| Objective:To explore the relationship between public health input and health output with Grossman health production function,according to the law of quantity and effect interaction between public health service input variable and health output variable,and to discusses how to improve the benefit of public health input to health output in Shenzhen,and provides feasible suggestions for health decision making in Shenzhen.Methods:Life expectancy was collected as dependent variable and the health input variables were selected based on Grossman health demand model from 1989 to2016 in Shenzhen to construct health production function of residents.Economy,social,health,environment and life variable were included in health input variables.Vector auto regressive(VAR)model was applied to single static analysis and impulse response function was used for single dynamic analysis.Limited distributed lag model was applied to multiple factors analysis.Results:1.In study period from 1989 to 2016,GDP per capita,disposable income per capita every month,the proportion of health expense accounts for local fiscal expenditure,health expense per capita,the number of doctors per thousand population,the hospital beds per thousand population,living space per capita,the proportion of female population,these 8 factors have a positive effect on improving life expectancy.Nevertheless,junior high school graduation rates and average PM10in these two factors for the influence of residents’life expectancy has negative effects.2.The increase of the proportion of health expense accounts for local fiscal expenditure and health expense per capita can improve life expectancy in seven years later in Shenzhen.The increase of GDP per capita and disposable income per capita every month can improve life expectancy,which had positive effect,but it had negative effect sometimes for data fluctuation.The increase of the number of doctors per thousand population,the hospital beds per thousand population and living space per capita can improve life expectancy always.The increase of the proportion of female population can improve life expectancy in the first year to the eighth year but reduce life expectancy after nine years.3.The R~2 of health production function in Shenzhen we constructed was 0.961,suggests that the health production function to explain ability of residents’health reached 96.1%.The elasticity of the GDP per capita for health is 0.019,when GDP per capita increased by 1%,residents life expectancy can be increased by0.019%.The elasticity of health service fee per capita for health is 0.016,when GDP per capita increased by 1%,residents life expectancy can be increased by 0.016%.GDP per capita contribution to the life expectancy was 62%,health expense per capita contribution to the life expectancy was 38%.In our study,the contribution of the growth of GDP per capita to improve life expectancy is greater than health expense per capita.4.Life expectancy that were predicted by health production function in Shenzhen in 2015 and 2016 were 80.75 and 80.99,prediction accuracy were 0.001and 0.002.Conclusions:1.The increase of GDP per capita,disposable income per capita every month,the number of doctors per thousand population,the hospital beds,living space per capita,the proportion of female population can improve life expectancy.The increase of junior high school graduation rates and average PM10 can reduce life expectancy.2.The influence factors of health in Shenzhen mainly represented by GDP per capita of economic variables and health expense per capita of health care variables.Life expectancy can be predicted well by Shenzhen health production function we constructed.The growth of the input of GDP and health expense per capita GDP per capita had positive effect on health output and improve health of residents,which suggested that the government should increase investment in health care. |