Background: Air pollution levels in Beijing have been improving through implementation of the Air Pollution Prevention and Control Action Plan (2013-2017) but implication for respiratory morbidity have not been directly investigated. Methods: Daily city-wide mean concentrations of PM10, PM2·5, NO2, SO2, CO and Ozone in 2013-17 were averaged from 35 monitoring stations across the city. PMcoarse was derived by subtracting PM2·5 from PM10. A generalized additive Poisson time-series model was applied to estimate the percentage change with 95% confidence interval (CI) on hospitalisations for acute exacerbation of chronic obstructive pulmonary disease (AECOPD) in Beijing. Findings: During 18/January/2013-31/December/2017, 161,613 AECOPD hospitalisations were recorded. Ambient concentrations of SO2 and PM2·5 decreased by 68% and 33% over this five-year period. AECOPD hospitalisations increased by 0·3%(95%CI:0·3%-0·4%), 0·4%(95%CI:0·3%-0·5%), 0·7%(95%CI:0·5%-0·9%), 1·2%(95%CI:0·9%-1·4%), 1·2%(95%CI:0·9%-1·6%), 3·0%(95%CI:2·3%-3·6%) and 0·3%(95%CI:0·1%-0·5%) for same day PM10, PM2·5, PMcoarse, NO2, SO2, CO and Ozone (warm season) respectively per 10 µg/m3 (except per 1 mg/m3 for CO) increment. Females and patients aged ≥65 years were more susceptible. In 2013, there were 12,679 AECOPD cases advanced by PM2·5 pollution over expected rates if daily levels had not exceeded the World Health Organisation target, whereas in 2017 the respective figure was 7,377 cases. Interpretation: Increased acute air pollution exposures were significantly associated with increased hospitalisations for AECOPD in Beijing despite improvement in overall air quality. Our findings however highlight the importance and effectiveness of stringent air pollution control policy on reducing COPD morbidity and provide rationale for long-term multidimensional policy to safeguard public health. Funding: None. Declaration of Interest: None declared. Ethical Approval: This study was approved by the Research Ethics Board of Beijing Chaoyang Hospital (IR9511).