The Geographic/Environmental Predicting Models Of PaCO₂ Reference Values In Healthy Middle-aged And Elderly People From China

Carbon dioxide partial pressure in arterial blood(PaCO2)is an index which is commonly used in clinical blood gas analysis.Its reference value plays important roles in the diagnosis and treatment of common respiratory disease.For instance,the status of oxygenation,ventilation and acid-base balance of respiratory system in clinical patience with chronic obstructive pulmonary disease can be determined by observing the values of carbon dioxide partial pressure in arterial blood.In the field of health geography-reference value of carbon dioxide partial pressure in arterial blood is mainly concentrated on carbon dioxide partial pressure in arterial blood distribution in different geographical environments.However,studies based on this point of view are poorly conducted.Moreover,there is lack of standard in reference values of carbon dioxide partial pressure in arterial blood for healthy people in different geographical backgrounds in China.Therefore,it is of great practical significance in the research on the distribution of carbon dioxide partial pressure in arterial blood under the perspective of geographical environments and in establishing carbon dioxide partial pressure in arterial blood standard ranges for healthy people in different regions in China.This study mainly focuses on the geographical distribution of carbon dioxide partial pressure in arterial blood reference values for healthy middle-aged and elderly people in China,and establishes appropriate carbon dioxide partial pressure in arterial blood reference value prediction model for healthy middle-aged and elderly people,so as to provide scientific basis for carbon dioxide partial pressure in arterial blood prediction in China.In this study,the carbon dioxide partial pressure in arterial blood data of 5677 healthy middle-aged(40-59 years old)and 5691 elderly(over 60 years old)in different regions of China were firstly collected through literature search.Total 20 factors including topography,climate and soil indices(altitude,longtitude,latitude,annual sunshine duration,annual average temperature,annual average relative humidity,annual rainfull,annual range of temperature,annual average wind speed,sand percentage,clay percentage,silt percentage,soil pH,soil salinity,soil bulk density,soil organic matter content,clay cation exchange capacity,powder soil cation exchange capacity,total carbonate content and total sulfate content)were selected as basis data of geographical environments.The data of geographical and environmental factors were combined with measured PaCO2 reference values to make model analysis.On the basis of spatial autocorrelation,correlation analysis and redundancy analysis,multiple linear regression model,curve estimation and standard deviation model,and BP network model were established,respectively.The optimal models of each model were obtained by means of statistical evaluation.The three optimal models were combined and changed into combination models by methods of average,standard deviation,simple weighted average and inverse of prediction error sum of squares.The combination model assessments were conducted by means of the evaluation of mean square error(MSE),sum of square error(SSE),mean absolute error(MAE)and mean relative error(MRE).Then,the optimal combination prediction model of PaCO2 reference values for healthy middle-aged and elderly people was obtained.The two combination models were also the final optimal equations for predicting PaCO2 reference values of Chinese healthy middle-aged and elderly people in this study.The main conclusions are as follows:(1)According to analyses of correlation and redundancy,the reference values of PaCO2 in healthy middle-aged and elderly people showed strong correlations with altitude,longtitude,annual average temperature,soil salinity,soil organic matter content,annual average wind speed,annual average relative huminity,total sulfate content and soil bulk density.Altitude is the most important factor in affecting the PaCO2 in healthy middle-aged and elderly people.The influence intensity of each geographical environment factor on values of middle-aged people was as follows:altitude>soil organic matter content>annual average temperature>longtitude>soil salinity>soil bulk density>total sulfate content>annual average wind speed>annual average relative huminity;while elderly was:altitude>longtitude>annual average temperature>soil organic matter content>soil salinity>soil bulk density>total sulfate content>annual average relative huminity>annual average wind speed.(2)The optimal equation for predicting PaCO2 reference values of Chinese healthy middle-aged people was:Y=0.044Y1+0.033Y2+0.92373The optimal equation for predicting PaCO2 reference values of Chinese healthy elderly people was:Y=0.2487Y1,+0.1379Y,+0.6134Y3Y1,Y2 and Y3 mean predicted values from multiple linear regression model,curve estimation and standard deviation model,and BP network model,respectively.(3)The predicted reference values of PaCO2 in healthy middle-aged and elderly people in China were generally low in the southwest,high in the northwest and moderate level in other regions.The ranges of PaCO2 can be directly reflected by charted distribution maps,which provide convenience for relevant medical workers.