Perceptions, secondhand smoke exposure and smoking among Hong Kong children

Abstract

Background: Although perception predicts behaviours in general, it is less clear whether perceived peer smoking prevalence is associated with smoking among children. Although the harmful effects of smoking and secondhand smoke (SHS) exposure on health are well documented, little is known about the influence of SHS exposure on smoking behaviours, and the effects of smoking on perceived health.

Objectives: This thesis aims to investigate the associations of (1) overestimation of peer smoking prevalence and smoking initiation, (2) self-rated health (SRH) with smoking and SHS exposure, and (3) SHS exposure at home with smoking initiation, tobacco addiction and smoking cessation, among an under-studied population of Chinese children in Hong Kong.

Methods: A prospective study among primary school students (n = 2171) was conducted in 2006 (baseline) and followed up in 2008. A cross-sectional study on youth smoking (n = 36225) was conducted in 2003/4. Self-administered, anonymous questionnaires were used to collect information on socio-demographic characteristics, smoking status of the subjects, parents, siblings and other family members and SHS exposure at home and outside home. Other detailed information on smoking perceptions and behaviours were recorded among adolescents using standard questions adapted from the Global Youth Tobacco Survey. Self-reported smoking and SHS exposure were validated using hair nicotine tests. Logistic and linear regressions were used to yield adjusted odds ratio (AOR) and β-coefficients for poor SRH, smoking initiation, tobacco addiction and smoking cessation adjusting for socio-demographic characteristics and school clustering effects. Mediation of SHS exposure at home on the associations of smoking family with smoking initiation, tobacco addiction and smoking cessation were evaluated using standard mediation procedures.

Results: Experimental smoking, ex-smoking and current smoking were associated with higher AORs for poor SRH than never-smoking among adolescents. Similarly, SHS exposure among never-smoking adolescents was linked to poor SRH. Overestimation of peer smoking prevalence significantly predicted smoking initiation (AOR: 1.79, 95% CI 1.03-3.13) among primary school students after 2 years, although SHS exposure did not predict overestimation. Living in smoking families and SHS exposure at home were significantly associated with smoking initiation and tobacco addiction, and hindered smoking cessation among adolescents (All p<0.05). The association between smoking family and child smoking was mediated by SHS exposure at home.

Conclusions: This study provided the first evidence that overestimation of peer smoking predicted smoking initiation among young children, and smoking was associated with poor SRH among Chinese adolescents. The significant associations of SHS exposure at home with smoking behaviours should stimulate further investigations into the underlying psychosocial and physiological mechanisms. Future prospective studies with more comprehensive measurements of smoking and SHS exposure are warranted to confirm these results. Evidence-based interventions to correct misperceptions and promote smokefree homes are needed to protect children from smoking and exposure to SHS.