Ecotoxicology and Environmental Safety 170 (2019) 286–292
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv
Acute effect of ambient air pollution on hospitalization in patients with hypertension: A time-series study in Shijiazhuang, China
Jie Songa,b, , Mengxue Luc, Jianguo Lud, Ling Chaoa, Zhen Ana,b, Yue Liue, Dongqun Xue, Weidong Wua,b ⁎
School of Public Health, Xinxiang Medical University, Xinxiang 453003, China Henan International Collaborative Laboratory for Air Pollution Health Effects and Intervention, Xinxiang 453003, China c Xinxiang Medical University, Xinxiang 453003, China d The First Affiliated Hospital of Xinxiang Medical University, Xinxiang 453100, China e National Institute of Environmental Health, Chinese Center for Disease Control and Prevention, Beijing 100021, China b
Keywords: Air pollution Hypertension Hospitalization Time-series study
Although numerous studies have investigated the association between air pollution and hospitalization, few studies have focused on the health effect of air pollution on populations with hypertension. In this study, we conducted a time-series study to investigate the acute adverse effect of six criteria ambient air pollutants (fine particulate matter [PM2.5], inhalable particulate matter [PM10], nitrogen dioxide [NO2], sulfur dioxide [SO2], ozone [O3], and carbon monoxide [CO]) on hospitalization of patients for hypertension in Shijiazhuang, China, from 2013 to 2016. An over-dispersed Poisson generalized addictive model adjusting for weather conditions, day of the week, and long-term and seasonal trends was used. In addition, we evaluated the effect of modification by season, sex, and age. A total of 650,550 hospitalization records were retrieved during the study period. A 10 μg/ m3 increase of PM2.5 (lag06), PM10 (lag06), NO2 (lag03), O3 (lag6), and CO (lag04) corresponded to 0.56% (95% confidence interval [CI]: 0.28–0.83%), 0.31% (95% CI: 0.12–0.50%), 1.18% (95% CI: 0.49–1.87%), 0.40% (95% CI: 0.09–0.71%), and 0.03% (95% CI: 0.01–0.05%) increments in hospitalization of patients for hypertension, respectively. We observed statistically significant associations with PM2.5, PM10, NO2, O3, and CO, while positive but insignificant associations with SO2. The effects of PM2.5, PM10, NO2, O3, and CO were robust when adjusted for co-pollutants. We found stronger associations in the cool season than in the warm season. Moreover, there were non-significant differences in the associations between air pollution and sex or age group. This study suggests that patients with hypertension had an increased risk of hospital admission when exposed to air pollution.
1. Introduction Air pollution is recognized as a global public health issue. It was responsible for 6.7 million deaths worldwide in 2016, during which approximately 2.3 million of cardiovascular deaths were attributed to ambient air pollution (Hadley et al., 2018;). Most of this burden is shouldered by developing countries. Almost 89% of ambient air pollution-related deaths occur in these regions, especially in Asia (Hadley et al., 2018; Landrigan et al., 2018). As one of the most polluted cities in China, Shijiazhuang is experiencing the worst air pollution problems in the world. In 2013, the annual average concentrations of six criteria
ambient air pollutants (fine particulate matter [PM2.5], inhalable particulate matter [PM10], nitrogen dioxide [NO2], sulfur dioxide [SO2], ozone [O3] and carbon monoxide [CO]) were 141, 303, 101, 62, 79 and 1924 μg/m3, respectively. These values for PM2.5, PM10, NO2 and SO2 are approximately 2.0, 8.7, 1.6, and 1.7 times higher than the Chinese National Ambient Air Quality Standard (PM2.5, 70 μg/m3; PM10, 35 μg/ m3; NO2, 40 μg/m3; SO2, 60 μg/m3), and only 5 days were below the standard. However, few studies have been conducted to elucidate adverse health effects of ambient air pollution in Shijiazhuang, China. Strong evidence of associations between ambient air pollution and cardiovascular disease has been provided in recent years (Anderson,
Abbreviations: AIC, Akaike information criterion; CO, carbon monoxide; E-R, exposure-response; GCV, generalized cross validation; ICD10, International Classification of Diseases, revision 10; NO2, nitrogen dioxide; PACF, partial autocorrelation function; PM10, inhalable particulate matter; PM2.5, fine particulate matter; SEMRC, Shijiazhuang Electronic Medical Record Data Center; SHFPC, Shijiazhuang Health and Family Planning Commission; SO2, sulfur dioxide ⁎ Corresponding author at: School of Public Health, Xinxiang Medical University, Xinxiang 453003, China. E-mail addresses: [email protected]
, [email protected]
(J. Song). https://doi.org/10.1016/j.ecoenv.2018.11.125 Received 21 September 2018; Received in revised form 20 November 2018; Accepted 28 November 2018 0147-6513/ © 2018 Elsevier Inc. All rights reserved.
Ecotoxicology and Environmental Safety 170 (2019) 286–292
J. Song et al.
Fig. 1. Location of the study and geographical distribution of 7 air pollution monitors.
2017; Brook et al., 2010; Cosselman et al., 2015; Mazidi and Speakman, 2018). Acute exposure has been associated with increased hospitalizations or deaths due to heart failure, coronary syndrome, or cardiac arrest (Dai et al., 2015; H. Liu et al., 2018; Xia et al., 2017). Moreover, the risk might be higher in certain susceptible subpopulations, such as individuals of advanced age and lower socioeconomic status, and those with traditional cardiovascular risk factors such as hypertension and diabetes (Hadley et al., 2018; Pitchika et al., 2017; Qiu et al., 2018; Stachyra et al., 2017). In 2010, 335.8 million Chinese adults had hypertension (Bundy and He, 2016). With the acceleration of population aging, the number of patients with hypertension is still increasing. However, studies evaluating the associations between ambient air pollution and hospitalizations of patients with hypertension are lacking. Therefore, we aimed to explore the acute effects of six ambient air pollutants (PM2.5, PM10, NO2, SO2, O3, and CO) on hospital admissions of patients in this region who are hypertensive.
Unlike the top hospitals in Shijiazhuang that serve a large proportion of nonlocal patients, those hospitals under SHFPC serve permanent residents in this city and cover approximately 90% of permanent residents’ medical service and thus may be highly representative of the Shijiazhuang population. The diseases are coded according to the International Classification of Diseases, revision 10 (ICD10): A00-R99 for non-accidental diseases and I10 for hypertension. The data were also classified according to season, sex, and age. The cool season lasts from October to March, and the warm season lasts from April to September. Patients were also classified into two age groups: younger (18–64 years) and elderly (≥65 years). 2.2. Air pollutants and meteorological data Daily concentrations of six ambient air pollutants (PM2.5, PM10, NO2, SO2, O3, and CO) in Shijiazhuang were obtained from the website of the Chinese Ministry of Environmental Protection (http://106.37. 208.233:20035/). The daily average air pollutant concentrations were based on seven fixed-site monitoring stations operated under the Ministry of Ecology and Environment of the People's Republic of China. These stations are distributed in urban districts of Shijiazhuang (Fig. 1) and located away from major roads and buildings and industrial or residential sources of emissions from burning coal, oil, or waste; thus, these monitoring results reflect the general urban background level rather than local sources. For PM2.5, PM10, SO2, and NO2, daily concentrations represent 24-h averages, and the O3 concentration is the maximal 8-h average from all valid monitoring sites in this study. Two meteorological parameters (daily mean relative humidity and temperature) for Shijiazhuang were collected from the China's Meteorological Data Sharing Service System. Those two values were introduced into the model to adjust for confounders.
2. Materials and methods Shijiazhuang, located 273 km south of Beijing, is the capital of Hebei Province with a population of approximately 2.79 million in 2016. The major air pollution source is emissions of coal combustion. Approximately 86.7 millions of tons of standard coal was burned in 2016 (S. Yang et al., 2018). Shijiazhuang has a temperate climate, with a high temperature and heavy rain in summer and a cold and dry winter. 2.1. Hospital admission data The daily number of non-accident hospital admissions of patients with hypertension between January 1, 2013 and December 31, 2016 were collected from Shijiazhuang Electronic Medical Record Data Center (SEMRC), which is administered by the Shijiazhuang Health and Family Planning Commission (SHFPC). SHFPC is a government agency that administers governmental hospitals of Shijiazhuang. Those hospitals cover all residents of Shijiazhuang and are under contract with SEMRC. Each record contains hospitalization causality diseases and chronic non-communicable diseases such as hypertension or diabetes. Computerized records of hospital admissions are maintained at each hospital and sent to SEMRC through an internal computer network.
2.3. Statistical analysis A time-series regression was applied to examine the acute effect of air pollutants on hospital admissions of people with hypertension. An over-dispersed generalized additive model (GAM) was constructed as previously described (Song et al., 2018). Several covariates were introduced into the main model: (1) a natural cubic regression smooth function of calendar time with 7 degrees of freedom (df) per year to 287
Ecotoxicology and Environmental Safety 170 (2019) 286–292
J. Song et al.
described as the percent changes and 95% CI in daily hospital admissions for patients with hypertension per 10 μg/m3 increase of PM2.5, PM10, SO2, NO2, O3, and CO.
Table 1 The summary of descriptive statistics during the study period (January 1, 2013 to December 31, 2016).
Air pollutant concentration (μg/m3) PM2.5 PM10 SO2 NO2 O3 CO Meteorological measures Temperature (°C) Humidity (%) Hospitalizations of nonaccidental patients (A00-R99) Hospitalizations of patients for hypertension (I10) Gender (N) Male Female Age (N) 18–64 ≥65 Season (N) Warm (Apr–Sep) Cool (Oct–Mar)
114 207 63 56 79 1603
95 143 59 29 53 1342
6 17 3 9 2 137
49 106 23 35 36 764
88 167 43 50 73 1153
145 273 83 70 114 1857
771 867 319 188 262 10,457
15 57 446
11 20 158
−10 12 123
5 41 305
16 58 441
24 73 537
35 98 974
3. Results After excluding non-local hospitalizations, 650,550 hospitalizations for non-accidental disease were extracted from SEMRC during our study period (2013–2016), of which 172,966 patients had hypertension (accounting for 26.6%). Table 1 summarizes the descriptive statistics on daily non-accidental hospitalizations and hospitalizations of patients with hypertension, air quality, and weather conditions. We recorded 89,299 hospitalizations of men with hypertension, and the elderly (≥65 years) accounted for 53.9%. Hospital admissions for patients with hypertension were higher in the cool season (52.2%) than in the warm season (47.8%). The annual average values of daily mean concentrations of air pollutants were 114 μg/m3 for PM2.5, 207 μg/m3 for PM10, 63 μg/m3 for SO2, 56 μg/m3 for NO2, 79 μg/m3 for O3, and 1603 μg/m3 for CO, respectively. The concentrations of PM2.5, PM2.5, SO2 and NO2 were 3.3, 3.0, 1.1, and 1.4 times the of Chinese Ambient Air Quality Standards limit. The annual mean temperature was 15 ℃ and the mean relative humidity was 57%. Air pollutants except O3 were strongly correlated with each other (Spearman's correlation coefficient distributed from 0.63 to 0.92) and moderately correlated with temperature and humidity (-0.62–0.28). Table 2 shows the effect estimates of air pollutant concentrations on daily hospital admissions for non-accidental diseases and daily hospital admissions of patients for hypertension. Fig. 2 shows the effect estimates of single-pollutant models using different lag days. The estimates using moving average lags were much higher than those using singleday lags. According to the five model fit statistics, lags of 0–6 days (lag06) for PM2.5 and PM10, 0–4 days (lag04) for SO2 and CO, 0–3 days (lag03) for NO2, and 6 days (lag6) for O3 were selected as the best lag structure as they could produce the smallest AIC/GCV/PACF value. Overall, air pollutants except SO2 were statistically significantly associated with total hospitalizations and hospital admissions of patients for hypertension. Increases of 10 μg/m3 in the concentrations of PM2.5, PM10, SO2, NO2, O3, and CO were associated with increases of 0.56% (95% CI: 0.28–0.83%), 0.31% (95% CI: 0.12–0.50%), 0.29% (95% CI: −0.31% to 0.89%), 1.18% (95% CI: 0.49–1.87%), 0.40% (95% CI: 0.09–0.71%), and 0.03% (95% CI: 0.01–0.05%), respectively, in daily hospitalizations of patients for hypertension. Furthermore, we found higher effect estimates (1.21–14.5 times) of air pollutants on hospitalization for hypertension than hospitalizations for non-accidental. However, the differences were statistically insignificant. The E-R curves for the associations between air pollutants and daily hospital admissions of patients for hypertension varied (Fig. 3). Generally, the curves of six pollutants were obviously positive. The E-R curves of PM2.5 and PM10 were almost S-shaped, rising sharply for concentrations ≥ 50 μg/m3 and becoming flat for concentrations ≥ 400 μg/m3. The E-R curve of SO2 showed a flat slope at low concentrations and then a slight increase at concentrations ≥ 150 μg/m3. The E-R curves of NO2 and CO were almost J-shaped, increasing sharply at concentrations ≥ 50 μg/m3 and 1000 μg/m3, respectively. The curve for O3 showed a steep slope at concentrations < 150 μg/m3 and then became flat. According to the results of stratification analyses (Table 3), the effect estimates for patients with hypertension were higher in the cool season. In addition, the differences between the seasons were statistically significant for PM2.5, PM10, NO2, and O3. The associations were similar in those 18–64 years of age and those of 65 years of age and older. In men, the associations of PM2.5 and PM10 were higher, whereas the association of O3 was lower. The sensitivity analysis showed robust hospitalization effects of air pollutants (Table S1). Co-pollutants with Spearman's correlation coefficient of < 0.6 were added into two-pollutant models for adjustment.
exclude unmeasured long-term and seasonal trends longer than 2 months (F. Chen et al., 2018); (2) natural smooth functions of the mean temperature (6 df) and relative humidity (3 df) to control for the nonlinear confounding effects of weather conditions (Li et al., 2015; Liu et al., 2017); (3) indicator variables for day of the week and public holidays. We plotted the exposure-response (E-R) relationship curves between air pollutants and hospitalizations of people with hypertension as Chen et al. described (C. Chen et al., 2018). In brief, a natural spline function with 3 df was added into the above mentioned model. Given the uncertainty in determining the best lag days for estimation, we used multiple lag structures including single-day lags from 0 to 7 and moving average exposure of multiple days, including lag0–1, 0–2, 0–3, 0–4, 0–5, 0–6, 0–7. To determine the best lag structure, we calculated the model fit based on three statistics: Akaike information criterion (AIC), generalized cross validation (GCV), and partial autocorrelation function (PACF). Five sensitivity analyses were conducted to check the stability of the model. First, we checked the df value from 4 to 10 per year. Second, two-pollutant models were constructed to examine the robustness of the effect estimates. Third, we analyzed the short-term association between air pollution and daily hospital admissions for accident that were not biologically related to air pollution using the same basic model. Besides that, we controlled longer lag day's temperature and relative humidity (up to 21 days), as previous studies have found that the temperature may have a prolonged health effect (Ma et al., 2015; Wang et al., 2018). In addition, three stratification analyses were conducted according to sex, age (18–64 years and ≥65 years), and season (warm: April to September; cool: October to March). The statistical significance of differences between effect estimates of the strata were tested by calculating the 95% confidence interval (CI) as (Qˆ 1 Qˆ 2 ) ± 1.96 (SÊ1)2 + (SÊ2) 2 , where Qˆ 1 and Qˆ 2 are the estimates for two categories, and SÊ1 and SÊ2 are their respective standard errors (Lin et al., 2016a). The statistical tests were two-sided, and associations with P < 0.05 were considered statistically significant. All statistical models were run in R software (version 3.3.3) using the MGCV package. The effects are 288
Ecotoxicology and Environmental Safety 170 (2019) 286–292 0.02(0.01–0.03) 0.03(0.01–0.05) 0.01(−0.01 to 0.03) 0.03(−0.01 to 0.06) 0.33(0.07–0.60) 0.40(0.09–0.71) 0.31(0.04–0.58) 0.4(−0.12 to 0.92)
Fig. 2. Percent change of hospital admissions (mean and 95% confidence internal) of patients with hypertension associated with a 10 μg/m3 increase in various air pollutant concentrations using different lag structures.
The associations of all six pollutants with hospitalizations of patients for hypertension remained robust after adjustment for co-pollutants. Fig. S1 shows that those effects didn’t change substantially with the adjustment of df from 4 to 10 per year. When we further adjusted for temperature or relative humidity of longer lag days (up to 21 days), the effects remained consistent (Fig. S2). For a further check of the stability, the association between air pollution and accidental hospitalizations were analyzed. On average, there were 12 hospital admissions for accidental causes per day. We found no effect of air pollution on hospitalizations due to accidents (data not shown), suggesting our results are robust and probably not attributable to chance. 4. Discussion
The statistically significant estimates are highlighted in bold. a Moving average of lag06 pollutants concentrations were used. b Moving average of lag04 pollutants concentrations were used. c Moving average of lag03 pollutants concentrations were used. d Single-day lag of lag6 pollutants concentrations were used.
0.78(0.17–1.38) 1.18(0.49–1.87) 0.63(0.02–1.24) 1.23(−0.08 to 2.55) 0.02(−0.50 to 0.53) 0.29(−0.31 to 0.89) -0.07(−0.59–0.44) -0.28(−1.37–0.81) 0.24(0.08–0.40) 0.31(0.12–0.50) 0.22(0.06–0.38) 0.23(−0.11 to 0.57) Hospitalizations Hospitalizations Hospitalizations Hospitalizations
of of of of
non-accidental patients patients for hypertension patients without hypertension accidental
0.45(0.21–0.69) 0.56(0.28–0.83) 0.41(0.17–0.65) 0.25(−0.26 to 0.76)
NO2c SO2b PM10a PM2.5a
Table 2 Percent change (95% CI) in hospital admission of patients for hypertension with a 10 μg/m3 increase in air pollutants concentrations in Shijiazhuang.
J. Song et al.
To our knowledge, this analysis is the first attempt to demonstrate associations between air pollution and hospitalizations of patients for hypertension in China. Specifically, our study suggested that short-term exposure to air pollutants (PM2.5, PM10, NO2, O3, and CO) was significantly associated with an increased risk of hospital admissions for individuals with hypertension. The effect estimates were higher for individuals with hypertension than for the whole population. The associations between those five pollutants (PM2.5, PM10, NO2, O3, and CO) and hospitalizations of patients for hypertension were robust when adjusted for co-pollutants in two-pollutant model. The associations were stronger in the cool season than in the warm season. The associations between air pollutants and hospital admissions of patients for hypertension varied by sex or age, but the differences were not statistically significant. This analysis added to the limited evidence of health effects of air pollution in developing countries. Shijiazhuang was one of the most polluted cities in China. During our study period, the mean concentrations of PM2.5 (114 μg/m3) and PM10 (207 μg/m3) far exceeded the Chinese National Ambient Air Quality Standards (35 and 70 μg/m3, respectively). The heavy pollution 289
Ecotoxicology and Environmental Safety 170 (2019) 286–292
J. Song et al.
Fig. 3. The exposure-response relationship curves of PM2.5, PM10, SO2, NO2, O3 and CO with hospital admissions of patients with hypertension.
might have been driven by the rapid expansion of the city, fast development of industrial production, and growing use of motor vehicles. Combined with an aging population, hypertension has become the leading risk factor of cardiovascular disease-related death in Chinese adults (Bundy and He, 2016). Recent estimates indicate that 33.6% (35.3% in men and 32% in women), or 335.8 million (178.6 million men and 157.2 million women), of the Chinese adult population had hypertension in 2010 (Bundy and He, 2016). Even if the adverse effect attributable to exposure to air pollutants in an individual might be small, the overall attributable risk might be considerably higher, given the higher pollutant concentrations and large proportion of the population with hypertension in China. The number of studies on the association between air pollution and risk of hypertension has increased in the past few years. We observed significant associations between air pollutants (PM2.5, PM10, NO2, O3, and CO) and hospitalizations for hypertension. These findings are generally consistent with those of previous studies (Chen and Yang, 2018; Liu et al., 2017; Sanidas et al., 2017). Systematic reviews showed that short-term exposure to four air pollutants (PM2.5, PM10, NO2, and O3) were significantly associated with hypertension (ORs: 1.05–1.10) (C. Yang et al., 2018). Guo et al. found that 10 μg/m3 increases in
PM2.5, PM10, and NO2 were significantly associated with emergency hospital visits for hypertension, with odds ratios (ORs) of 1.084 (95% CI: 1.028–1.139), 1.060% (95% CI: 1.020–1.101), and 1.101 (95% CI: 1.038–1.168), respectively (Guo et al., 2010a, 2010b). Significant correlation was also observed between hospitalizations for hypertension and O3 with an OR of 1.2 (95% CI: 1.03–1.4) on warm (> 23 °C) days and an OR of 1.2 (95%CI: 1.02–1.42) on cool (< 23 °C) days (Chen and Yang, 2018). Several previous studies found a statistically significant association between SO2 and hypertension (Cai et al., 2016; Guo et al., 2010a). For example, Guo et al. (2010a) observed a 1.037 (95% CI: 1.004–1.071) increase in emergency visits corresponded to a 10 μg/m3 increase in SO2. However, we only observed a positive but non-significant association. The inconsistency might be due to differences in the proportions of pollutants, climate, and population characteristics (such as age, body mass index, education and economic level, and lifestyle). A weak but statistically significant association between CO and hospital admissions of patients for hypertension was observed in the present study. No previous study reported a significant association, but several studies indicated that high levels of CO were associated with increased arterial blood pressure (C. Liu et al., 2018; Quinn et al., 2016). In addition, Liu et al. (2017) observed a 1.12% (95% CI:
Table 3 Percent change (95% CI) in hospitalizations with a 10 μg/m3 increase in air pollutants concentrations by season, gender and age among patients had hypertension in Shijiazhuang.
Seasone Gender Age
Cool Warm Female Male 18–64 ≥65
0.62(0.28–0.95)* −0.13(−0.76 to 0.50)* 0.46(0.15–0.77) 0.65(0.35–0.95) 0.54(0.22–0.85) 0.57(0.28–0.87)
0.34(0.11–0.57)* -0.22(-0.65–0.21)* 0.22(0.01–0.43) 0.40(0.19–0.60) 0.28(0.06–0.49) 0.34(0.14–0.54)
0.52(−0.21 0.17(−1.20 0.38(−0.29 0.20(−0.46 0.21(−0.48 0.35(−0.30
The statistically significant estimates are highlighted in bold. a Moving average of lag06 pollutants concentrations were used. b Moving average of lag04 pollutants concentrations were used. c Moving average of lag03 pollutants concentrations were used. d Single-day lag of lag6 pollutants concentrations were used. e Cool season: from October to March; Warm season: from April to September. * Statistically significant for between-group difference. 290
to to to to to to
1.24) 1.54) 1.05) 0.86) 0.91) 0.99)
1.28(0.42–2.14)* -0.52(-2.05–1.02)* 1.18(0.41–1.95) 1.18(0.42–1.94) 1.16(0.36–1.96) 1.19(0.46–1.93)
1.41(0.77–2.04)* 0.16(−0.16 to 0.48)* 0.43(0.09–0.78) 0.37(0.03–0.71) 0.40(0.03–0.76) 0.41(0.08–0.74)
0.03(0.01–0.05) −0.01(−0.08 to 0.06) 0.03(0.01–0.05) 0.03(0.01–0.05) 0.03(0.01–0.05) 0.03(0.01–0.05)
Ecotoxicology and Environmental Safety 170 (2019) 286–292
J. Song et al.
0.42–1.83) increment in cardiovascular mortality corresponded to a 1000 μg/m3 increase in CO, which may partially explain the differences. For particulate matter, the effect of PM2.5 was higher than that of PM10, which was consistent with the findings of several previous studies (Liu et al., 2013; Meng et al., 2013; Wang et al., 2018). Several mechanisms proposed previously might explain that difference. First, PM10 can enter only the respiratory tract, whereas PM2.5 can reach all the way to the alveoli in the lungs and can also penetrate into the systemic circulation (Godleski et al., 2000; Gordana and Ivana, 2018). Second, PM2.5 had a higher particle number and larger active surface area than PM10 (Brown et al., 2001; Lin et al., 2016b). Third, the larger surface area could carry a larger amount of toxic air pollutants, including oxidant gas pollutants, transition metals, organic compounds, and micro-organisms, which have been identified as generating a pro-inflammatory response through the action of reactive oxygen species (Brown et al., 2001; Lin et al., 2016b). Another interesting finding of our study is that all of the effect estimates for the six air pollutants in the population with hypertension were higher than those in the whole population. All previous studies of both short-term and long-term exposure were designed to study the effects of air pollution on the morbidity or mortality of hypertension. Our results indicate that for people with hypertension, more attention should be paid to the health hazards of air pollution. However, Yang et al. (2017) found that exposure to ambient air pollution was more strongly associated with prehypertension than with hypertension. The exact reason for the difference needs to be elucidated in future studies. Stronger effects were observed in the cool season in this study. The differences of effects for PM2.5, PM10, NO2, O3, and CO between the warm season and the cool season were statistically significant. One possible explanation is related to increased blood pressure and viscosity in the cold season, which could be important causal factors in increasing the health effects of air pollution on people with hypertension. Another possible reason is varying concentrations, sources and composition of air pollution in different seasons. Seasonal variation of hypertension prevalence might be a third cause of this difference. One study reported that the hypertension prevalence was highest in winter (35.9%) and lowest in summer (27.8%) in China in 2010 (Li et al., 2017). The higher association in the cool season is consistent with the findings of the Brook et al. study (Brook and Kousha, 2015), but contrary to the findings of the Chen et al. study (Chen and Yang, 2018; Chiu et al., 2017) and Lin et al. (2016a) study. The difference might also be attributed to different population structure and pollutant characteristics. Generally higher association effects were found in the older group (≥65 years of age), which is consistent with the findings of previous reports (Brook and Kousha, 2015; Lin et al., 2016a; C. Liu et al., 2018; B.Y. Yang et al., 2018). The government has reported that the prevalence of non-communicable disease has been increasing steeply with increasing age in China (Commission, 2015). Specifically, the hypertension prevalence in those 18–44, 45–59, and ≥ 60 years of age was 11.3%, 36.6%, and 60.6%, respectively (Commission, 2015). Furthermore, the diabetes prevalence in those 18–29, 30–39, 40–49, 50–59, 60–69, and ≥ 70 years of age was 3.4%, 5.5%, 9.6%, 14.6%, 18.9%, and 20.5%, respectively (Commission, 2015). All of those noncommunicable diseases greatly reduce the body's resistance to environmental hazards. We hypothesize that preexisting diseases (noncommunicable diseases such as diabetes mellitus and cancer) may make older people more likely to develop elevated blood pressure or hypertension (Wang et al., 2017; B.Y. Yang et al., 2018). For patients with hypertension, the effect of each air pollutant on hospital admission varied by sex in this study. Particulate matter (PM2.5 and PM10) had stronger effects in men, whereas gaseous pollutants (SO2 and O3) had stronger effects in women. These findings are consistent with previous studies (Brook and Kousha, 2015). Yang et al. suggested that the sex-specific effects might be caused by both sex-linked
biological differences and socially derived gendered exposure (B.Y. Yang et al., 2018). For example, lung size and growth, deposition of particles, gas absorption, gas-blood barrier permeability, airway hyperresponsiveness, vascular response, and inflammation all differ by sex (B.Y. Yang et al., 2018). Meanwhile, the prevalence of hypertension is higher in men (34.5%, 95% CI: 33.2–35.9) than in women (29.5%, 95% CI: 28.3–30.7) in China (Li et al., 2017). In addition, men participate more outdoor activities and have lower personal protective intentions (such as wearing particulate-filtering respirators), which may lead to increased exposure to particulate matter (Shi et al., 2017; B.Y. Yang et al., 2018). Exploration of the shape of E-R relationships is crucial for public health assessment, especially for those populations with preexisting non-communicable diseases. In this study, we observed a linear and non-threshold relationship between air pollution and the risk of hospitalization for populations with hypertension. The S-shaped curves for particulate matter (PM2.5 and PM10) tend to plateau at high levels of pollutant. This may be a consequence of the “harvesting effect” in that susceptible people might have been hospitalized before air pollutant concentrations reached a reasonably high level (Chen et al., 2017). That might also explain the saturation effect of O3. Although E-R relationships may vary by air pollution mixture, climatic characteristics, and population sensitivity (C. Chen et al., 2018), there are still public health implications such that air pollution levels should be further limited to protect those populations with hypertension. Although the exact mechanisms behind the association of air pollution with increased hospitalization of patients with hypertension have not been determined, there is growing evidence suggesting that the hypertension effect of air pollution might be caused by three potential pathways: 1) classic pathway, provocation of systemic inflammation and oxidative stress. A variety of inflammatory/prooxidative factors (cytokines such as interleukin-6, interleukin-1β and tumor necrosis factor TNF-α; activated immune cells such as T- and B-lymphocytes; or platelets) are released into the systemic circulation (Sanidas et al., 2017; Vidale and Campana, 2018); 2) alternative pathway, translocation into the circulation directly. Some components of pollutants might be able to pass through the alveolar capillary membrane directly into the systemic circulation and impair vasomotor activity (Sanidas et al., 2017; Vidale and Campana, 2018); 3) central pathway, activation of autonomic nervous system. Previous studies have shown that the activation of receptor-mediated autonomic reflexes in the pulmonary tree by pollutants influences blood pressure and heart rate variability and rhythm (Sanidas et al., 2017; Vidale and Campana, 2018). These mechanisms might overlap and/or be generated at different times, and the type and size of pollutants would determine the relative importance of the mechanisms (Sanidas et al., 2017; Vidale and Campana, 2018). Well-designed studies are needed to clarify the detailed mechanisms in the future. This study had some limitations. First, as in most previous studies, we used fixed-site monitor measurements as a proxy for personal exposure, resulting in exposure errors. However, the resultant non-differential error was reported to produce an underestimate of the associations of ambient air pollution (Samet et al., 2000). Second, owing to the limitation of the available data, we did not obtain data on several factors such as time of diagnosis and control measures for hypertension. The sensitivity might be different between new cases and current cases (Yang et al., 2017). Third, other preexisting diseases and unhealthy factors such as diabetes, cancer, and obesity were not calculated owing to data limitations. Those preexisting diseases may reduce the body's tolerance of pollutants. Fourth, there might be few family aggregation cases, which may affect the population sensitivity (Y. Yang et al., 2018), but we did not collect data on the relationships between patients. Fifth, we could not obtain socioeconomic status and education data, and these variables might be highly related to lifestyle and awareness of health protection and therefore affect health status. Sixth, this study collected data from only one highly polluted city, and the generalizability of our 291
Ecotoxicology and Environmental Safety 170 (2019) 286–292
J. Song et al.
results might be limited.
ambient air particles. Res Rep. Health Eff. Inst. 5–88 (discussion 89-103). Gordana, P., Ivana, J., 2018. Carcinogenic potency of airborne polycyclic aromatic hydrocarbons in relation to the particle fraction size. Int. J. Environ. Res. Public Health 15, 2485. Guo, Y., et al., 2010a. Gaseous air pollution and emergency hospital visits for hypertension in Beijing, China: a time-stratified case-crossover study. Environ. Health 9, 57. Guo, Y., et al., 2010b. The relationship between particulate air pollution and emergency hospital visits for hypertension in Beijing, China. Sci. Total Environ. 408, 4446–4450. Hadley, M.B., et al., 2018. Air pollution and cardiovascular disease: a window of opportunity. Nat. Rev. Cardiol. 15, 193–194. Landrigan, P.J., et al., 2018. The Lancet Commission on pollution and health. Lancet 391, 462–512. Li, H., et al., 2015. Short-term exposure to ambient air pollution and coronary heart disease mortality in 8 Chinese cities. Int. J. Cardiol. 197, 265–270. Li, Y., et al., 2017. Burden of hypertension in China: a nationally representative survey of 174,621 adults. Int. J. Cardiol. 227, 516–523. Lin, H., et al., 2016a. Mortality burden of ambient fine particulate air pollution in six Chinese cities: results from the Pearl River Delta study. Environ. Int. 96, 91–97. Lin, H., et al., 2016b. Particle size and chemical constituents of ambient particulate pollution associated with cardiovascular mortality in Guangzhou, China. Environ. Pollut. 208, 758–766. Liu, C., et al., 2017. Associations between ambient fine particulate air pollution and hypertension: a nationwide cross-sectional study in China. Sci. Total Environ. 584–585, 869–874. Liu, C., et al., 2018. Ambient carbon monoxide and cardiovascular mortality: a nationwide time-series analysis in 272 cities in China. Lancet Planet Health 2, e12–e18. Liu, H., et al., 2018. Effect of ambient air pollution on hospitalization for heart failure in 26 of China's largest cities. Am. J. Cardiol. 121, 628–633. Liu, L., et al., 2013. Size-fractioned particulate air pollution and cardiovascular emergency room visits in Beijing, China. Environ. Res. 121, 52–63. Ma, W., et al., 2015. The temperature-mortality relationship in China: an analysis from 66 Chinese communities. Environ. Res. 137, 72–77. Mazidi, M., Speakman, J.R., 2018. Impact of obesity and ozone on the association between particulate air pollution and cardiovascular disease and stroke mortality among US adults. J. Am. Heart Assoc. 7. Meng, X., et al., 2013. Size-fractionated particle number concentrations and daily mortality in a Chinese city. Environ. Health Perspect. 121, 1174–1178. Pitchika, A., et al., 2017. Long-term associations of modeled and self-reported measures of exposure to air pollution and noise at residence on prevalent hypertension and blood pressure. Sci. Total Environ. 593–594, 337–346. Qiu, H., et al., 2018. Long-term exposure to fine particulate matter air pollution and type 2 diabetes mellitus in elderly: a cohort study in Hong Kong. Environ. Int. 113, 350–356. Quinn, A.K., et al., 2016. Association of carbon monoxide exposure with blood pressure among pregnant women in rural Ghana: evidence from GRAPHS. Int. J. Hyg. Environ. Health 219, 176–183. Samet, J.M., et al., 2000. The National Morbidity, Mortality, and Air Pollution Study. Part II: morbidity and mortality from air pollution in the United States. Res. Rep. Health Eff. Inst. 94, 5–70 (discussion 71-79). Sanidas, E., et al., 2017. Air pollution and arterial hypertension. A new risk factor is in the air. J. Am. Soc. Hypertens. 11, 709–715. Shi, J., et al., 2017. Cardiovascular benefits of wearing particulate-filtering respirators: a randomized crossover trial. Environ. Health Perspect. 125, 175–180. Song, J., et al., 2018. Acute effects of ambient particulate matter pollution on hospital admissions for mental and behavioral disorders: a time-series study in Shijiazhuang, China. Sci. Total Environ. 636, 205–211. Stachyra, K., et al., 2017. Air pollution and atherosclerosis - a brief review of mechanistic links between atherogenesis and biological actions of inorganic part of particulate matter. Folia Med Cracov. 57, 37–46. Vidale, S., Campana, C., 2018. Ambient air pollution and cardiovascular diseases: from bench to bedside. Eur. J. Prev. Cardiol. 25, 818–825. Wang, C., et al., 2017. Acute inflammation following personal exposure to fine-particulate air pollution. Am. J. Epidemiol. Wang, X., et al., 2018. Estimating the acute effects of fine and coarse particle pollution on stroke mortality of in six Chinese subtropical cities. Environ. Pollut. 239, 812–817. Xia, R., et al., 2017. Ambient air pollution and out-of-hospital cardiac arrest in Beijing, China. Int. J. Environ. Res. Public Health 14. Yang, B.Y., et al., 2018. Global association between ambient air pollution and blood pressure: a systematic review and meta-analysis. Environ. Pollut. 235, 576–588. Yang, B.Y., et al., 2017. Is prehypertension more strongly associated with long-term ambient air pollution exposure than hypertension? Findings from the 33 Communities Chinese Health Study. Environ. Pollut. 229, 696–704. Yang, S., et al., 2018. Characteristics and formation of typical winter haze in Handan, one of the most polluted cities in China. Sci. Total Environ. 613–614, 1367–1375. Yang, Y., et al., 2018. Ambient fine particulate pollution associated with diabetes mellitus among the elderly aged 50 years and older in China. Environ. Pollut. 243, 815–823.
5. Conclusions This time-series study suggested that air pollution could significantly increase the risk of hospitalization for individuals with hypertension in Shijiazhuang, China. Even though the air pollution level has been sharply reduced in 2017, implementation of policies that seek to improve air pollution should still be a government priority. Acknowledgement The study was supported by the National Natural Science Foundation of China (21677136), the Ph.D. Research Project of Xinxiang Medical University (XYBSKYZZ201804), Key Scientific Research Projects In Universities Of Henan Province (19B330004) and Peak Subject Project of Public Health in Xinxiang Medical University. This study does not involve experimental animals or individual information of human subjects. Disclosures The authors declare they have no actual or potential competing financial interests. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ecoenv.2018.11.125. References Anderson, H.R., 2017. Implications for the science of air pollution and health. Lancet Respir. Med. 5, 916–918. Brook, R.D., Kousha, T., 2015. Air pollution and emergency department visits for hypertension in Edmonton and Calgary, Canada: a case-crossover study. Am. J. Hypertens. 28, 1121–1126. Brook, R.D., et al., 2010. Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association. Circulation 121, 2331–2378. Brown, D.M., et al., 2001. Size-dependent proinflammatory effects of ultrafine polystyrene particles: a role for surface area and oxidative stress in the enhanced activity of ultrafines. Toxicol. Appl. Pharmacol. 175, 191–199. Bundy, J.D., He, J., 2016. Hypertension and related cardiovascular disease burden in China. Ann. Glob. Health 82, 227–233. Cai, Y., et al., 2016. Associations of short-term and long-term exposure to ambient air pollutants with hypertension: a systematic review and meta-analysis. Hypertension 68, 62–70. Chen, C., et al., 2018. Ambient air pollution and daily hospital admissions for mental disorders in Shanghai, China. Sci. Total Environ. 613–614, 324–330. Chen, C.C., Yang, C.Y., 2018. Association between gaseous air pollution and hospital admissions for hypertension in Taipei, Taiwan. J. Toxicol. Environ. Health A. 81, 53–59. Chen, F., et al., 2018. The effects of PM2.5 on asthmatic and allergic diseases or symptoms in preschool children of six Chinese cities, based on China, Children, Homes and Health (CCHH) project. Environ. Pollut. 232, 329–337. Chen, R., et al., 2017. Fine particulate air pollution and daily mortality. A nationwide analysis in 272 Chinese cities. Am. J. Respir. Crit. Care Med. 196, 73–81. Chiu, H.F., et al., 2017. Short-term effects of fine particulate air pollution on hospital admissions for hypertension: a time-stratified case-crossover study in Taipei. J. Toxicol. Environ. Health A. 80, 258–265. Commission, N. H. a. F. P., 2015. Report on Nutrition and chronic Diseases of Chinese residents. Cosselman, K.E., et al., 2015. Environmental factors in cardiovascular disease. Nat. Rev. Cardiol. 12, 627–642. Dai, X., et al., 2015. Short-term effects of air pollution on out-of-hospital cardiac arrest in Shenzhen, China. Int. J. Cardiol. 192, 56–60. Godleski, J.J., et al., 2000. Mechanisms of morbidity and mortality from exposure to