluded that after 12 to 24 months, intensive lifestyle counseling for individuals selected because of risk factors reduced TC levels by an average of 0.12 mmol/L, LDL-C levels by 0.09 mmol/L, SBP by 2.03 mm Hg, DBP by 1.38 mm Hg, fasting glucose by 0.12 mmol/L, DM incidence by an RR of 0.58, and weight outcomes by a standardized difference of 0.25.⁴²

• Results from NHANES 2011 to 2014 demonstrated that the prevalence of low HDL-C was higher among adults who reported not meeting PA guidelines (21.0%) than among adults meeting guidelines (17.7%).⁴³

• Engaging in active transport to work has been associated with lower cardiovascular risk factors.

  — In a large Swedish cohort of 23 732 individuals, bicycling to work at baseline was associated with a lower odds of developing incident obesity, hypertension, hypertriglyceridemia, and impaired glucose tolerance at 10 years’ follow-up than among those using passive modes of transportation.44

• A total of 120 to 150 min/wk of moderate-intensity activity, compared with none, can reduce the risk of developing metabolic syndrome.⁶

• Even lighter-intensity activities, such as yoga, were reported to improve BMI, BP, triglycerides, LDL-C, and HDL-C but not fasting blood glucose in a meta-analysis of 32 RCTs comparing yoga to nonexercise control groups.⁴⁵

• In CARDIA, females who maintained high PA through young adulthood gained 6.1 fewer kilograms of weight and 3.8 fewer centimeters in waist circumference in middle age than those with lower activity. Highly active males gained 2.6 fewer kilograms and 3.1 fewer centimeters than their lower-activity counterparts.⁴⁶

• In 3 US cohort studies, males and females who increased their PA over time gained less weight in the long term, whereas those who decreased their PA over time gained more weight.⁴⁷

• Among US males and females, every hour per day of increased television watching was associated with 0.3 lb of greater weight gain every 4 years.⁴⁷

• In a sample of 466 605 participants in the China Kadoorie Biobank study, a 1-SD (1.5 h/d) increase in sedentary time was associated with a 0.19-U higher BMI, a 0.57-cm larger waist circumference, and 0.44% more body fat. Both higher sedentary leisure time and lower PA were independently associated with an increased BMI.⁴⁸

Cardiovascular Events

• In the WHI observational study (N=71 018), sitting for ≥10 h/d compared with ≤5 h/d was associated with increased CVD risk (HR, 1.18) in multivariable models that included PA. Low PA was also associated with higher CVD risk. It was concluded that both low PA and prolonged sitting augment CVD risk.⁴⁹

• In a meta-analysis of longitudinal studies among females, RRs of incident CHD were 0.83 (95% CI, 0.69–0.99), 0.77 (95% CI, 0.64–0.92), 0.72 (95% CI, 0.59–0.87), and 0.57 (95% CI, 0.41–0.79) across increasing quintiles of PA compared with the lowest quintile.⁵⁰

• A study of the factors related to declining CVD among Norwegian adults ≥25 years of age found that increased PA (≥1 hour of strenuous PA per week) accounted for 9% of the decline in hospitalized and nonhospitalized fatal and nonfatal CHD events.⁵¹

• In the Health Professionals Follow-Up Study, for every 3-hour-per-week increase in vigorous-intensity activity, the multivariate RR of MI was 0.78 (95% CI, 0.61–0.98) for males. This 22% reduction of risk can be explained in part by beneficial effects of PA on HDL-C, vitamin D, apolipoprotein B, and HbA_1c.⁵²

• A 2003 meta-analysis of 23 studies on the association of PA with stroke indicated that compared with low levels of activity, high (RR, 0.73; 95% CI, 0.67–0.79) and moderate (RR, 0.80; 95% CI, 0.74–0.86) levels of activity were inversely associated with the likelihood of developing total stroke (ischemic and hemorrhagic).⁵³

• In a study involving 1.1 million females without prior vascular disease and followed up over an average of 9 years, those who reported moderate activity were found to be at lower risk of CHD, a cerebrovascular event, or a first thrombotic event. However, strenuous PA was not found to be as beneficial as moderate PA.⁵⁴

• Domains of PA, other than leisure time, are understudied and often overlooked. A meta-analysis reported a protective relation of transportation activity to cardiovascular risk, which was greater in females.⁵⁵ However, occupational PA has recently been associated with higher MI incidence in men 19 to 70 years old.^30,56 These relationships require further investigation, because a protective association of occupational activity with MI has been reported in young men (19–44 years).⁵⁶

• In a Swedish cohort of 773 925 young males without history of VTE, cardiorespiratory fitness was associated with a reduced risk of VTE (HR, 0.81; 95% CI, 0.78–0.85) at ≥20 years of follow-up.⁵⁷

• In 5962 veterans, lower exercise capacity was associated with a higher risk of developing AF. For every 1-MET increase in exercise capacity, the risk of developing AF was 21% lower (HR, 0.79; 95% CI, 0.76–0.82).⁵⁸

• In a large clinical trial (NAVIGATOR) involving 9306 people with impaired glucose tolerance, ambulatory activity as assessed by pedometer at baseline and 12 months was found to be inversely associated with risk of a cardiovascular event.⁵⁹

• Self-reported low lifetime recreational activity has been associated with increased PAD.⁶⁰

Secondary Prevention

• A Cochrane systematic review of 63 studies concluded that exercise-based cardiac rehabilitation programs for CHD patients reduced cardiovascular mortality and hospital admissions, but not overall mortality.⁶¹

• In a prospective study that monitored 902 HF patients (with preserved or reduced EF) for 3 years, reporting participation in any PA (≥1 min/wk) was associated with a lower risk of cardiac death and all-cause death than no PA. Less television screen time (<2 versus >4 h/d) was also associated with lower all-cause death.⁶²

• Using data from a registry of stable outpatients with symptomatic coronary disease, cerebrovascular disease, or PAD, the mortality rate of patients with a recent MI was significantly lower in patients who participated in supervised (N=593) versus unsupervised (N=531) exercise programming.⁶³

• Early mortality after a first MI was lower for patients who had higher exercise capacity before the MI event. Every 1-MET-higher exercise capacity before the MI was associated with an 8% to 10% lower risk of mortality at 28 days, 90 days, and 365 days after MI.⁶⁴

Costs

• The economic consequences of physical inactivity are substantial. Using data derived primarily from WHO publications and data warehouses, one study estimated that economic costs of physical inactivity account for 1.5% to 3.0% of total direct healthcare expenditures in developed countries such as the United States.⁶⁵

• A global analysis of 142 countries (93.2% of the world’s population) concluded that physical inactivity cost healthcare systems $53.8 billion in 2013, including $9.7 billion paid by individual households.⁶⁶

• A study of American adults reported that inadequate levels of aerobic PA (after adjusting for BMI) were associated with an estimated 11.1% of aggregate healthcare expenditures (including expenditures for inpatient, outpatient, ED, office-based, dental, vision, home health, prescription drug, and other services).⁶⁷

• An evaluation of healthcare costs based on the cardiovascular risk factor profile (including ≥30 minutes of moderate to vigorous PA ≥5 times per week) found that among adults aged ≥40 years with CVD, the highest marginal expenditures ($2853 in 2012) were for those not meeting the PA guidelines. Healthcare costs included hospitalizations, prescribed medications, outpatient visits (hospital outpatient visits and office-based visits), ED visits, and other expenditures (dental visits, vision aid, home health care, and other medical supplies).⁶⁸

• A systematic review of population-based interventions to encourage PA found that improving biking trails, distributing pedometers, and school-based PA were most cost-effective.⁶⁹

• Interventions and community strategies to increase PA have been shown to be cost-effective in terms of reducing medical costs⁷⁰:

  	— Nearly $3 in medical cost savings is realized for every $1 invested in building bike and walking trails.
  	— Incremental cost and incremental effectiveness ratios range from $14 000 to $69 000 per QALY gained from interventions such as pedometer or walking programs compared with no intervention, especially in high-risk groups.

Strategies to Prevent Physical Inactivity

The US Surgeon General has introduced “Step It Up!, a Call to Action to Promote Walking and Walkable Communities” in recognition of the importance of PA.⁷¹ There are roles for communities, schools, and worksites.

Communities

• Community-level interventions have been shown to be effective in promoting increased PA. Communities can encourage walking with street design that includes sidewalks, improved street lighting, and landscaping design that reduces traffic speed to improve pedestrian safety. Moving to a walkable neighborhood was associated with a lower risk for incident hypertension in the Canadian Community Health Survey.⁷²

• Community-wide campaigns: Such campaigns include a variety of strategies, such as media coverage, risk factor screening and education, community events, and policy or environmental changes.

• Educating the public on the recommended PA guidelines may increase adherence. In a study examining awareness of current US PA guidelines, only 33% of respondents had direct knowledge of the recommended dosage of PA (ie, frequency/duration).⁷³

Schools

• Schools can provide opportunities for PA through physical education, recess, before- and after-school activity programs, and PA breaks.⁷⁴

• According to the School Health Policies and Practices Study, <5% of elementary schools and junior and senior high schools required daily physical education in 2014.³⁸

• In 2012, the School Health Policies and Practices Study also reported that 58.9% of school districts required regular elementary school recess.³⁸

• Healthy afterschool programs and active school day policies have been shown to be cost-effective solutions to increase PA and prevent childhood obesity.⁷⁵

Worksites

• Worksites can offer access to on-site exercise facilities or employer-subsidized off-site exercise facilities to encourage PA among employees.

• Worksite interventions for sedentary occupations, such as providing “activity-permissive” workstations and e-mail contacts that promote breaks, have reported increased occupational light activity, and the more adherent individuals observed improvements in cardiometabolic outcomes.⁷⁶

Global Burden

• Physical inactivity is responsible for 12.2% of the global burden of MI after accounting for other CVD risk factors such as cigarette smoking, DM, hypertension, abdominal obesity, lipid profile, excessive alcohol intake, and psychosocial factors.⁷⁷

• Worldwide, the prevalence of physical inactivity (35%) is now greater than the prevalence of smoking (26%). On the basis of the HRs associated with these 2 behaviors (1.57 for smoking and 1.28 for inactivity), it was concluded that the PAR was greater for inactivity (9%) than for smoking (8.7%). Inactivity was estimated to be responsible for 5.3 million deaths compared with 5.1 million deaths for smoking.⁷⁸

• The GBD 2015 Study used statistical models and data on incidence, prevalence, case fatality, excess mortality, and cause-specific mortality to estimate disease burden for 315 diseases and injuries in 195 countries and territories. Mortality rates attributable to low PA are high in the Pacific Island countries (Chart 4-11).⁷⁹

5. Nutrition

See Table 5-1 and Charts 5-1 through 5-8

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Chart 5-8. Age-standardized global mortality rates attributable to dietary risks per 100 000, both sexes, 2015. Please click here to view the chart and its legend.

This chapter of the Update highlights national dietary habits, focusing on key foods, nutrients, dietary patterns, and other dietary factors related to cardiometabolic health. It is intended to examine current intakes, trends and changes in intakes, and estimated effects on disease to support and further stimulate efforts to monitor and improve dietary habits in relation to cardiovascular health.

Prevalence and Trends in the AHA 2020 Healthy Diet Metrics

(See Table 5-1 and Charts 5-1 through 5-4)

The AHA’s 2020 Impact Goals prioritize improving cardiovascular health,¹ which includes following a healthy diet pattern characterized by 5 primary and 3 secondary metrics (Table 5-1) that should be consumed within the context that is appropriate in energy balance and consistent with a DASH-type eating plan.¹

The AHA scoring system for ideal, intermediate, and poor diet patterns uses a binary-based scoring system, which awards 1 point for meeting the ideal target for each metric and 0 points otherwise.² For better consistency with other dietary pattern scores such as DASH, an alternative continuous scoring system has been developed to measure small improvements over time toward the AHA ideal target levels (Table 5-1). The dietary targets remain the same, and progress toward each of these targets is assessed by use of a more granular range of 1 to 10 (rather than 0 to 1).

Using the alternative scoring system, the mean AHA healthy diet score improved between 2003 to 2004 and 2011 to 2012 in the United States in both children and adults (Chart 5-1). The prevalence of an ideal healthy diet score (>80) increased from 0.2% to 0.6% in children (Chart 5-2) and from 0.7% to 1.5% in adults (Chart 5-3). The prevalence of an intermediate healthy diet score (40–79) increased from 30.6% to 44.7% in children and from 49.0% to 57.5% in adults.³ These improvements were largely attributable to increased whole grain consumption and decreased SSB consumption in both children and adults.³ Among adults, other significant changes between 1999 and 2012 included increased consumption of nuts, seeds, and legumes (from 0.54 to 0.81 servings/d) and decreased consumption of 100% fruit juice (from 0.43 to 0.32 servings/d) and white potatoes (from 0.39 to 0.32 servings/d) among adults between 2003 to 2004 and 2011 to 2012 (Chart 5-4).³ No major trends were evident in progress toward the targets for consumption of sodium, fish, fruits and vegetables, processed meats, and saturated fat.

Although AHA healthy diet scores tended to improve in all categories of age, race/ethnicity, income, and education level, many disparities present in earlier years widened over time, with generally smaller improvements seen in minority groups and those with lower income or education.³ For example, among Americans with the highest family income (income-to-poverty ratio ≥3.0), the proportion with a poor diet decreased from 50.5% to 35.7%, whereas among those with the lowest family income (income-to-poverty ratio <1.3), the proportion with a poor diet decreased from 67.8% to only 60.6%.

Global Trends in Key Dietary Factors

Globally, between 1999 and 2010, SSB intake increased in several countries.⁴ Among adults, mean global intake was highest in males aged 20 to 39 years, at 1.04 8-oz servings per day. In comparison, globally, females >60 years of age had the lowest mean consumption at 0.34 servings per day. SSB consumption was highest in the Caribbean, with adults consuming on average 2 servings per day, and lowest in East Asia, at 0.20 servings per day. Adults in the United States had the 26th-highest consumption among 187 countries.

Worldwide in 2010, mean sodium intake among adults was 3950 mg/d, which corresponds to salt intake of ≈10 g/d.⁵ Across world regions, mean sodium intakes were highest in Central Asia (5510 mg/d) and lowest in Eastern sub-Saharan Africa (2180 mg/d). Across countries, the lowest observed mean national intakes were ≈1500 mg/d. Between 1990 and 2010, global mean sodium intake appeared to remain relatively stable, although data on trends in many world regions were suboptimal.⁵

In a systematic review of population-level sodium initiatives (10 initiatives, 64 798 participants with sufficient data), reduction in mean sodium intake occurred in 5 initiatives, whereas there was no change in 3 initiatives and an increase in 2 initiatives.⁶ Successful population-level sodium initiatives tended to use multiple strategies and included structural activities, such as food product reformulation. For example, Finland initiated a nationwide campaign in the late 1970s through public education, collaboration with the food industry, and salt labeling legislation. From 1979 to 2002, mean 24-hour urine sodium excretion in population-based samples decreased in Finnish males (5.1 g/d to 3.9 g/d) and females (4.1 g/d to 3.0 g/d); mean SBP and prevalence of hypertension also decreased over that time period.^7,8 Similarly, the United Kingdom initiated a nationwide salt reduction program in 2003 to 2004 that included consumer awareness campaigns, progressively lower salt targets for various food categories, clear nutritional labeling, and working with industry to reformulate foods. The UK initiative appears to have been successful in reducing the average sodium intake of the population by 15% from 2003 to 2011.⁶ Over the same time period, there were also parallel decreases in blood pressure of 3.0/1.4 mm Hg in patients not taking antihypertensive medication (P<0.001), in stroke mortality by 42% (P<0.001), and in IHD mortality by 40% (P<0.001); these findings remained statistically significant after adjustment for changes in demographics, BMI, and other dietary factors.

Dietary Habits in the United States: Current Intakes

Foods and Nutrients

Adults

(See Chart 5-5)

The average dietary consumption by US adults of selected foods and nutrients related to cardiometabolic health based on data from 2011 to 2012 NHANES data is detailed below:

• Consumption of whole grains was 1.1 servings per day by NH white males and females, 0.8 to 0.9 servings per day by NH black males and females, and 0.6 to 0.8 servings by Mexican American males and females. For each of these groups, <10% of adults in 2011 to 2012 met guidelines of ≥3 servings per day.

• Fruit consumption ranged from 1.0 to 1.6 servings per day in these sex and racial or ethnic subgroups: ≈9% of NH whites, 7% of NH blacks, and 6% of Mexican Americans met guidelines of ≥2 cups per day. When 100% fruit juices were included, the number of servings increased and the proportions of adults consuming ≥2 cups per day nearly doubled in NH whites, doubled in NH blacks, and more than doubled in Mexican Americans.

• Nonstarchy vegetable consumption ranged from 1.7 to 2.7 servings per day. Across all racial/ethnic subgroups, NH white females were the only group meeting the target of consuming ≥2.5 cups per day.

• Consumption of fish and shellfish was lowest among Mexican American females and white females (0.8 and 1.0 servings per week, respectively) and highest among NH black females and NH black and Mexican American males (1.9 and 1.7 servings per week, respectively). Generally, only 15% to 27% of adults in each sex and racial or ethnic subgroup consumed at least 2 servings per week.

• Weekly consumption of nuts and seeds was ≈3.5 servings among NH whites and 2.5 servings among NH blacks and Mexican Americans. Approximately 1 in 4 whites, 1 in 6 NH blacks, and 1 in 8 Mexican Americans met guidelines of ≥4 servings per week.

• Consumption of unprocessed red meats was higher in males than in females, up to 4.8 servings per week in Mexican American males.

• Consumption of processed meats was lowest among Mexican American females (1.1 servings per week) and highest among NH black and NH white males (≈2.5 servings per week). Between 57% (NH white males) and 79% (Mexican American females) of adults consumed ≤2 servings per week.

• Consumption of SSBs ranged from 6.8 servings per week among NH white females to nearly 12 servings per week among Mexican American males. Females generally consumed less than males. Some adults, from 33% of Mexican American males to 65% of NH white females, consumed <36 oz/wk.

• Consumption of sweets and bakery desserts ranged from 3.9 servings per week (Mexican American males) to >7 servings per week (white females). Approximately 1 in 3 adults (1 in 2 Mexican American males) consumed <2.5 servings per week.

• Consumption of eicosapentaenoic acid and docosahexaenoic acid ranged from 0.058 to 0.117 g/d in each sex and racial or ethnic subgroup. Fewer than 8% of NH whites, 14% of NH blacks, and 11% of Mexican Americans consumed ≥0.250 g/d.

• One third to one half of adults in each sex and racial or ethnic subgroup consumed <10% of total calories from saturated fat, and approximately one half to two thirds consumed <300 mg of dietary cholesterol per day.

• Only ≈7% to 10% of NH whites, 4% to 5% of blacks, and 13% to 14% of Mexican Americans consumed ≥28 g of dietary fiber per day.

• Only approximately 6% to 8% of adults in each age and racial or ethnic subgroup consumed less than the recommended 2.3 g of sodium per day. Sodium is widespread in the US food supply, with top food categories of sodium intake listed in Chart 5-5. Top sources of sodium intake varied by race/ethnicity, with the largest contributor being yeast breads for NH whites, sandwiches for NH blacks, burritos and tacos for Hispanics, and soups for NH Asians.⁹

Foods and Nutrients

Children and Teenagers

• The average dietary consumption by US children and teenagers of selected foods and nutrients related to cardiometabolic health is detailed below:

  	— Whole grain consumption was low, <1 serving per day in all age and sex groups, with <5% of all children in different age and sex subgroups meeting guidelines of ≥3 servings per day.10
  	— Fruit consumption was low and decreased with age: 1.7 to 1.9 servings per day in younger boys and girls (5–9 years of age), 1.4 servings per day in adolescent boys and girls (10–14 years of age), and 0.9 to 1.3 servings per day in teenage boys and girls (15–19 years of age). The proportion meeting guidelines of ≥2 cups per day was also low and decreased with age: ≈8% to 14% in those 5 to 9 years of age, 3% to 8% in those 10 to 14 years of age, and 5% to 6% in those 15 to 19 years of age. When 100% fruit juices were included, the number of servings consumed increased by ≈50%, and proportions consuming ≥2 cups per day increased to nearly 25% of those 5 to 9 years of age, 20% of those 10 to 14 years, and 15% of those 15 to 19 years of age.
  	— Nonstarchy vegetable consumption was low, ranging from 1.1 to 1.5 servings per day, with <1.5% of children in different age and sex subgroups meeting guidelines of ≥2.5 cups per day.
  	— Consumption of fish and shellfish was low, ranging between 0.3 and 1.0 servings per week in all age and sex groups. Among all ages, only 7% to 14% of youths consumed ≥2 servings per week.
  	— Consumption of nuts, seeds, and beans ranged from 1.1 to 2.7 servings per week among different age and sex groups, and generally <15% of children in different age and sex subgroups consumed ≥4 servings per week.
  	— Consumption of unprocessed red meats was higher in boys than in girls and increased with age, up to 3.6 and 2.5 servings per week in 15- to 19-year-old boys and girls, respectively.
  	— Consumption of processed meats ranged from 1.4 to 2.3 servings per week, and the majority of children consumed no more than 2 servings per week of processed meats.
  	— Consumption of SSBs was higher in boys than in girls in the 5- to 9-year-old (7.7±6.2 versus 6.0±3.8 servings per week) and 10- to 14-year-old (11.6±5.3 versus 9.7±7.9 servings per week) groups, but it was higher in girls than in boys in the 15- to 19-year-old group (14±6.0 versus 12.4±5.8 servings per week). Only about half of children 5 to 9 years of age and one quarter of boys 15 to 19 years of age consumed <4.5 servings per week.
  	— Consumption of sweets and bakery desserts was higher among 5- to 9-year-old and 10- to 14-year-old boys and girls and modestly lower (4.7 to 6 servings per week) among 15- to 19-year-olds. A minority of children in all age and sex subgroups consumed <2.5 servings per week.
  	— Consumption of eicosapentaenoic acid and docosahexaenoic acid was low, ranging from 0.034 to 0.065 g/d in boys and girls in all age groups. Fewer than 7% of children and teenagers at any age consumed ≥250 mg/d.
  	— Consumption of saturated fat was ≈11% of calories in boys and girls in all age groups, and average consumption of dietary cholesterol ranged from ≈210 to 270 mg/d, increasing with age. Approximately 25% to 40% of youths consumed <10% energy from saturated fat, and ≈70% to 80% consumed <300 mg of dietary cholesterol per day.
  	— Consumption of dietary fiber ranged from ≈14 to 16 g/d. Fewer than 3% of children in all age and sex subgroups consumed ≥28 g/d.
  	— Consumption of sodium ranged from 3.1 to 3.5 g/d. Only 2% to 11% of children in different age and sex subgroups consumed <2.3 g/d.
  	— In children and teenagers, average daily caloric intake is higher in boys than in girls and increases with age in boys.

Impact on US Mortality

(See Chart 5-6)

A recent report used comparable risk assessment methods and nationally representative data to estimate the impact of 10 specific dietary factors on cardiometabolic mortality (eg, deaths attributable to heart disease, stroke, type 2 DM) in the United States in 2002 and 2012 (Chart 5-6).¹¹ In 2012, 318 656 (45.4%) of 702 308 cardiometabolic deaths were estimated to be attributable to poor dietary habits. The largest numbers of deaths attributable to diet were estimated to be from high sodium intake (66 508; 9.5% of all cardiometabolic deaths), low consumption of nuts/seeds (59 374; 8.5%), high consumption of processed meats (57 766; 8.2%), low intake of seafood omega-3 fats (54 626; 7.8%), low consumption of vegetables (53,410; 7.6%) and fruits (52 547; 7.5%), and high consumption of SSBs (51 694; 7.4%) (Chart 5-5). Between 2002 and 2012, population-adjusted US cardiometabolic deaths decreased by 26.5%, with declines in estimated diet-associated cardiometabolic deaths for polyunsaturated fats (−20.8%), nuts/seeds (−18.0%), and SSBs (−14.5%) and increases in diet-associated cardiometabolic deaths for sodium (5.8%) and unprocessed red meats (14.4%). Estimated cardiometabolic mortality related to whole grains, fruits, vegetables, seafood, omega-3 fats, and processed meats remained relatively stable.

Cost

(See Chart 5-7)

The US Department of Agriculture reported that the Consumer Price Index for all food decreased 0.3% from August 2016 to July 2017.¹² Prices for foods eaten at home decreased by 1.3% in 2016, whereas prices for foods eaten away from home increased by 2.6%.¹² Total food expenditures as a share of personal disposable income have remained relatively stable from 2004 (9.5%) to 2014 (9.7%).¹³ The share of foods eaten away from home increased from 47.2% in 2004 to 50.1% in 2014, eclipsing the share of foods eaten at home for the first time (Chart 5-7).¹³

Cost of a Healthy Diet

• A meta-analysis of price comparisons of healthy versus unhealthy diet patterns found that the healthiest diet patterns cost, on average, approximately $1.50 more per person per day to consume.¹⁴

• In an assessment of snacks served at YMCA afterschool programs from 2006 to 2008, healthful snacks were ≈50% more expensive ($0.26 per snack) than less healthful snacks.¹⁵ Higher snack costs were driven by serving fruit juice compared with water; serving refined grains without trans fat compared with refined grains with trans fat; and serving fruit and canned or frozen vegetables. Serving fresh vegetables (mostly carrots or celery) or whole grains did not alter price.

Cost-Benefit of a Healthy Diet

• An overview of the costs of various strategies for primary prevention of CVD determined that the estimated costs per year of life gained were between $9800 and $18 000 for statin therapy, $1500 for nurse screening and lifestyle advice, $500 to $1250 for smoking cessation, and $20 to $900 for population-based healthy eating.¹⁶

• Each year, more than $33 billion in medical costs and $9 billion in lost productivity resulting from HD, cancer, stroke, and DM are attributed to poor nutrition.^10,17–19

• Two separate cost-effectiveness analyses estimated that population reductions in dietary salt would not only be cost-effective but actually cost-saving. In one analysis, a 1.2-g/d reduction in dietary sodium was projected to reduce US annual cases of incident CHD by 60 000 to 120 000, stroke by 32 000 to 66 000, and total mortality by 44 000 to 92 000.²⁰ If accomplished through a regulatory intervention, estimated savings in healthcare costs would be $10 to $24 billion annually.²⁰ Such an intervention would be more cost-effective than using medications to lower BP in all people with hypertension.

Secular Trends

Trends in Dietary Patterns

In addition to individual foods and nutrients, overall dietary patterns can be another useful tool for assessing diet quality.²¹ Different dietary patterns have been defined, such as Mediterranean, DASH-type, Healthy Eating Index–2010, Alternate Healthy Eating Index, Western, prudent, and vegetarian patterns. The original DASH diet was low fat; a higher-monounsaturated-fat DASH-type diet is even more healthful and similar to a traditional Mediterranean dietary pattern.^22,23 The Healthy Eating Index–2010, which reflects compliance with the 2010 US Dietary Guidelines, exhibits a wide distribution among the US population, with a 5th percentile score of 31.7 and a 95th percentile score of 70.4 based on NHANES 2003 to 2004 data (theoretical maximum=100).²⁴ Average diet quality is worse in males (score=49.8) than in females (52.7), in younger adults (45.4) than in older adults (56.1), and in smokers (45.7) than in nonsmokers (53.3).

Between 1999 and 2010, the average Alternate Healthy Eating Index–2010 score of US adults improved from 39.9 to 46.8.²⁵ This was related to reduced intake of trans fat (accounting for more than half of the improvement), SSBs, and fruit juice, as well as an increased intake of whole fruit, whole grains, polyunsaturated fatty acids, and nuts and legumes. Adults with greater family income and education had higher scores, and the gap between low and high SES widened over time, from 3.9 points in 1999 to 2000 to 7.8 points in 2009 to 2010.

Worldwide, 2 separate, relatively uncorrelated dietary patterns can be characterized, 1 by greater intakes of health-promoting foods (eg, fruits, vegetables, nuts, fish) and 1 by lower intakes of less optimal foods (eg, processed meats, SSBs).²⁶ In 2010, compared with low-income nations, high-income nations had better diet patterns based on healthful foods but substantially worse diet patterns based on unhealthful foods. Between 1990 and 2010, both types of dietary patterns improved in high-income Western countries but worsened or did not improve in low-income countries in Africa and Asia. Middle-income countries showed the largest improvements in dietary patterns based on healthful foods but the largest deteriorations in dietary patterns based on unhealthful foods. Overall, global consumption of healthy foods improved but was outpaced by increased intake of unhealthy foods in most world regions.

Trends in Energy Intake

Until 1980, total energy intake remained relatively constant²⁷; however, data from NHANES indicate that average energy intake among US adults increased from 1971, peaked at 2004, and has since stabilized. Average total energy intakes among US white adults in 1971, 2004, and 2010 were 1992 kcal/d, 2283 kcal/d, and 2200 kcal/d, respectively. Average total energy intakes among US black adults in 1971, 2004, and 2010 were 1780, 2169, and 2134 kcal/d, respectively. This rise in energy intake was primarily attributable to increased carbohydrate intake, particularly of starches, refined grains, and sugars.^28–30

In a study using 4 nationally representative US Department of Agriculture surveys of food intake among US children,³¹ average portion sizes among US children increased for many foods between 1977 and 2006. For example, pizza portion size increased from 406 to 546 calories, much of this increase occurring from the 1990s to the 2000s. Portion sizes for other foods increased, including Mexican food (from 373 to 512 calories), cheeseburgers (from 380 to 473 calories), soft drinks (from 121 to 155 calories), fruit drinks (from 106 to 133 calories), and salty snacks (from 124 to 165 calories). French fry portion sizes increased at fast food locations but not stores and restaurants. Soft drink and pizza portion sizes increased at all food sources (stores, restaurants, and fast foods).

In a quantitative analysis using various US surveys between 1977 and 2010, the relationships of national changes in energy density, portion sizes, and number of daily eating/drinking occasions to changes in total energy intake were assessed.^27,32 Changes in energy density were not consistently linked to energy intake over time, whereas increases in both portion size and number of eating occasions were linked to greater energy intake.

Among US children 2 to 18 years of age, increases in energy intake between 1977 and 2006 (179 kcal/d) were entirely attributable to substantial increases in energy eaten away from home (255 kcal/d).³³ The percentage of energy eaten away from home increased from 23.4% to 33.9% during this time, with a shift toward energy from fast food as the largest contributor to foods eaten away from home for all age groups.

In a study using NHANES and Nielsen Homescan data to examine disparities in calories from store-bought consumer packaged goods over time, beverages from stores decreased between 2003 to 2006 and 2009 to 2012. However, the decline in calories from consumer packaged goods was slower for NH blacks, Mexican-Americans, and lowest-income households.³⁴

Cardiovascular Health Impact

Cardiovascular and Metabolic Risk

Overweight and Obesity

• For short-term (up to 1–2 years) weight loss among overweight and obese individuals, the macronutrient composition of the diet has much less influence than compliance with the selected diet.³⁵

• In ad libitum (not energy-restricted) diets, a low-carbohydrate (high-fat) diet demonstrated better weight loss and reduced fat mass than a low-fat (high-carbohydrate) diet at 1 year.³⁶

• In ad libitum diets, intake of dietary sugars was positively linked to weight gain.³⁷ However, isoenergetic exchange of dietary sugars with other carbohydrates had no relationship with body weight,³⁷ which suggests that all refined carbohydrates (complex starches and simple sugars) might be similarly obesogenic.

• In pooled analyses across 3 prospective cohort studies of US males and females, increased glycemic index and glycemic load were independently associated with greater weight gain over time.³⁸

• Across types of foods, energy density (total calories per gram of food) is not consistently linked with weight gain or obesity. For example, nuts have relatively high energy density and are inversely linked to weight gain, and cheese has high energy density and appears relatively neutral, whereas sugar-sweetened beverages have low energy density and increase obesity.³⁸ National changes in energy density over time are not consistently linked to changes in energy intake.²⁷

• In analyses of >120 000 US males and females in 3 separate US cohorts followed up for up to 20 years, changes in intakes of different foods and beverages were linked to long-term weight gain in different ways.^38,39 Foods and beverages linked to increased weight gain included high-glycemic carbohydrates such as potatoes, white bread, white rice, low-fiber breakfast cereals, sweets/desserts, and SSBs, as well as red and processed meats. In contrast, increased consumption of several other foods, including nuts, whole grains, fruits, vegetables, legumes, fish, and yogurt, was linked to relative weight loss over time. These findings suggest that attention to food-based dietary quality, not simply counting total calories, is crucial for long-term weight homeostasis.³⁹

• In both adults and children, intake of SSBs has been linked to weight gain and obesity.⁴⁰ Randomized trials in children demonstrate reductions in obesity when SSBs are replaced with noncaloric beverages.⁴⁰

• Dietary factors that stimulate hepatic de novo lipogenesis, such as rapidly digestible grains, starches, and sugars, as well as trans fat, appear more strongly related to weight gain.^38,39,41

• Diet quality might also influence energy expenditure. After intentional weight loss, isocaloric diets higher in fat and lower in rapidly digestible carbohydrates produced significantly smaller declines in total energy expenditure than low-fat, high-carbohydrate diets, with a mean difference of >300 kcal/d.³⁵

• Other possible nutritional determinants of positive energy balance (more calories consumed than expended), as determined by adiposity or weight gain, include larger portion sizes, skipping breakfast, consumption of fast food, and eating foods prepared outside the home, although the evidence for long-term relevance of these factors has been inconsistent.^42–45

• Societal and environmental factors independently associated with diet quality, adiposity, or weight gain include education, income, race/ethnicity, and (at least cross-sectionally) neighborhood availability of supermarkets.^46–48

• Other local food-environment characteristics, such as availability of grocery stores (ie, smaller stores than supermarkets), convenience stores, and fast-food restaurants, are not consistently associated with diet quality or adiposity and could be linked to social determinants of health for CVD.⁴⁹