1991—2015年中国九省社区城镇化水平与成人超重发生风险关系的时空演变

Spatiotemporal variations in the association of community urbanization levels with obesity incidence among Chinese adults in nine provinces

目的考察社区城镇化水平与成人超重发生风险关系的时期和地区差异。方法采用分层多阶段整群随机抽样方法,选择1991—2015年"中国健康与营养调查"中来自沿海、中部、东北和西部地区的九省(自治区)的8684位20~64岁成人为研究对象,通过城镇化指数测量社区城镇化水平,并将社区划分为高城镇化水平和低城镇化水平社区。采用分层Logistic回归模型考察城镇化水平与超重发生关系的时期和地区差异。结果 38.08%的受访者在随访过程中发生超重。高城镇化水平与沿海(OR=0.45,95%CI 0.34~0.60)、中部(OR=0.67,95%CI 0.52~0.86)、东北(OR=0.70,95%CI 0.52~0.94)的交互项有统计学意义。高城镇化水平与时期的交互项也有统计学意义(OR=0.97,95%CI 0.96~0.99)。在1991年时,除沿海地区外(OR=0.99,95%CI 0.77~1.26),高城镇化水平与超重发生在中部(OR=1.45,95%CI 1.15~1.83)、东北(OR=1.53,95%CI 1.16~2.02)和西部(OR=2.18,95%CI 1.68~2.83)呈正相关且有统计学意义。在2015年时,高城镇化水平与超重发生在沿海(OR=0.53,95%CI 0.41~0.68)和中部(OR=0.77,95%CI 0.62~0.97)呈负相关且有统计学意义;在东北(OR=0.82,95%CI 0.63~1.05)和西部地区(OR=1.17,95%CI 0.91~1.49)的关系无统计学意义。结论在中国九省(自治区)成人中,社区城镇化水平与成人超重发生风险的关系因时、因地而异。

OBJECTIVE To explore whether the association of community urbanization levels with overweight incidence changes over time and varies by region. METHODS The data is from the China Health and Nutrition Survey(1991-2015). The survey applied a multistage, random cluster design to draw samples. A total of 8684 adults aged 20-64 years were confined in nine provinces(autonomous regions). Based on an urbanicity score, communities were divided into two groups: low-and high-urbanization-level groups. The nine provinces(autonomous regions) were divided into four regions, including the Coast, Central, Northeast and West. Multilevel Logistic regression models was applied to examine whether the association between urbanization levels and overweight incidence changed over time and varied by region. RESULTS 38.08% of respondents became overweight during the follow-up surveys.The interactions between the high urbanization level with the Coast(OR=0.45, 95%CI 0.34-0.60), the Central(OR=0.67, 95%CI 0.52-0.86), and the Northeast(OR=0.70, 95%CI 0.52-0.94) were statistically significant. The interaction between the high urbanization level and period was also statistically significant(OR=0.97, 95%CI 0.96-0.99). In 1991, the Coast(OR=0.99, 95%CI 0.77-1.26), the high urbanization level was positively associated with overweight incidence in the Central(OR=1.45, 95%CI 1.15-1.83), Northeast(OR=1.53, 95%CI 1.16-2.02) and West(OR=2.18, 95%CI 1.68-2.83), with statistical significance. In 2015, the high urbanization level was significantly negatively associated with overweight incidence in the Coast(OR=0.53, 95%CI 0.41-0.68) and Central(OR=0.77, 95%CI 0.62-0.97), while the association in the Northeast(OR=0.82, 95%CI 0.63-1.05) and West(OR=1.17, 95%CI 0.91-1.49) was statistically insignificant. CONCLUSION There were spatiotemporal variations in the association between urbanization levels and overweight incidence among adults in the nine provinces(autonomous regions).