Objective: We planned to evaluate the effectiveness of moving epidemic method (MEM) in calculating influenza epidemic threshold of 7 climatic zones in China mainland. Methods: The positive rate of influenza virus was obtained from the National Influenza Surveillance Network System from 2010/2011 to 2017/2018. We divided the 31 provinces into 7 climatic zones according to previous literatures and applied MEM to calculate the influenza epidemic threshold of 2018/2019 influenza season for these climatic zones. Sensitivity, specificity, positive predictive value and negative predictive value were calculated to evaluate the effectiveness of MEM. Results: Pre-epidemic threshold (the positive rate of influenza virus) varied from 9.66% (temperate zone) to 16.36% (subtropical zone) for 2018/2019 influenza season. The gap between pre-epidemic and post-epidemic thresholds was less than 5% except for plateau zone. The sensitivity was 86.16% (95CI:66.81%-98.23%), the specificity was 94.92% (95CI: 91.13%-98.41%), the positive predictive value was 89.87% (95%CI: 84.39%-94.38%), the negative predictive value was 92.96% (95%CI: 84.46%-99.17%). Conclusion: Overall, moving epidemic Method performs well in calculating influenza epidemic threshold in China, much better than the previous study.
目的： 评价移动流行区间法（MEM）在中国7个气候区流感流行阈值研究中的应用效果。 方法： 流感病毒阳性率数据来源于国家流感监测网络系统中2010—2011年至2017—2018年共8个流感流行季节的监测数据，参考既往文献将中国31个省份划分为7个气候区，利用移动流行区间法为7个气候区制定2018—2019年流感流行阈值及强度阈值。计算各气候区应用此方法制定流感流行阈值的灵敏度、特异度、阳性预测值和阴性预测值，以评估方法的应用效果。 结果： 7个气候区中，2018—2019年流行开始阈值（流感病毒阳性率）最小的为中温带（9.66%），最大的为亚热带（16.36%），除高原地区外，其他气候区的流行前与流行后阈值差别均小于5%。该方法在7个气候区应用的灵敏度为86.16%（95%CI：66.81%~98.23%）；特异度为94.92%（95%CI：91.13%~98.41%）；阳性预测值为89.87%（95%CI：84.39%~94.38%）；阴性预测值为92.96%（95%CI：84.46%~99.17%）。 结论： 移动流行期间法在我国流感流行阈值研究中的应用效果较好，相比既往研究有较大改善。.
Keywords: Epidemics; Influenza, human; Moving epidemic method; Threshold limit values.