Perceived peer smoking prevalence and its association with smoking behaviours and intentions in Hong Kong Chinese adolescents

Abstract

Background: Among the many personal, social and environmental risk factors of adolescence smoking, normative beliefs stand out for their potential to be modified with factual information on smoking prevalence. Aims: To study the perceived peer smoking prevalence and its association with smoking behaviours in Hong Kong Chinese adolescents. Design and setting: Cross-sectional territorial-wide school-based survey conducted in 64 randomly selected secondary schools in Hong Kong. Participants: A total of 13 280 forms 1-3 students (equivalent to grades 7-9 in the United States) aged 12-16 years. Measurements: Perceived peer smoking prevalence, smoking status, intention to smoke in future, other smoking-related factors and demographic information. Findings: Overestimation of peer smoking prevalence was observed regardless of gender and smoking status, and was more common in girls (69.4%) than boys (61.0%), and in experimental (74.3%) and current smokers (85.4%) than in never smokers (60.7%). Boys who overestimated and grossly overestimated (over two times) peer smoking were more likely to be current smokers, with adjusted odds ratios and 95% confidence intervals (95% CI) of 1.95 (1.24-3.07) and 3.52 (2.37-5.24) (P for trend <0.001). Similarly, boys who grossly overestimated peer smoking were 76% (95% CI: 41-120%) more likely to have ever smoked. Conclusion: Overestimation of peer smoking prevalence was common in Hong Kong Chinese boys and girls, and was associated with current and ever smoking in boys. These findings have important implications on normative education in adolescence smoking prevention programmes.