The authors have declared that no competing interests exist.
Conceived and designed the experiments: VK OP. Performed the experiments: CE RO RK ZM DG A. Boyd MC SL VK OP A. Bitfoi. Analyzed the data: A. Boyd. Contributed reagents/materials/analysis tools: VK OP A. Boyd. Wrote the paper: VK DP ES OP.
To ascertain patterns of parental smoking in the vicinity of children in Eastern and Western Europe and their relation to Tobacco Control Scale (TCS) scores.
Data on parental smoking patterns were obtained from the School Child Mental Health Europe (SCMHE), a 2010 cross-sectional survey of 5141 school children aged 6 to 11 years and their parents in six countries: Germany, Netherlands, Lithuania, Romania, Bulgaria and Turkey ranked by TCS into three level categories toward tobacco control policies.
A slightly higher proportion of Eastern compared to Western European mothers (42.4 vs. 35.1%) were currently smoking in but the difference was not statistically significant after adjusting for maternal age and maternal educational attainment. About a fifth (19.3%) and a tenth (10.0%) of Eastern and Western European mothers, respectively, smoked in the vicinity of their children, and the difference was significant even after adjustment for potential confounders (p<0.001). Parents with the highest educational attainment were significantly less likely to smoke in the vicinity of their children than those with the lowest attainment. After control of these covariates lax tobacco control policies, compared to intermediate policies, were associated with a 50% increase in the likelihood of maternal smoking in the vicinity of children adjusted odds ratio (AOR) = 1.52 and 1.64. Among fathers, however, the relationship with paternal smoking and TCS seems more complex since strict policy increases the risk as well AOR = 1,40. Only one country, however belongs to the strict group.
Tobacco control policies seem to have influenced maternal smoking behaviors overall to a limited degree and smoking in the vicinity of children to a much greater degree. Children living in European countries with lax tobacco control policies are more likely to be exposed to second hand smoking from maternal and paternal smoking.
Cigarette smoking is a major determinant of health and longevity
Passive or second hand smoking (SHS) is also detrimental to health
Article 8 of the World Health Organization Framework Convention on Tobacco Control (2005) requires all signatory countries to adopt measures to protect people from tobacco smoke in indoor workplaces, indoor public places, public transport and other public places as appropriate
Parental smoking is the most important predictor of SHS for children and children from impoverished households are more likely to be exposed
To address this gap, we used data from the School Children Mental Health Evaluation project (SCHME), a multisite school-based survey of children aged 6–11 in two Western (Netherlands and Germany) and four Eastern (Romania, Bulgaria, Lithuania, and Turkey) countries in the European region in 2010
We also examined the relation of parental smoking patterns to tobacco control policies for countries in the European region. Tobacco control policies aim at decreasing tobacco consumption using a variety of approaches. Although these policies have been implemented in many countries, very few studies have focused on associations between these policies and parental smoking in the vicinity of children. This is important because many countries are engaged in costly campaigns to decrease smoking in the population, and parents are an important target of these campaigns.
The SCHME was a cross-sectional survey of schoolchildren aged 6 to 11 years and their parents in six countries: Germany, Netherlands, Lithuania, Romania, Bulgaria and Turkey. Grade schools were randomly selected in each participating country, classes were randomly selected in each school, and 6 children were randomly selected in each class. Approximately 48 children were randomly selected in each school, except in the Netherlands, where a lesser number of schools participated and therefore a greater number of children, about 120, were randomly selected in each school. To interview approximately 1000 parents, children and teachers, it was necessary to sample from 45 to 49 schools in each country. Additional information about sampling methods was included in the final SCMHE report
Of the children invited to participate, 72.2% and 61.3% participated in the survey in Western and Eastern Europe, respectively. In most cases the corresponding parental respondent was the mother. Parental respondents were asked questions about smoking in the vicinity of the child in reference to both parents. Overall smoking patterns, however, were only asked in reference to the respondent. To reduce heterogeneity, we restricted these analyses to the mother’s report, and therefore, overall smoking patterns are only reported for mothers. Further information on participation, such as by country and by parental respondent, is available in the SCHME report.
The SCHME was designed to allow for broad comparisons of Western and Eastern Europe. For this purpose, SCHME classifies participants in the former West Germany and Netherlands as Western, and classifies participants in the former East Germany, Lithuania, Romania, Bulgaria and Turkey as Eastern. The division of Germany into former West and East was made because of their distinctive historical experience.
Data were collected from three informants: the child, the teacher and a parent. The responding parent (usually the mother) was asked to report about the other parent, regardless of marital/cohabitation status. Parental self-reports included a demographic and social questionnaire concerning household composition (including age, gender and familial link for each member), parental education (highest level completed), marital status, occupational level, rural/urban type of residence, as well as a questionnaire focusing on tobacco use. In the Netherlands the same questions were completed electronically using a secured website, though paper questionnaires were made available upon request.
The questionnaire on tobacco use included questions from a periodically administered tobacco use survey known as “Eurobarometer”
The “Tobacco Control Scale” (TCS) was used to measure tobacco control policies
We compared demographic characteristics of participants in Eastern and Western Europe using Chi Square analyses to estimate the significance of the differences with respect to the categorical variables listed in
Demographic Characteristics | Subcategory | East Europe(% (n)) | West Europe(% (n)) |
|
Number of children in family | Total Sample Size | ( |
( |
<0.001 |
1 | 32.9 (1402) | 10.3 (91) | ||
2 or 3 | 52.3 (2225) | 78.2 (691) | ||
≥4 | 14.8 (630) | 11.5 (102) | ||
Marital Status | Total Sample Size | ( |
( |
<0.001 |
Single/Never Married/Separated/Divorced/widowed | 16.7 (659) | 9.1 (77) | ||
Married/Remarried/Cohabitation/Other | 83.4 (3298) | 90.9 (768) | ||
Mother's highest level of education | Total Sample Size | ( |
( |
<0.001 |
Some primary or secondary | 19.4 (715) | 2.2 (17) | ||
Secondary completed | 39.8 (1469) | 30.2 (235) | ||
College or technical school completed | 40.8 (1504) | 67.6 (526) | ||
Age of mother | Total Sample Size | ( |
( |
|
Years |
35.7 (5.6) | 40.4 (4.7) | <0.0001 | |
>35 | 52.4 (2116) | 14.0 (122) | <0.001 | |
35–40 | 29.5 (1190) | 34.1 (297) | ||
>40 | 18.2 (734) | 51.8 (451) | ||
Age of father | Total Sample Size | ( |
( |
|
Years |
38.0 (5.9) | 41.9 (5.1) | <0.0001 | |
>35 | 36.8 (1323) | 8.5 (71) | <0.001 | |
35–40 | 33.5 (1203) | 28.8 (241) | ||
>40 | 29.7 (1065) | 62.8 (526) |
West Europe: the Netherlands and West Germany.
East Europe: Bulgaria, Lithuania, Romania, Turkey, East Germany.
East Europe | West Europe | ||||||||||
Bulgaria | Lithuania | Romania | Turkey | EastGermany | Total | The Netherlands | West Germany | Total | |||
Mother’s current smoking status | Total Sample Size | ( |
( |
( |
( |
( |
|
( |
( |
|
<0.001 |
Daily current smoker | 47.9 | 17.3 | 33.2 | 27.5 | 30.4 |
|
35.3 | 15.2 |
|
||
Occasional current smoker | 11.4 | 12.6 | 8.8 | 11.0 | 10.9 |
|
3.9 | 7.1 |
|
||
Former current smoker | 18.5 | 18.7 | 13.2 | 14.7 | 15.2 |
|
8.2 | 32.0 |
|
||
Never smoker | 22.2 | 51.4 | 44.8 | 46.8 | 43.5 |
|
52.6 | 45.7 |
|
||
% of mothers currently smoking |
Total Sample Size | ( |
( |
( |
( |
( |
|
( |
( |
|
|
Unadjusted | 58.4 | 29.4 | 41.6 | 39.5 | 42.2 |
|
39.2 | 22.3 |
|
<0.001 | |
Adjusted |
57.6 | 30.5 | 39.4 | 33.8 | 37.6 |
|
44.8 | 23.2 |
|
0.2 |
Adjusted for mother’s age as a continuous variable, and for mother’s education status; All unadjusted values are among participants without missing information on covariates.
Includes daily and occasional current smokers.
East Europe | West Europe | ||||||||||
Bulgaria | Lithuania | Romania | Turkey | East Germany | Total | The Netherlands | WestGermany | Total | |||
% of mothers smoking in vicinity of child | Total Sample Size | ( |
( |
( |
( |
( |
|
( |
( |
|
|
Unadjusted | 32.4 | 10.0 | 20.3 | 16.1 | 19.5 |
|
8.8 | 3.9 |
|
<0.001 | |
Adjusted |
31.8 | 10.8 | 18.6 | 12.1 | 20.2 |
|
11.9 | 3.8 |
|
<0.001 | |
% of fathers smoking in vicinity of child | Total Sample Size | ( |
( |
( |
( |
( |
|
( |
( |
|
|
Unadjusted | 32.0 | 21.5 | 25.8 | 23.8 | 7.9 |
|
10.4 | 6.8 |
|
<0.001 | |
Adjusted |
31.4 | 22.6 | 24.0 | 18.7 | 10.1 |
|
13.5 | 6.5 |
|
<0.001 |
Western and Eastern Europe.
Adjusted for mother’s age as a continuous variable, and for maternal educational status. All unadjusted values are among participants without missing information on covariates.
Multivariate models were used to identify the determinants of (i) current smoking and (ii) smoking in the vicinity of the child for the mother (
Mother smoking in vicinity of child | Mother current smoking | ||||
East Europe/Turkey | West Europe | East Europe/Turkey | West Europe | ||
OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | ||
Sociodemographic Characteristics | Total Sample Size | (n = 2887) | (n = 757) | (n = 2997) | (n = 762) |
Presence of other children | No other children | 1.00 | 1.00 | 1.00 | 1.00 |
Children ≤3 years old | 0.74 (0.56–0.98) | 0.28 (0.06–1.27) | 0.77 (0.62–0.96) | 0.57 (0.29–1.13) | |
Children 4–10 years old | 0.82 (0.67–1.02) | 0.64 (0.35–1.17) | 0.92 (0.78–1.09) | 0.59 (0.43–0.83) | |
Children 11–18 years old | 0.97 (0.77–1.22) | 0.74 (0.37–1.46) | 0.84 (0.70–1.01) | 0.74 (0.51–1.08) | |
Mother's age | ≤35 years | 1.00 | 1.00 | 1.00 | 1.00 |
>35, ≤40 years | 0.84 (0.67–1.05) | 0.62 (0.16–2.39) | 0.76 (0.64–0.91) | 0.67 (0.40–1.10) | |
>40 years | 0.73 (0.55–0.96) | 0.30 (0.08–1.14) | 0.72 (0.58–0.89) | 0.94 (0.57–1.56) | |
Mother’s education level | None/some secondary | 1.00 | 1.00 | 1.00 | 1.00 |
Secondary completed | 0.91 (0.72–1.17) | 0.59 (0.15–2.32) | 1.11 (0.90–1.38) | 1.32 (0.45–3.89) | |
College completed | 0.42 (0.32–0.55) | 0.25 (0.06–0.98) | 0.57 (0.46–0.71) | 0.84 (0.29–2.45) | |
Living with partner | 0.57 (0.45–0.72) | 0.51 (0.24–1.08) | 0.57 (0.47–0.70) | 0.87 (0.51–1.48) |
Based on a multivariate logistic regression that included all variables in the table.
OR = Odds Ratio.
CI = Confidence Interval.
Father smoking in vicinity of child | |||
East Europe/Turkey ( |
West Europe ( |
||
Sociodemographic characteristics | Subcategory | OR (95% CI) | OR (95% CI) |
Presence of other children | No other children | 1.00 | 1.00 |
Children ≤3 years old | 0.78 (0.59–1.02) | 0.83 (0.30–2.24) | |
Children 4–10 years old | 1.17 (0.95–1.43) | 0.76 (0.45–1.28) | |
Children 11–18 years old | 1.15 (0.92–1.43) | 1.04 (0.58–1.87) | |
Father's age | ≤35 years | 1.00 | 1.00 |
>35, ≤40 years | 0.93 (0.75–1.16) | 1.01 (0.43–2.39) | |
>40 years | 0.81 (0.64–1.04) | 0.60 (0.26–1.43) | |
Father’s education level | None/some secondary | 1.00 |
|
Secondary completed | 0.95 (0.74–1.22) |
|
|
College completed | 0.40 (0.31–0.53) |
|
|
Living with partner | 0.84 (0.61–1.15) | 0.41 (0.20–0.85) |
Data not available.
Based on a multivariate logistic regression that included all variables in the table.
OR = Odds Ratio.
CI = Confidence Interval.
Lastly, we examined the relation of country level tobacco control policies to the following variables: mother smoking in vicinity of the child, mother currently smoking, and father smoking in vicinity of the child (
Mother currentlySmoking(n = 3759) | Mother smoking invicinity of child(n = 3644) | Father smoking invicinity of child(n = 3627) | ||||
TCP |
OR(95% CI) | p | OR(95% CI) | p | OR(95% CI) | p |
High | 0.88(0.69–1.11) | 0.3 | 0.82(0.60–1.11) | 0.2 | 1.40(1.08–1.83) | 0.01 |
Middle | 1.00 | 1.00 | 1.0 | |||
Low | 1.17(1.01–1.35) | 0.04 | 1.52(1.25–1.85) | <0.001 | 1.64(1.38–1.95) | <0.001 |
Based on multivariate logistic regression models for the four smoking patterns shown above. These models were adjusted for the presence of other children in the household (none/children ≤3 years/4–10 years and 11–18 years), mother’s or father’s age (≤35; >35,≤40;and.40), mothers’ educational level (none to some secondary; secondary completed; and college completed), living with partner (yes/no).
Tobacco control policies rank (upper third, middle third and lowest third tobacco control scores rank). Thus, low refers to the countries with the laxest tobacco control policies. Note that only one country is in the high category.
OR = Odds Ratio.
CI = Confidence Interval.
All participating countries had the support of their governments, including their ministers of education and health and received ethical approval from the corresponding authority.
As shown in
As shown in
With respect to mothers, those with a higher educational attainment were less likely to smoke in the vicinity of the child in both Western and Eastern Europe (
As shown in
We have reported data on maternal and paternal smoking in the vicinity of children in Eastern and Western Europe. There were three main findings. First, Eastern European parents were about twice as likely to smoke in the vicinity of their children as their Western European counterparts. Current maternal smoking prevalence was similar, however, in Eastern and Western Europe (adjusted analysis: 41.2% and 40.2%, respectively). Thus, Western European mothers were specifically restricting their smoking in the presence of children. Second, a strong relationship was observed between parental education and smoking in the vicinity of the child, i.e. college completers in both regions were less likely to smoke in the presence of their children than those with the least education. Third, in countries with lax tobacco control policies, compared to those with intermediate policies, mothers were more likely to smoke in the vicinity of the child. The relationship of tobacco control policies to current smoking of mothers, although significant, was much weaker, suggesting that these policies may have specifically restricted maternal smoking in the vicinity of children.
The prevalence of SHS has been estimated differently for adults and children. In the case of adults it is often defined as having a spouse who smokes or being exposed to tobacco smoke at work; for children as having one or both parents who smoke
A study of five European countries (Ireland, Sweden, France, Italy and the Czech Republic) demonstrated that adult women living in Sweden and Ireland were more likely to have quit smoking in the 5 years leading to a 2008 survey than those living in the Czech Republic
Overall, 72.2% and 61.3% of invited children participated in Western and Eastern Europe, respectively. This level of participation is similar, however, to most other contemporary surveys. We do not know whether there was an association between participation and parental smoking status (e.g. children of smoking parents participating less than children of non-smoking parents) but it is likely that there were multiple other reasons for non-participation, e.g. privacy concerns, and lack of child interest in filling out questionnaires. The sample from East Germany was smaller than in other countries because sampling was designed for Germany as a whole, and we report separately on the prevalence of smoking in western and eastern regions of Germany. Additionally, the response rates for two important items (“current smoking” and “smoking in vicinity of child”) were particularly low in East Germany.
Although we found a relationship between tobacco control policies and smoking behaviour, laws are more likely to be enacted when supported by constituents and when they reflect societal beliefs and norms. Thus, reverse causation cannot be ruled out. Nevertheless, the association between lax tobacco control legislation and smoking in the vicinity of children is noteworthy.
The inclusion of specific information about smoking in the vicinity of children is a major strength of this study. Because mothers specifically limited smoking in the vicinity of their children, survey data on maternal current smoking patterns would not have revealed the patterns described here.
Tobacco control measures seem to have influenced maternal smoking behavior overall to a limited degree and smoking in the vicinity of children to a greater degree, leading to a lower prevalence of SHS exposure among European children, and should be expanded and continued. Thus far, these policies do not appear to have had a similar impact on paternal smoking in the vicinity of children. Thus, fathers should be specifically targeted in anti-smoking campaigns, with an emphasis on the importance of children’s exposure to SHS. Furthermore, parents with low educational attainment should be targeted by messages appropriate to their social context, literacy and comprehension level by anti-smoking campaigns.
As suggested by Winickoff et al., pediatricians and family doctors should ask parents about their smoking patterns and specifically about smoking in the vicinity of children, and counsel parents about the impact of SHS exposure on children’s health
We would like to thank the SCMHE Project group and everybody who contributed to the survey and the production of this European report. Great thanks are also due to the interviewers who worked on the survey in the participating countries. Most importantly, we would like to thank all the parents, young people, and teachers for their cooperation.