Impact of Smoking Bans on Infant and Child Health

As evidence of the negative effects of environmental tobacco smoke (ETS) has mounted, an increasingly popular public policy response has been to impose restrictions on smoking through 100% smoke-free bans. However, a key link necessary for a comprehensive evaluation of the effectiveness of smoking bans has been missing from the literature including the CDC and IOM reports.  Specifically, it is not known whether and to what extent smoking bans at the state and local levels impact the health of children and infants.  The rationale for smoke-free laws implicitly presumes there are public health gains from reducing adult cigarette consumption and declines in adult ETS exposure that extend to children.  The presumption that these gains, a priori, will extend to infants and children may or may not be correct.   On the one hand, decreases in adult smoking prevalence as a result of the smoking bans may reduce ETS exposure among children, infants, and pregnant women.  However, if smokers shift from consuming cigarettes at 100% smoke free work and other public places to smoking at home (when children, infants, or pregnant women are present) then these policies may have a harmful effect on children and infants.  Such compensatory behavior by smokers can potentially increase the prevalence of smoking at home and lead to higher ETS exposures among children. 

This study addresses this critical gap in the literature and provides key estimates of how 100% smoke-free regulations impact the venue of smoking and how they impact the health of children and infants.  We match comprehensive information on all local and state-level 100% smoke-free legislation by type (workplace, restaurant, and freestanding bar) to parent and child records from the National Health Interview Surveys (NHIS) and birth records from the U.S. Natality Files (NF). Exploiting the panel structure of the data, fixed effects estimates suggest that smoking bans have improved both infant and children’s health, though most of these effects are realized from more comprehensive bans.  Limited bans are not found to be effective. Decomposing this improvement in infant health, we find that the effects are being realized from a decrease in prenatal smoking (among pregnant women who smoked) as well as a decrease in ETS exposure (among non-smoking expectant mothers).  We further assess whether this improvement in health is being moderated by a shift in smoking from the banned venues to inside the home, using the NHIS.  We find no evidence of such compensatory behaviors among smokers (both smokers with and without children in the household), and actually find that the bans had a positive spillover effect in terms of reducing smoking inside the home – an effect which may further explain the improvement in infant and children’s health.  We propose an explanation for this positive spillover effect.  Overall, the estimates from the proposed research address the missing link in the literature and provide valuable inputs towards a comprehensive evaluation of state and local restrictions on smoking in various venues.