The impact of carbon monoxide on years of life lost and modified effect by individual- and city-level characteristics: Evidence from a nationwide time-series study in China

Ambient carbon monoxide (CO) has been linked with mortality and morbidity. Little evidence is available regarding the relation between CO and years of life lost (YLL). Using data from 48 major cities in China from 2013 to 2017, we applied generalized additive models and random effects meta-analyses to explore the effects of CO on YLL from various diseases. Stratified analyses and meta-regression were performed to estimate potential effect modifications of demographic factors, regions, meteorological factors, co-pollutants, urbanization rate, economic level and health service level. Additional life gains due to avoidable YLL under certain scenario were also evaluated. Results indicated that a 1-mg/m(3) increase of CO concentrations (lagged over 0-3 d), was associated with 2.08% (95% confidence interval [CI], 1.35%, 2.80%), 2.35% (95% CI: 1.39%, 3.30%), 1.47% (95% CI:-0.01%, 2.93%), 2.28% (95% CI: 1.09%, 3.47%), 2.42% (95% CI: 1.31%, 3.54%), 2.09% (95% CI: 0.47%, 3.72%) increments in daily YLL from non-accidental causes, cardiovascular diseases, respiratory diseases, coronary heart disease, stroke and chronic obstructive pulmonary disease, respectively. These associations were robust to the adjustment of co-pollutants and varied substantially by geography and demographic characteristics. Associations were stronger in the elder people (>= 65 years), females, population with low education attainment, and lived in south region, than younger people, males, high educated populations and those lived in north region. Moreover, the harmful impact of increasing CO concentration could be attenuated by city-level characteristics, including the growth of urbanization rate, gross domestic product (GDP), GDP per capita, number of hospital beds, doctors and hospitals. Finally, an estimated life of 0.081 (95% CI:-0.027, 0.190) years would be gained per deceased people if CO concentration could fall to 1 mg/m(3). In conclusions, this nationwide analysis showed significant associations between short-term CO exposure and cause-specific YLL. The heterogeneity of both in dividual and city-level characteristics should be considered for relevant intervention. These findings may have significant public health implications for the reduction of CO-attributed disease burden in China.