论文部分内容阅读
目的基于潜在剖面模型识别广州市居民健康知识知晓模式,从多维度识别异质性人群的特征。方法通过分层整群抽样选取广州市1179名居民进行健康知识和健康行为的问卷调查,采用潜在剖面模型依据健康知识得分对人群分类,并与传统聚类方法进行比较。分析不同健康知识知晓模式的人口学特征和健康行为。结果(1)潜在剖面分析将被调查居民分成高、中、低三类健康知识知晓模式,分别占62.2%、27.4%和10.4%,其分类效果优于传统聚类法。(2)除性别外,健康知识知晓高、中、低三类人群的年龄(χ~2=10.431,P=0.034)、文化程度(χ~2=49.510,P<0.001)、职业类型(χ~2=20.781,P<0.001)差异均有统计学意义,较低水平健康知识知晓人群以18~44岁、初中及以下文化程度、体力型劳动者居多。(3)除吸烟外,健康知识知晓高、中、低三类人群在是否吃早餐(χ~2=25.763,P<0.001)、睡眠时间>7 h(χ~2=7.483,P=0.024)、是否运动量足够(χ~2=8.317,P=0.016)和是否健康体检(χ~2=6.909,P=0.032)等方面差异有统计学意义。结论潜在剖面模型应用到健康教育领域,可有效识别出不同健康知识知晓情况的异质亚群体,揭示健康教育的重点人群和内容。
OBJECTIVE: To identify the patterns of health knowledge awareness of residents in Guangzhou based on potential profile models and to identify the characteristics of heterogeneous populations from multiple dimensions. Methods The stratified cluster sampling was used to select 1179 residents in Guangzhou for questionnaire survey on health knowledge and health behaviors. The potential cross-sectional model was used to classify the population according to health knowledge scores and compared with the traditional clustering method. Demographic characteristics and health behaviors of different patterns of health knowledge awareness were analyzed. Results (1) Potential profile analysis will be divided into high, medium and low levels of health awareness among residents surveyed, accounting for 62.2%, 27.4% and 10.4% respectively, and the classification results are better than the traditional clustering method. (2) In addition to gender, the health knowledge of age, age, education level (χ ~ 2 = 10.431, P = 0.034) ~ 2 = 20.781, P <0.001). There was a statistically significant difference between the groups with lower level of health knowledge and those with 18-44 years of age, junior middle school and below, with most of the manual laborers. (3) In addition to smoking, health knowledge of high, middle and low groups of people eat breakfast (χ ~ 2 = 25.763, P <0.001), sleep time> 7 h (Χ ~ 2 = 8.317, P = 0.016), and whether or not physical examination (χ ~ 2 = 6.909, P = 0.032) were statistically significant. Conclusion The potential profile model is applied to the field of health education, which can effectively identify heterogeneous sub-groups with different health knowledge and reveal the key population and content of health education.