Louisiana State University, Baton Rouge, Louisiana, USA.
Obesity (Silver Spring). 2014 Mar;22(3):868-74. doi: 10.1002/oby.20502. Epub 2014 Feb 6.
This study used spatial statistical methods to test the hypotheses that county-level adult obesity prevalence in the United States is (1) regionally concentrated at significant levels, and (2) linked to local-level factors, after controlling for state-level effects.
Data were obtained from the Centers for Disease Control and Prevention and other secondary sources. The units of analysis were counties. The dependent variable was the age-adjusted percentage of adults who were obese in 2009 (body mass index >30 kg/m2).
The prevalence of county-level obesity varied from 13.5% to 47.9% with a mean of 30.3%. Obesity prevalence across counties was not spatially random: 15.8% belonged to high-obesity regions and 13.5% belonged to low-obesity regions. Obesity was positively associated with unemployment, outpatient healthcare visits, physical inactivity, female-headed families, black populations, and less education. Obesity was negatively correlated with physician numbers, natural amenities, percent ≥65 years, Hispanic populations, and larger population size. A number of variables were notable for not reaching significance after controlling for other factors, including poverty and food environment measures.
The findings demonstrate the importance of local-level factors in explaining geographic variation in obesity prevalence, and thus hold implications for geographically targeted interventions to combat the obesity epidemic.
本研究采用空间统计方法检验了以下两个假设：在美国，县级成年人肥胖患病率（1）在区域上集中到显著水平，以及（2）在控制州级效应后与地方级因素相关。
数据来自疾病控制与预防中心和其他二级来源。分析单位为县。因变量为 2009 年年龄调整后肥胖成年人的比例（身体质量指数>30kg/m2）。
县级肥胖患病率从 13.5%到 47.9%不等，平均为 30.3%。各县之间的肥胖患病率分布不均匀：15.8%属于高肥胖地区，13.5%属于低肥胖地区。肥胖与失业率、门诊医疗就诊次数、身体活动不足、女性为户主家庭、黑人人口和教育程度较低呈正相关。肥胖与医生人数、自然美景、≥65 岁人口比例、西班牙裔人口和较大的人口规模呈负相关。一些变量在控制其他因素后，包括贫困和食品环境措施，都没有达到显著水平。
研究结果表明，地方因素在解释肥胖患病率的地域差异方面具有重要意义，因此对于针对地理区域的干预措施以对抗肥胖流行具有影响。
