A study published in this week’s Canadian Medical Association Journal implicates secondhand smoke exposure among adolescents as a cause of smoking initiation in this population (see: Becklake MR, Ghezzo H, Ernst P. Childhood predictors of smoking in adolescence: a follow-up study of Montreal schoolchildren. CMAJ 2005;173:377-9).
The study followed 191 Montreal schoolchildren (average age 9 at baseline and 13 at follow-up) for four years, examining the relationship between secondhand smoke exposure at baseline (measured by salivary cotinine levels) and smoking initiation (defined as having smoked at least one cigarette per week for one month or more). The analysis revealed that controlling for socioeconomic status and household smoking, secondhand smoke exposure was a significant predictor of progression to smoking.
The study also found that increased lung capacity (forced vital capacity) was a significant predictor of smoking initiation among the older children. The paper concludes that “lung size (or some associated characteristic) increases the uptake of environmental tobacco smoke, maximizes the influence of passive smoking in childhood and induces smoking in adolescence.”
Overall, the paper concludes that “enhanced susceptibility to environmental tobacco smoke in childhood increases the risk of nicotine-seeking behavior in adolescence.” The media reported this study as having suggested that exposure to secondhand smoke as children may lead adolescents to later take up smoking.
The Rest of the Story
Before adding smoking to the list of things caused by secondhand smoke exposure, I think one must strongly consider the possibility that the observed association between secondhand smoke and risk of smoking was a spurious one, due to a third factor (known as a confounding variable) that is related both to secondhand smoke exposure and to smoking initiation.
There are a number of probable confounding variables in this study. To start, parental approval or disapproval of smoking and household smoking policies may well explain the observed relationship. Although the study controlled for parental smoking, it is not the case that all parents who smoke have the same attitudes about smoking and transmit those same attitudes to their children. Some parents who smoke may still send negative messages about smoking to their kids and some parents who don’t smoke may fail to express a disapproval of smoking.
In addition, parents who smoke may or may not institute rules about smoking in the household. These rules send kids an important message about how smoking is viewed in the household.
It is very likely that the levels of disapproval of smoking and the nature of household smoking policies are strongly related to the level of secondhand smoke exposure experienced by youths. For example, if parents do not allow smoking in the household, their kids are more likely to get the message that they disapprove of smoking. But they will also be less likely to be exposed to secondhand smoke. So levels of secondhand smoke exposure are almost certainly related both to household smoking policies and to parental disapproval of smoking.
And in turn, each of these two factors is known to be related to the risk of smoking initiation. So the observed relationship between secondhand smoke and smoking initiation could well be due to the fact that secondhand smoke exposure implies a lower level of parental disapproval of smoking and more lenient household smoking policies, rather than to the hypothesis that secondhand smoke exposure somehow causes kids to start smoking.
Another likely confounding variable is peer smoking. Children whose parents’ smoke are also more likely to have friends who smoke and it has been well-established that peer pressure is a significant factor in smoking initiation.
As a reviewer of this paper, I would go so far as to say that I don’t think it provides any solid evidence that exposure to secondhand smoke as a child increases one’s risk of starting to smoke. I find it to be a huge stretch to conclude that secondhand smoke exposure is a cause of youth smoking and I think it is also a stretch to suggest that biologic susceptibility to secondhand smoke exposure (such as increased lung capacity) increases the risk of smoking in adolescence.
The rest of the story suggests that secondhand smoke exposure has been implicated as a cause of active smoking when in fact well-known causes of smoking, such as parental attitudes towards smoking and peer smoking are the more likely explanations of this observed relationship. Secondhand smoke exposure is, in my opinion, related to a number of adverse health conditions, but there is presently no reason to believe that smoking is one of them.