研究背景：宫颈癌在中国的发病率及死亡率逐年上升。研究表明人群筛查能够有效降低宫颈癌死亡率。如何提高适龄妇女参与度及筛查率是当下一大挑战。目的意义：从宫颈癌筛查供需双方视角入手，了解影响其参加筛查的因素，提出改进措施以提高适龄妇女及医务人员的参与度。研究方法：于2022年10月~2023年1月，在深圳市宝安区及南山区采用方便抽样选取312名社区女性开展调查。问卷包含人口学、认知态度、社会支持、对医疗的信任程度四部分。用SPSS进行分析，频数及百分比对人口学资料进行描述性统计；中位数和四分位数间距描述女性参加筛查的社会心理因素；根据有无宫颈癌筛查史，将女性分为已参与筛查及未参与筛查，用?2 检验和Mann-Whitney U比较两组女性差异；使用二元Logistic回归分析筛查参与的影响因素。供方通过扎根理论法对深圳市宝安区内专家访谈资料进行分析，得出参与筛查影响因素；专家对因素重要性评分，确定最终因素集；最后邀请12名专家对因素互相的影响程度进行打分，利用DEMATEL、ISM等方法分析打分结果。研究结果：（1）参与研究的312名女性中，有48.7%有过筛查经历，有59.6%知晓相关的筛查项目；（2）年龄、婚姻状况、户籍地、学历、在职情况、家庭人均月收入、医保类型、吸烟频率、饮酒频率、体育锻炼频率、健康体检频率、对宫颈癌认知态度、感知社会支持、对医疗服务信任度、对筛查的政策知晓度对筛查行为的影响有统计学差异（P<0.05）；（3）年龄、婚姻状况、医保类型、体育锻炼频率、健康体检频率、对医疗服务信任度、对筛查政策的知晓度可预测筛查参与情况（P<0.05），最终回归模型对筛查行为预测率为90.1%；（4）影响供方持续参与筛查的因素共12个，涵盖外部环境、流程、资源、参与人四个维度。使用DEMATEL、ISM等方法综合分析后，得出12个因素的重要性顺序，并将其划分为五层递阶模型，因素间逐层向上传递影响，最终影响供方的参与意愿。研究结论：（1）年龄、婚姻状况、医保类型、体育锻炼频率、健康体检频率、对医疗服务信任度、对筛查政策的知晓度可正向预测女性筛查参与率，在制定干预措施时应重点关注；（2）改善交流合作、参与人员感知成就感和心理风险可快速改善供方参与意愿；改善政策支持和社群氛围会持续影响供方参与意愿；改善指南内容质量、加强人员保障会对供方参与意愿造成较大影响。

Background: The incidence and mortality rates of cervical cancer are increasing in China and population screening can effectively reduce cervical cancer mortality. How to improve the participation rate and screening rate among eligible women is a major challenge at present.Objective: From the perspectives of supply and demand for cervical cancer screening, this study aimed to understand the factors that influence participation in screening and to propose improvement measures to increase the willingness of participation of eligible women and healthcare professionals.Methods: From October 2022 to January 2023, 312 community women were conveniently selected in Baoan and Nanshan Districts of Shenzhen city, and a questionnaire containing demographic, cognitive, social support, and trust in medical services was used. Descriptive statistics were used to describe the demographic data by frequency and percentage. The social and psychological factors of women participating in screening were described by median and interquartile range. According to whether they had a history of cervical cancer screening or not, women were divided into those who had participated in screening and those who had not, and the differences between the two groups were compared using chi-square test and Mann-Whitney U. Binary logistic regression analysis was used to analyze the factors affecting screening participation. From the perspective of supply, a grounded theory method was used to analyze expert interview data in Baoan District of Shenzhen and determine the most important factors by expert ratings. Finally, 12 experts were invited to score the influence of factors on each other, and DEMATEL、ISM was used to analyze the expert score results. All analyses were used SPSS.Results: Among the 312 women who participated in the study, 48.7% had a history of screening, and 59.6% were aware of relevant screening projects. Age, marital status, household registration, education level, employment status, average monthly household income, medical insurance type, smoking frequency, alcohol consumption frequency, physical exercise frequency, health examination frequency, attitude towards cervical cancer, perception of social support, trust in medical services, and awareness of screening policies had significant differences in screening behavior （P＜0.05）. Age, marital status, medical insurance type, physical exercise frequency, health examination frequency, trust in medical services, and awareness of screening policies could predict screening participation （P＜0.05）. The final regression model had a predictive rate of 90.1% for screening behavior. There were 12 factors that affected the continued participation of the supply side in screening, covering four dimensions of external environment, process, resources, and participants. After comprehensive analysis using methods such as DEMATEL and ISM, a sequence of importance for 12 factors was obtained and divided into a five-layer hierarchical model. The factors transmit influence upward between layers, ultimately affecting the supplier‘s willingness to participate.Conclusions: Age, marital status, medical insurance type, physical exercise frequency, health examination frequency, trust in medical services, and awareness of screening policies can positively predict the screening participation rate among women. These factors should be given priority in the development of intervention measures. Improving communication and cooperation, enhancing the sense of achievement and psychological risk perception of participants, and improving policy support and community atmosphere can rapidly improve the willingness of the supply side to participate. Improving guideline content quality and strengthening personnel protection will have a significant impact on the willingness of the supply side to participate.