我国居民健康状况的时空特征及预测研究

Study on the Spatial and Temporal Characteristics and Prediction of Chinese Residents′Health Status

目的构建居民健康指数,研究我国居民健康水平的时空变化及其影响因素并进行预测。方法收集2005-2019年31个省份的死亡率、围产儿死亡率、孕产妇死亡率、传染病发病率以及相关影响因素等数据。使用熵值法、变异系数法、因子分析法对相关指标进行单一赋权,对上述三种方法求得的权数进行线性组合,然后采用加权求和法计算不同省份的居民健康指数。采用空间杜宾模型进行居民健康水平的影响因素分析。分别使用ArcGis 10.2和Stata 15.0软件绘制居民健康水平分布地图及进行空间分析。使用R 4.0.5和SPSS 22.0软件对我国居民健康水平进行求和自回归移动平均模型预测分析。结果(1)2005-2019年居民健康指数从0.60上升为0.68;(2)2005年和2019年中国居民健康指数的Moran′s I值分别为0.234和0.290(P<0.05);(3)空间回归模型中,人均GDP、65岁以上人口所占比重和PM 2.5系数分别为0.086、-0.156和-0.035(P<0.05)。(4)预测得出2020~2022年居民健康水平呈现上升趋势(0.68到0.70)。结论(1)2005-2019年居民健康水平总体呈上升趋势。(2)居民健康水平分布存在空间聚集性。(3)随着人均收入水平以及环境质量的提高,65岁以上人口所占的比重降低,居民的健康水平将升高。(4)未来居民健康水平将稳定上升。

Objective To build a health index and study the temporal and spatial changes of the health level of Chinese residents and its influencing factors and make predictions.Methods We collected data on mortality,perinatal mortality,maternal mortality,infectious disease incidence and related influencing factors in each province from 2005 to 2019.We used entropy method,coefficient of variation method and factor analysis method to weight relevant indicators.Linear combination weight was used to determine the final weight.Spatial Durbin Model was used to analyse the affecting factors of health.ArcGIS 10.2 software and Stata 15.0 software were used to draw the residents′health level distribution map and spatial analysis.R 4.0.5 and SPSS 22.0 software were used to predicte and analyse the health level of Chinese residents by Autoregressive Integrated Moving Average model.Results(1)The resident health index rose from 0.60 to 0.68 from 2005 to 2019;(2)Moran′s I values of the Chinese resident health index in 2005 and 2019 were 0.234 and 0.290(P<0.05),respectively;(3)The per capita GDP、proportion of the population over the age of 65 and PM 2.5 coefficient in spatial regression model were 0.086、-0.156 and-0.035(P<0.05).(4)It is predicted that the health level of residents will show an upward trend in 2020-2022(from 0.68 to 0.70).Conclusion(1)From 2005 to 2019,the health level of Chinese residents has shown an overall upward trend.(2)The distribution of health levels in China has certain spatial agglomeration characteristics.(3)The per capita GDP、proportion of the population over the age of 65 and PM 2.5 have significant impacts on the health of residents.(4)In the prediction of the health level of residents,the health index rises,and the health level of the residents is in a stable state.