Related Experiment Video
Updated: May 26, 2026

Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Multimethod geospatial modeling of life expectancy determinants in China
Ke Hu1, Xingjin Yang2, Shuiping Ou3
1Xiamen Haicang Hospital, Xiamen, Fujian, China.
Introduction:
Although China has achieved sustained growth in life expectancy, significant spatial disparities persist across regions, necessitating investigation into their underlying determinants.
Methods:
This study systematically examines provincial-level life expectancy patterns in 2020 using multiple analytical approaches, including multiple linear regression (MLR), spatial lag model (SLM), geographically weighted regression (GWR), multiscale geographically weighted regression (MGWR), and Geodetector. The analysis incorporates comprehensive datasets covering socioeconomic indicators, education levels, healthcare resources, and environmental factors.
Results:
The findings reveal a pronounced east-west gradient, with coastal provinces exhibiting significantly higher life expectancy than inland regions. Spatial analysis identified distinct clustering patterns (Global Moran's I = 0.5339, p < 0.001), with high-high clusters in eastern areas and low-low clusters in western regions. MGWR demonstrated superior performance by accounting for spatial variations in factor influences. Education level emerged as a fundamental determinant (PD = 0.850, p = 0.012), showing particularly strong effects in western China. Economic development also showed substantial explanatory power (PD = 0.661, p = 0.030), with greater influence in western provinces. Household consumption expenditure per capita (PD = 0.586, p = 0.027) and urbanization rate (PD = 0.633, p = 0.036) were additionally identified as significant factors. Household size showed a negative association exclusively in certain western regions. Notably, healthcare resource allocation demonstrated no significant effect (PD = 0.344, p = 1.000), while PM2.5 also lacked statistical significance (PD = 0.344, p = 1.000). Furthermore, Geodetector interaction analysis revealed that economic development and education level exhibited particularly strong synergistic effects (interaction PD = 0.944).
Conclusion:
The findings provide a scientific basis for formulating region-specific health policies and offer valuable insights for advancing the Healthy China initiative.
Related Concept Videos
Applications of Life Tables
Statistical Methods for Analyzing Epidemiological Data
Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
Life Tables
Parametric Survival Analysis: Weibull and Exponential Methods
Weibull Distribution
The Weibull distribution is a flexible model used in parametric survival analysis. It can handle both increasing and decreasing hazard rates, depending on its shape parameter...
Mechanistic Models: Compartment Models in Individual and Population Analysis