<p>In China, public health currently faces a significant threat from air pollution, and it is a crucial moment for air pollution management and control. In this study, we conducted PM_2.5, PM₁₀ and O₃ estimations by integrating machine learning with a land-use regression model to analyze the spatiotemporal characteristics of the Beijing-Tianjin-Hebei (BTH) region. The log-linear exposure‒response function (ERF) model and willingness to pay (WTP) method were used to evaluate the health impacts, namely, all-cause, cardiovascular disease and respiratory disease mortalities, of and related economic losses from PM_2.5, PM₁₀ and O₃ concentrations. Our results show that meteorology and the aerosol optical depth (AOD) greatly influence the PM_2.5 and PM₁₀ models, while meteorology and road length play important roles in the O₃ model. Consequently, the estimated PM_2.5, PM₁₀ and O₃ concentrations reached 56.59, 98.76 and 143.18&#xa0;µg/m³, respectively, in the BTH region. The PM_2.5 and PM₁₀ concentrations decreased during this period in 13 cities, whereas an increasing trend was confirmed for O₃ in 11 cities, except for Beijing and Zhangjiakou. Specifically, Handan, Hengshui and Xingtai should devote more attention to air pollution. In addition, the health impacts and economic losses from O₃ were much greater than those from PM_2.5 and PM₁₀. For the same pollutant, CVD mortality and its associated economic losses were greater than those associated with RD. Decreases in health impacts and economic losses were also observed in this study, and differences were detected at the city level. Beijing, Tianjin, Chengde, Handan and Shijiazhuang should receive more attention because of their health impacts and economic losses. Moreover, there were differences not only between mortalities and those as a percentage of the total population but also between economic losses and those as a percentage of GDP. There were inconsistent changes in air pollution, health impacts and economic losses; therefore, considering these three aspects comprehensively, Beijing and Handan were key cities for developing coordinated control measures. Our study pinpoints the spatiotemporal characteristics, health impacts and related economic losses of PM_2.5, PM₁₀ and O₃ at the city level and provides important information for epidemiological research and custom-made pollution control policies for each city.</p>