In a repeated large population-based survey from the Capital Region of Denmark we found marked and persistent social inequalities in children’s exposure to SHS; the lower the education of the parent/adult the higher the probability of smoking in the home. There was significant temporal decrease in domestic smoking from 2007 to 2010, independent of education level. Thus, social disparities in the level of children’s exposure to SHS at home have persisted over time, but not increased significantly.
It has been documented that children’s exposure to SHS has declined over recent decades
] and it is also known that there are large social disparities in children’s exposure to SHS at home. Parents with a low SES and single parents have a higher prevalence of smoking and are less likely to have implemented smoking bans at home
]. However, to our knowledge, only four studies have investigated the temporal changes in social disparities in children’s exposure to SHS
]. A recent Australian study found that the proportion of children who lived with a smoker had declined in all social groups except the most disadvantaged households
]. A study from Wales found that reductions in SHS exposure were limited to children from higher SES households
]. A Scottish study found that the introduction of smoke-free legislation had reduced exposure to SHS among all children
]. Results also indicated that inequalities in cotinine concentration in children increased after legislation. According to an English study, however, the decline in cotinine over time tended to be greatest in children who were most exposed, indicating that absolute inequalities in exposure to SHS have fallen from 1996 to 2006
]. In our study, no significant increase or decrease in social inequalities over time was found. The different findings may be explained by national or temporal differences or differences in the measurement of exposure to SHS or measurement of socioeconomic status.
On August 15, 2007, a few weeks before the first health survey of the Capital Region of Denmark, a national smoking ban was implemented. Our data represents the time immediately after implementation of the legislation and three years later. The temporal reductions in smoking at home might therefore reflect an effect of the legislation, but might also reflect a secular trend towards smoke-free homes-a longer observational period before legislation would be needed to answer that. Reduction in children’s exposure to SHS, and an increase in the proportion of children reporting a ban on smoking in their household have been reported after implementation of the smoke-free legislation in several western countries
]. However, in Hong Kong it seems that a comprehensive smoke-free legislation might have displaced smoking into the homes of children
], which underlines the importance of strong simultaneous support for smoking cessation and comprehensive information about children’s health hazards when exposed to SHS.
Smoking outdoors with the door closed does not constitute total protection, but is the most effective way to protect children from environmental tobacco-smoke exposure. Other modes of action has been shown to have a minor effect
]. The low levels of knowledge of the adverse health effects of SHS, especially among smokers, and the known relationship between knowledge and SHS-protective behaviours, suggest that greater efforts to educate smokers about the risks associated with SHS are worthwhile
]. An important factor in reducing inequalities in SHS exposure among children lies in educating people, especially parents with less education/income, about the health benefits of keeping their homes smoke-free. It is an important task for health professionals, teachers, nursery teachers and so on, but it is difficult, expensive and time consuming. A recent Australian study reported that child health services in almost eight out of ten cases did not assess the SHS exposure of any child
]. Many counselling interventions for parents have tried to reduce children’s exposure to SHS at home. Controlled trials of clinicians’ one time counselling services have shown null results
]. A review by Cochrane found that in only 11 of 36 studies was there a statistically significant intervention effect, and that there was limited support for more intensive counselling interventions for parents
]. Another review concluded that studies with more rigorous study designs, interventions of greater intensity and duration, and those based on sound behaviour change theory have yielded the most promising results
]. Another approach, interventions to achieve cessation among parents, for the sake of the children, can help protect vulnerable children from harm due to tobacco smoke exposure. However, most parents do not quit, and additional strategies to protect children are needed
Home smoking bans are surrounded by social, legal, and political challenges and so far no state or country in the world has dared to implement legislation banning smoking in homes with children. In Australia, however, in 2009, smoking was banned in cars if children are present
]. In many countries around the world corporal punishment in the home is outlawed. Even though there is no evidence that outlawing smoking in the home will reduce SHS in the home and that this will be equal across SES it seems reasonable to suggest legislative protection of children from tobacco smoke inside their home. In general, legislation has shown to be a very strong instrument in tobacco control. Home smoking bans may also contribute towards a reduced risk of children becoming smokers, particularly when their parents smoke
], thereby further reducing social disparities in smoking-related morbidity and mortality in the longer term.
Strengths of this study include the large study size, and the random sample of a general population. It is also a strength that the statistical analyses were weighted for the size of municipalities and for non-responses, which increases the generalizability of the study. Information about those co-habiting with a child/children, and about education levels, is from the National Central Registers, which have a high validity.
A limitation of the study is that information on smoking in the home is self-reported. It would have strengthened our results if we could have provided objective data on children’s exposure to SHS, e.g. via salivary cotinine measurements, but this was not possible in the large scale survey. More detailed information on where and how many persons smoked in the home would have been useful. It would have strengthened our study if we had used other measures of socioeconomic status, such as income or the employment status of the parent/adult. Measuring SES is very complex and each measurement has different strengths and weaknesses. There is no single best indicator of SES
]. Level of education can be defined and applied regardless of working circumstances, it is a strong determinant of employment and income and it generally reflects knowledge
]. The responding adult has, for the sake of convenience been called the parent, but in some cases it will be a step-parent, guardian, older sibling or another adult person living in a household with a child. The definition of a smoke-free home is not really a smoke-free home, as we have included ‘almost never’ and ‘less than weekly’ in our definition. Finally, other predictors for exposure, such as age of youngest child, could have been included. It has been reported that persons with infants in the home are more likely to have a smoke-free home than those with older children
]. However, we believe this is consistent over time.