江西农村地区传染病症状监测数据报告质量及其影响因素研究

目的 分析农村地区传染病症状监测系统中不同数据源数据上报质量及其影响因素。方法 采用多阶段整群抽样和目的抽样的方法在江西省两县抽取162家医疗机构、62家小学和30家药店作为监测点。连续两年收集患者主诉、非处方药物销量和小学生因病缺勤记录并进行质量控制，采用自行设计的调查表对数据报告员开展上报数据质量及影响因素的调查。结果 监测期间共收集到医疗机构门诊症状报告记录504 082条，药店非处方药物销售记录916 669条，小学缺勤记录18 402人次。医疗机构、药店、小学的及时报告率从第1个月的78.38%、71.27%和81.52%上升到了第24个月的99.50%、94.94%和99.00%。村级数据报告员及时上报率增幅为57.4%高于县乡级报告员的38.6%（OR=2.90，95%CI：1.28～6.57）；学历较低的数据报告员及时上报率改善明显（OR=2.36，95%CI：1.07～5.21）。参与计算机技术培训可以提高报告人员的及时上报率（OR=2.72，95%CI：1.41～5.24）。结论 在中国农村建立多数据源症状监测系统有一定的可行性，症状监测系统数据报告质量随时间推移逐步提升并稳定。常规数据报告的质量控制、必要的计算机操作技术培训有助于改善数据报告质量。

Evaluation of data quality and influencing factors of integrated syndromic surveillance in rural Jiangxi Province,China

Objective To assess the data reporting quality, and to explore potential influencing factors. Methods Multistage sampling and purposive sampling were used to select 162 health facilities, 62 primary schools and 30 pharmacies as study sites in two counties of Jiangxi Province. Chief complaints, over-the-count (OTC) drug sales and absenteeism were recorded and quality supervision was conducted. Self-developed questionnaires were used to investigate the data reporting quality and influencing factors. Results During the two years' implementation, the Integrated Surveillance System in Rural China (ISSC) collected 504 082 records of outpatients with specific syndromes, 916 669 records of OTC-drug sales and 18 402 records of school absenteeism. The timely report rates of the three data sources raised from 78.38%, 71.27% and 81.52% in the first month to 99.50%, 94.94% and 99.00% in the last month. Data reporters with a higher timely report improvement were those from village areas (OR=2.90, 95% CI:1.28-6.57), less educated (OR=2.36, 95% CI:1.07-5.21) and having been educated with computer operation skills during the research program (OR=2.72, 95% CI:1.41-5.24). Conclusions Syndromic surveillance data is proved to be collectable in resource limited settings. Quality of data improved greatly with the help of a set of supervision measures. The timely report rate stabled above 90% across all three data sources after one years' implementation suggests that it is feasible to establish an integrated syndromic surveillance system in the long run.