Sign in Subscribe
Exercise

Machine Learning Identifies Key Risk Factors for Subjective Life Expectancy

Strategic Insights

— 2 min read

Research conducted at East China Normal University in Shanghai, People's Republic of China, has utilized machine learning methods to identify key risk factors influencing subjective life expectancy among middle-aged and older adults. The study analyzed data from the China Health and Retirement Longitudinal Study (CHARLS) 2018 survey, which included 10,945 participants. Five machine learning models were constructed to predict subjective life expectancy in active and inactive individuals, with the Support Vector Machine (SVM) model achieving the best performance in the inactive group and the Light Gradient Boosting Machine (LightGBM) model achieving the best performance in the active group. The study found that age and perceived health were the most important variables in predicting subjective life expectancy.

Key Takeaways:

  • The study used machine learning methods to identify key risk factors influencing subjective life expectancy among middle-aged and older adults.
  • Data from the China Health and Retirement Longitudinal Study (CHARLS) 2018 survey was used, with 10,945 participants included in the analysis.
  • Five machine learning models were constructed, including Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM).
  • The Support Vector Machine (SVM) model achieved the best performance in the inactive group, with an AUC of 0.797 and an accuracy of 0.722.
  • The Light Gradient Boosting Machine (LightGBM) model achieved the best performance in the active group, with an AUC of 0.775 and an accuracy of 0.745.
  • Feature importance analysis indicated that 'age' was the most important variable in the Support Vector Machine (SVM) model, while 'perceived health' was the most important variable in the Light Gradient Boosting Machine (LightGBM) model.
  • The study concluded that machine learning methods can effectively identify key risk factors influencing subjective life expectancy among middle-aged and older adults, providing valuable guidance for targeted health management strategies.

Statistics:

  • 10,945 participants were included in the analysis.
  • 4,707 men and 4,885 women were in the active group.
  • 662 men and 691 women were in the inactive group.
  • The active group had a mean age of 59.76 years.
  • The inactive group had a mean age of 63.00 years.
  • The Support Vector Machine (SVM) model achieved an AUC of 0.797 in the inactive group.
  • The Light Gradient Boosting Machine (LightGBM) model achieved an AUC of 0.775 in the active group.

Sources:

  • Identifying subjective life expectancy risk factors in physically active and inactive middle-aged and older adults using machine learning models. BMC Public Health, 2025;25(1):3506.
  • China Health and Retirement Longitudinal Study (CHARLS) 2018 survey.
  • East China Normal University.

Read more