Objective To understand the associations between different smoking behavior characteristics and sleep quality among Chinese male residents aged 15 years and above, thus providing scientific evidence for developing relevant intervention measures.

Methods From March to October 2024, a multi-stage stratified cluster random sampling method was used to select survey subjects from 70 counties/districts in 7 provinces/autonomous regions/municipalities across China. Face-to-face interviews were conducted with electronic questionnaires, and 23 333 individuals were ultimately included in the analysis. The Rao-Scott χ² test based on complex sampling design was conducted to compare demographic characteristics of male residents of different smoking status, and univariate analysis was performed to examine the associations between different smoking behavior characteristics and sleep quality. Multivariate logistic regression was adopted to analyze the associations of different smoking status and smoking behavior characteristics (smoking index, daily cigarette consumption, and smoking duration) with sleep quality among males.

Results Among the 22 333 male residents aged 15 and above in China, 44.92% (8 561), 42.84% (9 905), and 12.24% (3 867) were non-smokers, current smokers, and former smokers, respectively. The prevalence of poor sleep quality was 20.42%. Univariate analysis showed that the prevalence of poor sleep quality differed among male residents of different smoking status (χ² = 83.893, P < 0.000 1). In both current smokers and former smokers, there were statistically significant differences in the associations of daily cigarette consumption (χ²_{current smoker} = 26.821; χ²_{former smoker} = 10.958), smoking duration (χ²_{current smoker} = 39.380; χ²_{former smoker} = 9.078), and smoking index (χ²_{current smoker} = 26.177; χ²_{former smoker} = 9.389) with sleep quality (P < 0.05). Cochran-Armitage trend tests showed that as daily cigarette consumption (χ²_{current smoker} = –6.741, χ²_{former smoker} = –3.955), smoking duration (χ²_{current smoker} = –13.775, χ²_{former smoker} = –5.536), and smoking index (χ²_{current smoker} = –9.294, χ²_{former smoker} = –5.081) increased, the prevalence of poor sleep quality showed a linear increasing trend (P < 0.000 1). Multivariate logistic regression analysis, adjusted for confounders including age, household registration, geographic region, educational background, occupation, annual household income, alcohol consumption, exercise frequency, BMI, and chronic diseases, revealed that compared to non-smoking, both current smoking (OR = 1.312, 95%CI: 1.107–1.556) and former smoking (OR = 1.475, 95%CI: 1.173–1.854) were associated with a higher risk of poor sleep quality. Compared to non-smoking, the smoking index of <10 pack-years (OR = 1.293, 95%CI: 1.047–1.598), 10 to <20 pack-years (OR = 1.472, 95%CI: 1.177–1.841), 20 to <30 pack-years (OR = 1.470, 95%CI: 1.085–1.992), and ≥30 pack-years (OR = 1.283, 95%CI: 1.086–1.514) were associated with a higher risk of poor sleep quality. Compared to non-smoking, daily cigarette consumption of ≤10 cigarettes (OR = 1.197, 95%CI: 1.001–1.431), 11–20 cigarettes (OR = 1.421, 95%CI: 1.169-1.727), 21–30 cigarettes (OR = 1.608, 95%CI: 1.166–2.216), and >30 cigarettes (OR = 1.681, 95%CI: 1.208-2.338) were associated with a higher risk of poor sleep quality. Compared to non-smoking, smoking duration of 10 to <31 years (OR = 1.339, 95%CI: 1.109–1.618), 31 to <40 years (OR = 1.296, 95%CI: 1.093–1.536), and ≥40 years (OR = 1.327, 95%CI: 1.116–1.578) were associated with a higher risk of poor sleep quality.

Conclusions Smoking is associated with high risk of poor sleep quality, and the detrimental effect of smoking on sleep quality exhibits a dose-response relationship, suggesting that the cumulative effect of long-term nicotine exposure may be a key pathway for disrupting sleep quality.