pediatrics
November 2015, VOLUME136 /ISSUE 5

Protecting Children From Tobacco, Nicotine, and Tobacco Smoke

  1. Harold J. Farber, MD, MSPH, FAAP,
  2. Judith Groner, MD, FAAP,
  3. Susan Walley, MD, FAAP,
  4. Kevin Nelson, MD, PhD, FAAP,
  5. SECTION ON TOBACCO CONTROL

    Abstract

    This technical report serves to provide the evidence base for the American Academy of Pediatrics’ policy statements “Clinical Practice Policy to Protect Children From Tobacco, Nicotine, and Tobacco Smoke” and “Public Policy to Protect Children From Tobacco, Nicotine, and Tobacco Smoke.” Tobacco use and involuntary exposure are major preventable causes of morbidity and premature mortality in adults and children. Tobacco dependence almost always starts in childhood or adolescence. Electronic nicotine delivery systems are rapidly gaining popularity among youth, and their significant harms are being documented. In utero tobacco smoke exposure, in addition to increasing the risk of preterm birth, low birth weight, stillbirth, placental abruption, and sudden infant death, has been found to increase the risk of obesity and neurodevelopmental disorders. Actions by pediatricians can help to reduce children’s risk of developing tobacco dependence and reduce children’s involuntary tobacco smoke exposure. Public policy actions to protect children from tobacco are essential to reduce the toll that the tobacco epidemic takes on our children.

  1. Abbreviations:
    ALL
    acute lymphoblastic leukemia
    CDC
    Centers for Disease Control and Prevention
    CI
    confidence interval
    e-cigarette
    electronic cigarette
    FDA
    US Food and Drug Administration
    FTND
    Fagerström Test of Nicotine Dependence
    NHANES
    National Health and Nutrition Examination Survey
    NYTS
    National Youth Tobacco Survey
    OR
    odds ratio
    s-ICAM1
    soluble intercellular adhesion molecule 1
    SIDS
    sudden infant death syndrome
    RSV
    respiratory syncytial virus
    USPSTF
    US Preventive Services Task Force
  2. Introduction

    This technical report describes the evidence base for the American Academy of Pediatrics’ policy statements “Public Policy to Protect Children From Tobacco, Nicotine, and Tobacco Smoke” and “Clinical Practice Policy to Protect Children From Tobacco, Nicotine, and Tobacco Smoke”.1,2

    The goal of the present technical report is to document knowledge regarding the harms of tobacco to children and adolescents and to document the evidence for actions by clinicians and policy makers to reduce the toll that tobacco takes on children and adolescents. Because comprehensive literature reviews and evaluations are conducted by the Office of the Surgeon General, the present report focused on additional research findings subsequent to the Reports of the Surgeon General and topics not well covered in those reports. When multiple studies produced similar findings, the best quality and/or most recent are presented with reference to meta-analyses or authoritative statements (eg, Reports of the Surgeon General, US Public Health Service recommendations or guidelines) when available. Relevant literature was identified by using PubMed searches and references cited in relevant review articles or authoritative statements. Each article was assessed for quality and relevance. Reports of the Surgeon General were identified from the Web site http://www.surgeongeneral.gov/library/reports/index.html.

    Grading Evidence on the Harms of Tobacco

    Because it is unethical to conduct randomized controlled clinical trials among children with a substance that is harmful, the evidence related to the harms of tobacco has been graded as follows: strong quality—consistent findings from double-blind, randomized controlled clinical trials, large representative-sample epidemiologic studies, good-quality meta-analyses incorporating large representative studies with consistent results, and/or a Report of the Surgeon General of a “major conclusion” or “evidence is sufficient”; good quality—well-performed, generalizable case-control study, other well-performed epidemiologic study, other meta-analyses, and a Report of the Surgeon General of “evidence is suggestive”; or fair quality—other research study, small sample size, and findings not replicated.

    Grading Evidence for Clinical and Policy Recommendations

    Evidence quality and strength of recommendations were determined on the basis of guidelines of the American Academy of Pediatrics’ policy statement “Classifying Recommendations for Clinical Practice Guidelines” and are summarized in Fig 1 and Table 1.3

    FIGURE 1

    Evidence quality. RCT, randomized controlled trial.

    TABLE 1

    Guideline Definitions for Evidence-Based Statements

    Definitions

    • Tobacco product: any nicotine delivery product, currently regulated or unregulated by the US Food and Drug Administration (FDA), which is not approved for safe and effective tobacco dependence treatment.

    • Secondhand smoke: the smoke emitted from a tobacco product that is inhaled by a nonuser.

    • Thirdhand smoke: the tobacco smoke that is absorbed onto surfaces and exposes the nonuser by either direct contact and dermal absorption and/or off-gassing and inhalation. Thirdhand smoke may react with oxidants and other compounds in the environment to yield secondary pollutants.4

    • Involuntary tobacco smoke exposure: the tobacco smoke exposure of nonusers. Involuntary exposure includes both secondhand and thirdhand exposure.

    • Electronic nicotine delivery systems: handheld devices that produce an aerosol from a solution typically containing nicotine, flavoring chemicals, and carrier solvents such as propylene glycol and vegetable glycerin (glycerol) for inhalation by the user. Alternate names for these products include electronic cigarettes, e-cigarettes, e-cigs, electronic cigars, e-cigars, electronic hookah, e-hookah, hookah sticks, personal vaporizers, mechanical mods, vape pens, and vaping devices.

    Key Findings

    How Tobacco Harms Children

    1. Tobacco product use is common among youth. Quality of evidence: strong

    2. More than one-half of children in the United States have evidence of tobacco smoke exposure. Quality of evidence: strong

    3. Tobacco dependence creates a substantial economic burden for both civilian and military sectors. Quality of evidence: strong

    4. Tobacco kills people when used as intended. Quality of evidence: strong

    5. Tobacco smoke exposure harms children. Quality of evidence: strong

    6. Tobacco exposure harms the fetus. Quality of evidence: strong

    7. Tobacco increases infant mortality. Quality of evidence: strong

    8. Tobacco smoke exposure increases asthma prevalence and severity. Quality of evidence: strong

    9. The effects of tobacco smoke exposure on risk of asthma start in utero. Quality of evidence: strong

    10. Tobacco smoke exposure increases the severity of bronchiolitis. Quality of evidence: strong

    11. Tobacco smoke exposure increases risk for and severity of other respiratory illnesses. Quality of evidence: strong

    12. Tobacco smoke exposure increases the risk of middle ear disease. Quality of evidence: strong

    13. In utero tobacco smoke exposure increases the risk of being overweight in childhood. Quality of evidence: strong

    14. Tobacco smoke exposure increases the risk of learning and neurobehavioral problems. Quality of evidence: strong

    15. Tobacco smoke exposure of children leads to findings of preclinical atherosclerosis. Quality of evidence: strong

    16. Tobacco smoke exposure increases the risk of childhood cancers. Quality of evidence: good

    17. Smoke-free homes may reduce children’s tobacco smoke exposure. Quality of evidence: good

    18. Tobacco dependence almost always develops before reaching the age of majority. Quality of evidence: strong

    19. Tobacco dependence is a treatable chronic illness. Quality of evidence: strong

    Recommended Actions for Pediatricians

    1. Inquire about tobacco use and tobacco smoke exposure as part of health supervision visits and visits for diseases that may be caused or exacerbated by tobacco smoke exposure. Evidence quality: B; Recommendation strength: strong recommendation

    2. Include tobacco use prevention as part of anticipatory guidance. Evidence quality: B; Recommendation strength: strong recommendation

    3. Address parent/caregiver tobacco dependence as part of pediatric health care. Evidence quality: B; Recommendation strength: strong recommendation

      • 3a. Recommend tobacco dependence treatment of tobacco-dependent parents and caregivers. Evidence quality: B; Recommendation strength: strong recommendation

      • 3b. Implement systems to identify and offer counseling, treatment, treatment recommendations, and/or referral for tobacco-dependent parents. Evidence quality: C; Recommendation strength: recommendation

    4. Offer tobacco dependence treatment and/or referral to adolescents who want to stop smoking. Evidence Quality: B; Recommendation strength: strong recommendation

      • 4a. Tobacco dependence pharmacotherapy can be considered for moderate to severely tobacco-dependent adolescents who want to stop smoking. Evidence quality: D; Recommendation strength: option

    5. Offer tobacco-dependent individuals quitline referral. Evidence quality: A; Recommendation strength: strong recommendation

    6. Consider potential for neuropsychiatric symptoms with tobacco dependence treatment. Evidence quality: C; Recommendation strength: recommendation

    7. Do not recommend electronic nicotine delivery systems for tobacco dependence treatment. Evidence quality: B; Recommendation strength: strong recommendation

    8. If the sources of a child’s tobacco smoke exposure cannot be eliminated, provide counseling about strategies to reduce the child’s tobacco smoke exposure. Evidence quality: C, Recommendation strength: recommendation

    Public Policy Recommendations

    • 1. The FDA should regulate all tobacco products to protect the public health. Evidence quality: X; Recommendation strength: strong recommendation

    • 2. Tobacco control should be adequately funded. Evidence quality: A; Recommendation strength: strong recommendation

    Recommendations for Public Policy to Protect Children From Tobacco Use Initiation

    • 3. Tobacco product advertising and promotion in forms that are accessible to children and youth should be prohibited. Evidence quality: B; Recommendation strength: strong recommendation

    • 4. Point-of-sale tobacco product advertising and product placement that can be viewed by children should be prohibited. Evidence quality: B; Recommendation strength: strong recommendation

    • 5. Depictions of tobacco products in movies and other media that can be viewed by youth should be restricted. Evidence quality: B; Recommendation strength: strong recommendation

    • 6. The promotion and sale of electronic nicotine delivery systems to youth should be prohibited. Evidence quality: B; Recommendation strength: strong recommendation

    • 7. Tobacco control programs should change the image of tobacco by telling the truth about tobacco. Evidence quality: B; Recommendation strength: strong recommendation

    • 8. Tobacco product prices should be increased to reduce youth tobacco use initiation. Evidence quality: B; Recommendation strength: strong recommendation

    • 9. The minimum age to purchase tobacco should be increased to 21 years. Evidence quality: B; Recommendation strength: strong recommendation

    • 10. Flavoring agents, including menthol, should be prohibited in all tobacco products. Evidence quality: B; Recommendation strength: strong recommendation

    Recommendations to Protect Children From Tobacco Smoke and Nicotine Exposure

    • 11. Comprehensive smoking bans should be enacted. Evidence quality: B; Recommendation strength: strong recommendation

    • 12. Smoking in multi-unit housing should be prohibited. Evidence quality: B; Recommendation strength: strong recommendation

    • 13. Prohibitions on smoking and use of tobacco products should include prohibitions on use of electronic nicotine delivery systems. Evidence quality: B; Recommendation strength: strong recommendation

    Recommendations to Protect Children From Acute Nicotine Poisoning

    • 14. Children younger than 18 years should be legally prohibited from working on tobacco farms and in tobacco production. Evidence quality: C, Recommendation strength: recommendation

    • 15. Concentrated nicotine solution for electronic nicotine delivery systems should be sold in child-resistant containers with amounts limited to that which would not be lethal to a young child if ingested. Evidence quality: B, Recommendation strength: strong recommendation

    How Tobacco Harms Children

    1. Tobacco product use is common among youth.

    Quality of evidence: strong

    The 2012 Report of the Surgeon General concluded, “Almost one in four high school seniors is a current (in the past 30 days) cigarette smoker, compared with one in three young adults and one in five adults.” Approximately 1 in 10 high school senior male students is a current smokeless tobacco user, and ∼1 in 5 high school senior male students is a current cigar smoker. Significant disparities in tobacco use remain among young people nationwide. In the United States, the prevalence of cigarette smoking is highest among American-Indian and Alaska Native people compared with other ethnic groups and highest among youth of lower socioeconomic status compared with more affluent youth. The use of smokeless tobacco is increasing among self-identified white male high school students, and cigar smoking may be increasing among black female high school students.4

    Table 2 describes the different forms of tobacco currently available in the United States. The 2012 US National Youth Tobacco Survey (NYTS) described 6.7% of middle school students (5.6% of girls and 7.8% of boys) as currently (ie, within the last 30 days) using any tobacco products, with cigarettes (3.1%), cigars (2.8%), smokeless tobacco (1.7%), pipes (1.8%), hookahs (water pipes) (1.3%), electronic cigarettes (e-cigarettes) (1.1%), and Swedish snuff (snus) (0.8%) as the most commonly used products. Among high school students, 23.3% (18% of girls and 28% of boys) reported current (ie, within the last 30 days) use of any tobacco product, with cigarettes (14.0%), cigars (12.6%), smokeless tobacco (6.4%), pipes (4.5%), hookahs (5.4%), e-cigarettes (2.8%), and snus (2.5%) as the most commonly used products.5 Concurrent use of multiple tobacco products is common. In the 2012 NYTS, 9.9% of US high school students reported use of 2 or more different tobacco products.6

    TABLE 2

    Tobacco Products Currently Available in the United States

    Analyses of data from the 2011 and 2012 NYTS found that initiation rates for e-cigarette use doubled from 1.4% to 2.7% among middle school students and from 4.7% to 10.0% among high school students. Current (ie, within the last 30 days) e-cigarette use showed a similar pattern, with increases from 0.6% to 1.1% of middle school students and 1.5% to 2.8% of high school students in the 2011–2012 period.7 Current data show that rates of e-cigarette use are dramatically escalating among youth, with the 2014 NYTS reporting current (ie, within the last 30 days) e-cigarette use in 3.9% of middle school students and 13.4% of high school students, increases of 650% and 890%, respectively, compared with 2011 data.8 The Monitoring the Future survey reported current (ie, within the last 30 days) e-cigarette use among 16.2% of 11th graders and 17.1% of 12th graders in 2014.9 Hookah use among youth is also rapidly escalating, with 9.4% of high school students reporting hookah use according to the 2014 NYTS.8

    Analysis of data from the Growing Up Today Study, a large, longitudinal cohort of adolescents followed up from ages 12 to 24 years (N = 13 913), found smoking initiation at younger ages and greater prevalence of smoking and amount smoked among homosexual and bisexual youth. Odds of past-month smoking were 1.66 (95% confidence interval [CI]: 1.22–2.28) and 1.57 (95% CI: 1.28–1.94) for bisexual and gay male subjects, respectively, and 2.62 (95% CI: 2.31–2.97) and 2.12 (95% CI: 1.70–2.64) for bisexual and gay female subjects relative to heterosexual youth.10

    2. More than one-half of children in the United States have evidence of tobacco smoke exposure.

    Quality of evidence: strong

    The National Health and Nutrition Examination Survey (NHANES) assesses a nationally representative sample of the noninstitutionalized US civilian population. Tobacco smoke exposure is indicated by a measurable serum cotinine concentration. Cotinine, a nicotine metabolite, is a biomarker of recent tobacco exposure and can be measured in serum, urine, and saliva.11 In the 2007–2008 NHANES, 53.6% (95% CI: 46.2–61.0) of children 3 to 11 years of age had evidence of tobacco smoke exposure. This amount is greater than the level of exposure in adults aged 20 years or older, among whom 36.7% (95% CI: 32.0–41.3) were exposed. The 2007–2008 survey results were similar to findings of the 2001–2002 and 2005–2006 surveys and showed slightly less exposure than the 1999–2000 and 2003–2004 surveys.12

    Tobacco-dependent parents and caregivers are important sources of children’s tobacco smoke exposure. A total of 519 children aged 3 to 12 years with a history of asthma and tobacco smoke exposure were enrolled in a clinical trial of an intervention for reducing tobacco smoke exposure.13 The ratio of urine cotinine to creatinine was higher if either the mother or caregiver were tobacco smokers and highest if both were smokers.

    3. Tobacco dependence creates a substantial economic burden for both civilian and military sectors.

    Quality of evidence: strong

    Using rigorous methods, the 2014 Report of the Surgeon General determined that smoking-attributable economic costs were $289 to $332.5 billion per year in the United States.14 These costs include $132.5 to $175.9 billion for direct medical care of adults, $151 billion for lost productivity because of premature death, and $5.6 billion from lost productivity because of exposure to secondhand smoke.

    Tobacco use leads to a substantial burden for the armed services. A 2009 Institute of Medicine report quotes the US Department of Defense Assistant Secretary of Defense for Health Affairs, “Every year, tobacco use leads to unnecessary compromises in the readiness of our troops and costs the Department of Defense millions of dollars in preventable health care costs.” Tobacco use can impair troops both through the effects of tobacco smoke toxicants, including carbon monoxide, and through nicotine withdrawal. Analyses of data from TRICARE (the health care program serving US uniformed service members, retirees, and their families) identified excess medical costs of approximately $228 per tobacco user per year, with retirees and dependents incurring greater medical costs because of tobacco use ($321) than active-duty enrollees ($104) or their dependents ($106). Tobacco use in the military is also associated with failure to complete basic training and premature discharge from the armed forces.1517

    4. Tobacco kills people when used as intended.

    Quality of evidence: strong

    Tobacco use by youth and young adults has severe adverse health consequences. This evidence has been summarized in multiple Reports of the Surgeon General from 1964 onward.4,14,1820 The landmark 40-year prospective follow-up study of 34 439 male British physicians by Doll et al21 found that nearly one-half of all regular cigarette smokers died as a result of their addiction. The age at which one-half of subjects had died was 8 years younger for smokers than for nonsmokers. The risk of smoking-related disease was reduced by stopping smoking. Similar results were found in a 25-year follow-up study of 49 539 Norwegian adults.22 The same Norwegian study found a synergistic effect of smoking and obesity on mortality in middle age, with the combination of heavy smoking and obesity leading to much greater premature mortality than either factor alone.23 A study in Finland comparing changes in smoking rates with changes in lung cancer rates 20 years later found a correlation close to 1:1 (R2 = 0.95).24

    5. Tobacco smoke exposure harms children.

    Quality of evidence: strong

    The substantial harm of tobacco toxin exposure for children has been extensively documented, with evidence summarized in the 2006 Report of the Surgeon General.25 Because nicotine and other tobacco toxins cross the placenta, children are harmed from exposure to tobacco toxins starting in utero. This exposure can be both from the mother’s tobacco product use as well as her exposure (via inhalation or absorption) to the tobacco smoke of others. Children are harmed from secondhand tobacco smoke exposure by breathing in the smoke emitted by others who are using combustible tobacco products.

    Thirdhand tobacco smoke exposure is increasingly being recognized as another route of tobacco toxin exposure.26 Thirdhand tobacco smoke is the smoke that remains on surfaces and in dust, which may be re-emitted into the gas phase or may react with oxidants and other compounds in the environment to yield secondary pollutants.27 Thirdhand smoke includes nicotine, tobacco-specific carcinogens, and other toxicants.28,29 Children can absorb, ingest, and inhale these substances. An analysis of house dust samples collected from private homes of tobacco smokers in northeastern Spain found tobacco-related carcinogens in the dust at levels sufficient to increase the risk of cancer.30 Young children may have greater exposure because toddlers commonly explore by placing objects in the mouth. Children and adolescents can also be harmed from absorption of tobacco toxins when they participate in tobacco production.31,32

    6. Tobacco exposure harms the fetus.

    Quality of evidence: strong.

    The 2014 Report of the Surgeon General found that the evidence is sufficient to conclude that tobacco smoking in early pregnancy causes orofacial clefts.14

    In utero tobacco exposure from either maternal active tobacco product use or maternal secondhand tobacco smoke exposure increases the risk of stillbirth. The Missouri maternally linked cohort data set contains information on both live births and fetal deaths, with maternal smoking during pregnancy routinely documented on the birth certificate on the basis of the mother’s response after delivery.33 The large data set included 57 965 stillbirth cases and 51 436 413 live birth controls. There was a dose–response relationship, with greater risk of stillbirth relating directly to the amount the mother smoked. For deliveries at 20+ weeks’ gestation, the risk of stillbirth was increased compared with nonsmokers, with an odds ratio (OR) of 1.43 (95% CI: 1.31–1.57) if the mother smoked >1 pack per day and an OR of 1.31 (95% CI: 1.22–1.41) if the mother smoked one-half to 1 pack per day. A study using data from the Swedish Medical Birth Register (with 2322 stillbirths and 851 371 live births) found that, compared with mothers who did not use tobacco, the risk of stillbirth increased for mothers who were tobacco smokers (adjusted OR: 1.59 [95% CI: 1.40–1.80]) and snus users (adjusted OR: 1.43 [95% CI: 1.02–1.99]).34 A study of 1110 pregnant women in Mumbai, India, found a substantial increased risk of stillbirth associated with maternal use of smokeless tobacco (OR: 2.6 [95% CI: 1.4–4.8])35 and if the mother did not smoke but was exposed to the smoke of another inside the home.36 A study in Newfoundland, Canada, of 11 862 nonsmoking women with singleton gestations, of whom 11.1% reported exposure to secondhand tobacco smoke, found an increased risk of stillbirth with an adjusted OR of 3.35 (95% CI: 1.16–9.72).37 A recent meta-analysis found that maternal exposure to secondhand tobacco smoke during pregnancy increased the risk of stillbirth (OR: 1.23 [95% CI: 1.09–1.38]).38

    Maternal smoking also increases the risk of placenta-associated complications of pregnancy. A case-control study in Finland compared 175 placental abruption case subjects with 370 delivery time–matched control subjects.39 There were more smokers among case subjects (27.4% vs 14.3%; P < .001). Serum cotinine concentrations were greater in case subjects compared with control subjects (median: 229.5 vs 153.5 ng/mL; P = .002), and many more case subjects than control subjects had serum cotinine concentrations >15 ng/dL (30.3% vs 17.6%; P < .001). These findings provide biological confirmation of the greater tobacco exposure in case subjects compared with control subjects. An analysis of data from the Missouri Electronic Vital Records system of 1 312 505 singleton births at 20 to 44 weeks’ gestation found that if the mother was a smoker, the risk of placental abruption and placenta previa were substantially greater (0.71% vs 1.27% [P < .01] and 0.35% vs 0.48% [P < .01], respectively). A dose–response relationship was observed; the ORs for placental abruption and placenta previa were greater for those who smoked ≥20 cigarettes per day compared with those who smoked 0 to 9 cigarettes per day (OR compared with nonsmokers: 1.9 [95% CI: 1.7–2.0] and 1.7 [95% CI: 1.1–1.9] vs 1.5 [95% CI: 1.4–1.7] and 1.4 [95% CI: 1.2–1.5]).40

    Both active maternal smoking and secondhand maternal tobacco smoke exposure have been shown to reduce birth weight. A case-control study in Saudi Arabia compared birth outcomes of 1085 tobacco smoke–exposed women with those of 2341 unexposed women with term, singleton pregnancies. Mean birth weight was lower in the tobacco smoke–exposed women compared with the unexposed women (3.15 vs 3.21 kg; P = .002).41 In Rhode Island, a longitudinal follow-up study of 119 pregnant women enrolled in their third trimester of pregnancy found that birth weight was greater for nonsmoking women and women who stopped smoking during pregnancy compared with those who continued to smoke (mean birth weight: 3.46 and 3.56 kg vs 3.16 kg, respectively; P = .004).42 When biomarkers of tobacco exposure in the infant’s meconium were assayed, findings were similar, with greater mean birth weight if biomarkers of tobacco exposure were absent versus if those biomarkers were present (3.50 vs 3.20 kg; P < .001). A case-control study in Lucknow, India, of mothers aged 20 to 30 years who did not use tobacco compared those who had a low birth weight infant (<2.5 kg) versus those who did not have a low birth weight infant (≥2.5 kg).43 After adjusting for other factors associated with low birth weight, a history of tobacco smoke exposure (ie, active smoker in the home smoked in their presence) increased the odds for low birth weight (adjusted OR: 3.16 [95% CI: 1.9–5.3]). Similar findings were noted in a retrospective cohort study in Newfoundland, Canada, of nonsmoking women with singleton gestations.37 A total of 1202 women with a history of tobacco smoke exposure were compared with 10 650 women with no tobacco smoke exposure. Mean birth weight was lower for the smoke-exposed women (3.43 vs 3.51 kg; P < .0001), and the odds of low birth weight (<2.5 kg) were also greater for the smoke-exposed women (OR: 1.65 [95% CI: 1.29–2.09]).

    Maternal smoking and tobacco smoke exposure increase the risk of preterm birth. A birth cohort study of 10 095 nonsmoking women aged >18 years who had a singleton live birth in Lanzhou, China, found that maternal tobacco smoke exposure was associated with an increased risk of very preterm (<32 weeks’ gestation) birth, with a dose–response relationship demonstrated.44 For those exposed to smoke for <1 hour per day, the OR for a very preterm birth was 1.89 (95% CI: 1.26–2.84), and for those exposed to ≥1 hour per day, the OR for a very preterm birth was 2.61 (95% CI: 1.56–4.34) compared with mothers who were not exposed to smoke. An analysis of data from the Swedish Medical Birth Register found that the risk of preterm birth was increased for both maternal snus use (adjusted OR: 1.27 [95% CI: 1.14–1.41]) and maternal smoking at 1 to 9 cigarettes per day and ≥10 cigarettes per day (adjusted OR: 1.24 [95% CI: 1.17–1.32] and 1.56 [95% CI: 1.44–1.69], respectively).45 Analysis of data from the Generation R study, a longitudinal follow-up study from early pregnancy onward of 9778 mothers and their children living in Rotterdam, the Netherlands, found that continued maternal smoking after pregnancy was associated with low birth weight (adjusted OR: 1.75 [95% CI: 1.20–2.56]) and preterm birth (adjusted OR: 1.36 [95% CI: 1.04–1.78]) in the fully adjusted models.46 An ecologic study in Belgium found that after introduction of staged smoke-free legislation (first workplaces, then restaurants, then bars that serve food), rates of preterm birth decreased, with further decreases in preterm birth rates with each successive escalation of the smoking restrictions.47 Similar reductions in preterm birth were observed coinciding with the implementation of smoke-free legislation in Scotland.48

    7. Tobacco increases infant mortality.

    Quality of evidence: strong

    Using 2005–2009 data, the 2014 Report of the Surgeon General determined that tobacco smoking during pregnancy results in nearly 1000 infant deaths per year or ∼8% of all infant deaths and 17% of all cases of sudden infant death syndrome (SIDS).14

    Tobacco smoke exposure increases an infant’s risk of SIDS. This conclusion was a major finding of the 2006 Report of the Surgeon General.25 Both prenatal and postnatal exposure contributes to the risk. Association of in utero tobacco smoke exposure with abnormalities of sleep arousal responses has been demonstrated.4951 A recent Dutch case-control study compared 142 SIDS case subjects versus 2841 control subjects recruited from well-infant clinics. The study found that, compared with nonsmoking parents, parental smoking led to an increased risk of SIDS, with the risk greater if both parents were smokers (OR: 5.8 [95% CI: 2.2–15.5]) versus if 1 parent was a smoker (OR: 2.5 [95% CI: 1.2–5.0]), thus demonstrating a dose–response effect.52

    8. Tobacco smoke exposure increases asthma prevalence and severity.

    Quality of evidence: strong

    Tobacco smoke exposure increases the risk of asthma, wheezing, and asthma exacerbations in children. The 2006 Report of the Surgeon General concluded, “The evidence is sufficient to infer a causal relationship between parental smoking and cough, phlegm, wheeze, and breathlessness among children of school age. The evidence is sufficient to infer a causal relationship between parental smoking and ever having asthma among children of school age. The evidence is sufficient to infer a causal relationship between secondhand smoke exposure from parental smoking and the onset of wheeze illnesses in early childhood.”25 The Bogalusa Heart Study showed that asthma prevalence was consistently greater in children whose mothers were smokers, with nearly 3000 children surveyed on 3 occasions between 1983 and 1994; ORs ranged from 1.35 (95% CI: 1.01–1.81) to 1.51 (95% CI: 1.17–1.96) depending on survey year.53 Among children aged 1 to 16 years hospitalized for asthma, having detectable salivary cotinine levels was associated with increased odds of readmission within 12 months (adjusted OR: 2.35 [95% CI: 1.22–4.55]).54 Among 466 children enrolled in the CHIRAH (Chicago Initiative to Raise Asthma Health Equity) study, increases in salivary cotinine concentrations were associated with an increased risk of asthma exacerbations.55 Implementation of smoke-free legislation led to decreases in childhood asthma hospitalizations in England and Scotland.56,57

    9. The effects of tobacco smoke exposure on risk of asthma start in utero.

    Quality of evidence: strong

    Prenatal tobacco smoke exposure adversely affects lung development. In a cohort of 4574 mothers and their children prospectively followed up from pregnancy through the first 4 years of the child’s life, exposure to maternal smoking when in utero as well as secondhand smoke after birth were associated with increased risk for wheezing at 2 to 4 years of age.58 History of in utero tobacco smoke exposure was associated with greater rates of poor asthma control in 2481 Latino and African-American children with asthma when assessed at 8 to 17 years of age.59 A prospective follow-up study of 1129 children from birth to 14 years of age found that maternal smoking during pregnancy was associated with current asthma, current wheeze, and exercise-induced wheeze (ORs: 1.84 [95% CI: 1.16–2.92], 1.77 [95% CI: 1.14–2.75], and 2.29 [95% CI: 1.37–3.85], respectively).60 Analyses of data from the CAMP (Childhood Asthma Management Program) study found that a history of in utero tobacco smoke exposure markedly attenuated the benefit of inhaled corticosteroid response among children aged 5 to 12 years with persistent asthma and demonstrated airway hyperreactivity.61 A meta-analysis of 79 prospective epidemiologic studies published between 1997 and February 2011 assessed the association between tobacco smoke exposure and the incidence of wheeze or asthma in childhood; it found that prenatal maternal smoking and household secondhand tobacco smoke exposure were associated with an increased risk of asthma.62

    10. Tobacco smoke exposure increases the severity of bronchiolitis.

    Quality of evidence: strong

    In utero and secondhand tobacco smoke exposure of children leads to more severe episodes of bronchiolitis. The 2006 Report of the Surgeon General concluded, “Smoking by parents causes respiratory symptoms and slows lung growth in their children.” It also concluded, “The evidence is sufficient to infer a causal relationship between secondhand smoke exposure from parental smoking and lower respiratory tract illnesses in infants and children.”25 The increased risk of lower respiratory illnesses is greatest from smoking by the mother. A study in Liverpool, United Kingdom, of 378 infants hospitalized for bronchiolitis, of whom 299 (79%) had respiratory syncytial virus (RSV) infection, found that having a household tobacco smoker increased the odds of needing supplemental oxygen and needing mechanical ventilation during the hospitalization (ORs: 2.45 [95% CI: 1.60–3.74] and 5.49 [95% CI: 2.78–10.83], respectively).63 A case-control study of infants born at 32 to 35 weeks’ gestation found that smoking in the presence of the child and maternal smoking during pregnancy were more common in the cases with RSV hospitalization (ORs: 1.59 [95% CI: 1.12–2.26] and 1.62 [95% CI: 1.17–2.24]).64 According to a prospective follow-up study of 217 healthy newborn infants from a single pediatric center in Denmark, a history of smoking in the household was associated with a higher risk of hospitalization because of RSV in the first year of life (OR: 5.06 [95% CI: 1.36–18.76]).65 A systematic literature review of studies assessing the effect of tobacco smoke exposure on RSV bronchiolitis in children aged younger than 5 years identified 30 relevant articles published between 1990 and 2009.66 The review found a consistent impact of tobacco smoke exposure on risk of hospitalizations for RSV disease.

    11. Tobacco smoke exposure increases risk for and severity of other respiratory illnesses.

    Quality of evidence: strong

    Tobacco smoke exposure increases the risk of pneumonia and cough. The 2006 Report of the Surgeon General concluded, “The evidence is sufficient to infer a causal relationship between parental smoking and cough, phlegm, wheeze, and breathlessness among children of school age”; “The evidence is sufficient to infer a causal relationship between maternal smoking during pregnancy and persistent adverse effects on lung function across childhood”; and “The evidence is sufficient to infer a causal relationship between exposure to secondhand smoke after birth and a lower level of lung function during childhood.”25 A population survey in Vietnam of 24 781 households with a child younger than 5 years found that having 1 or more smokers in the household was associated with an increased risk of hospitalization for pneumonia in the previous 12 months (adjusted OR: 1.55 [95% CI: 1.25–1.92]).67 The Cincinnati Childhood Allergy and Air Pollution Study prospectively followed up children from the newborn period.68 At 7 years of age, lung function and hair cotinine data were available on 486 members of the cohort. Significant reductions in forced expiratory volume in 1 second (0.03 L, P < .05) and mean forced expiratory flow between 25% and 75% of the forced vital capacity (0.06 L/s, P < .01) were seen for every log-unit change in cotinine concentration. A survey of 1718 children in third and fourth grades in Guangzhou, China, found increased sneezing and coughing at night among those with tobacco smoke exposure inside the home.69 A study of 117 children younger than 15 years hospitalized for influenza found that the risk of ICU admission and length of stay were greater among children with a history of secondhand tobacco smoke exposure (adjusted OR of 4.7 [95% CI: 1.4–18.5] and adjusted incidence rate ratio of 1.7 [95% CI: 1.2–2.3], respectively).70 A meta-analysis of 60 studies published before November 2010 of infants aged ≤2 years confirmed that smoking by any household member, paternal smoking, maternal prenatal smoking, and maternal postnatal smoking all increased the risk of an infant’s lower respiratory tract infection (ORs: 1.54 [95% CI: 1.40–1.69], 1.22 [95% CI: 1.10–1.35], 1.58 [95% CI: 1.45–1.73], and 1.24 [95% CI: 1.11–1.38]).71

    12. Tobacco smoke exposure increases the risk of middle ear disease.

    Quality of evidence: strong

    The 2006 Report of the Surgeon General concluded, “The evidence is sufficient to infer a causal relationship between parental smoking and middle ear disease in children, including acute and recurrent otitis media and chronic middle ear effusion.”25 Using data from the 91 642 interviews completed for the 2007–2008 US National Survey of Children’s Health, an association of secondhand smoke exposure with recurrent ear infections for children older than 6 years was demonstrated (adjusted ORs of 1.48 [95% CI: 0.98–2.21] for age 6–11 years and 1.67 [95% CI: 1.02–2.72] for age 12–17 years).72 A meta-analysis published in 2012 found that maternal postnatal smoking (20 studies) and household smoking (49 studies) increased the risk for otitis media, with ORs of 1.62 (95% CI: 1.33–1.97) and 1.37 (95% CI: 1.25–1.50), respectively.73

    13. In utero tobacco smoke exposure increases the risk of being overweight in childhood.

    Quality of evidence: strong

    Both active smoking and secondhand tobacco smoke exposure of the mother during pregnancy increase the child’s later risk of being overweight. A prospective follow-up study included 7924 infants of nonsmoking mothers who were born in Hong Kong in April or May 1997, with maternal secondhand tobacco smoke exposure determined at the first postnatal visit.74 Of those infants, 6790 (86%) had their BMI measured at 7 to 11 years of age. Children of daily paternal smokers had higher mean BMI z scores at 7 to 11 years of age, with a mean difference in BMI of 0.10 (95% CI: 0.02–0.19) at 7 years and 0.16 (95% CI: 0.07–0.26) at 11 years. A retrospective cohort study of 1366 fourth grade students in Kumagaya City, Japan, found that those with a history of maternal smoking during pregnancy had a higher mean ± SD BMI (17.2 ± 2.7 vs 16.9 ± 2.5; P = .016) when assessed at 9 to 10 years of age.75 The Millennium Cohort Study prospectively followed up 18 296 healthy term infants in the United Kingdom.76 Maternal smoking during pregnancy was associated with an increased risk of the child being overweight at 3 years of age (adjusted OR: 1.33 [95% CI: 1.15–1.55]). A meta-analysis of studies published from 1990 to May 2011 identified 7 relevant studies and confirmed the effect of maternal smoking during pregnancy on the risk of the children being overweight (adjusted OR: 1.47 [95% CI: 1.26–1.73]).77

    14. Tobacco smoke exposure increases the risk of learning and neurobehavioral problems.

    Quality of evidence: strong

    The 2014 Report of the Surgeon General concluded, “The evidence is sufficient to infer that nicotine exposure during fetal development, a critical window for brain development, has lasting adverse consequences for brain development,” and “The evidence is suggestive but not sufficient to infer a causal relationship between maternal prenatal smoking and disruptive behavioral disorders, and attention deficit hyperactivity disorder in particular, among children.”14 Analysis of data from the 2007 US National Survey of Children’s Health found that having a history of someone who smokes inside the home increased the child’s risk of having attention-deficit/hyperactivity disorder, learning disabilities, and/or conduct disorders reported (adjusted ORs: 1.44 [95% CI: 1.21–1.72], 1.54 [95% CI: 1.27–1.85], and 1.78 [95% CI 1.44–2.21], respectively).78 A survey of 5494 preschool-aged children in Bavaria, Germany, confirmed that a history of secondhand tobacco smoke exposure was associated with an increase in conduct problems and hyperactivity/inattention that was independent of the effect of maternal smoking before and during pregnancy.79 The Generation R Study enrolled mothers early in pregnancy and observed them and their children prospectively.80 Maternal smoking and smoke exposure were assessed during pregnancy, and child behavior problems were assessed at 18 months for 4329 mother–child dyads. Maternal smoking during pregnancy and paternal smoking were both associated with increased behavior problems for the child; however, these findings were not robust to statistical adjustment for psychosocial variables. A study in Finland of administrative data on 175 869 children born 1987 through 1989 found that maternal smoking increased the risk of having a psychiatric diagnosis through 18 years of age in analyses controlling for maternal psychiatric diagnosis and child’s gender.81

    A birth cohort study in Brisbane, Australia, assessed maternal smoking starting at the first prenatal visit.82 At 14 years of age, the child’s academic performance was assessed according to mother’s report. Rate of academic performance below average was greater if the mother smoked during pregnancy (adjusted OR: 1.35 [95% CI: 1.07–1.70]), with findings robust to adjustment for multiple potential confounding variables, including maternal age, income, education, alcohol consumption, family communication, and behavior problems. An analysis of computerized population data for children born in Sweden between 1983 and 1987 merged results from birth registries, school and education registries, and census data.83 The study found that the risk of poor school performance at 15 years of age (mean score below passing) was greater if the mother smoked during pregnancy, with a dose–response effect demonstrated. The OR for poor school performance was 1.58 (95% CI: 1.53–1.62) if the mother smoked 1 to 9 cigarettes per day and 1.89 (95% CI: 1.83–1.96) if the mother smoked ≥10 cigarettes per day during pregnancy. These findings were robust to statistical adjustment for maternal and birth characteristics. The NHANES III (1988–1994) included both measurement of serum cotinine concentrations and assessments of academic performance for children 6 to 16 years of age in the United States.84 Children’s tobacco exposure, as assessed by using serum cotinine concentrations, was significantly associated with lower scores for reading, math, and visuospatial skills, with even very low cotinine concentrations seeming to have an effect. Findings were robust to statistical adjustment for potentially confounding variables.

    15. Tobacco smoke exposure of children leads to findings of preclinical atherosclerosis.

    Quality of evidence: strong

    A cohort of 545 children in Finland prospectively followed up from infancy through adolescence had serum cotinine concentrations measured annually between 8 and 13 years of age.85 Carotid intima-media thickness was greater and peak flow–mediated dilation of the brachial artery was lower in children in the highest tertile of tobacco smoke exposure. Similar results were found in a study of 16-year-old male adolescents (N = 610) in Lhasa City, Tibet,86 and in a case-control study of healthy young adults.87 Using pooled data from the Cardiovascular Risk in Young Finns study (Finland) and the Childhood Determinants of Adult Health study (Australia), exposure to parental smoking was assessed in 3416 children and carotid intima-media thickness was assessed in adulthood, 21 to 28 years later. If both parents smoked at baseline, the child’s carotid intima-media thickness in adulthood was greater (mean: 0.652 vs 0.637 mm; P = .003 in fully adjusted analyses).88 This study did not find that smoking by 1 parent had an effect; however, the study did not differentiate maternal from paternal smoking. This omission may be important because maternal smoking generally has a greater effect on a child’s level of tobacco smoke exposure.13

    Soluble intercellular adhesion molecule 1 (s-ICAM1) is a measure of endothelial stress, and hair cotinine is a biomarker of tobacco smoke exposure. In a sample of children aged 9 to 18 years in Columbus, Ohio, hair cotinine concentrations were positively correlated with s-ICAM1 levels.89 A substantial amount of the variance in s-ICAM1 was accounted for by hair cotinine level of the child (partial R2 = 0.26, P = .0001 for the association of log hair cotinine with s-ICAM1 level in multivariate models adjusting for BMI, age, mean blood pressure, and very low-density lipoprotein level).

    16. Tobacco smoke exposure increases the risk of childhood cancers.

    Quality of evidence: good

    The 2006 Report of the Surgeon General concluded, “The evidence is suggestive but not sufficient to infer a causal relationship between prenatal and postnatal exposure to secondhand smoke and childhood cancer.”25 Additional evidence of an association between childhood cancers and tobacco smoke exposure has accumulated since this 2006 report. Tobacco-specific carcinogens have been detected in the blood of children who have a tobacco smoker in the home.90 A case-control study of children with acute lymphoblastic leukemia (ALL) in Australia found that a history of paternal smoking of ≥15 cigarettes per day during the pregnancy year was associated with an increased risk of childhood ALL, with an OR of 1.46 (95% CI: 1.05–2.01).91 The authors then pooled their results with 9 other relevant studies and documented a modest increased risk of paternal prenatal smoking on childhood ALL, with an OR of 1.15 (95% CI: 1.06–1.24). A case-control study of childhood (<14 years) cancers in Shiraz, Iran, found that paternal smoking before and during pregnancy and maternal secondhand smoke exposure during pregnancy were associated with an increased cancer risk in children, with ORs of 1.8 (95% CI: 1.4–6.0), 3.0 (95% CI: 1.4–5.0), and 3.6 (95% CI: 1.3–5.0), respectively.92 The Australian Study of Childhood Brain Tumors, a national population-based case-control study conducted between 2005 and 2010, found that maternal smoking before and during pregnancy was associated with increased risk of brain tumors diagnosed before 2 years of age, with ORs of 5.06 (95% CI: 1.35–19.00) and 4.61 (95% CI: 1.08–19.63).93

    17. Smoke-free homes may reduce children’s tobacco smoke exposure.

    Quality of evidence: good

    Smoke-free homes and cars may reduce children’s tobacco smoke exposure but are unlikely to completely protect a child as long as household members are smokers. A randomized controlled trial of an intensive intervention to implement smoke-free homes in Ankara, Turkey, found substantial reductions in urine cotinine levels over 12 months of follow-up in the intervention group but not in the control group.94 A randomized controlled study among Latino families in Houston, Texas, in which an adult was a smoker compared provision of 2 culturally appropriate fotonovelas (illustrated storybooks) and 1 comic book, which were designed to promote a tobacco-free indoor environment, versus use of a standard smoking cessation guide published by the American Cancer Society. At the 12-month follow-up, there were more reported bans on in-home smoking for the intervention (fotonovela) condition (73% vs 56%). Although the results did not differ according to intervention group, those homes with an in-home smoking ban had reduced nicotine concentrations on the home surfaces sampled (0.04 ± 0.04 μg/m3 vs 0.47 ± 0.66 μg/m3; P < .01).95 A study in Columbus, Ohio, of children younger than 3 years found that when the mother was a smoker, hair cotinine concentrations were not significantly different if the mother reported smoking inside or outside.96

    18. Tobacco dependence almost always develops before reaching the age of majority.

    Quality of evidence: strong

    Tobacco is a substance of abuse. The 2012 Report of the Surgeon General concluded, “Given their developmental stage, adolescents and young adults are uniquely susceptible to social and environmental influences to use tobacco,” and “Among adults who become daily smokers, nearly all first use of cigarettes occurs by 18 years of age (88%), with 99% of first use by 26 years of age.”4

    Nicotine dependence develops early and drives the progression from intermittent to daily smoking. A study of 1246 English-speaking students from central Massachusetts enrolled students in sixth grade and monitored them prospectively for 4 years.97,98 Of the 370 subjects who had inhaled from a cigarette, the median age at first cigarette use was 12 years. At least 1 symptom of early nicotine dependence was reported by 33% of participants who had ever puffed on a cigarette. Experiencing any symptom of nicotine dependence increased the risk of progressing to monthly smoking (adjusted hazard ratio: 3.7 [95% CI: 2.4–5.5]) or daily smoking (adjusted hazard ratio: 6.8 [95% CI: 4.4–10.5]). Analyses of data from the 2004 NYTS found that among the 2580 adolescent smokers aged 12 to 18 years who participated in the survey, there was a strong correlation between nicotine withdrawal symptoms and both the amount and frequency of smoking.99

    19. Tobacco dependence is a treatable chronic illness.

    Quality of evidence: strong

    The 2008 US Public Health Service–sponsored clinical practice guideline advised that “Clinicians strongly recommend the use of effective tobacco dependence counseling and medication treatments to their patients who use tobacco, and that health systems, insurers, and purchasers assist clinicians in making such effective treatments available.”100 Even brief advice can increase quit rates.101

    A 2013 Cochrane review concluded that nicotine replacement therapy, bupropion, varenicline, and cytisine (a nicotine receptor partial agonist, not currently available in the United States) improved the chances of stopping smoking.102 Combination nicotine replacement therapy outperformed single formulations. Forms of nicotine replacement therapy that are approved by the FDA for tobacco dependence treatment are nicotine patches, nicotine gum, and nicotine lozenges (available over the counter in the United States) as well as nicotine nasal spray and nicotine inhalers (available only by prescription in the United States).

    The 2009 American College of Chest Physicians’ Tobacco Dependence Treatment ToolKit advises, “Approaching tobacco dependence as a chronic disease acknowledges the altered central nervous system (CNS) neurobiology in tobacco-dependent patients. The goal of therapy in tobacco dependence is to normalize brain function—so that the patient has minimal to no symptoms of nicotine withdrawal, thus allowing the patient to feel (near) normal while not using tobacco. The intensity of treatment should be based on the severity level of nicotine dependence. For highly nicotine-dependent patients, combination therapy is often needed.”103 Adhering to the model of asthma and other chronic diseases regarding exacerbations and remissions, the longer acting medications can be thought of as “controllers,” with the faster acting medications used as “relievers.”

    In studies of tobacco-dependent adults, initiation of nicotine replacement therapy before stopping smoking improves the effectiveness of treatment. A multisite, randomized clinical trial with parallel groups (placebo and active treatment–controlled) included 400 tobacco-dependent adult subjects in North Carolina.104 It provided an intervention of a 21-mg nicotine patch versus placebo daily for 2 weeks before cessation, with active nicotine patch (no placebos) provided for all subjects after the stop smoking date. Precessation treatment with the nicotine patch substantially improved quit rates, with the greatest benefit for those with lower levels of nicotine dependence (Fagerström Test for Nicotine Dependence [FTND] score <6). For smokers with lower FTND scores, 10-week continuous abstinence rates were 33.8% in the precessation nicotine patch condition versus 9.3% in the placebo patch condition. In contrast, for smokers with higher FTND scores, abstinence rates did not differ significantly between patch conditions (14.0% in the precessation nicotine patch condition vs 10.8% in the placebo patch condition). A meta-analysis identified 4 relevant studies completed before February 2007 and found that precessation therapy with a nicotine patch doubled abstinence rates at 6 weeks and 6 months (ORs: 1.96 [95% CI: 1.31–2.93] and 2.20 [95% CI: 1.39–3.48], respectively).105

    For further details on practical, evidence-based, expert consensus recommendations for tobacco dependence treatment, the reader is referred to the American College of Chest Physicians’ Tobacco Dependence Treatment ToolKit (http://tobaccodependence.chestnet.org).

    Recommended Actions for Pediatricians

    1. Inquire about tobacco use and tobacco smoke exposure as part of health supervision visits and visits for diseases that may be caused or exacerbated by tobacco smoke exposure.

    Evidence quality: B

    Recommendation strength: strong recommendation

    It is important to identify tobacco use so that appropriate interventions can be offered to prevent tobacco use initiation, recommend stopping smoking, and/or recommend treatment of tobacco dependence. It is important that tobacco smoke exposure be identified so that interventions can be offered to reduce this exposure. The Memphis Health Project, a longitudinal study of smoking in 5154 adolescents, found that participants who were both screened and advised had more knowledge about the health risks of smoking and more negative perceptions of the social value of smoking than participants with no intervention or screening.106 Recall of physician communication was associated by teenagers with improved perceptions of the dangers of smoking. The adolescents who were smokers reported more quit attempts and fewer intentions to continue to smoke when they reported recall of physician advice. Analyses of data from the 18 866 respondents of the 2011 NYTS found low rates of recalled health care provider screening for tobacco use (32.2% [95% CI: 30.2–34.1]).107 Among current youth smokers, receipt of health professional counseling was associated with having made an attempt to stop smoking (OR: 1.39 [95% CI: 1.15–1.68]). Similar results were found on analyses of data from the 24 573 participants in the 2000 NYTS.108

    2. Include tobacco use prevention as part of anticipatory guidance.

    Evidence quality: B

    Recommendation strength: strong recommendation

    The US Preventive Services Task Force (USPSTF) recommends that primary care clinicians provide interventions, including education or brief counseling, to prevent initiation of tobacco use in school-aged children and adolescents.109 The USPSTF report concluded, “The USPSTF found adequate evidence that behavioral counseling interventions, such as face-to-face or phone interaction with a health care provider, print materials, and computer applications, can reduce the risk for smoking initiation in school-aged children and adolescents.” A meta-analysis of behaviorally based interventions to prevent tobacco use initiation found 10 relevant trials with a pooled risk ratio of 0.81 (95% CI: 0.70–0.93).110

    3. Address parent/caregiver tobacco dependence as part of pediatric health care.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Because tobacco smoke exposure is harmful to the child, and parental tobacco smoking is an important source of a child’s tobacco smoke exposure, addressing parental and caregiver tobacco dependence is important in protecting the health of the child. According to the US Public Health Service–sponsored evidence-based guideline Treating Tobacco Use and Dependence: 2008 Update, “It is essential that clinicians and health care delivery systems consistently identify and document tobacco use status and treat every tobacco user seen in a health care setting.”100

    An assessment of parents of tobacco smoke–exposed children with asthma who enrolled in a clinical trial found that most of the primary caregivers who were tobacco smokers were interested in stopping smoking; 56.5% were at the contemplation or better stage of change.13

    Tobacco dependence treatment or recommendation for treatment by pediatricians is acceptable to most parents. A nationally representative (US) telephonic survey included 730 parents who had accompanied a child to a health care visit in the past year, of whom 21% were tobacco smokers.111 The survey found that most (59.7%) of the smokers thought it would be acceptable to have the child’s physicians prescribe or recommend a tobacco dependence treatment medication for them.

    3a. Recommend tobacco dependence treatment of tobacco-dependent parents and caregivers.

    Evidence quality: B

    Recommendation strength: strong recommendation

    According to the US Public Health Service–sponsored evidence-based guideline Treating Tobacco Use and Dependence: 2008 Update, “Counseling and medication are effective when used by themselves for treating tobacco dependence. The combination of counseling and medication, however, is more effective than either alone. Thus, clinicians should encourage all adults making an attempt to stop tobacco to use both counseling and medication.”100

    Behaviorally based interventions for parental tobacco dependence have limited benefit; interventions that include use of medications show greater efficacy. A clinical trial of cotinine feedback and behavioral counseling for parents of tobacco smoke–exposed children with asthma found no significant intervention effect; however, on subgroup analyses, children with high-risk asthma who received the intervention had a greater reduction in the ratio of urine cotinine to creatinine than did the control group.112 A meta-analysis of studies designed to protect children from tobacco smoke exposure through parental cessation or modification of parental smoking identified 18 relevant studies published through March 2011.113 Studies had substantial variation in the methods and interventions tested. The relative risk of parental smoking cessation from the pooled study results was 1.34 (95% CI: 1.05–1.71). The impact was greatest in the subgroup of 2 studies that included the use of nicotine replacement therapy in the intervention (risk ratio: 3.13 [95% CI: 1.19–8.21]).

    3b. Implement systems to identify and offer counseling, treatment, treatment recommendations, and/or referral for tobacco-dependent parents.

    Evidence quality: C

    Recommendation strength: recommendation

    Practical systems have been developed and validated to address parental tobacco dependence as part of the child’s health care. A summary of counseling approaches is described in Table 3. A randomized controlled clinical trial involving 22 practices in the Pediatric Research in Office Settings network tested the use of the Clinical Effort Against Secondhand Smoke Exposure (CEASE) program.114 The intervention included: (1) routine screening for parental tobacco use; (2) motivational messaging based on the parents' own concerns; and (3) recommendation and possible provision of nicotine patch and gum by the clinician and enrollment in the free state (telephone) quitline. On exit interview, parental smokers in the intervention practices reported a higher rate of discussing methods to stop smoking (24% vs 2%; P < .001), prescription of nicotine replacement medication (12% vs 0%; P < .001), and enrollment in the state (telephonic) quitline (10% vs 0%; P < .001).

    TABLE 3

    Counseling Approaches

    Before prescribing tobacco dependence treatment for parents, pediatricians should verify that their medical liability insurance provides coverage for care offered to adults. If the pediatrician elects to prescribe for parents, he or she should conduct an appropriate assessment of disease (tobacco dependence), consider possible contraindications to the medications, counsel about risks and benefits, offer recommendations for follow-up, and provide appropriate treatment. Follow-up is important to monitor for adherence to treatment recommendations, adverse effects of medications prescribed or used, correct technique for use of the recommended treatments, adequacy of treatment in controlling nicotine withdrawal symptoms, and stage of change in relation to stopping tobacco use and in relation to accepting tobacco dependence treatment. Pediatricians should follow state regulations and institutional policies for charting on care provided for parents and caregivers to benefit the health of the child. Electronic health record systems should facilitate adherence to documentation requirements without placing excessive burdens on the parent or pediatrician.

    4. Offer tobacco dependence treatment and/or referral to adolescents who want to stop smoking.

    Evidence quality: B

    Recommendation strength: strong recommendation

    4a. Tobacco dependence pharmacotherapy can be considered for moderate to severely tobacco-dependent adolescents who want to stop smoking.

    Evidence quality: D

    Recommendation strength: option

    Tobacco dependence treatment of adolescents has varying degrees of success in stopping smoking, with results contingent on the severity of the dependence. Behaviorally based programs for tobacco-dependent adolescents are effective and are most beneficial for those with minimal to mild degrees of dependence. The most effective of the behaviorally based programs are developmentally relevant and focus on contingency management skills, stage of change-based motivational interviewing, social support, or a combination of these methods.115 Although more benefit is observed with more intensive interventions, even brief counseling by physicians can be of assistance, with effects enhanced by repeated advice.101

    Although still beneficial compared with nonintervention, behaviorally based programs have much lower rates of smoking cessation among teenagers who are severely tobacco dependent. For example, Project EX (an 8-session, school-based clinic tobacco use cessation program for adolescents that includes enjoyable, motivating activities) found that 30-day abstinence from smoking on completion of the program was 42% for those with minimal to mild tobacco dependence but only 7% for those with severe tobacco dependence.116 Similar results were found in evaluation of the Not On Tobacco program; at 3 months’ follow-up, 24% of those with minimal to mild nicotine dependence reported not smoking, but only 9.4% of those with severe nicotine dependence reported not smoking.117 Clinical trials of motivational interviewing versus brief advice (without medication use) for tobacco-dependent adolescents yielded very low stop-smoking rates that did not differ between treatment groups.118,119

    There is emerging evidence documenting a benefit of pharmacotherapy for tobacco-dependent adolescents; adherence is challenging, however, and relapse after brief courses of treatment is common. Investigators compared the use of a nicotine patch versus nicotine gum versus placebo in a randomized, double-blind, placebo-controlled clinical trial with 120 adolescents who wanted to stop smoking and had moderate or greater tobacco dependence.120 Medication was initiated on the planned stop-smoking day and continued for 12 weeks. Group cognitive behavioral therapy was provided to all participants. At 1 week after the stop-smoking date, 26.5% of those receiving the nicotine patch, 17.4% of those receiving the nicotine gum, and 5.0% of the placebo group were not smoking (P = .02 for patch versus placebo). By 3 months after pharmacotherapy was discontinued, nonsmoking rates were 20.6% for the patch, 8.7% for the gum, and 5% for placebo (P = .06 for patch versus placebo). Adherence to daily use of the patch was acceptable at 78.4%; adherence to use of the nicotine gum was poor at 38.5%. Analyses of trial participants who had stopped smoking and then restarted found that those who restarted smoking after a period of abstinence had greater craving scores, suggesting that inadequately controlled withdrawal contributed to the lapse.121 A randomized, double-blind, placebo-controlled, parallel-group clinical trial included 6 weeks of bupropion plus counseling for tobacco-dependent adolescents.122 The authors found improved rates of smoking cessation with 150 mg of bupropion twice daily when the medication was being taken (29% abstinent at 6 weeks with bupropion vs 16% with placebo [P = .02]); the benefit was quickly lost after the medication was stopped, however.

    Because there has been limited research on tobacco dependence pharmacotherapy in adolescents, the FDA-approved labeling for these medications states, “Safety and effectiveness in the pediatric population have not been established.” Because tobacco dependence is a severe chronic illness that debilitates, harms offspring, and shortens life, it is reasonable to consider that pharmacotherapy documented as effective in adults is an option for the treatment of adolescents with moderate to severe tobacco dependence. Given the high rates of nonadherence during therapy and relapse after discontinuation of therapy among adolescents in the trials of these medications, close follow-up is recommended.

    5. Offer tobacco-dependent individuals quitline referral.

    Evidence quality: A

    Recommendation strength: strong recommendation

    Studies among adults found that free (to the user) telephonic tobacco dependence treatment is beneficial; however, research among teenagers is not yet available, and the services provided vary substantially. Callers to the California Smokers’ Helpline during periods of high call volume who were ready to stop smoking were randomized to be immediately assigned to a counselor (treatment) or asked to call back (control).123 All callers were sent written self-help materials. Twelve-month abstinence rates were 9.1% in the treatment group versus 6.9% in the control group (P < .001), with the entire difference in the control group attributable to those members who did not call back to receive assistance. A recent Cochrane review of telephonic counseling for smoking cessation found 77 trials that met inclusion criteria.124 Among smokers who contacted helplines, quit rates were higher for groups randomized to receive multiple sessions of proactive counseling (9 studies, >24 000 participants; relative risk for cessation at longest follow-up: 1.37 [95% CI: 1.26–1.50]).

    6. Consider potential for neuropsychiatric symptoms with tobacco dependence treatment.

    Evidence quality: C

    Recommendation strength: recommendation

    Pediatricians should be aware of the increased risk of suicidal ideation and suicide, both among continuing smokers and among those being treated for tobacco dependence. In studies among adults, suicide is associated with cigarette use in a dose-dependent manner.125127 Studies of adolescent population samples and youth with bipolar disorder demonstrate an association between tobacco use and suicide risk.128,129 In postmarketing surveillance, suicidal ideation and suicide have been reported among patients taking varenicline and bupropion, with a black box warning issued by the FDA for both medications.130 A study of 119 546 adults in England who used a smoking cessation product found no differences in rates of treated depression, nonfatal self-harm, and fatal self-harm within 3 months of the first smoking cessation prescription; the study compared those who received varenicline or bupropion with those who received nicotine replacement.131 Neuropsychiatric symptoms associated with tobacco dependence treatment may reflect inadequate control of nicotine withdrawal.103

    Comorbidities of substance abuse and psychiatric disorders may accompany tobacco dependence14 and can make tobacco dependence more difficult to treat. These comorbidities, when identified, should be addressed through appropriate referral.

    The risk of treatment-emergent neuropsychiatric symptoms should be balanced against the substantial harms of continued tobacco use. As with any other prescribing, the discussion of risks versus benefits should be documented in the health record.

    7. Do not recommend electronic nicotine delivery systems for tobacco dependence treatment.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Electronic nicotine delivery systems have not been shown to be effective for smoking cessation.132 There is currently no regulation on content or manufacturing standards for electronic nicotine delivery systems. In addition to nicotine, carcinogens, toxins, metals, and silicates have been found in the emissions from these devices.133135 The adverse effects of long-term inhalation of the flavoring agents used is not known, and some commonly used agents are known respiratory irritants.136

    Pediatricians should direct patients who want pharmacotherapy for tobacco dependence to therapies that have been documented to be effective and are approved as such by the FDA. In New Zealand, a clinical trial of e-cigarettes for smoking cessation among moderately to severely tobacco-dependent adults found low cessation rates and no statistically significant difference between the use of nicotine-containing e-cigarettes and placebo.137

    Among adolescents, the use of e-cigarettes is associated with decreased rates of stopping smoking. A nationally representative survey of middle and high school students in Korea found that among current adolescent smokers, although there were more attempts to stop smoking among e-cigarette users (OR: 1.67 [95% CI: 1.48–1.90]), current e-cigarette users were much less likely to have stopped using cigarettes (OR: 0.10 [95% CI: 0.09–0.12]) compared with smokers who never used e-cigarettes.138 An analysis of 2011 and 2012 NYTS data found that among smokers (having smoked at least 100 cigarettes in their lifetime) who had ever used e-cigarettes, both ever having used e-cigarettes and current e-cigarette use were associated with lower rates of abstinence from cigarette smoking (ORs of 0.32 [95% CI: 0.18–0.56] and 0.34 [95% CI: 0.13–0.87], respectively, for 1 year or greater abstinence; ORs of 0.61 [95% CI: 0.42–0.89] and 0.35 [95% CI: 0.18–0.69] for more than 30 days but less than 6 months’ abstinence from cigarettes).139

    8. If the sources of a child’s tobacco smoke exposure cannot be eliminated, provide counseling about strategies to reduce the child’s tobacco smoke exposure.

    Evidence quality: C

    Recommendation strength: recommendation

    A randomized, single-blind study of 352 children aged 8 to 16 years in Ankara, Turkey, tested a smoke-free home intervention aimed at increasing household smoking bans and reducing secondhand smoke exposure. Those randomized to the smoke-free home intervention had greater rates of in-home smoking restrictions and substantially lower urine cotinine levels at 2, 6, and 12 months of follow-up.94 A randomized, single-blind study in 91 Mexican-American households used low literacy fotonovelas and a comic book to promote a tobacco-free indoor air environment.95 The study found that a greater proportion of households in the intervention group implemented bans on smoking inside of the home. Although the study found no differences in ambient nicotine level according to intervention group, those households that implemented a ban on smoking indoors had lower ambient nicotine levels than households that did not implement such a ban.

    A randomized single-blind study compared an in-home motivational interviewing intervention versus provision of written information by mail in households with children aged <3 years and a parent/caregiver who is a smoker.140 Ambient nicotine levels measured in the kitchen and the room with a television were lower in the motivational interviewing intervention group than in the written information control group.

    A convenience sample of 291 children ages 2 weeks to 3 years was recruited from a clinic that serves predominantly low-income families in Columbus, Ohio.96 Maternal smoking was reported in 41% of the sample. The report of an in-home smoking ban was associated with decreased hair cotinine levels, although the child’s hair nicotine level did not differ regardless of whether the mother reported smoking outside only.

    A single-blind, randomized controlled clinical trial of cotinine feedback plus behavioral counseling to reduce tobacco smoke exposure among tobacco smoke–exposed children with asthma in Northern California found no overall intervention effect; however, the subgroup of children with high-risk asthma had lower urine cotinine levels on follow-up.112

    Public Policy Recommendations

    1. The FDA should regulate all tobacco products to protect the public health.

    Evidence quality: X

    Recommendation strength: strong recommendation

    The FDA is charged with the mission of protecting consumers and enhancing public health by maximizing compliance of FDA-regulated products and minimizing risks associated with those products. The FDA Center for Tobacco Products is responsible for carrying out the Family Smoking Prevention and Tobacco Control Act, passed in 2009 in an effort to protect the public and create a healthier future for all Americans.141 This act puts in place restrictions on marketing tobacco products to children and gives the FDA the authority to take action in the future to protect public health. Some of the agency’s responsibilities under the law include setting performance standards, reviewing premarketing applications for new and modified-risk tobacco products, and requiring new warning labels for tobacco products.142

    2. Tobacco control should be adequately funded.

    Evidence quality: A

    Recommendation strength: strong recommendation

    Tobacco dependence treatment should be available to tobacco-dependent individuals of all ages. Given the important benefits to society of reducing tobacco dependence, cost should not be a barrier to program participation and access to tobacco dependence treatment medications. The Best Practices for Comprehensive Tobacco Control Programs—2014 guidelines from the Centers for Disease Control and Prevention (CDC) should be implemented with funding at or near recommended levels.143 This funding will ensure that tobacco control programs are available to those who need them.

    The CDC’s Community Preventive Services Task Force evidence review found strong support for the effectiveness of comprehensive tobacco control programs in reducing tobacco use and secondhand smoke exposure, independent of increases in tobacco product prices or adoption of smoke-free policies.144 These programs reduce the prevalence of tobacco use among adults and young people, reduce tobacco product consumption, increase quitting, and contribute to reductions in tobacco-related diseases and deaths. Increases in program funding are associated with increases in program effectiveness, with the greatest impact seen if programs are funded at CDC-recommended levels.

    Tobacco control research should be considered a high priority and funded accordingly from both government and private sources. Tobacco industry funding should not be used for this purpose. The tobacco industry has a long history of using industry-funded programs to divert attention away from effective tobacco control programs and research, as well as misusing health care providers and academia to thwart attempts at tobacco control.14

    Recommendations for Public Policy to Protect Children From Tobacco Use Initiation

    3. Tobacco product advertising and promotion in forms that are accessible to children and youth should be prohibited.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Tobacco promotion is an important cause of initiation of tobacco use among youth. Major conclusions of the 2012 Report of the Surgeon General included, “Advertising and promotional activities by tobacco companies have been shown to cause the onset and continuation of smoking among adolescents and young adults.” The report also concluded, “In 2008, tobacco companies spent $9.94 billion on the marketing of cigarettes and $547 million on the marketing of smokeless tobacco. Spending on cigarette marketing is 48% higher than in 1998,” and “The evidence is suggestive but not sufficient to conclude that tobacco companies have changed the packaging and design of their products in ways that have increased these products’ appeal to adolescents and young adults.”4

    4. Point-of-sale tobacco product advertising and product placement that can be viewed by children should be prohibited.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Point-of-sale advertising increases tobacco initiation and tobacco product use among youth. Analysis of data from the 2011 NYTS found that youth who reported higher levels of exposure to advertisements in retail establishments had higher rates of trying snus and e-cigarettes (adjusted ORs: 3.33 [95% CI: 2.66–4.18] and 1.71 [95% CI: 1.21–2.41], respectively) comparing most of the time or always viewing retail tobacco advertisements versus never or rarely.145 Students in Tracy, California, were surveyed in grades 6 through 8 with follow-up surveys at approximately 12 months and 30 months later. Convenience stores, small markets, and liquor stores typically contain the most cigarette advertising. These cues were noticeable to students who had never smoked, with 82.1% of the sample reporting seeing cigarette advertisements in stores. The study found that a moderate (0.5–1.9 per week) and high (2.0–18 per week) number of visits to these stores were associated with increased rates of smoking initiation at 12 months (ORs: 1.64 [95% CI: 1.06–2.55] and 2.58 [95% CI: 1.68–3.97]) and at 30 months’ follow-up (ORs: 1.19 [95% CI: 1.00–1.41]) and 1.42 [95% CI: 1.19–1.69]).146 The results from the 2005–2006 California Student Tobacco Survey were matched to retailer licensing data about the location of tobacco outlets and with observations regarding the quantity of cigarette advertising in a random sample of those stores. The study found that the prevalence of current smoking was 3.2 percentage points higher at schools in neighborhoods with the highest tobacco outlet density (>5 outlets) than in neighborhoods with no tobacco outlets.147 Analyses of data from the 1999–2003 Monitoring the Future surveys matched to data on retail cigarette marketing found that higher levels of advertising, lower cigarette prices, and greater availability of cigarette promotions were associated with smoking uptake.148 A virtual store experiment found that youth aged 13 to 17 years were substantially less likely to try purchasing tobacco products when tobacco products were not displayed (OR: 0.30 [95% CI: 0.13−0.67]).149

    5. Depictions of tobacco products in movies and other media that can be viewed by youth should be restricted.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Depictions of smoking in the movies have been repeatedly shown to increase rates of smoking initiation among adolescents both in the United States and globally. The 2012 Report of the Surgeon General concluded, “The evidence is sufficient to conclude that there is a causal relationship between depictions of smoking in the movies and the initiation of smoking among young people.”4 In a prospective study of a nationally representative sample of US adolescents aged 10 to 14 years recruited in 2003 and followed up for 24 months, the adjusted hazard ratios for smoking initiation were 1.90 (95% CI: 1.47–2.45), 1.91 (95% CI: 1.49–2.44), and 2.02 (95% CI: 1.52–2.67) for views of tobacco use by negative, mixed, and positive movie characters, respectively.150 A school-based prospective follow-up study of 9987 children (mean ± SD age: 13.15 ± 1.10 years) in 6 European countries documented an adjusted incidence rate ratio for smoking initiation of 1.13 (95% CI: 1.08–1.17) for each additional 1000 occurrences of movie smoking exposure.151 A school-based cross-sectional survey of 4943 adolescents 12 to 16 years of age in New Delhi, India, found that the adjusted odds of ever having used tobacco among adolescents with high exposure to smoking in movies was 2.3 (95% CI: 1.3–3.9) compared with those with low exposure.152

    6. The promotion and sale of electronic nicotine delivery systems to youth should be prohibited.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Nicotine is a highly addictive substance. The 2010 Report of the Surgeon General produced a major conclusion: “Nicotine is the key chemical compound that causes and sustains the powerful addicting effects of commercial tobacco products.”20 Electronic nicotine delivery systems have the potential to serve as introductory products and to escalate levels of nicotine dependence among youth. An analysis of 2011 and 2012 NYTS data found that among cigarette experimenters (ever smoked a puff), a history of both ever and current e-cigarette use was associated with being a current cigarette smoker (ORs: 5.96 [95% CI: 5.67–6.27] and 7.88 [95% CI: 6.01–10.32], respectively).139

    7. Tobacco control programs should change the image of tobacco by telling the truth about tobacco.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Tobacco control programs that denormalize tobacco use by changing the image that youth have about tobacco and nicotine products can be effective. Tobacco industry–sponsored programs have been ineffective. The 2012 Report of the Surgeon General concluded, “The evidence is sufficient to conclude that school-based programs with evidence of effectiveness, containing specific components, can produce at least short-term effects and reduce the prevalence of tobacco use among school-aged youth,” and “The tobacco companies’ activities and programs for the prevention of youth smoking have not demonstrated an impact on the initiation or prevalence of smoking among young people.”4

    The Best Practices for Comprehensive Tobacco Control Programs—2014 of the CDC concluded the following: “Mass-reach health communication interventions can be powerful tools for preventing the initiation of tobacco use, promoting and facilitating cessation, and shaping social norms related to tobacco use.”143 The Community Preventive Services Task Force (2013) recommended mass-reach health communication interventions based on strong evidence of effectiveness in decreasing the prevalence of tobacco use, increasing cessation and use of available services such as quitlines, and decreasing the initiation of tobacco use among young people.153

    Adolescents and young adults are very sensitive to perceived social norms and media presentations of smoking behavior. Florida appropriated $23 million in fiscal year 1997 and $70 million in fiscal year 1998 to fund the Florida Tobacco Pilot Program to prevent and reduce tobacco use among Florida’s youth.154 The program’s major component was a youth-oriented, countermarketing media campaign developed to reduce the allure of smoking; the other program components comprise community partnerships in all 67 Florida counties, an education and training initiative, and enhanced enforcement of youth tobacco access laws. From 1998 to 1999, the prevalence of current cigarette use declined from 18.5% to 15.0% (P < .01) among middle school students and from 27.4% to 25.2% (P = .02) among high school students. Current cigar use declined among middle school students from 14.1% in 1998 to 11.9% in 1999 (P < .01). Smokeless tobacco use declined among middle school students from 6.9% in 1998 to 4.9% in 1999. This decline has been larger than any annual decline observed nationally among youth since 1980. In January 2008, Panama adopted Law No. 13, which intensified tobacco control measures by prohibiting pro-tobacco statements on cigarette packages; requiring complete prohibition of any form of pro-tobacco advertising, promotion, or sponsorship in all venues, including sports venues; prohibiting tobacco consumption in all enclosed work environments; and requiring the integration of content on the health consequences of tobacco consumption into the curricula of general education and basic secondary education.155 From 2002 to 2008, the prevalence of current cigarette smoking among students 13 to 15 years of age in Panama decreased from 13.2% to 4.3%.

    Pictorial health warnings improve adolescents’ awareness of the harms of smoking and decrease their perceptions of the social appeal of smoking. A survey of 4482 adolescents in Melbourne, Australia, found that adolescents had more accurate perceptions of the health risks of smoking if they had seen the graphic warnings.156 Focus groups of adolescents in Auckland, New Zealand, found that graphic warning labels clearly prompted a more severe judgement of smokers’ social appeal.157

    8. Tobacco product prices should be increased to reduce youth tobacco use initiation.

    Evidence quality: B

    Recommendation strength: strong recommendation

    The US Task Force on Community Preventive Services (2001) strongly recommended increasing the unit price for tobacco products to reduce smoking initiation and reduce consumption of tobacco products.158 Data from the Global Youth Tobacco Survey were matched to data on cigarette prices and estimated overall price elasticity (the relationship between demand for a product and its price) at –1.5; for low- and middle-income countries, however, price elasticity was greater at –2.2, suggesting that a 10% price increase would decrease youth smoking by 22%.159 Increasing the tobacco tax has the benefit of both raising the price and providing a source of funds that can be used for tobacco control programs.

    9. The minimum age to purchase tobacco should be increased to 21 years.

    Evidence quality: B

    Recommendation strength: strong recommendation

    In 2005, Needham, Massachusetts, raised the minimum age for the purchase of tobacco to 21 years; by 2010, the youth smoking rate had dropped by nearly one-half (12.9% to 6.7%), a change not observed in surrounding communities.160 The minimum age for the legal purchase of tobacco increased from 16 to 18 years in England, Scotland, and Wales on October 1, 2007. Data from the 2003–2008 Smoking, Drinking and Drug Use Among Young People in England, an annual survey of youths aged 11 to 15 years, reported a substantial reduction in regular smoking among 11- to 15-year-olds after the minimum age was increased (adjusted OR: 0.67 [95% CI: 0.55–0.81]).161 In Massachusetts cities in which compliance with underage purchase was enforced and verified, focus groups of adolescent smokers found that for high school–aged smokers, teenage store clerks and friends who are aged ≥18 years are major sources of attaining tobacco products.162 An analysis of Canada’s 1994–1995 National Population Health Survey found that most smokers began daily smoking in their teenage years, with 16% of 21- to 39-year-olds who had ever smoked daily reporting that they had started to do so at ≤13 years of age; 55% reported starting smoking at 14 to 17 years of age; and 15% reported starting smoking at 18 or 19 years of age. Only 14% had started daily smoking at ≥20 years of age. Younger age at starting smoking was associated with a lower probability of stopping smoking. For example, 18% of smokers who started at ≤13 years of age had stopped within 10 years, compared with 42% of those who started at ≥20 years of age.163

    Enforcement activities are important for age-of-purchase laws to be effective. A 2005 Cochrane review on interventions for preventing tobacco sales to minors found that active enforcement, including media coverage of that enforcement, was much more efficacious than educational programs alone.164 A 2011 comprehensive literature review found that enforcement programs which disrupted the sale of tobacco to minors reduced smoking among youth, whereas merely enacting a law without sufficient enforcement had minimal, if any, impact on youth tobacco use.165

    10. Flavoring agents, including menthol, should be prohibited in all tobacco products.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Flavoring agents increase the appeal of tobacco products to youth. Tobacco use initiation and progression to tobacco dependence are more common with the use of the flavored products. The only flavoring agent currently allowed in cigarettes is menthol, but small cigars and electronic nicotine delivery systems are not subject to the same regulations and are commonly flavored. Analysis of data from the National Youth Smoking Cessation Survey and the Assessing Hardcore Smoking Survey found that flavored cigarette use was more common among 17-year-olds than among older smokers (23% for 17-year-old smokers vs 9% for 22- to 26-year-old smokers).166 In the American Legacy Longitudinal Tobacco Use Reduction Study, middle school and high school students were surveyed in 3 waves from 2000 to 2003.167 Of those who initiated smoking, 39% used menthol cigarettes. This percentage was higher for African-American respondents (52%) compared with all other ethnic groups. Initiating smoking with menthol cigarettes was associated with greater risk of progression to established smoking (OR: 1.80 [95% CI: 1.02–3.16]). An analysis of data from the 2000 and 2002 NYTS found that menthol cigarettes were most popular among younger and newer smokers.168 Teenagers in middle school who had been smoking for less than 1 year were more likely to smoke menthol cigarettes than middle school students who had been smoking for more than 1 year (62.4% vs 53.3%; P = .002). Smokers of menthol cigarettes were significantly less likely to be “seriously thinking of quitting within the next 30 days” (adjusted OR: 0.79 [P = .012]). A survey of 1800 adolescents in New York City found that the likelihood of current smoking was greater for teenagers who had tried flavored tobacco products (OR: 2.70 [95% CI: 1.47–4.98]) or menthol cigarettes (OR: 15.16 [95% CI: 8.34–27.57]).169 A 2010–2011 survey of Canadian high school students found that 52% of these young tobacco consumers reported use of flavored tobacco products.170

    Recommendations to Protect Children From Tobacco Smoke and Nicotine Exposure

    11. Comprehensive smoking bans should be enacted.

    Evidence quality: B

    Recommendation strength: strong recommendation

    The 2006 Report of the Surgeon General concluded, “Workplace smoking restrictions are effective in reducing secondhand smoke exposure. Total bans on indoor smoking in hospitals, restaurants, bars, and offices substantially reduce secondhand smoke exposure. Exposures of non-smokers to secondhand smoke cannot be controlled by air cleaning or mechanical air exchange. Evidence from multiple peer-reviewed studies shows that smoke-free policies and regulations do not have an adverse economic impact on the hospitality industry.”25 Scotland implemented a comprehensive ban on smoking in public places in March 2006.57 Before the legislation was implemented, admissions for asthma in children younger than 15 years were increasing at a mean rate of 5.2% per year (95% CI: 3.9–6.6); after implementation, there was a mean reduction in the rate of admissions of 18.2% per year relative to the rate on March 26, 2006 (95% CI: 14.7–21.8). Enactment and enforcement of a law regarding smoke-free public places in Lexington-Fayette County, Kentucky, led to a reduction in emergency department visits for asthma for all ages, with an 18% decline for children and a 24% decline in adults.171 In Toronto, Ontario, Canada, asthma hospitalizations decreased after implementation of comprehensive smoke-free policies; no such changes were observed in nearby cities that did not implement similar policies, and they were also not observed for common gastrointestinal (ie, non–tobacco-related) conditions.172

    Smoking in motor vehicles leads to substantial tobacco smoke exposure for nonsmokers. Laws prohibiting smoking in vehicles with children present improved reports of not allowing smoking inside of the car. Studies of tobacco smoking in automobiles found that although open windows in a moving vehicle substantially improve air exchange rates, a significant amount of tobacco smoke (measured as particulate mass concentrations) remains in the vehicle even with the windows open.173 A study of nonsmokers sitting in a parked car with an open window while a smoker smoked 3 cigarettes over 1 hour found substantially elevated levels of cotinine as well as other tobacco-related toxicants and carcinogens after the in-vehicle tobacco smoke exposure.174,175 The levels of polycyclic aromatic hydrocarbons in the vehicles after smoking 3 cigarettes with a window open was higher than measurements made in highly polluted urban traffic or in bars and restaurants where smoking was allowed.176 After passage of legislation in Maine prohibiting smoking in vehicles when a child is present, smoke-free car policies among households with children and a current smoker increased substantially, from 19.2% to 30.7% (Behavioral Risk Factor Surveillance System data).177

    12. Smoking in multi-unit housing should be prohibited.

    Evidence quality: B

    Recommendation strength: strong recommendation

    Smoking in multi-unit housing exposes children to tobacco smoke. A survey of Minnesota residents of common interest communities (eg, condominiums, cooperatives) found that 28% of households reported secondhand smoke incursion into their unit in the preceding 6 months; 59% of those surveyed said this infiltration bothered them a lot.178 A survey of 5936 residents of multi-unit housing who participated in the New York State Adult Tobacco Survey between May 2007 and May 2009 found that among respondents with a smoke-free home policy, 46.2% reported secondhand smoke entering their home in the past year, with 9.2% reporting daily incursions.179 Passive nicotine monitors in low-income multi-unit residences in the greater Boston area detected nicotine in 89% (17 of 19) of nonsmoking homes, indicating secondhand tobacco smoke infiltration.180 An analysis of data from the 2001–2006 NHANES results for children who lived in nonsmoking residences found greater mean cotinine concentrations among children who lived in apartments than those who lived in attached or detached houses (0.075, 0.053, and 0.031 ng/mL, respectively; P < .01).181

    13. Prohibitions on smoking and use of tobacco products should include prohibitions on use of electronic nicotine delivery systems.

    Evidence quality: B

    Recommendation strength: strong recommendation

    The aerosol emitted from electronic nicotine delivery systems contains toxic and carcinogenic substances in addition to nicotine. A laboratory study of indoor air quality after indoor e-cigarette use by volunteers found substantial amounts of 1,2-propanediol, glycerin, and nicotine as well as high concentrations of particulate matter less than 2.5 micrometers in diameter (PM2.5) (mean: 197 μg/m3). Polyaromatic hydrocarbons (carcinogens) increased by 20% to 147 ng/m3.182 An analysis of the content of the aerosol from e-cigarettes found toxic, irritating, and carcinogenic substances, including formaldehyde, acetaldehyde, and acrolein; volatile organic compounds such as toluene and m,p-xylene; tobacco-specific nitrosamines; and the heavy metals cadmium, nickel, and lead.134

    Recommendations to Protect Children From Acute Nicotine Poisoning

    14. Children younger than 18 years should be legally prohibited from working on tobacco farms and in tobacco production.

    Evidence quality: C

    Recommendation strength: recommendation

    Green tobacco sickness is a well-described entity. Dermal absorption of nicotine from moist tobacco plants can lead to symptoms of severe nicotine poisoning, including weakness, headache, nausea, vomiting, dizziness, abdominal cramps, breathing difficulty, pallor, diarrhea, chills, fluctuations in blood pressure or heart rate, seizures, and increased perspiration and excessive salivation.183 A report using Kentucky Regional Poison Control Center records in 1991 found 104 cases of green tobacco sickness; one-quarter of the reports were in children under 17 years.184 A case series of 5 children with green tobacco sickness reported symptoms that included a seizure, bradycardia, vomiting, dizziness, headache, pallor, and/or muscle weakness. One child stated that it made him feel, “like I was going to die.”32

    15. Concentrated nicotine solution for electronic nicotine delivery systems should be sold in child-resistant containers with amounts limited to that which would not be lethal to a young child if ingested.

    Evidence quality: B

    Recommendation strength: strong recommendation

    The colorful fruit- and candy-flavored concentrated nicotine solutions for use in electronic nicotine delivery systems can appeal to young children. The oral lethal dose of nicotine by body weight that is estimated to kill 50% of adults is projected to be between 0.8 and 13 mg/kg.185 Severe nicotine toxicity in children has been reported with doses of nicotine as low as 2 mg. Calls to poison control centers for exposures to electronic nicotine delivery systems (with the majority occurring in children under 5 years) increased from 1 exposure call per month in September 2010 to 215 calls in February 2014.186 The concentrated nicotine solution used in these devices is a poisoning risk for young children, and at least 1 child has already died of its accidental ingestion.187

    Conclusions

    Tobacco dependence starts in childhood. The tobacco epidemic takes a substantial toll on children’s health, and the harm starts in utero. There is strong evidence to support actions by pediatricians and public policy decision makers that can reduce the development of tobacco dependence and reduce the tobacco smoke exposure of children.

    Lead Author

    Harold J. Farber, MD, MSPH, FAAP

    Judith A. Groner, MD, FAAP

    Contributing Authors

    Kevin E. Nelson, MD, PhD, FAAP

    Susan C. Walley, MD, FAAP

    Section on Tobacco Control, 2015–2016

    Ruth A. Etzel, MD, PhD, FAAP, Co-chairperson

    Karen M. Wilson, MD, MPH, FAAP, Co-chairperson

    Harold J. Farber, MD, MSPH, FAAP, Policy Chairperson

    Sophie J. Balk, MD, FAAP

    Judith A. Groner, MD, FAAP

    John E. Moore, MD, FAAP

    Staff

    Janet Brishke, MPH

    Regina Shaefer, MPH

    Footnotes

    • This document is copyrighted and is property of the American Academy of Pediatrics and its Board of Directors. All authors have filed conflict of interest statements with the American Academy of Pediatrics. Any conflicts have been resolved through a process approved by the Board of Directors. The American Academy of Pediatrics has neither solicited nor accepted any commercial involvement in the development of the content of this publication.

    • Technical reports from the American Academy of Pediatrics benefit from expertise and resources of liaisons and internal (American Academy of Pediatrics) and external reviewers. However, technical reports from the American Academy of Pediatrics may not reflect the views of the liaisons or the organizations or government agencies that they represent.

    • The guidance in this report does not indicate an exclusive course of treatment or serve as a standard of medical care. Variations, taking into account individual circumstances, may be appropriate.

    • All technical reports from the American Academy of Pediatrics automatically expire 5 years after publication unless reaffirmed, revised, or retired at or before that time.

    References