Abstract
BACKGROUND: Many countries, including China, have implemented nationwide smoking cessation programs in accordance with Article 14 of the World Health Organization (WHO) Framework Convention on Tobacco Control (FCTC). However, the use of assisted smoking cessation services remains low, while unassisted smoking cessation (USC) methods are widely preferred. Studies have shown that many individual and environmental factors are associated with USC adoption. However, no studies have examined the association between social networks and USC adoption. OBJECTIVE: This study examined the effects of social networks on USC choice among male smokers in China. METHODS: A cross-sectional multivariable sampling design was used to interview subjects from six selected cities in China. The study sample included only male participants. A standardized questionnaire was used to obtain information about sociodemographic characteristics, social networks, and USC choice. Multivariable logistic regression models were used to examine the association between social networks and USC choice. Furthermore, quantitative analysis was conducted to demonstrate the dose-response relationship between the city-level social networks and the prevalence of USC. Structural equation modeling (SEM) was used to establish the mechanisms by which reference group norms affect USC through social networks. RESULTS: We identified 2852 smokers, of whom 1647 (57.7%) had attempted to or had quit smoking. Among them, 91.6% (n=1509; 95% CI 90.9%-97.5%) reported quitting without assistance, and 42.1% (n=58; 95% CI 32.4%-61.3%) of the remaining 138 (8.4%) participants who used USC methods achieved abstinence. Multiple logistic regression analysis found that although the higher social network (HSN) is not significantly associated with USC adoption, both the ordinary social network (OSN) and the lower social network (LSN) are significantly negatively associated with USC adoption. Compared to the low-size group of the OSN, the odds ratios (ORs) for the medium-, high-, and very high-size groups were 0.23 (95% CI 0.10-0.80), 0.22 (95% CI 0.07-0.60), and 0.25 (95% CI 0.08-0.72), respectively. Similarly, compared to the low- and medium-size groups of the LSN, the OR for the high-size group was 0.32 (95% CI 0.23-0.45). The analysis revealed a significant positive dose-response relationship of city-level OSN and LSN sizes with the probability of USC adoption, as indicated by regression coefficients β=0.2784 (P<.01) and β=2.2269 (P<.01), respectively. SEM analysis indicated that the size of OSNs and LSNs exerts an indirect effect on the relationship between reference group norms and USC (β=0.2315, P<.01; β=0.4613, P<.01). CONCLUSIONS: Both OSNs and LSNs are significantly negatively associated with USC adoption. The findings underscore the significance of USC and its potential role in reducing smoking prevalence within the Chinese population.