Social norms change and cigarette smoking: a longitudinal study in mainland China

Cigarette smoking is a global epidemic, and is particularly pronounced in China, the world's largest manufacturer and consumer of cigarettes. Price-based approaches are effective for tobacco control, but can falter in settings with weak policy enforcement. Changing social norms around cigarette smoking may reduce smoking through the exertion of anti-smoking social influence even in the absence of strong regulatory conditions. However, evidence of the social norms-smoking relationship over time at a population level in China is severely lacking.

We fit Bayesian latent growth curve models to understand how smoking-related descriptive norms (perceptions of others' smoking), injunctive norms (perceptions of others' approval of smoking), and cigarette smoking unfold over time. We used longitudinal data from adults who smoke in 4 cities over 9 years (n = 5400 observations), provided by the International Tobacco Control Policy Evaluation Project in Mainland China. Self-reported smoking status was defined as smoking ≥100 cigarettes in one's lifetime and currently smoking at least 1 cigarette per week.

Higher pro-smoking descriptive and injunctive norms were associated with increases in cigarette smoking over time, even after controlling for known confounders such as addiction level, exposure to household smoking, and exposure to pro-smoking media. Higher pro-smoking descriptive norms were associated with both higher initial level of cigarette smoking and slower rates of smoking reduction over time.

Social norms impact smoking over time among middle-aged people who smoke in China. Tobacco control efforts should support anti-smoking descriptive norms through group-based interventions and creating social networks of those who do not smoke.

This is the first study to examine the dynamics of change related to social norms and smoking over time among people who smoke in Mainland China. We find that increases in pro-smoking social norms are associated with increases in cigarette smoking, even after controlling for known confounders such as exposure to household smoking, addiction level, tobacco-related knowledge, and more. While both increases in pro-smoking injunctive and descriptive norms are associated with slower smoking reduction, descriptive norms appear consistently consequential, with reductions in pro-smoking descriptive norms associated with greater cigarette smoking reduction over time. Group-based interventions leveraging the salient reduction in proximal referents' smoking appear especially promising for tobacco control efforts.