{"title":"A measurement study of the environmental quality and medical expenditures of elderly individuals: causal inference based on machine learning.","authors":"Yu Zhang, Sheng Chen, Dewen Liu","doi":"10.1186/s13690-024-01386-2","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>The global surge of environmental pollution exacerbates health issues, disease incidence, and economic strain. In China, the increasing healthcare costs of the elderly population necessitate addressing this challenge as part of the \"Healthy China\" strategy. We explore the impact of environmental quality on elderly healthcare expenses.</p><p><strong>Methods: </strong>This study devised a comprehensive environmental quality index for 30 Chinese provinces, excluding Tibet, which was correlated with medical expenses for individuals older than 60 years, using China Family Panel Studies (CFPS) data. Because the traditional econometric model cannot solve the endogeneity problem and the selection of instrumental variables is subjective, a new machine learning algorithm is adopted based on the traditional ordinary least squares (OLS) model and the fixed effect model to conduct causal analysis to ensure the reliability of the results. Finally, heterogeneity analysis was conducted based on the generalized random forest algorithm.</p><p><strong>Results: </strong>Southern provinces such as Jiangxi and Guangxi exhibited superior environmental qualities. A regional analysis revealed a gradient where environmental quality decreased from west to east and from south to north. Both conventional and machine learning methodologies underscored a pivotal finding: enhanced environmental qualities significantly curtail elderly healthcare expenses. A heterogeneity assessment revealed that such improvements predominantly benefit elderly people in the eastern and central regions, with marginal impacts in the west. For different groups, the improvement of environmental quality can significantly reduce the medical expenditure of people aged 60 to 75, with bedtime hours between 9 and 11 PM and a lower household income.</p><p><strong>Conclusions: </strong>This study, employing machine learning and traditional models, demonstrates that enhancements in environmental quality significantly reduce medical costs for the elderly in China, especially in the eastern and central regions, and among demographics such as individuals aged 60-75 and low-income households. These findings underscore the potential of environmental policies to lower medical costs within the \"Healthy China\" initiative framework. However, the study's scope is limited by the environmental quality index and the extent of data coverage, indicating a need for further research expansion.</p>","PeriodicalId":48578,"journal":{"name":"Archives of Public Health","volume":"82 1","pages":"195"},"PeriodicalIF":3.2000,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11523846/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Archives of Public Health","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s13690-024-01386-2","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0
Abstract
Background: The global surge of environmental pollution exacerbates health issues, disease incidence, and economic strain. In China, the increasing healthcare costs of the elderly population necessitate addressing this challenge as part of the "Healthy China" strategy. We explore the impact of environmental quality on elderly healthcare expenses.
Methods: This study devised a comprehensive environmental quality index for 30 Chinese provinces, excluding Tibet, which was correlated with medical expenses for individuals older than 60 years, using China Family Panel Studies (CFPS) data. Because the traditional econometric model cannot solve the endogeneity problem and the selection of instrumental variables is subjective, a new machine learning algorithm is adopted based on the traditional ordinary least squares (OLS) model and the fixed effect model to conduct causal analysis to ensure the reliability of the results. Finally, heterogeneity analysis was conducted based on the generalized random forest algorithm.
Results: Southern provinces such as Jiangxi and Guangxi exhibited superior environmental qualities. A regional analysis revealed a gradient where environmental quality decreased from west to east and from south to north. Both conventional and machine learning methodologies underscored a pivotal finding: enhanced environmental qualities significantly curtail elderly healthcare expenses. A heterogeneity assessment revealed that such improvements predominantly benefit elderly people in the eastern and central regions, with marginal impacts in the west. For different groups, the improvement of environmental quality can significantly reduce the medical expenditure of people aged 60 to 75, with bedtime hours between 9 and 11 PM and a lower household income.
Conclusions: This study, employing machine learning and traditional models, demonstrates that enhancements in environmental quality significantly reduce medical costs for the elderly in China, especially in the eastern and central regions, and among demographics such as individuals aged 60-75 and low-income households. These findings underscore the potential of environmental policies to lower medical costs within the "Healthy China" initiative framework. However, the study's scope is limited by the environmental quality index and the extent of data coverage, indicating a need for further research expansion.
期刊介绍:
rchives of Public Health is a broad scope public health journal, dedicated to publishing all sound science in the field of public health. The journal aims to better the understanding of the health of populations. The journal contributes to public health knowledge, enhances the interaction between research, policy and practice and stimulates public health monitoring and indicator development. The journal considers submissions on health outcomes and their determinants, with clear statements about the public health and policy implications. Archives of Public Health welcomes methodological papers (e.g., on study design and bias), papers on health services research, health economics, community interventions, and epidemiological studies dealing with international comparisons, the determinants of inequality in health, and the environmental, behavioural, social, demographic and occupational correlates of health and diseases.