广州市妇女参加宫颈癌筛查影响因素的结构方程模型分析

目的 探讨影响城市社区妇女参加宫颈癌筛查的因素,为有效开展宫颈癌筛查提供依据.方法 采用整群随机抽样方法,自行设计宫颈癌认知、参与筛查的态度和行为调查问卷对广州市基层社区的813名妇女进行调查.采用结构方程模型对调查数据进行统计分析 结果 被调查的813名妇女近3年宫颈癌筛查率为48.3％.结构方程模型分析结果表明,“客观因素”、“危险因素认知”、“防治知识”及“筛查支持感知”等4个外生潜变量对内生潜变量“筛查积极态度”有正向影响作用,其标准化路径系数分别为0.21、0.13、0.21、0.22,均有统计学意义(均有P＜0.05);“筛查积极态度”进而影响妇女参加宫颈癌筛查的时间间隔,其标准化路径系数为0.16(P=0.001),即态度积极者参加筛查的时间间隔较短.外生潜变量“筛查支持感知”还直接影响筛查的时间间隔,是妇女参加宫颈癌筛查的重要促进因素.模型的拟合效果良好,其卡方自由度比(x2/df)为2.99、拟合优度指数(GFI)为0.914、近似误差均方根(RMSEA)为0.057.结论 提高妇女参加宫颈癌筛查的积极性,除了要加强对妇女自身的健康教育外,还要提高其对筛查支持的感知度,比如同时也对其家庭、所在社区或工作单位普及宫颈癌防治知识.

Analysis of the factors affecting cervical cancer screening in the community population of Guangzhou City via astructural equation model

Objective To investigate the influencing factors of cervical cancer screening participation and to point out the key directions for the health education in view of cervical cancer prevention and control. Methods 813 women were collected as respondents from some communities of Guangzhou City via cluster random sampling method. Structural e- quation model （SEM） was used to explore the influencing factors of cervical cancer screening. Results In this study, the screening rate for cervical cancer at least once every three years was 48.3%. The analysis results of structural equation model indicate that the exogenous latent variables including ＂objective conditions＂, ＂cognition of risk factors＂, ＂knowl- edge of prevention＂ and ＂perceptions of support for screening＂ had positive impact on the endogenous latent variable ＂posi- tive attitude toward screening＂. The standardized path coefficients were 0. 21,0. 13, 0. 21 and 0. 22, respectively, and all of them were statistically significant （ all P 〈 0. 05 ）. Positive attitude had further impact on screening intervals. Its stand- ardized path coefficient was 0. 16 （P = 0. 001 ）. The latent variable ＂perceptions of support for screening＂ also had direct impact on screening intervals, which indicated an important promotive effect on cervical cancer screening. The main good- ness-of-fit indices were, x2/df = 2.99, GFI = 0. 914 and RMSEA = 0. 057, which demonstrated a good imitative effect of the structural equation model. Conclusions It is necessary to implement health education focusing on the women with poor knowledge of cervical cancer and their families, and the entire communities where they live in order to elevate women＇ s perceptions of support for screening.