TY - JOUR
AU - O’Loughlin,, Jennifer
AB - Abstract Introduction We investigated whether secondhand smoke (SHS) exposure is associated with depressive symptoms in a population-based sample of children. Methods Never-smoking students from 29 French-language elementary schools in Greater Montréal, Canada, were followed from 5th to 11th grade (2005–2011) in five waves: 1 (5th grade), 2 (spring 6th grade), 3 (7th grade), 4 (9th grade), and 5 (11th grade). Associations between depressive symptoms and SHS exposure at home and in cars were examined in cross-sectional and longitudinal gamma generalized regression models adjusted for sex, maternal education, and neighborhood socioeconomic status. Results The sample comprised 1553 baseline never-smokers (mean [SD] age = 10.7 [0.5] years; 44% male; 89% French-speaking). SHS exposure at home and in cars was associated with higher depressive symptom scores in cross-sectional analyses pooled across grades and adjusted for demographics (B [95% confidence interval (CI)] = 0.041 [0.017 to 0.068] for home exposure; 0.057 [0.030 to 0.084] for car exposure). In longitudinal analyses from fifth to sixth grade, B (95% CI), adjusted for demographics and baseline depressive symptoms, was 0.042 (0.003 to 0.080) for home exposure and 0.061 (0.019 to 0.103) for car exposure. From sixth to seventh grade, B (95% CI) was 0.057 (0.003 to 0.110) for home exposure and 0.074 (0.015 to 0.133) for car exposure. SHS exposure at any age did not predict depressive symptoms 2 years later. Conclusions SHS exposure is associated with depressive symptoms in young persons, both concurrently and 1 year later. This finding adds to the evidence base supporting that children should be protected from SHS exposure. Implications SHS exposure has deleterious effects on physical health and results of this study raise concerns that such exposure might also affect the mental health of young persons. Clearly, protecting children from SHS in all locations is a critical public health priority. Although research is needed to determine if the association between SHS exposure and depressive symptoms is causal, continued implementation of smoking bans and educational efforts to discourage smoking in vehicles when children are present are warranted. Introduction In 2006, the US Surgeon General concluded that there is no risk-free level of secondhand smoke (SHS) exposure and that only complete elimination of indoor smoking assures no household exposure.1 Although SHS exposure at home and in cars declined among youth between 2000 and 2009,2 80% of Canadian youth aged 12–19 years reported some SHS exposure in the past month in 2015.3 Exposure is inversely related to socioeconomic status (SES); the proportion of children aged 12–17 years exposed daily or almost daily to SHS in homes and cars is much higher among those living in neighborhoods of lower compared to higher SES (33.5% vs. 15.9% in homes; 27.1% vs. 15.5% in cars).4 Significant cumulative SHS exposure is indicated by levels of blood cotinine, a byproduct of nicotine metabolism, between 0.05 and 10 ng/mL.5 In 2011 in the United States, 40.6% of nonsmoking children aged 3–11 years and 33.8% of nonsmoking adolescents aged 12–19 years had levels in this range.5 In Canada in 2008–2009, nonsmoking children (aged 6–11 years) and adolescents (aged 12–19 years) exposed to SHS most days or every day had urinary cotinine levels twice to almost three times as high as SHS-exposed nonsmoking adults.6 In addition to risks to physical health (eg, childhood asthma and reduced lung function, a 20%–30% increase in lifetime risk of lung cancer, a 25%–30% increase in risk of coronary heart disease in adulthood),1 recent research suggests that SHS exposure may also be associated with mental health. Cross-sectional data from a population-based sample of Canadian adults indicated that adults exposed to SHS were 40% more likely than those not exposed to be diagnosed with a major depressive episode, with prevalence rates of 6.1% versus 4.0%, respectively.7 In contrast, two studies in the Netherlands found no relationship between urinary cotinine levels and symptoms of either depression or anxiety among nonsmoking adults.8 Findings in children are also contradictory. SHS exposure and Beck Depression Inventory scores were positively correlated among Korean male, but not female high school freshmen.9 SHS exposure was associated with internalizing symptoms (eg, depression) and behaviors in boys, but not girls with asthma,10 but was unrelated to “emotional symptoms” (assessed by the Strengths and Difficulties Questionnaire) in a population-based sample of Spanish children aged 4–12 years.11 Finally, Bandiera et al.12 reported cross-sectional associations between biologically verified SHS exposure and DSM-IV symptoms of major depressive disorder, attention deficit hyperactivity disorder, and generalized anxiety disorder in children aged 8–15 years. The association with depression was statistically significant among all participants and was greater among boys than girls.12 A limitation of this study, however, was that intermittent smokers could have been included in the sample because, if they had not smoked within 36 hours of the measurement of cotinine, their blood levels would have been indistinguishable from those of nonsmokers. To the best of our knowledge, there are no longitudinal studies investigating the association between SHS and depressive symptoms in children or adolescents. Our objective in this current study was to examine the associations between SHS exposure at home and in cars and reports of depressive symptoms in a population-based sample of children. We studied the association in both cross-sectional and longitudinal analyses. Methods Participant Recruitment and Data Collection Participants were drawn from the AdoQuest I Study. Children (n = 1801) were recruited from 29 French language elementary schools in greater Montréal with more than 90 fifth-grade students and were followed for up to 6 years, from 5th to 11th grades (2005–2011).13 Representation of students of high, middle, and low SES was balanced by stratifying eligible schools into groupings defined by tertile of a school deprivation indicator incorporating parental employment, a measure of family income that accounts for family size and area of residence, and maternal education. An equal number of schools was randomly selected in each grouping; 10 schools in the first, 10 in the second, and nine in the third grouping (72.5% of schools invited) agreed to participate. All students in all fifth-grade classes in participating schools were eligible for recruitment. Participants provided assent and their parents and/or guardians provided informed consent. The study received approval from the ethics review boards of Concordia University and the Centre de Recherche du Centre Hospitalier de l’Université de Montréal. Characteristics of the “AdoQuest I” sample were comparable to those of two provincially representative samples of similarly aged Québec youth.14,15 AdoQuest data were collected in classroom-administered self-report questionnaires in spring 2005 (fifth grade) and in fall 2005 and spring 2006 (sixth grade). Follow-up in 2006–2007 (seventh grade), 2008–2009 (ninth grade), and 2010–2011 (11th grade) used self-report questionnaires mailed to participants’ homes. Because depressive symptoms were not measured in fall 2005 (sixth grade), we drew data for the current study on sixth-grade SHS exposure and depressive symptoms from the spring cycle only. Thus, the data collection waves included in this analysis were 1 (5th grade), 2 (spring 6th grade), 3 (7th grade), 4 (9th grade), and 5 (11th grade). Parents provided data via mailed self-report questionnaires when their offspring were in fifth to sixth and ninth grades. One parent, typically the mother, completed the survey and, where appropriate, answered questions about the other parent (eg, level of education). To assure inclusion of only never users of tobacco in our analyses, we examined responses to four cigarette smoking items (ie, ever smoked, recall of number of days on which participant had smoked in each of the past 3 months, usual number of cigarettes smoked per day, and number of cigarettes smoked in the previous week) and one item assessing ever use of other tobacco products (ie, cigars, pipes, bidi, chewing tobacco, and powdered sniff tobacco) at each data collection wave. If use of any tobacco product had been reported in a previous cycle or if use was reported on any of these items in the current cycle, the participant was excluded. Study Variables Depressive symptoms were measured with the Depressive Symptoms Scale,16 which includes six items: “During the past three months, how often have you (i) felt too tired to do things?; (ii) had trouble going to sleep or staying asleep?; (iii) felt unhappy, sad, or depressed?; (iv) felt hopeless about the future?; (v) felt nervous or tense?; (vi) worried too much about things?” Response options were never, rarely, sometimes, and often. Responses to the six items were averaged, yielding a scale ranging from 1 to 4. The scale has adequate psychometric properties,16 including invariance across sex and time.17 Four bilingual public health professionals translated the scale into French and then back-translated to English; discrepancies were resolved by consensus. SHS exposure at home was assessed with “Excluding yourself, how many people smoke inside your home every day or almost every day?” with response options None, 1 person, 2 persons, 3 persons, 4 persons, and 5 or more persons. We collapsed responses into none versus any persons. SHS exposure in cars was assessed with “During the last seven days, how many times did you get into a car with someone who smoked during the trip?” with response options 0, 1–2, 3–4, 5–6, and 7 days. We collapsed the responses into 0 versus any days. To assess reliability of participants’ reports of exposure at home and in cars, we compared them to parental responses to questionnaires18 administered when participants were in sixth to seventh grade and again in ninth grade. Agreement between parental and participant reports was high: 94%/95% that nobody smoked at home and 88%/86% that nobody smoked in the car in sixth and ninth grades, respectively. However, agreement on SHS exposure was lower: 64%/76% that anyone smoked in the home, and 57%/65% that anyone smoked in the car in sixth and ninth grades, respectively. Lack of agreement about home exposure likely relates to offspring not living or living part time with the parent who responded. Lack of agreement about car exposure possibly relates to participants riding in cars with smokers other than their parents. Covariates for the multivariable analyses were selected based on demonstrated associations with SHS exposure and/or depressive symptoms,16,19 and included sex, mother’s education, and neighborhood SES. Data on maternal education were drawn from the parents’ questionnaires and categorized as high–moderate (postsecondary education) versus low (high school or less). Neighborhood SES was assessed with the Pampalon index of material deprivation, which accounts for the ratio of the number of persons employed to the total population, the average personal income of persons in the enumeration area, and the proportion of persons with no high school diploma in a dissemination area.20 For analysis we created a binary variable in which the lowest and middle tertile groupings of deprivation represented high–moderate SES and the highest tertile represented low SES. Participants were also classified as susceptible or not susceptible to smoking based on responses to three questions regarding smoking intentions;21 translation of these questions paralleled the process for the Depressive Symptoms Scale. Statistical Analyses Analyses were undertaken in Stata version 14.2 (September 2016; Stata Corp., College Station, TX). We examined patterns and predictors of missing data (Supplementary Table 1). At wave 1, the proportion of missing values ranged from 0.1% for sex to 23.4% for maternal education (median [interquartile range ]) = 3.6% (2.3, 5.2)). Corresponding median (interquartile range) proportions of missing data were 1.8% (0.1, 4.4) at wave 2, 1.5% (0.1, 3.8) at wave 3, 1.2% (1.0, 3.2) at wave 4, and 1.7% (1.5, 3.6) at wave 5. Missing values were imputed via chained equations.22,23 The distribution of the Depressive Symptoms Scale was positively skewed and values were bounded by 1 and 4, so imputation was performed using predictive mean matching with 10 nearest neighbor donors.24 Because 23.4% of values of maternal education were missing, we created 25 imputed datasets.25 Participants with any data in a specific wave were included in analyses of that wave, with missing values imputed, whereas participants missing entire waves were omitted from analyses of those waves. For example, if a participant had data in wave 3, her information was included in analyses of wave 3; if she had no data in wave 3, her information was omitted from analyses of wave 3. To assess selection bias attributable to attrition, which could affect estimates in the longitudinal analyses and compromise external generalizability of findings,26 we compared never-smokers retained throughout follow-up to those lost to follow-up using χ2 and Wilcoxon rank sum tests. Scatter plots revealed that the relationship between SHS exposure and depressive symptoms was linear in all grades. Residuals from test linear regression models were skewed and homoskedastic with only the exposure variable in the models; they became heteroskedastic (ie, the variance of errors was not constant over observations) when covariates were added. We tested three generalized linear models,27 normal, log-normal, and gamma (with log link). Deviance residuals from the gamma model were approximately normally distributed, indicating good model fit.28 Gamma models are useful in analyzing skewed and heteroskedastic data,28 especially when paired with the sandwich (robust HC1) variance estimator to account for heteroskedasticity,29,30 which can bias estimates of standard errors and confidence intervals (CIs). We estimated the cross-sectional association between SHS exposure and depressive symptoms in gamma regression analyses, pooled over waves 1–4, within a generalized estimating equation framework with an unstructured working correlation matrix to account for repeated measurements within individuals.31,32 Wave 5 could not be included in the cross-sectional analyses because the wording of the items measuring SHS exposure was substantially different from that used in other waves. We estimated the association in both unadjusted and adjusted (for maternal education, neighborhood SES, and sex) models. Longitudinal associations were estimated in a series of gamma generalized regression models using the robust variance estimator. We examined 1- and 2-year intervals between SHS exposures and depressive symptoms: (1) SHS exposures in fifth grade predicting depressive symptoms in sixth grade; (2) SHS exposures in sixth grade predicting symptoms in seventh grade; (3) SHS exposures in fifth grade predicting symptoms in seventh grade; (4) SHS exposures in seventh grade predicting symptoms in ninth grade; and (5) SHS exposures in ninth grade predicting symptoms in 11th grade. In each dataset, we estimated unadjusted and adjusted (for demographics) models as well as models adjusted for demographics and baseline depression (at wave 1). Because others9,10,12 have found sex differences in the relationship between SHS exposure and depression, we explored potential differences by conducting each of the aforementioned analyses again including an additional sex by SHS interaction term. Results The analytic sample comprised 1553 never-smokers in wave 1. Figure 1 illustrates retention of participants across the five data collection waves. The AdoQuest sample was approximately 90% Caucasian. Table 1 presents selected characteristics of the sample at each wave. The proportion of participants susceptible to smoking decreased from 36% at wave 1 to 28% at wave 5, as more susceptible participants initiated smoking and became ineligible for inclusion in the analytic database. Susceptible participants reported more depressive symptoms at baseline (M [95% CI] = 2.06 [2.01 to 2.12]) than did non-susceptible participants (M [95% CI] = 1.80 [1.76 to 1.84]; p < .001). Four hundred participants initiated smoking prior to 11th grade. Among the 1153 who did not initiate, 432 (37.5%) provided data in all five waves, 268 (23.2%) in four waves, 204 (17.7%) in three waves, 197 (17.1%) in two waves, and 52 (4.5%) only at wave 1. There were no differences in baseline variables between the 52 participants lost to follow-up after wave 1 and the 1153 never-smokers retained. Figure 1. Open in new tabDownload slide Participant retention, AdoQuest, 2005–2011. Figure 1. Open in new tabDownload slide Participant retention, AdoQuest, 2005–2011. Table 1. Characteristics of Nonsmokers Retained in the Grade 5, 6, 7, 9, and 11 Analytic Samples . Grade/wave . 5/1 (n = 1553) . 6/2 (n = 1399) . 7/3 (n = 836) . 9/4 (n = 965) . 11/5 (n = 765) . na . % . na . % . na . % . na . % . na . % . Male 684 44.1 612 43.8 362 43.3 403 41.8 350 45.8 Speak French at home 1341 89.1 1220 89.8 737 90.0 841 89.7 662 88.6 Maternal education ≥ Bachelor’s 354 29.8 323 28.9 240 32.2 293 32.5 229 34.0 Neighborhood SES high 664 45.2 603 44.8 393 49.0 455 48.9 334 45.6 Father smokes 364 23.5 328 23.5 182 21.8 184 19.1 131 17.1 Mother smokes 314 20.3 284 20.3 139 16.7 145 15.0 118 15.4 Complete home smoking ban 681 47.3 602 46.3 361 46.9 451 50.6 359 50.9 ≥1 smoker at home 512 33.6 405 31.8 199 25.1 128 18.5 b — ≥1 day/wk exposed to smoking in car 353 23.4 329 25.8 140 17.6 147 21.3 b — Susceptible to smokingc 552 36.0 503 36.4 274 33.1 309 32.3 209 27.6 Depressive symptoms, Mdn (IQR) 1.83 (1.33, 2.33) 1.83 (1.33, 2.33) 1.67 (1.17, 2.17) 1.83 (1.33, 2.50) 2.00 (1.50, 2.67) . Grade/wave . 5/1 (n = 1553) . 6/2 (n = 1399) . 7/3 (n = 836) . 9/4 (n = 965) . 11/5 (n = 765) . na . % . na . % . na . % . na . % . na . % . Male 684 44.1 612 43.8 362 43.3 403 41.8 350 45.8 Speak French at home 1341 89.1 1220 89.8 737 90.0 841 89.7 662 88.6 Maternal education ≥ Bachelor’s 354 29.8 323 28.9 240 32.2 293 32.5 229 34.0 Neighborhood SES high 664 45.2 603 44.8 393 49.0 455 48.9 334 45.6 Father smokes 364 23.5 328 23.5 182 21.8 184 19.1 131 17.1 Mother smokes 314 20.3 284 20.3 139 16.7 145 15.0 118 15.4 Complete home smoking ban 681 47.3 602 46.3 361 46.9 451 50.6 359 50.9 ≥1 smoker at home 512 33.6 405 31.8 199 25.1 128 18.5 b — ≥1 day/wk exposed to smoking in car 353 23.4 329 25.8 140 17.6 147 21.3 b — Susceptible to smokingc 552 36.0 503 36.4 274 33.1 309 32.3 209 27.6 Depressive symptoms, Mdn (IQR) 1.83 (1.33, 2.33) 1.83 (1.33, 2.33) 1.67 (1.17, 2.17) 1.83 (1.33, 2.50) 2.00 (1.50, 2.67) IQR = interquartile range; SES = socioeconomic status. ans differ because of missing data. bNot measured. cAssessed at wave 1 using criteria established by Choi et al.21 Open in new tab Table 1. Characteristics of Nonsmokers Retained in the Grade 5, 6, 7, 9, and 11 Analytic Samples . Grade/wave . 5/1 (n = 1553) . 6/2 (n = 1399) . 7/3 (n = 836) . 9/4 (n = 965) . 11/5 (n = 765) . na . % . na . % . na . % . na . % . na . % . Male 684 44.1 612 43.8 362 43.3 403 41.8 350 45.8 Speak French at home 1341 89.1 1220 89.8 737 90.0 841 89.7 662 88.6 Maternal education ≥ Bachelor’s 354 29.8 323 28.9 240 32.2 293 32.5 229 34.0 Neighborhood SES high 664 45.2 603 44.8 393 49.0 455 48.9 334 45.6 Father smokes 364 23.5 328 23.5 182 21.8 184 19.1 131 17.1 Mother smokes 314 20.3 284 20.3 139 16.7 145 15.0 118 15.4 Complete home smoking ban 681 47.3 602 46.3 361 46.9 451 50.6 359 50.9 ≥1 smoker at home 512 33.6 405 31.8 199 25.1 128 18.5 b — ≥1 day/wk exposed to smoking in car 353 23.4 329 25.8 140 17.6 147 21.3 b — Susceptible to smokingc 552 36.0 503 36.4 274 33.1 309 32.3 209 27.6 Depressive symptoms, Mdn (IQR) 1.83 (1.33, 2.33) 1.83 (1.33, 2.33) 1.67 (1.17, 2.17) 1.83 (1.33, 2.50) 2.00 (1.50, 2.67) . Grade/wave . 5/1 (n = 1553) . 6/2 (n = 1399) . 7/3 (n = 836) . 9/4 (n = 965) . 11/5 (n = 765) . na . % . na . % . na . % . na . % . na . % . Male 684 44.1 612 43.8 362 43.3 403 41.8 350 45.8 Speak French at home 1341 89.1 1220 89.8 737 90.0 841 89.7 662 88.6 Maternal education ≥ Bachelor’s 354 29.8 323 28.9 240 32.2 293 32.5 229 34.0 Neighborhood SES high 664 45.2 603 44.8 393 49.0 455 48.9 334 45.6 Father smokes 364 23.5 328 23.5 182 21.8 184 19.1 131 17.1 Mother smokes 314 20.3 284 20.3 139 16.7 145 15.0 118 15.4 Complete home smoking ban 681 47.3 602 46.3 361 46.9 451 50.6 359 50.9 ≥1 smoker at home 512 33.6 405 31.8 199 25.1 128 18.5 b — ≥1 day/wk exposed to smoking in car 353 23.4 329 25.8 140 17.6 147 21.3 b — Susceptible to smokingc 552 36.0 503 36.4 274 33.1 309 32.3 209 27.6 Depressive symptoms, Mdn (IQR) 1.83 (1.33, 2.33) 1.83 (1.33, 2.33) 1.67 (1.17, 2.17) 1.83 (1.33, 2.50) 2.00 (1.50, 2.67) IQR = interquartile range; SES = socioeconomic status. ans differ because of missing data. bNot measured. cAssessed at wave 1 using criteria established by Choi et al.21 Open in new tab In cross-sectional analyses, after adjusting for demographics, exposure to SHS at home and in cars was associated with reporting more depressive symptoms (Table 2). Table 2. Cross-sectional Association Between Secondhand Tobacco Smoke Exposure and Depressive Symptoms (n = 4292 Observations Pooled Across Four Waves) Exposure . Depressive symptoms . Unadjustedb . Adjustedc . Mdn (IQR)a B (95% CI) B (95% CI) Secondhand smoke at home No 1.67 (1.33, 2.33) 0.041*** (0.017 to 0.070) 0.041** (0.014 to 0.068) Yes 1.83 (1.33, 2.40) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.063*** (0.036 to 0.089) 0.057*** (0.030 to 0.084) Yes 2.00 (1.50, 2.50) Exposure . Depressive symptoms . Unadjustedb . Adjustedc . Mdn (IQR)a B (95% CI) B (95% CI) Secondhand smoke at home No 1.67 (1.33, 2.33) 0.041*** (0.017 to 0.070) 0.041** (0.014 to 0.068) Yes 1.83 (1.33, 2.40) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.063*** (0.036 to 0.089) 0.057*** (0.030 to 0.084) Yes 2.00 (1.50, 2.50) IQR = interquartile range. aDescriptive statistics were calculated using non-imputed data. bCoefficients are unstandardized and represent the log of the mean response. cModels were adjusted for sex, maternal education, and neighborhood socioeconomic status. **p < .01; ***p ≤ .001. Open in new tab Table 2. Cross-sectional Association Between Secondhand Tobacco Smoke Exposure and Depressive Symptoms (n = 4292 Observations Pooled Across Four Waves) Exposure . Depressive symptoms . Unadjustedb . Adjustedc . Mdn (IQR)a B (95% CI) B (95% CI) Secondhand smoke at home No 1.67 (1.33, 2.33) 0.041*** (0.017 to 0.070) 0.041** (0.014 to 0.068) Yes 1.83 (1.33, 2.40) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.063*** (0.036 to 0.089) 0.057*** (0.030 to 0.084) Yes 2.00 (1.50, 2.50) Exposure . Depressive symptoms . Unadjustedb . Adjustedc . Mdn (IQR)a B (95% CI) B (95% CI) Secondhand smoke at home No 1.67 (1.33, 2.33) 0.041*** (0.017 to 0.070) 0.041** (0.014 to 0.068) Yes 1.83 (1.33, 2.40) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.063*** (0.036 to 0.089) 0.057*** (0.030 to 0.084) Yes 2.00 (1.50, 2.50) IQR = interquartile range. aDescriptive statistics were calculated using non-imputed data. bCoefficients are unstandardized and represent the log of the mean response. cModels were adjusted for sex, maternal education, and neighborhood socioeconomic status. **p < .01; ***p ≤ .001. Open in new tab In 1-year longitudinal analyses, after adjusting for demographics and baseline depression, exposure to SHS at home and in cars in fifth grade was associated with depressive symptoms in sixth grade, and SHS exposure at home and in cars in sixth grade was associated with depressive symptoms in seventh grade (Table 3). Table 3. Association Between Secondhand Tobacco Smoke Exposure in Fifth or Sixth Grade and Depressive Symptoms 1 Year Later Exposure . Depressive symptoms Mdn (IQR)a . Model 1bBc (95% CI) . Model 2 B (95% CI) . Model 3 B (95% CI) . Exposure in fifth grade predicting symptoms in sixth grade (n = 1250) Secondhand smoke at home No 1.67 (1.33, 2.17) 0.058** (0.017 to 0.099) 0.052* (0.010 to 0.094) 0.042* (0.003 to 0.080) Yes 1.83 (1.50, 2.33) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.084*** (0.039 to 0.128) 0.076*** (0.030 to 0.121) 0.061** (0.019 to 0.103) Yes 2.00 (1.50, 2.33) Exposure in sixth grade predicting symptoms in seventh grade (n = 798) Secondhand smoke at home No 1.67 (1.17, 2.17) 0.067* (0.009 to 0.124) 0.065* (0.007 to 0.122) 0.057* (0.003 to 0.110) Yes 1.83 (1.33, 2.33) Secondhand smoke in cars No 1.67 (1.17, 2.00) 0.108*** (0.045 to 0.171) 0.099** (0.036 to 0.162) 0.074* (0.015 to 0.133) Yes 1.83 (1.33, 2.33) Exposure . Depressive symptoms Mdn (IQR)a . Model 1bBc (95% CI) . Model 2 B (95% CI) . Model 3 B (95% CI) . Exposure in fifth grade predicting symptoms in sixth grade (n = 1250) Secondhand smoke at home No 1.67 (1.33, 2.17) 0.058** (0.017 to 0.099) 0.052* (0.010 to 0.094) 0.042* (0.003 to 0.080) Yes 1.83 (1.50, 2.33) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.084*** (0.039 to 0.128) 0.076*** (0.030 to 0.121) 0.061** (0.019 to 0.103) Yes 2.00 (1.50, 2.33) Exposure in sixth grade predicting symptoms in seventh grade (n = 798) Secondhand smoke at home No 1.67 (1.17, 2.17) 0.067* (0.009 to 0.124) 0.065* (0.007 to 0.122) 0.057* (0.003 to 0.110) Yes 1.83 (1.33, 2.33) Secondhand smoke in cars No 1.67 (1.17, 2.00) 0.108*** (0.045 to 0.171) 0.099** (0.036 to 0.162) 0.074* (0.015 to 0.133) Yes 1.83 (1.33, 2.33) aDescriptive statistics were calculated using non-imputed data. bThree models were estimated for each exposure. Model 1 is unadjusted. Model 2 is adjusted for sex, maternal education, and neighborhood socioeconomic status (SES). Model 3 is adjusted for sex, maternal education, neighborhood SES, and baseline depression. cCoefficients are unstandardized and represent the log of the mean response. *p < .05; **p ≤ .01; ***p ≤ .001. Open in new tab Table 3. Association Between Secondhand Tobacco Smoke Exposure in Fifth or Sixth Grade and Depressive Symptoms 1 Year Later Exposure . Depressive symptoms Mdn (IQR)a . Model 1bBc (95% CI) . Model 2 B (95% CI) . Model 3 B (95% CI) . Exposure in fifth grade predicting symptoms in sixth grade (n = 1250) Secondhand smoke at home No 1.67 (1.33, 2.17) 0.058** (0.017 to 0.099) 0.052* (0.010 to 0.094) 0.042* (0.003 to 0.080) Yes 1.83 (1.50, 2.33) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.084*** (0.039 to 0.128) 0.076*** (0.030 to 0.121) 0.061** (0.019 to 0.103) Yes 2.00 (1.50, 2.33) Exposure in sixth grade predicting symptoms in seventh grade (n = 798) Secondhand smoke at home No 1.67 (1.17, 2.17) 0.067* (0.009 to 0.124) 0.065* (0.007 to 0.122) 0.057* (0.003 to 0.110) Yes 1.83 (1.33, 2.33) Secondhand smoke in cars No 1.67 (1.17, 2.00) 0.108*** (0.045 to 0.171) 0.099** (0.036 to 0.162) 0.074* (0.015 to 0.133) Yes 1.83 (1.33, 2.33) Exposure . Depressive symptoms Mdn (IQR)a . Model 1bBc (95% CI) . Model 2 B (95% CI) . Model 3 B (95% CI) . Exposure in fifth grade predicting symptoms in sixth grade (n = 1250) Secondhand smoke at home No 1.67 (1.33, 2.17) 0.058** (0.017 to 0.099) 0.052* (0.010 to 0.094) 0.042* (0.003 to 0.080) Yes 1.83 (1.50, 2.33) Secondhand smoke in cars No 1.67 (1.33, 2.33) 0.084*** (0.039 to 0.128) 0.076*** (0.030 to 0.121) 0.061** (0.019 to 0.103) Yes 2.00 (1.50, 2.33) Exposure in sixth grade predicting symptoms in seventh grade (n = 798) Secondhand smoke at home No 1.67 (1.17, 2.17) 0.067* (0.009 to 0.124) 0.065* (0.007 to 0.122) 0.057* (0.003 to 0.110) Yes 1.83 (1.33, 2.33) Secondhand smoke in cars No 1.67 (1.17, 2.00) 0.108*** (0.045 to 0.171) 0.099** (0.036 to 0.162) 0.074* (0.015 to 0.133) Yes 1.83 (1.33, 2.33) aDescriptive statistics were calculated using non-imputed data. bThree models were estimated for each exposure. Model 1 is unadjusted. Model 2 is adjusted for sex, maternal education, and neighborhood socioeconomic status (SES). Model 3 is adjusted for sex, maternal education, neighborhood SES, and baseline depression. cCoefficients are unstandardized and represent the log of the mean response. *p < .05; **p ≤ .01; ***p ≤ .001. Open in new tab SHS exposure at home was not associated with depressive symptoms in any of the 2-year analyses, including fifth-grade SHS exposure in relation to seventh-grade depressive symptoms (B [95% CI] = 0.040 [−0.016 to 0.047]); seventh-grade exposure in relation to ninth-grade symptoms (B [95% CI] = 0.030 [−0.049 to 0.109]); and ninth-grade exposure in relation to 11th-grade symptoms (B [95% CI]= 0.016 [−0.067 to 0.100]). In the unadjusted model, SHS exposure in cars in fifth grade was associated with depressive symptoms in seventh grade (B [95% CI] = 0.069 [0.049 to 0.133]), but the association was not statistically significant after adjusting for demographics (B [95% CI] = 0.052 [−0.012 to 0.116]). There was no association between SHS exposure in cars in seventh grade and depressive symptoms in ninth grade (B [95% CI] = −0.029 [−0.127 to 0.068]) or between SHS exposure in ninth grade and depressive symptoms in 11th grade (B [95% CI] = 0.004 [−0.036 to 0.124]). None of the sex by SHS interaction terms were statistically significant. Discussion In this longitudinal study of a large population-based sample of children in Québec, we found that SHS exposure, whether at home or in cars, is associated with depressive symptoms in the short-term (ie, contemporaneously or up to 1 year later) but not in the long-term (ie, 2 years later). That SHS exposure is associated with self-reports of depressive symptoms in the cross-sectional analyses in AdoQuest aligns with previous cross-sectional findings in children, adolescents, and adults.7,9,10,12 A novel contribution of this study is that early SHS exposure (in fifth or sixth grade), both at home and in cars, was independently associated with depressive symptoms 1 year later, although SHS exposure at any age was not associated with depressive symptoms 2 years later. Even after controlling for baseline depressive symptoms, the 1-year associations between SHS exposure and depressive symptoms were significant. An alternative explanation, that parents with depression smoked more than nondepressed parents,33 thereby exposing their children to SHS more often, and also passed along to their children a propensity to experience depressive symptoms, seems less plausible once baseline depressive symptoms are taken into account. Nevertheless, we did not assess parental mental health, which limits our ability to explore this alternative hypothesis. Nicotine exerts its acute effects by first occupying nicotinic acetylcholine receptors throughout the brain. Taking one to two puffs on a cigarette results in occupation of approximately 50% of the receptors within a few seconds.34 In comparison, an average of 19% of these receptors are occupied when nonsmokers are exposed to moderate levels of SHS.35 In young adolescent never-smokers (average age 13 years), higher levels of social exposure to parents’, siblings’, and friends’ smoking were associated with experiencing more nicotine-dependence symptoms, and the association became more pronounced with greater pharmacological exposure to nicotine via SHS, as measured by both salivary cotinine and airborne nicotine.36 Possible mechanisms by which SHS exposure might influence depressive symptoms are suggested by animal studies. Compared to rats exposed to nicotine as adults, rats exposed as adolescents showed a “depression-like state, manifested in decreased sensitivity to natural reward, and enhanced sensitivity to stress- and anxiety-eliciting situations later in life.”37(p1617) Furthermore, this depressed state was induced after only a single day of exposure. A meta-analysis of 15 studies of adult smokers found consistent significant decreases in depression and mixed depression and anxiety months to years after quitting smoking,38 suggesting that continued nicotine exposure was exacerbating mood-related symptoms. The social impacts of living with smokers and being exposed to SHS are also unknown. It is possible that children of smokers experience depressive symptoms because they feel less able to socialize with friends (ie, reluctance to invite them home) or embarrassed because their clothing smells like tobacco smoke, because they are in conflict with the smoking caretaker about the person’s smoking and/or worried about the caretaker’s health, or because they feel their own health is not being protected by parents. Future research investigating the mechanistic underpinnings of the association between SHS exposure and depressive symptoms in humans is warranted. Several explanations are plausible for our finding that SHS exposure was associated with depressive symptoms one but not 2 years later. First, this could reflect that younger nonsmokers are more vulnerable to the effects of SHS exposure. Early adolescent mice are more sensitive to the effects of nicotine than mice in middle to late adolescence.39 Brain circuitry develops and changes rapidly from childhood through adolescence, and these changes continue into adulthood as individuals adapt to both environmental (external) challenges and developmental (internal) needs. Nicotine also acts on and remodels circuits involved in regulation of mood and anxiety,40 particularly the prefrontal cortex,41 potentially making those exposed more vulnerable to dysregulation. A second explanation is that the effects of SHS exposure might be short term, dissipating over a longer span, or SHS exposure itself might fluctuate over time, making estimation of its effects detectable over 1 year but not more. Third, because children who were susceptible to smoking reported more depressive symptoms at baseline and dropped out after initiating smoking, it is conceivable that the range of depression scores was restricted in later waves, reducing the likelihood of detecting a relationship. Finally, the questionnaires were administered at school in fifth and sixth grades, but completed at home in seventh through 11th grades. It is possible that respondents felt more or less able to answer honestly depending on the setting. We cannot differentiate among these explanations, but future research could be designed to address them. Limitations of this analysis include that measures of SHS exposure are based on self-report, which may have resulted in misclassification. However, the agreement between parental and participant reports of exposure suggests that misclassification was likely minimal. Misclassification may also have resulted from participants having been exposed to the residue of tobacco smoke (third-hand smoke) at home or in cars, which was not measured in this study. Selection bias attributable to attrition may have influenced estimates of the association, although never-smokers retained did not differ from participants lost to follow-up on any measured variable. That the sample comprised French-speaking adolescents primarily may have affected external generalizability. Residual confounding by unknown or unmeasured confounders (eg, parental depression) could have biased the estimates. In conclusion, results of this study raise concerns that in addition to physical health, SHS exposure might also affect the mental health of young persons. Clearly, protecting children from SHS is a critical public health priority, and children should be protected in all locations including private homes and vehicles, daycares, schools, multi-unit housing,42 foster and group homes, and outdoor spaces. Complete bans on smoking in public places, already widespread,43,44 reduce SHS exposure directly45 and, by changing smoking-related social norms, help promote voluntary bans at home.46,47 Smoking bans in private vehicles, however, have been less widely implemented,44,48 and both legislative and educational efforts are warranted to discourage smoking in vehicles when children are present. In addition, access to mental health care is critical for children and adolescents with symptoms of depression or anxiety, and providers should be aware that this risk is higher among youth with household smokers. Finally, health care providers and social welfare advocates have an important role in raising awareness and educating their clientele about the dangers of SHS exposure in children. Funding This project was funded by the Canadian Tobacco Control Research Initiative and the Institut national de santé publique du Québec (INSPQ) through a financial contribution from the Québec Ministry of Health and Social Services to the INSPQ. Declaration of Interests Views expressed in this document do not necessarily reflect those of the Québec Ministry of Health and Social Services. JO holds a Canada Research Chair in the Early Determinants of Adult Chronic Disease. EO is supported by a doctoral fellowship from the Fonds de Recherche du Québec—Santé. The funders were not involved in the design or conduct of the study, data collection, management, analysis, or interpretation, or preparation, review, or approval of the manuscript. The authors declare no conflicts of interest. References 1. U.S. Department of Health and Human Services . The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General . Atlanta, GA : U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health ; 2006 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 2. McIntire RK , Macy JT , Seo DC , Nelson AA , Kolbe LJ . Secondhand smoke exposure in cars and rooms: trend comparisons among subpopulations of nonsmoking U.S. middle and high school students . Nicotine Tob Res . 2014 ; 16 ( 6 ): 663 – 671 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Reid JL , Hammond D , Rynard VL , Madill CL , Burkhalter R . Tobacco Use in Canada: Patterns and Trends, 2017 Edition . Waterloo, ON : Propel Centre for Population Health Impact, University of Waterloo . https://uwaterloo.ca/tobacco-use-canada/ Accessed June 26, 2018 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 4. Lasnier B , Leclerc BS , Hamel D. Les Inégalités Sociales de Santé en Matière de Tabagisme et d’Exposition à la Fumée de Tabac dans l’environnement au Québec . Montréal, Canada : Institut national de santé publique du Québec ; 2012 . https://www.inspq.qc.ca/publications/1487. Accessed June 26, 2018 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 5. Homa DM , Neff LJ , King BA , et al. ; Centers for Disease Control and Prevention (CDC) . Vital signs: disparities in nonsmokers’ exposure to secondhand smoke–United States, 1999-2012 . MMWR Morb Mortal Wkly Rep . 2015 ; 64 ( 4 ): 103 – 108 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 6. Wong SL , Malaison E , Hammond D , Leatherdale ST . Secondhand smoke exposure among Canadians: cotinine and self-report measures from the Canadian Health Measures Survey 2007-2009 . Nicotine Tob Res . 2013 ; 15 ( 3 ): 693 – 700 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Patten SB , Williams JVA , Lavorato DH , et al. Major depression and secondhand smoke exposure . J Affect Disord . 2018 ; 225 : 260 – 264 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Bot M , Vink JM , Willemsen G , et al. Exposure to secondhand smoke and depression and anxiety: a report from two studies in the Netherlands . J Psychosom Res . 2013 ; 75 ( 5 ): 431 – 436 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Kim NH , Park JH , Choi DP , Lee JY , Kim HC . Secondhand smoke exposure and depressive symptoms among Korean adolescents: JS high school study . PLoS One . 2016 ; 11 ( 12 ): e0168754 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Yolton K , Khoury J , Hornung R , Dietrich K , Succop P , Lanphear B . Environmental tobacco smoke exposure and child behaviors . J Dev Behav Pediatr . 2008 ; 29 ( 6 ): 450 – 457 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Padrón A , Galán I , García-Esquinas E , Fernández E , Ballbè M , Rodríguez-Artalejo F . Exposure to secondhand smoke in the home and mental health in children: a population-based study . Tob Control . 2016 ; 25 ( 3 ): 307 – 312 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Bandiera FC , Richardson AK , Lee DJ , He JP , Merikangas KR . Secondhand smoke exposure and mental health among children and adolescents . Arch Pediatr Adolesc Med . 2011 ; 165 ( 4 ): 332 – 338 . Google Scholar Crossref Search ADS PubMed WorldCat 13. O’Loughlin J , O’Loughlin EK , Wellman RJ , et al. Predictors of cigarette smoking initiation in early, middle, and late adolescence . J Adolesc Health . 2017 ; 61 ( 3 ): 363 – 370 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Elton-Marshall T , Leatherdale ST , Manske SR , Wong K , Ahmed R , Burkhalter R . Research methods of the Youth Smoking Survey (YSS) . Chronic Dis Inj Can . 2011 ; 32 ( 1 ): 47 – 54 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 15. Paradis G , Lambert M , O’Loughlin J , et al. The Québec child and adolescent health and social survey: design and methods of a cardiovascular risk factor survey for youth . Can J Cardiol . 2003 ; 19 ( 5 ): 523 – 531 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 16. Kandel DB , Davies M . Epidemiology of depressive mood in adolescents: an empirical study . Arch Gen Psychiatry . 1982 ; 39 ( 10 ): 1205 – 1212 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Brunet J , Sabiston CM , Chaiton M , et al. Measurement invariance of the depressive symptoms scale during adolescence . BMC Psychiatry . 2014 ; 14 ( 1 ): 95 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Avila-Tang E , Elf JL , Cummings KM , et al. Assessing secondhand smoke exposure with reported measures . Tob Control . 2013 ; 22 ( 3 ): 156 – 163 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Wellman RJ , Sylvestre MP , O’Loughlin EK , et al. Socioeconomic status is associated with the prevalence and co-occurrence of risk factors for cigarette smoking initiation during adolescence . Int J Public Health . 2018 ; 63 ( 1 ): 125 – 136 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Pampalon R , Raymond G . A deprivation index for health and welfare planning in Quebec . Chronic Dis Can . 2000 ; 21 ( 3 ): 104 – 113 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 21. Choi WS , Gilpin EA , Farkas AJ , Pierce JP . Determining the probability of future smoking among adolescents . Addiction . 2001 ; 96 ( 2 ): 313 – 323 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Azur MJ , Stuart EA , Frangakis C , Leaf PJ . Multiple imputation by chained equations: what is it and how does it work ? Int J Methods Psychiatr Res . 2011 ; 20 ( 1 ): 40 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat 23. White IR , Royston P , Wood AM . Multiple imputation using chained equations: issues and guidance for practice . Stat Med . 2011 ; 30 ( 4 ): 377 – 399 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Morris TP , White IR , Royston P . Tuning multiple imputation by predictive mean matching and local residual draws . BMC Med Res Methodol . 2014 ; 14 ( 1 ): 75 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Bodner TE . What improves with increased missing data imputations ? Struct Equ Modeling . 2008 ; 15 ( 4 ): 651 – 675 . Google Scholar Crossref Search ADS WorldCat 26. Gustavson K , von Soest T , Karevold E , Røysamb E . Attrition and generalizability in longitudinal studies: findings from a 15-year population-based study and a Monte Carlo simulation study . BMC Public Health . 2012 ; 12 ( 1 ): 918 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Lindsey JK , Jones B . Choosing among generalized linear models applied to medical data . Stat Med . 1998 ; 17 ( 1 ): 59 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Ng V , Cribbie RA . Using the gamma generalized linear model for modeling continuous, skewed and heteroscedastic outcomes in psychology . Current Psychol . 2017 ; 36 ( 2 ): 225 – 235 . Google Scholar Crossref Search ADS WorldCat 29. White HL , Jr . A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity . Econometrica . 1980 ; 48 ( 4 ): 817 – 838 . Google Scholar Crossref Search ADS WorldCat 30. Long JS , Ervin LH . Using heteroskedasticity consistent standard errors in the linear regression model . Am Stat . 2000 ; 54 ( 3 ): 217 – 224 . OpenURL Placeholder Text WorldCat 31. Hanley JA , Negassa A , Edwardes MD , Forrester JE . Statistical analysis of correlated data using generalized estimating equations: an orientation . Am J Epidemiol . 2003 ; 157 ( 4 ): 364 – 375 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Grace-Martin K . Unstructured covariance matrix: when it does and doesn’t work . The Analysis Factor .http://www.theanalysisfactor.com/unstructured-covariance-matrix-when-it-does-and-doesn%E2%80%99t-work. Accessed June 26, 2018 . 33. Hall SM , Muñoz RF , Reus VI , Sees KL . Nicotine, negative affect, and depression . J Consult Clin Psychol . 1993 ; 61 ( 5 ): 761 – 767 . Google Scholar Crossref Search ADS PubMed WorldCat 34. Brody AL , Mandelkern MA , London ED , et al. Cigarette smoking saturates brain alpha 4 beta 2 nicotinic acetylcholine receptors . Arch Gen Psychiatry . 2006 ; 63 ( 8 ): 907 – 915 . Google Scholar Crossref Search ADS PubMed WorldCat 35. Brody AL , Mandelkern MA , London ED , et al. Effect of secondhand smoke on occupancy of nicotinic acetylcholine receptors in brain . Arch Gen Psychiatry . 2011 ; 68 ( 9 ): 953 – 960 . Google Scholar Crossref Search ADS PubMed WorldCat 36. McGrath JJ , Racicot S , Okoli CTC , Hammond SK , O’Loughlin J . Airborne nicotine, secondhand smoke, and precursors to adolescent smoking . Pediatrics . 2018 ; 141 ( suppl 1 ): S63 – S74 . Google Scholar Crossref Search ADS PubMed WorldCat 37. Iñiguez SD , Warren BL , Parise EM , et al. Nicotine exposure during adolescence induces a depression-like state in adulthood . Neuropsychopharmacology . 2009 ; 34 ( 6 ): 1609 – 1624 . Google Scholar Crossref Search ADS PubMed WorldCat 38. Taylor G , McNeill A , Girling A , Farley A , Lindson-Hawley N , Aveyard P . Change in mental health after smoking cessation: systematic review and meta-analysis . BMJ . 2014 ; 348 ( Feb. 13 1 ): g1151 . Google Scholar Crossref Search ADS PubMed WorldCat 39. Adriani W , Macrì S , Pacifici R , Laviola G . Peculiar vulnerability to nicotine oral self-administration in mice during early adolescence . Neuropsychopharmacology . 2002 ; 27 ( 2 ): 212 – 224 . Google Scholar Crossref Search ADS PubMed WorldCat 40. Picciotto MR , Lewis AS , van Schalkwyk GI , Mineur YS . Mood and anxiety regulation by nicotinic acetylcholine receptors: a potential pathway to modulate aggression and related behavioral states . Neuropharmacology . 2015 ; 96 ( pt B ): 235 – 243 . Google Scholar Crossref Search ADS PubMed WorldCat 41. Goriounova NA , Mansvelder HD . Nicotine exposure during adolescence alters the rules for prefrontal cortical synaptic plasticity during adulthood . Front Synaptic Neurosci . 2012 ; 4 : 3 . Google Scholar Crossref Search ADS PubMed WorldCat 42. Tobacco Control Legal Consortium . HUD’s rule to restrict smoking in public housing: an overview . http://www.publichealthlawcenter.org/sites/default/files/resources/HUD-Final-Rule-Smoke-Free-Public-Housing-2016.pdf. Accessed June 26, 2018 . 43. American Nonsmokers’ Rights Foundation . Chronological table of U.S. population protected by 100% smokefree state or local laws , January 2, 2018. http://www.no-smoke.org/pdf/EffectivePopulationList.pdf. Accessed June 26, 2018 . 44. Campaign for Tobacco-free Kids . Tobacco control laws: legislation by country—Canada . https://www.tobaccocontrollaws.org/legislation/country/canada/summary. Accessed June 26, 2018 . 45. Hoffman SJ , Tan C . Overview of systematic reviews on the health-related effects of government tobacco control policies . BMC Public Health . 2015 ; 15 ( 1 ): 744 . Google Scholar Crossref Search ADS PubMed WorldCat 46. Cheng KW , Glantz SA , Lightwood JM . Association between smokefree laws and voluntary smokefree-home rules . Am J Prev Med . 2011 ; 41 ( 6 ): 566 – 572 . Google Scholar Crossref Search ADS PubMed WorldCat 47. Nazar GP , Lee JT , Glantz SA , Arora M , Pearce N , Millett C . Association between being employed in a smoke-free workplace and living in a smoke-free home: evidence from 15 low and middle income countries . Prev Med . 2014 ; 59 : 47 – 53 . Google Scholar Crossref Search ADS PubMed WorldCat 48. American Nonsmokers’ Rights Foundation . Get the Facts; Going Smoke Free; Smokefree cars . http://www.no-smoke.org/learnmore.php?id=616. Accessed June 26, 2018 . © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Secondhand Smoke Exposure and Depressive Symptoms in Children: A Longitudinal Study
JF - Nicotine and Tobacco Research
DO - 10.1093/ntr/nty224
DA - 2020-01-27
UR - https://www.deepdyve.com/lp/oxford-university-press/secondhand-smoke-exposure-and-depressive-symptoms-in-children-a-FTfv1M1Hfg
SP - 32
VL - 22
IS - 1
DP - DeepDyve
ER -