论文部分内容阅读
目的应用线性回归移动平均预警模型,探讨广西肺结核登记趋势;通过肺结核登记警戒线及行动线的设置,及时发现异常情况并采取有效的防控措施。方法将2009-2013年的月度肺结核登记数拟合线性回归方程,将残差移动平均处理后拟合预测曲线,按照1倍和2倍移动平均后标准差设置警戒线和行动线;评价模型拟合效果;根据实际登记情况核实异常登记情况并采取相应的防控措施。结果预测模型拟合优度较佳(R方=0.544,平稳的R方=0.092,正态化BIC=11.663);全年仅7月登记数超过了行动值,经网络、电话和专项调查发现,引起7月高值的原因为高考体检集中报告和全区规划督导数据补充登记。结论该预测和预警系统对可能出现的结核病流行和暴发及时采取有效的预测,减少主观判断,为干预行动建立客观标准,实现低成本的报告登记监测和评价,可在实际工作中参考运用。
Objective To apply the linear regression moving average early warning model to investigate the trend of tuberculosis registration in Guangxi. To detect the anomalies and take effective prevention and control measures through the setting of cordon and action lines of tuberculosis registration. Methods The linear regression equations were fitted to the monthly tuberculosis registration numbers from 2009 to 2013. The prediction curves were fitted after the residuals were moving averaged, and the warning lines and action lines were set according to the standard deviation of 1 and 2 times the moving average. The evaluation model Combined effect; according to the actual registration verification of abnormal registration and take appropriate prevention and control measures. Results The predicted goodness of fit of the model was better (R = 0.544, R = 0.092, normalized BIC = 11.663). For the full year, the number of registrations in July surpassed the value of action, and was found by internet, telephone and special surveys The reason for the high value in July was the concentration of college entrance examination and the centralized reporting of supervisory examinations in the whole district. Conclusions The prediction and early warning system can make effective and timely prediction of potential tuberculosis outbreaks and outbreaks, reduce subjective judgments, establish objective standards for interventions, and implement low-cost registration monitoring and evaluation of reports, which can be used for reference in practical work.