Estimating health burden and economic loss attributable to short-term exposure to multiple air pollutants in China

Background Existing studies focused on the evaluation of health burden of long-term exposure to air pollutants, whereas limited information is available on short-term exposure, particularly in China. Methods Air pollutants concentrations in 338 Chinese cities in 2017 were used to estimate the air pollutants related health burden which was defined as premature mortalities from all-cause, cardiovascular and respiratory disease as well as hospital admissions (HAs) for cardiovascular and respiratory disease. Log-linear model was used as the exposure-response function to estimate the health burden attributable to each air pollutant. The value of statistical life and cost of illness methods were used to estimate economic loss of the premature mortalities and HAs, respectively. Results The national all-cause premature mortalities attributable to all air pollutants was 1.35 million, accounting for 17.2% of reported deaths in China in 2017. Among all-cause premature mortality, contributions of PM10, SO2, NO2, CO, and O3 were 16.3%, 9.6%, 28.9%, 22.2% and 23.0%, respectively. The national cardiovascular and respiratory premature mortalities were 0.78 and 0.21 million, respectively. About 6.79 million cardiovascular and respiratory disease HAs were attributed to short-term exposure to PM10, SO2, NO2, and O3. The economic loss of the overall health burden was 2057.66 billion Yuan, which was equivalent to 2.5% of the national GDP in 2017. Conclusions The health burden and economic loss attributable to short-term exposure to ambient air pollutant are substantial in China. It suggested that the adverse health effects attributable to short-term exposure to air pollutant should not be neglected in China. PM2.5: particles ≤ 2.5 μm in aerodynamic diameter, PM10: particles ≤ 10 μm in aerodynamic diameter, SO2: sulfur dioxide, NO2: nitrogen oxide, CO: carbon monoxide, O3: ozone, VSL: the value of statistical life, COI: cost of illness, GDP: gross domestic product, CPI: consumer price index, HA: hospital admission.

This study comprehensively investigated the health burden and economic loss attributable to short-term exposure to PM 10 , SO 2 , NO 2 , CO, and O 3 in China in 2017. We are going to (1) estimate the premature deaths and HAs attributable to short-term exposure to above ambient air pollutants for 338 cities in China; (2) to evaluate the economic loss of premature death and HA using the value of statistical life (VSL) method and cost of illness (COI), respectively. The results will provide a scientific basis for the policy maker to formulate relevant policies for air pollution control projects. 2.1.2 Daily cause-specific health endpoint incidence rate The health endpoints were defined as "all-cause mortality", "cardiovascular disease mortality", "respiratory disease mortality", "cardiovascular disease HAs", and "respiratory disease HAs" in this study. The city-specific annual all-cause mortality rates in 2017 were obtained from the National Statistical Yearbook in each Province and Statistical Bulletin of National Economic and Social Development in each city (Fig. S1).
The annual cardiovascular and respiratory disease mortality rate for each city was calculated by multiplying the annual all-cause mortality rate by the proportion of corresponding cause, which was shown in Fig. S2 The monthly cause-specific mortality rate was calculated by multiplying the annual causespecific mortality rate by the proportion of mortality rate in different months. The province-specific proportions of mortality rates in different months were obtained from the Six National Population Census (http://www.stats.gov.cn/tjsj/pcsj/rkpc/6rp/indexch.htm).
The monthly mortality rate was evenly converted into daily rate, and the annual HA incidence rate was evenly converted into daily rate.

Health burden assessment
We used the log-linear exposure-response function to estimate the health burden attributable to short-term exposure to air pollutant [10-13]. Specifically, the health burden for each health endpoint was calculated as follows, (1) where HIi,j denotes the health burden j attributed to air pollutant i, and BIj is the daily baseline incidence rate of health endpoint j, and EP is the exposure population.is the exposure-response function for health endpoint j exposed to air pollutant i, which is an estimate of the percentage increase in daily health endpoint and can be obtained from recent epidemiological studies [14-20]{Lai, 2013 #2600;Ma, 2016 #2675} (Table 1). Ci is the daily concentration of air pollutant i, and Coi is the daily threshold concentration for air pollutant i which is assumed to be zero in our study [6,[21][22][23]. Because the study on CO-related cardiovascular and respiratory HAs did not existed in China, we did not estimate HAs attributed to CO. where COHj is the average medical treatment cost for a disease j for each case, DGDP is the GDP per capita per day, Tjis the average labor time lost due to disease j for each case, COIj is the average cost of illness of hospitalization for a disease j for each case.

Premature mortality attributable to air pollutant
The all-cause premature mortality attributed to short-term air pollutant exposure in 338 Chinese cities are shown in Fig. 2 The city-specific contributions of air pollutants to all-cause mortality are shown in Fig. 3.

HAs attributable to air pollutant
The air-pollutant related HAs in 31 provinces are shown in Table S2 and

Discussion
In this study, we firstly estimated premature mortalities and HAs attributed to short-term ambient air pollutants exposure and corresponding economic loss in 338 Chinese cities in 2017. About 1.35 million premature deaths were estimated attributable to short-term air pollutants exposure, accounting for 17.2% of total deaths in China in 2017. About 6.79 million HAs were attributable to short term exposure to PM 10 , SO 2 , NO 2 , and O 3 . The national economic loss of mortality and HAs associated with short-term exposure to air pollutants was 2057.66 billion Yuan, which was equivalent to 2.5% of the total GDP in China in 2017. To the best of our knowledge, this is the first study providing a comprehensive assessment of acute health burden and its economic losses attributable to short-term exposure to PM 10 , SO 2 , NO 2 , CO, and O 3 in China. This study would fill an important gap in the assessment of acute health burden associated with air pollution in China.
Our findings suggested that the health burden and economic loss attributable to shortterm air pollutants exposure are substantial. To date, few studies have been conducted on health burden attributable to short-term exposure to air pollutants in China. The possible reason is that the exposure-response coefficients of short-term exposure to air pollutants are far less than those of long-term exposure [14-18, 26-28]. However, China is the most polluted area and have the largest population worldwide so that the ultimate health burden may not negligible [6,7]. In our study, we found approximately 1.35 million premature deaths and 6.79 million HAs attributable to short-term exposure to air pollutants in China, respectively. We should not ignore such dramatic health burden and economic losses attributed to short-term air pollution exposure. . However, we also found that some cities in Xinjiang, Henan, and Anhui provinces also had heavy health burden, which were not in the priority regions mentioned above. The possible reason is that these provinces have higher air pollutant concentrations or huge population. From the perspective of health protection, the additional air pollution control measures in nonpriority regions also should be strengthened. In addition, the Guangdong province located in the coastal region with low air pollutants concentrations also had a huge health burden possibly due to a large population. All these suggested that it is of great importance for each city to continue to draw up pertinent air pollutant control policies to further reduce the health burden caused by air pollution.
We found that the health burden associated with gaseous pollutants (SO 2 , NO 2 , CO, and O 3 ) are significantly greater than that of PM 10 . In recent years, the particulate matter (PM) is typically considered to be the primary air pollutants, which contributes the most health impacts to Chinese population [31]. There are many studies focused on estimating the PMrelated health burden [1,3,32]. However, we observed that NO 2 , not PM 10 , contributed the most to the premature mortality among the five air pollutants in our study. In addition, the contributions of CO and O 3 to premature mortality were also significantly greater than that of PM 10 . For the HAs, the contributions of O 3 , NO 2 and SO 2 were also significantly greater than that of PM 10 . Similarly, Chen et al. [6] found that the highest contribution to premature mortality was made by NO 2 across China. Lu et al.
[7] also found that the contributions of NO 2 and SO 2 to all-cause mortality, outpatient visits and HAs were significantly greater than that of PM 10 in the Pearl River Delta region in China.
The possible reason is that the exposure-response coefficients of NO 2 , SO 2 , CO, and O 3 were all larger than PM 10 [16,33]. These results suggested that we should not only focus on the PM-related health burden, but also pay attention to the health burden of gaseous pollutants in China.
Existing studies have shown no evidence on a definite threshold concentration in the exposure-response relationships between health effects and air pollutants. For instance, studies conducted in China [23, 34,35] and North America [36][37][38][39] found that an threshold did not existed in the acute effect of short-term O 3 exposure. In addition, the threshold concentrations of PM 10 , SO 2 , and NO 2 for acute health effects are also not determined [21, [40][41][42]. Therefore, we set the threshold concentration for each air pollutant to be zero. This study has several limitations. First, we assumed that the exposure-response coefficients were the same among different cities without taking their spatial variations into account, however, it may differ due to the different age structure [43]; in addition, PM 10 in different cities may have different chemical compositions, sizes of distribution and sources, which would also trigger different health effects [44][45][46]. Second, for the exposure-response coefficients obtained from existing studies, which did not consider the synergistic effect of multiple air pollutants, we separately estimated the health burden of a certain air pollutant. Underestimation would exist because populations were usually exposed to multiple environmental risk factors possibly causing synergistic effects [47,48]. Third, due to inaccessibility to city-specific baseline incidence rates of hospitalization for respiratory and circulatory disease, we assumed these rates being the same among all cities in our study, while substantial disparities may exist due to the differentiations in social economics, social environments and health services.

Conclusions
To the best of our knowledge, this is the first work in China to evaluate the health burden and economic loss attributable to short-term exposure to ambient air pollutantsat national level. Substantial health burden and economic loss attributable to short-term air pollution exposure were ascertained in our study. The health burden and economic loss associated with gaseous pollutants were significantly higher than that with particulate matter. We also found some provinces not included in the air pollution priority control areas were also urgently needed to adopt a series of strict air pollution control measures to reduce the health burden. This study has contributed to the limited research concerning the health burden assessment of short-term air pollution exposure.    This map has been provided by the authors. Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.