无烟烟草
人口学
多项式logistic回归
烟草制品
年轻人
烟草控制
医学
人口
烟草使用
环境卫生
心理学
老年学
公共卫生
数学
统计
护理部
社会学
作者
Melissa D. Blank,Nicholas A. Turiano,Bethany C. Bray,Andrea R. Milstred,Margaret G. Childers,Geri Dino,Katelyn F. Romm
摘要
Abstract Background and Objectives This study examined young adults’ tobacco use transitions based on their past 30‐day use states, and identified factors associated with their transitions. Methods Participants ( N = 12377) were young adults aged 18‐29 years at Wave 1 of the Population Assessment of Tobacco and Health (PATH) study. Self‐reported tobacco use states were categorized by the number of past‐month use days (0, 1–4, 5–8, 9–12, 13–30 days) for cigarettes, electronic cigarettes [e‐cigarettes], traditional cigars, filtered cigars, cigarillos, smokeless tobacco (SLT), and hookah. Multistate Markov models examined transitions between use states across Waves 1–5 of unweighted PATH data and multinomial logistic regressions examined predictors of transitions. Results Most young adults remained nonusers across adjacent waves for all products (88%–99%). Collapsed across waves, transitioning from use at any level to nonuse (average 46%–67%) was more common than transitioning from nonuse to use at any level (average 4%–10%). Several factors that predicted riskier patterns of use (i.e., transitioning to use and/or remaining a user across adjacent waves) were similar across most products: male, Black, Hispanic, lower education levels, and lower harm perceptions. In contrast, other factors predicted riskier patterns for only select products (e.g., e‐cigarette and SLT use among Whites). Discussion and Conclusions Few sampled young adults escalated their tobacco use over time, and escalations for many products were predicted by similar factors. Scientific Significance Prevention and regulatory efforts targeted towards adolescents should continue, but also be expanded into young adulthood. These same efforts should consider both shared and unique factors that influence use transitions.
科研通智能强力驱动
Strongly Powered by AbleSci AI