Description

A study of air pollutants influencing life expectancy and longevity from spatial perspective in China

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Related Documents

Share

Transcript

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institutionand sharing with colleagues.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:http://www.elsevier.com/authorsrights
Author's personal copy
A study of air pollutants in
ﬂ
uencing life expectancy and longevity fromspatial perspective in China
Li Wang
a,b
, BingganWei
b
, Yonghua Li
b,
⁎
, Hairong Li
b
, Fengying Zhang
a,c
, Mark Rosenberg
d
, Linsheng Yang
b
, Jixia Huang
b
, Thomas Krafft
a,b,e
, Wuyi Wang
b
a
Department of International Health, Faculty of Health, Medicine and Life Sciences, Maastricht University, The Netherlands
b
Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
c
China National Environmental Monitoring Center, Beijing 100012, China
d
Department of Geography, Queen's University, Kingston K7L 3N6, Canada
e
Institute of Environmental Education and Research, Bharati Vidyapeeth University, Pune, India
H I G H L I G H T S
•
Using GWR to investigate the spatial correlations between health and air pollutants.
•
Difference of 10
μ
g/m3 in SO
2
can cause adjusted 0.28 year in life expectancy.
•
Difference of 10
μ
g/m3 in PM
10
can lead to a difference of 2.23 in longevity ratio.
•
Per capita GDP was positively associating with life expectancy in China.
a b s t r a c ta r t i c l e i n f o
Article history:
Received 24 January 2014Received in revised form 28 March 2014Accepted 31 March 2014Available online 23 April 2014Editor name: P. Kassomenos
Keywords:
Life expectancyLongevityAir pollutantsPer capita GDPGeographically Weighted RegressionStepwise Regression
Life expectancy and longevity are in
ﬂ
uenced by air pollutants and socioeconomic status, but the extend andsigni
ﬁ
cance are still unclear. Better understanding how the spatial differences of life expectancy and longevityare affected by air pollutants is needed for generating public health and environmental strategies since thewhole of China is now threatened by deteriorated air quality. 85 major city regions were chosen as researchareas. Geographically Weighted Regression (GWR) and Stepwise Regression (SR) were used to
ﬁ
nd the spatialcorrelationsbetween health indicators and air pollutants, adjusted by per capita GDP
1
. The results were, re-gions with higher life expectancy were mainly located in the east area and areas with good air quality, a re-gional difference of 10
μ
g/m
3
in ambient air SO
22
could cause adjusted 0.28 year's difference in lifeexpectancy, a regional difference of 10
μ
g/m
3
in ambient air PM
103
could lead to a longevity ratio differenceof 2.23, and per capita GDP was positively associating with life expectancy but not longevity ratio, with aregional difference of 10,000 RMB
4
associating with adjusted 0.49 year's difference in life expectancy.This research also showed the evidences that there exist spatially differences for ambient air PM
10
andSO
2
in
ﬂ
uencing life expectancy and longevity in China, and this in
ﬂ
uences were clearer in south China.© 2014 Elsevier B.V. All rights reserved.
1. Background
Life expectancy is considered as one of the three major parametersfor calculating the Human Development Index (HDI), which is used bytheUnited NationsDevelopmentProgramtorankhumandevelopmentlevels of countries (UNDP, 2011). It has also been used to assess thehealth impact of air pollution (Chen et al., 2013; Pope et al., 2009b;Popeet al.,2013;Wangetal., 2013).Lifeexpectancyisaffected by mul-tiple factors, and the social environment is considered as one of themost important factors (Blum, 1974). Ambient air SO
2
, PM, and NO
x
have been proved to lead to multiple diseases and diminished life ex-pectancy (Chen et al., 2013; Cao et al., 2011; Wang et al., 2013; Pope
et al., 2002, 2003, 2009a,b, 2013). The elderly population is found tobe more vulnerable to air pollutants except ozone and more sensitiveto air pollutants because of their depressed immune systems, existingdiseases, and the accumulation of toxic agents in their bodies (Fischeret al., 2003; Sun and Gu, 2008). Most of the studies on air pollution af-fecting the elderly focused on the population of 65 or 75 years old and
Science of the Total Environment 487 (2014) 57
–
64
⁎
Corresponding author at: Institute of Geographic Sciences and Natural ResourcesResearch, Chinese Academy of Sciences, 11 A Datun Road, Beijing 100101, China.Tel.: +86 10 64889198; fax: +86 10 64856504.
E-mail address:
yhli@igsnrr.ac.cn (Y. Li).
1
GDP, gross domestic product.
2
SO
2
, sulfur dioxide.
3
PM
10
, particular matter with diameter
b
10
μ
m.
4
RMB, renminbi.http://dx.doi.org/10.1016/j.scitotenv.2014.03.1420048-9697/© 2014 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
Author's personal copy
over(Fischeretal.,2003;WenandGu,2012).Duetothetoxicaccumu-
lation effect, the 100 years old and over population is more sensitive toairpollutants.Aclearunderstandingoftheeffectsofairpollutantstothelongevity group (people living to be 100 and over) is critically neededfor the county with the increasing elderly population. The negative as-sociationsbetweenairpollutantsandhealthoutcomeshavebeenlarge-ly proved through time-series or cohort based studies (Zhang et al.,2011; Cao et al., 2011; Gouveia and Fletcher, 2000). But there was nostudy considering the regional differences of population's sensitivity toair pollutants. A research regarding spatial differences of health out-comesassociatingwithairpollutantsisurgentlyneededforformulatingregional healthy coping strategies against air pollutants.In thispaper, we used lifeexpectancyandlongevityaspublichealthoutcomes.SocioeconomicrepresentedbypercapitaGDP(grossdomes-tic product), and air pollutants, including PM
10
and SO
2
, which areregarded as themajor ambient air pollutantschallengingChina's publichealth, were analyzed to
ﬁ
nd out how they in
ﬂ
uence life expectancyand longevity. We hypothesized that spatial differences of PM
10
andSO
2
were associated with changes in life expectancy and longevity. Re-gional differences in life expectancy and longevity could also be partlyattributed to socioeconomic status. This research was conducted at theprefecture-city level.Themajorobjectivesofthispaperwere:(1)todemonstratethespa-tial distribution of life expectancy and longevity at the prefecture-citylevel; (2) to analyze the spatial relationship between air pollutantsandthetwohealthindicators;(3)andtoanalyzethesocioeconomicef-fect on life expectancy and longevity. The research
ﬁ
ndings lead to pol-icy recommendations for decision-makers to use in developing healthstrategies to help the elderly population cope with the deteriorated airquality and to
ﬁ
nd a balance between environmental protection andsocioeconomic development from a public health perspective.
2. Method and data
2.1. Method and explanatory variables for models
Using ArcGIS software, we constructed a geographic and attributedatabase of the longevity group using an administrative map and thesixth national population census of China. Geographic distributionmaps of the longevity group, life expectancy, PM
10
, SO
2
, per capitaGDP were generated on prefecture-city level. In order to
ﬁ
nd out thepossibleeffectofpollutantsandsocioeconomicfactorsonlifeexpectan-cyandlongevity,wecollecteddataonPM
10
,SO
2
,andpercapitaGDPforour analysis. Life expectancy and longevity ratio were analyzed againstair pollutants and socioeconomic indicator using the Stepwise Regres-sion (SR) by SPSS 18.0 and Geographically Weighted Regression(GWR) by ArcGIS 10.1.In SR, PM
10
, SO
2
and per capita GDP were used as independent var-iables, while life expectancy and longevity ratio were dependent vari-ables, respectively. After we conducted SR, we used the variablesincludedbySRforGWR.GWRgeneratesaseparateregressionequationfor every feature analyzed in a sample dataset as a means to addressspatial variation. A general version of the model can be expressed as:
y
i
¼
β
0
u
i
;
v
i
ð Þ þ
X
n z
¼
1
β
z
u
i
;
v
i
ð Þ
x
iz
þ
ε
i
ð
1
Þ
Where
y
i
denotesthedependentvariable,inthiscasethelifeexpec-tancy or longevity i at location i,
β
0
(
u
i
,
v
i
) denotes the intercept coef
ﬁ
-cient at location i,
x
iz
is the value of the zth explanatory variable atlocation i and
β
z
(
u
i
,
v
i
) is the location regression coef
ﬁ
cient for the zthexplanatory variable. Furthermore, (
u
i
,
v
i
) denotes Cartesian x and ypoint coordinates and
ε
i
is the random location speci
ﬁ
c error term.When GWR was used, the parameters can be estimated by solving:
β
f
u
i
;
v
i
ð Þ ¼
X
T
W u
i
;
v
i
ð Þ
X
−
1
X
T
W u
i
;
v
i
ð Þ
y
ð
2
Þ
where
β
f
(
u
i
,
v
i
) is the estimate of the location-speci
ﬁ
c parameter,
W
(
u
i
,
v
i
) is an n by n spatial weight matrix whose off-diagonal elements arezero and the diagonal elements denote the geographical weights of ob-served data alocationi. The geographic weightstructure(
u
i
,
v
i
)is basedon a Gaussian Kernel function such that the in
ﬂ
uence of data points inthe proximity of i is given larger weights in the estimation.This paper used anadaptive bi-square function to generate the geo-graphicweights.Anadaptivefunction
ﬁ
ttedthedemographicdataana-lyzed in this paper since the research points clustered in some regions.The spatial context (the Gaussian kernel) is a function of a speci
ﬁ
ednumber of neighbors. Where feature distribution is dense, the spatialcontext is smaller; where feature distribution is sparse, the spatial con-text is larger (Charlton et al., 2009).Thebandwidthmay beeither de
ﬁ
ned bya given distance, or a
ﬁ
xednumber of nearest neighbors from the analysis location. In this case weused AICs, that the optimal number of nearest neighbors was deter-mined throughselecting themodelwiththelowestAkaike InformationCriterion (AIC) score (Hurvich et al., 1998), given as:
AICc
¼
2nln
σ
ð Þ þ
nln 2
π
ð Þ þ
n
n
þ
tr s
ð Þ
n
−
2
−
tr s
ð Þ
:
ð
3
Þ
Heretr(s)isthetraceofthehatmatrix.TheAIC methodcanbeusedtoselectbetweenanumberofcompetingmodelsbytakingintoaccountdifferences in model complexity (Fotheringham et al., 2002).
2.2. Data
Life expectancy data was calculated from demographic data, whichwere obtained from the demographic database of the six national pop-ulation census of China (National Bureau of Statistics of China, 2010b).The formula for calculating life expectancy was illustrated in Table 1.From left to right, where, x represents age. l(x) is
“
the survivorshipfunction
”
:thenumberofpersonsaliveatagex.Forexampleoftheorig-inal 100,000 people in the hypothetical cohort, l(14
–
19) = 98,989 (or98.989%) live to age 14
–
19. These values are computed recursivelyfrom the d(x) values using the formula l(x + i) = l(x)
−
d(x), withl(0), the
“
radix
”
of the table, arbitrarily set to 100,000. For example,l(1
–
4) = l(0)
−
d(0) = 100,000
−
550 = 99,450. d(x) is the numberof deaths in the interval (x,x + i) for persons alive at age x, computedas d(x) = q(x)
∗
I(x). For example, the I(10
–
14) = 99,127 personsalive at age 10
–
14, d(10
–
14) = 0.00139
∗
99,127 = 138 died prior toage 10
–
14. q(x) is probability of dying at age x. Also known as the(age-speci
ﬁ
c) risk of death. Generally these are derived using the for-mulaq(x)=1
−
exp[
−
m(x)],undertheassumptionthattheinstanta-neousmortalityrate,orforceofmortality,remainsconstantthroughouttheage intervalfrom x to x + i(here i = 5).m(x): themortality rate atage x. Generally these quantities are obtained between P(x) and D(x).By construction, m(x) = D(x) / P(x). D(x) and P(x) are death and livenumber of population at x year, which are the inputs in the life table.Thedataareobtained fromnationalcensus.L(x)istotalnumberofper-son
–
yearsalivebythecohortfromagextox+I(herei=5).Thisisthesum of the years lived by the l(x) and I(x + i) persons who survive theinterval.L(x)=i/2[I(x)+I(x+i)].ForI(0),thisisthesumoftheyearslived by the L(1) persons who survive, and the d(x) persons who dieduring the interval, L(0) = I(1) + 0.5
∗
d(0).T(x) is the total numberof person
–
years lived by the cohort from age x until all members of the cohort have died. This is the sum of numbers in the L(x) columnfrom agex to thelastrow in thetable. e(x):the (remaining) lifeexpec-tancy of persons alive at age x, computed as e(x) = T(x) / l(x).
58
L. Wang et al. / Science of the Total Environment 487 (2014) 57
–
64
Author's personal copy
The longevity ratio was de
ﬁ
ned as the number of centenarians per10,000 inhabitants. This indicator re
ﬂ
ects the feature of longevity. Thedata were also obtained from demographic database of the six nationalpopulation census of China (National Bureau of Statistics of China,2010b). Per capita GDP data were from the provincial statistical yearbook (National Bureau of Statistics of China, 2010a). PM
10
and SO
2
data were obtained from the data center of Ministry of EnvironmentalProtection of the People's Republic of China (Chinese NationalEnvironmental Monitoring Center, 2010). PM
10
concentration wasmonitored daily according to the Gravimetric method (GB 6921-86),and SO
2
concentration was monitoreddaily according to formaldehydeabsorbing-pararosaniline spectrophotometry (GB/T 15262) or thetetrachloromercurate (TCM)-pararosaniline method (GB 8970). All themonitoring equipments were calibrated regularly, and all thesemethods meet the National Ambient Air Quality Standard (GB 3095-1996). Month concentrations of the two pollutants were obtained by
Table 1
Abbreviated decennial life table for one city region in China.Age (yr) I(x) d(x) q(x) m(x) L(x) T(x) e(x) P(x) D(x)0 100,000 550 0.00550 0.00550 99533 7987240 79.87 47647 2621
–
4 99,450 185 0.00187 0.00047 397429 7887707 79.31 188449 885
–
9 99,264 137 0.00138 0.00028 495980 7490278 75.46 217311 6010
–
14 99,127 138 0.00139 0.00028 495292 6994298 70.56 283046 7914
–
19 98,989 217 0.00220 0.00044 494402 6499005 65.65 326943 144
… … … … … … … … … …
80
–
84 58,180 16,974 0.29175 0.06831 248468 568788 9.78 34019 232485+ 41,206 41,206 1 0.12864 320320 320320 7.77 16542 2128
Fig.1.
Distributionofresearchregions.Thebluelineisprovinceboundaryandthegraylineiscityregionboundary.Thelocationsofthecityregionsincludedinthisstudyareshowningray.Thoseregionsarecodedbynumbersasfollows:1
—
Beijing,2
—
Tianjing,3
—
Shijiazhuang,4
—
Qinghuangdao,5
—
Taiyuan,6
—
Datong,7
—
Yangquan,8
—
Changzhi,9
—
Huhehaote,10
—
Chifeng,11
—
Shenyang, 12
—
Dalian, 13
—
Anshan, 14
—
Fushun, 15
—
Changchun, 16
—
Haerbing, 17
—
Qiqihaer, 18
—
Mudanjiang, 19
—
Shanghai, 20
—
Nanjing, 21
—
Suzhou, 22
—
Nantong, 23
—
Lianyungang,24
—
Yangzhou, 25
—
Zhenjiang, 26
—
Hangzhou, 27
—
Ningbo, 28
—
Wenzhou, 29
—
Huzhou, 30
—
Shaoxing, 31
—
Hefei, 32
—
Wuhu, 33
—
Fuzhou, 34
—
Xiamen, 35
—
Quanzhou, 36
—
Nanchang,37
—
Jiujiang, 38
—
Jinan, 39
—
Qingdao, 40
—
Zibo, 41
—
Zaozhuang, 42
—
Yantai, 43
—
Weifang, 44
—
Jining, 45
—
Taian, 46
—
Zhenzhou, 47
—
Kaifeng, 48
—
Pingdingshan, 49
—
Wuhan, 50
—
Jingzhou,51
—
Changsha, 52
—
Changde, 53
—
Zhangjiajie, 54
—
Guangzhou, 55
—
Shaoguan, 56
—
Shenzhen, 57
—
Zhuhai, 58
—
Shantou, 59
—
Zhanjiang, 60
—
Nanning, 61
—
Liuzhou, 62
—
Guilin, 63
—
Beihai,64
—
Haikou, 65
—
Chongqing, 66
—
Chengdu, 67
—
Zigong, 68
—
Luzhou, 69
—
Deyang, 70
—
Mianyang, 71
—
Nanchong, 72
—
Guiyang, 73
—
Kunming, 74
—
Qujing, 75
—
Lhasa, 76
—
Xi'an, 77
—
Baoji,78
—
Weinan, 79
—
Lanzhou, 80
—
Jiayuguan, 81
—
Jinchang, 82
—
Baiyin, 83
—
Xining, 84
—
Yinchuan, and 85
—
Shizhuishan.59
L. Wang et al. / Science of the Total Environment 487 (2014) 57
–
64
Author's personal copy
averagingdayconcentrations,andtheyearconcentrationswereobtain-ed byaveragingmonth concentrations. Theair pollutantdatainthisre-search were the year average concentrations from 2004 to 2010.
2.3. Study area
85 major city regions were chosen as our research areas, including25 provincial capital city regions, 4 municipalities, and 56 other sub-major city regions. We excluded Xinjiang autonomous region becausethe demographic data in Xinjiang autonomous region were deemed tobe unreliable by most demographers in China. City regions includedthe urban area and rural area under the city's administrative jurisdic-tion. The 85 research regions were showed in Fig. 1.
3. Results
3.1. Spatial distribution characteristics
Summary statistics for key variables (e.g. life expectancy, longevity,PM
10
, SO
2
, and per capita GDP) in the research regions were listed inTable 2 and the spatial characteristics of the variables were illustratedin Fig. 2.Life expectancy is an index widely used in evaluating the nationalpublic health situation. The average life expectancy at birth of Chinesepeople was 76 years for both sexes (74 years for male and 77 yearsfor female) in China in 2010, which was 5 years longer than that in2000 and 7 years longer than that in 1990 (Mundi index, 2010). As itcanbeseeninTable2,theaveragelifeexpectancyandaveragelongevityratio of 85 city regions were 79.44 years and 23.98, respectively, whilethe average concentrations of PM
10
and SO
2
were 90.16
μ
g/m
3
and46.31
μ
g/m
3
, and the average per capita GDP was 42,230 RMB.Fig.2aillustratedlifeexpectancyinthe85regions.Lifeexpectancyinthe regions from the east China was obviously higher than that inthe central regions. The region with the longest life expectancywas Shenzhen of 86.31 years, followed by Beihai (84.84 yrs),Zhuhai (84.31 yrs), Jiayuguan (83.75 yrs), Taiyuan (82.87 yrs),Shanghai (82.69 yrs), Beijing (82.36 yrs), Haikou (82.23 yrs),Tianjing (82.17 yrs), and Lhasa (82.17 yrs). The regions with higherlife expectancy (81.72 yrs
–
86.31 yrs) mainly clustered in threeareas, which were Beijing-Tianjing, Yangtze River Delta (Shanghai)and Pearl River Delta (Zhuhai and Shenzhen). Those regions werealso China's three major economic centers with well-developed econo-mies. Taiyuan also had higher life expectancy with 46,144 RMB for percapita GDP. Meanwhile, Lhasa and Jiayuguan having higher life expec-tancywerepartiallyuntappedregionswithlowPM
10
andSO
2
concentra-tions. The other regions with higher life expectancy were Haikou andBeihai, which were tourism regions with relatively clean air. Amongthe ten regions with higher life expectancy,
ﬁ
ve regions had lowerPM
10
concentration,sixregionshadlowerSO
2
concentration,andsixre-gionshadhigherpercapitaGDP.Qujinghadthelowestlifeexpectancyof 71.67 years, followed by Shizuishan (74.94 yrs), Kunming (75.09 yrs),Changzhi (75.48 yrs), Luzhou (75.48 yrs), Yangquan (76.70 yrs), Yantai(76.86 yrs),Chifeng(76.95 yrs),Yinchuan(77.06 yrs),andShijiazhuang(77.20 yrs). Among these regions, there were no regions with lowerPM
10
or SO
2
, one region with higher per capita GDP.Fig.2bdemonstratedthelongevityratioin85regions.Beihaihadthehighest longevity ratio of 10.33, followed by Haikou (9.48), Nantong(6.65), Shaoguan (5.73), Zhanjiang (5.56), Liuzhou (5.00), Guilin(4.76), Zaozhuang (4.72), Zigong (4.63), and Yantai (4.5). Amongthese 10 regions with higher longevity ratios, there were
ﬁ
ve regionswith lower PM
10
concentrations, two with lower SO
2
concentrations,and one with higher per capita GDP. Jinchang had the lowest longevityratio of 0.22, followed by Changzhi (0.42), Jiayuguan (0.43), Yangquan(0.51), Chifeng (0.55), Baoji (0.57), Huhehaote (0.59), Huzhou (0.63),Nanchang (0.69), Yinchuan (0.70), and Shenzhen (0.70). Among thoseregions, there was no region with lower PM
10
concentration, two re-gionswith lowerSO
2
concentration,two regionswith higher percapitaGDP, and four regions with lower life expectancy.Fig.2cshowedthePM
10
concentrationdistribution.PM
10
ishigherinthe central China, while lower in the southeast part. Haikou had thelowest PM
10
concentration of 39.13
μ
g/m
3
, followed by Zhanjiang(46.11
μ
g/m
3
), Zhuhai (47.13
μ
g/m
3
), Guilin (48.11
μ
g/m
3
), Beihai(48.44
μ
g/m
3
), and Lhasa (51.29
μ
g/m
3
). Lanzhou had the highestPM
10
concentration of 148.44
μ
g/m
3
, followed by Xi'an (126.44
μ
g/m
3
),Weinan (125.78
μ
g/m
3
), Jinan (121.44
μ
g/m
3
), Beijing (121.00
μ
g/m
3
),Pingdingshan (118.33
μ
g/m
3
), and Shijiazhuang (115.89
μ
g/m
3
).Fig. 2d showed the SO
2
distribution. Most of the regions with lowerSO
2
concentration were located in the southeastern China and thewestern China, while central China had higher SO
2
concentration.Lhasa had the lowest SO
2
concentration of 5.78
μ
g/m
3
, followed byHaikou (7.88
μ
g/m
3
), Zhanjiang (12.78
μ
g/m
3
), Beihai (13.00
μ
g/m
3
),Fuzhou (14.75
μ
g/m
3
), and Zhuhai (17.13
μ
g/m
3
). Huhehaote hasthehighestSO
2
concentrationof118.00
μ
g/m
3
,followedbyYanquan(93.38
μ
g/m
3
), Shizuishan (74.13
μ
g/m
3
), Liuzhou (73.89
μ
g/m
3
),Taiyuan (72.44
μ
g/m
3
), and Zibo (65.00
μ
g/m
3
).Fig. 2e illustrated spatial distribution of per capita GDP respectively.Shenzhen ranked
ﬁ
rstwhile Jingzhou rankedlast. Regionswith highestper capita GDP were mainly located in the east coastal area.
3.2. Correlation analysis between health outcomes and air pollutants
First we did SR. Life expectancy and longevity ratio were set as de-pendent variables separately. SO
2
, PM
10
and per capita GDP were setas independent variables. For the life expectancy related regression,SO
2
and per capita GDP were included, and PM
10
was excluded. Forthe longevity ratio related regression, only PM
10
was included. Usingthe independent variables in included SR, we did GWR in ArcGIS. Allthe data were normally distributed. Summary parameters for GWR and SR were shown in Table 3.Table 3 showed the GWR and SR results. R-square adjust values forlife expectancy associated with SO
2
and per capita GDP were 0.389and 0.367 for GWR and SR, respectively, which indicated the correla-tionsbetweentwofactorsandlifeexpectancy. Estimatesoftheassocia-tions between SO
2
and life expectancy with the use of both regressionswere sensitive to the inclusion of per capita GDP. The association be-tween SO
2
and life expectancy was stronger without the adjustmentforpercapitaGDP.OnthebasisofregressionmodelswithoutpercapitaGDP, every 10
μ
g/m
3
increase in ambient SO
2
concentrations couldshorten life expectancy by 0.35 year for GWR and by 0.50 year for SR,while life expectancy could be shorten by 0.28 year for GWR and by0.47 year for SR if included per capita GDP for both regressions. Percapita GDP was positively, signi
ﬁ
cantly associated with life expectancy,which had been documented by other researches (Matthews et al.,2006; Bossuyt et al., 2004). In this research, we found that every10,000 RMB increase in per capita GDP could prolong life expectancyby 0.53 year for GWR and by 0.52 year for SR. PM
10
was associatedwith diminished likelihood of longevity ratio by 2.23 for GWR and by3.43 for SR with every 10
μ
g/m
3
increase of pollutant concentration.We did not
ﬁ
nd association between per capita GDP and longevity.
Table 2
Summary characteristics of the 85 city regions.Variable Arithmetic mean Standard deviation Max MinLife expectancy (yr) 79.44 2.25 86.31 71.67Longevity (ratio 100+) 23.98 18.50 103.30 2.15PM
10
(
μ
g/m
3
) (2004
–
2010) 90.16 20.79 148.44 39.13SO
2
(
μ
g/m
3
) (2004
–
2010) 46.31 19.87 118.00 5.78Per capita GDP (1000 RMB) 42.23 20.51 94.30 13.2260
L. Wang et al. / Science of the Total Environment 487 (2014) 57
–
64

Search

Similar documents

Related Search

Study on Dispersion of Air Pollutants on CoasA Study of Benefits in Object based Storage SA Study of Job Satisfaction and Work AttitudeA Study of the Recruitment and Selection procWomen in Urban Informal sector-a study of domA Study of Myth In the Later Poetry of W.B.YeLitewrature and Empowerment: A Study Of TribaA Study of Urdu Punjabi and English ProverbsPhilosophy as a way of lifePhilosophy as a Form of Life

We Need Your Support

Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks