Table 1_The impact of smart city construction on urban public health: evaluation using dual machine learning models.xlsx

IntroductionGlobal public health is confronted with multiple challenges such as the frequent occurrence of infectious diseases, the increasing burden of chronic diseases, climate change, and the transformation of population structure. These social and environmental factors have exacerbated the complexity and difficulty of public health governance. Digital technologies represented by artificial intelligence and big data are developing rapidly, providing new tools and paradigms for addressing these challenges.
MethodsThis paper takes the pilot policies of smart cities in China as the standard natural experiment, and uses the panel data of 270 prefecture-level cities in China from 2007 to 2020. It empirically examines the impact of smart city construction (SCC) on public health level (PHL) by adopting a dual machine learning model.
ResultsThe results show that SCC can significantly improve PHL, and the conclusion remains valid after robustness tests such as resetting the machine learning model, instrumental variable method, and multi-dimensional fixation. Mechanism tests show that SCC mainly enhances PHL by improving the quality of the ecological environment, enhancing the level of innovative services, and optimizing the structure of healthy human resources. Heterogeneity analysis revealed that the promoting effect of SCC on PHL was more pronounced in regions such as western China, large cities and key environmental protection cities.
DiscussionThe research conclusions can not only enrich the theoretical accumulation in the fields of smart cities and public health, but also provide replicable and scalable practical experience for other developing countries to enhance their public health governance capabilities through digital means under resource constraints.

引言：全球公共卫生面临多重挑战，包括传染病频发、慢性病负担持续攀升、气候变化以及人口结构转型。此类社会与环境因素进一步加剧了公共卫生治理的复杂性与实施难度。以人工智能（Artificial Intelligence）与大数据为代表的数字技术正快速发展，为应对上述公共卫生挑战提供了全新的工具与治理范式。
研究方法：本文以中国智慧城市试点政策为标准自然实验，选取2007年至2020年中国270个地级市的面板数据，采用双重机器学习模型实证检验智慧城市建设（Smart City Construction, SCC）对公共卫生水平（Public Health Level, PHL）的影响。
研究结果：结果显示，智慧城市建设可显著提升公共卫生水平，该结论在重新设定机器学习模型、工具变量法以及多维固定效应等稳健性检验后依然成立。机制检验表明，智慧城市建设主要通过改善生态环境质量、提升创新服务水平以及优化健康人力资源结构来提升公共卫生水平。异质性分析发现，智慧城市建设对公共卫生水平的提升效应在中国西部地区、大城市以及重点环保城市等区域更为显著。
讨论：本研究结论不仅丰富了智慧城市与公共卫生领域的理论积累，更为其他发展中国家在资源约束条件下通过数字化手段提升公共卫生治理能力提供了可复制、可推广的实践经验。