论文部分内容阅读
目的基于对中国疾病预测研究的发展沿革、预测方法及研究瓶颈分析,旨在通过优化慢性病趋势预测模式为中国慢性病防治提供一定的理论依据。方法通过文献荟萃分析,系统梳理中国慢性病预测发展现状及瓶颈,分析优化预测模式。结果中国慢性病预测重视度不足,人群发病率预测较匮乏,方法学应用仍停留在线性或多元回归层面。从人口、经济、社会3个范畴筛选出影响因素变量构建状态空间模型,该优化模式比其他的时间序列自回归模型的拟合优度更高。结论状态空间模型用于构建特定区域的慢性病趋势预测模型,可大大提高长期预测的精度和灵敏性,为循证决策提供强有力支撑。
OBJECTIVE Based on the development of Chinese disease prediction research, prediction methods and research bottlenecks, this paper aims to provide some theoretical basis for the prevention and treatment of chronic diseases in China by optimizing the trend prediction model of chronic diseases. Methods By meta-analysis of literature, systematically combing the status quo and bottleneck of the prediction of chronic disease in China, analysis and optimization of prediction model. Results In China, the degree of attention to chronic disease was not enough, and the incidence of population was scarce. The application of methodology still remained at the level of linear or multiple regression. From the population, economy and society, we selected three influential variables to construct the state space model, which has a higher goodness of fit than other time series autoregressive models. Conclusion The state space model is used to construct the trend prediction model of chronic diseases in a specific area, which can greatly improve the accuracy and sensitivity of long-term prediction and provide strong support for evidence-based decision-making.