BACKGROUND: Evidence on the acute effect of short-term exposure to nitrogen dioxide (NO(2)) on years of life lost (YLL) is rare, especially in multicity setting. METHODS: We conducted a time series study among 48 major Chinese cities covering more than 403 million people from 2013 to 2017. The relative percentage changes of NO(2)-YLL were estimated by generalized additive models in each city, then were pooled to generate average effects using random-effect models. In addition, stratified analyses by individual demographic factors and temperature as well as meta-regression analyses incorporating city-specific air pollutant concentrations, meteorological conditions, and socioeconomic indicators were performed to explore potential effect modification. RESULTS: A 10 ?g/m(3) increase in two-day moving average (lag01) NO(2) concentration was associated with 0.64% (95% CI: 0.47%, 0.81%), 0.47% (95% CI: 0.27%, 0.68%), and 0.68% (95% CI: 0.34%, 1.02%) relative increments in YLL due to nonaccidental causes, cardiovascular diseases (CVD), and respiratory diseases (RD), respectively. These associations were generally robust to the adjustment of co-pollutants, except for NO(2)-CVD that might be confounded by fine particulate matter. The increased YLL induced by NO(2) were more pronounced in elderly people, hotter days, and cities characterized by less severe air pollution or higher temperature. CONCLUSIONS: Our results demonstrated robust evidence on the associations between NO(2) exposure and YLL due to nonaccidental causes, CVD, and RD, which provided novel evidence to better understand the disease burden related to NO(2) pollution and to facilitate allocation of health resources targeting high-risk subpopulation.