ArticlePDF Available

Impacts of Social Inequality, Air Pollution, Rural–Urban Divides, and Insufficient Green Space on Residents’ Health in China: Insight from Chinese General Social Survey Data Analysis

Authors:

Abstract and Figures

Attention to physical and mental health is becoming more intensive. In China, factors and mechanisms are now a focus of research. We used dynamic air quality monitoring data and the Chinese General Social Survey (CGSS) to assess the spatial differences and the coupling between subjective and objective air pollution. In addition, a logistic model was used to explore the impact mechanisms of social inequality, air pollution, food safety, and lack of green space on health. The results show that (1) the impact of subjective and objective air pollution on the health level of the population is significant; (2) income inequality, air pollution, food pollution, and travel behavior significantly affect the residents’ health; and (3) environmental health has a significant differentiation mechanism between urban and rural areas. The negative health effects of air pollution and insufficient green space are more significant in cities; food pollution is more likely in rural areas. In terms of socioeconomic inequality, gender, family size, travel, and physical exercise had no significant effect on rural health. Health improvement was higher in the low-income group than in the high-income group. The adverse effect of travel behavior on environmental pollution is conducive to improving health. Therefore, social equality, strictly controlled environmental pollution, exercise, and travel can help narrow the gap between rich and poor, promote urban–rural health equity, and improve human health.
Content may be subject to copyright.
Citation: Zhou, P.; Sun, S.; Chen, T.;
Pan, Y.; Xu, W.; Zhang, H. Impacts of
Social Inequality, Air Pollution,
Rural–Urban Divides, and
Insufficient Green Space on
Residents’ Health in China: Insight
from Chinese General Social Survey
Data Analysis. Int. J. Environ. Res.
Public Health 2022,19, 14225.
https://doi.org/10.3390/
ijerph192114225
Academic Editors: Jiaxing Cui,
Jing Luo, Ying Jing and Liqun Sun
Received: 26 September 2022
Accepted: 28 October 2022
Published: 31 October 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
International Journal of
Environmental Research
and Public Health
Article
Impacts of Social Inequality, Air Pollution, Rural–Urban
Divides, and Insufficient Green Space on Residents’ Health
in China: Insight from Chinese General Social Survey
Data Analysis
Peng Zhou 1, Siwei Sun 1, Tao Chen 2 ,3 ,*, Yue Pan 1, Wanqing Xu 1and Hailu Zhang 1
1School of Civil Engineering and Architecture, Wuhan Institute of Technology, Wuhan 430074, China
2School of Arts and Communication, China University of Geosciences, Wuhan 430074, China
3Hubei Planning, Design and Research Institute, Wuhan 430064, China
*Correspondence: 2202210342@cug.edu.cn
Abstract:
Attention to physical and mental health is becoming more intensive. In China, factors
and mechanisms are now a focus of research. We used dynamic air quality monitoring data and the
Chinese General Social Survey (CGSS) to assess the spatial differences and the coupling between
subjective and objective air pollution. In addition, a logistic model was used to explore the impact
mechanisms of social inequality, air pollution, food safety, and lack of green space on health. The
results show that (1) the impact of subjective and objective air pollution on the health level of the
population is significant; (2) income inequality, air pollution, food pollution, and travel behavior
significantly affect the residents’ health; and (3) environmental health has a significant differentiation
mechanism between urban and rural areas. The negative health effects of air pollution and insufficient
green space are more significant in cities; food pollution is more likely in rural areas. In terms of
socioeconomic inequality, gender, family size, travel, and physical exercise had no significant effect
on rural health. Health improvement was higher in the low-income group than in the high-income
group. The adverse effect of travel behavior on environmental pollution is conducive to improving
health. Therefore, social equality, strictly controlled environmental pollution, exercise, and travel
can help narrow the gap between rich and poor, promote urban–rural health equity, and improve
human health.
Keywords:
health; social inequality; air pollution; travel behavior; urban–rural differentiation
mechanism
1. Introduction
In the current era of comprehensive health, China’s demand for physical and mental
health has stimulated the national demand for health research, health improvement, and
research on influencing factors and their mechanisms of action based on Chinese samples.
Many international scholars have used walkability as a mediating variable to discover the
health effects of social injustice and green space and the effects of environmental pollution
on physical and mental health, gradually forming a complete system that covers sociology,
economics, geography, behavior, public health, and psychology. However, a complete
interdisciplinary system has yet to be developed in China to explore the factors influenc-
ing human health. Therefore, the relationship between social inequality, environmental
pollution, insufficient green space, and health still needs to be explored.
Although social and wealth inequality have often been key elements in studying hu-
man health influences, environmental pollution, lack of blue-green space, and preferences
for travel behavior have been neglected. This situation occurs because dynamic monitoring
of environmental quality, blue-green space scale, quality assessment, behavioral tracking
Int. J. Environ. Res. Public Health 2022,19, 14225. https://doi.org/10.3390/ijerph192114225 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2022,19, 14225 2 of 17
surveys, monitoring in micro-social surveys, and macro panel data analysis have remained
limited by spatial technology and monitoring techniques. However, the importance of the
impact of environmental pollution on health has been empirically demonstrated in the
fields of sociology, environmental studies, urban planning, and geography [
1
3
]. Studies
have been conducted to overcome the limitations of data acquisition and scale accuracy
to include air pollution in the independent variable indicator system of regression mod-
els to explore the negative environmental effects of air pollution on physical and mental
health
[46]
. Since 2010, the effect of green spaces on health has been a popular research
topic, and the number of such studies has grown. Many international scholars have identi-
fied the health improvement effects of green spaces using walkability and the frequency of
physical activity as mediating variables [
7
9
]. However, in China, research on the mecha-
nism of the effects of blue and green elements and behavioral preferences on health remains
in the early stages of development. Gradual empirical evidence of the impact of behavioral
characteristics on health has also become a current research hotspot in behavioral and
health geography. Scholars in related fields have actively explored the impact of dietary
behavior on obesity signs [
10
], the health impact of travel behavior and exposure to pollu-
tion [
11
], the impact of environmental behavior on health economic effects and pollution
sensitivity [
12
], and the avoidance effect of travel behavior on environmental pollution [
13
].
Combining the relevant literature on the exploration of health-influencing factors
shows that mathematical statistics and spatial models including correlation analysis [
14
],
multiple linear regression models [
15
], logit [
16
] and logistic [
17
] models, probit mod-
els [
18
], and other artificial intelligence models [
19
] are primarily used to dissect the
coupling relationship between social inequality, air pollution, inadequate green space,
and health, and their mechanisms of action. Simple multivariate linear models can no
longer meet the needs of discontinuous categorical variables as explanatory or explained
variables, which has led to a series of logit and logistic models and empirical evidence
of their robustness and feasibility of multivariate marginal utility exploration. Shi and
Yanwei [
20
] constructed ordered logit models to explore the mechanism of leisure behavior
characteristics on women’s leisure satisfaction. Recently, another improved model with a
multistage and multi-breakpoint regression design was applied to explore the associated
mechanisms and feedback effects among environmental pollution, social inequality, and
health and well-being. Multistage associative least squares (3SLS) have been applied to the
coupling between environmental quality and wealth inequality from a health perspective.
Liu Cong et al.
[
21
] empirically tested the effect of air quality deterioration due to air pollu-
tion on the urban–rural income gap using a system of associative equations. The two-stage
probit model is frequently applied to explore intrinsic links between different variables. Ma
Xiaojun et al. [
22
] used this model to discover the factors that influence happiness and the
feedback effect of happiness on environmental behavior. The categorical variables of the
social surveys further require applied spatial regression models to explain the mechanisms
of their explanatory variables as fully as logistic models in the factors exploration of the
influencing health and their mechanisms of action.
Many previous studies have focused on objective environmental pollution but have
neglected subjective environmental pollution. In recent years, some studies have gradually
revealed the endogeneity of subjective and objective air pollution and well-being, finding
that subjective air pollution has a more significant influence on well-being, life satisfaction,
and psychological health [
23
]. When exploring the effects of environmental pollution
variables such as air pollution on human health, a clear distinction should be made be-
tween subjective and objective levels of environmental pollution. Different studies have
used various subjective and objective indicators to measure health levels, but significant
heterogeneity has also been observed between self-rated and objective health. Self-rated
health is a subjective assessment that focuses on physical and mental health. Meanwhile,
objective health can be reflected in many public health statistics such as mortality, mor-
bidity, mortality from air pollution-related diseases, and neonatal mortality. This study
identified the variability between subjective and objective air pollution. Furthermore, this
Int. J. Environ. Res. Public Health 2022,19, 14225 3 of 17
study explored the mechanisms of social inequality, air pollution, and insufficient green
space, among other factors, on health using logistic models. This study aims to promote the
exploration of environmental pollution heterogeneity and its impact on health and life from
subjective and objective perspectives in the fields of environmental studies and sociology.
2. Material and Methods
2.1. Data Sources
This study was based on the Chinese General Social Survey (CGSS) Database. Given
that the CGSS environmental survey on the ‘severity of the following types of environ-
mental problems in your area’ has not been available since 2013, this study used data from
the 2013 CGSS to investigate the impact of the mechanisms of environmental problems
and social inequalities on health. The data were cleaned to exclude those who answered
‘do not know’ to the environmental questions related to air pollution, lack of green space,
degradation of arable land quality, and food pollution (i.e., excluding samples with missing
items in the environmental pollution assessment variables) and removed extreme outliers.
The final study was conducted using 5966 valid samples.
2.2. Variable Selection
Individual socioeconomic characteristics have historically had a significant influence
on health [
24
]. However, annual income, gender, and age have been primarily used to dis-
tinguish disadvantaged groups from the general population, with some studies neglecting
the influence of household characteristics and living space. Class and educational levels
are often missing from panel data studies due to their limitations. In recent years, wealth
and class gaps have received increasing attention. In the past, the simple use of individual
income to measure class disparity was inaccurate. This study used individual self-rated
class levels from social surveys to assess the impact of social class disparity on health.
Considering that the dependent variable ‘health level’ was selected based on the
‘subjective self-assessment variable’, two variables were available in the social survey:
survey question a15, ‘How do you feel your current physical health status’, and question
a16, ‘How often in the last four weeks have health problems affected your work or other
daily activities’. As the latter is a more accurate indicator of the impact of health problems
on daily work and life and a more accurate assessment of health levels, it was chosen as the
dependent variable in this study. Furthermore, 0 = always, often, and sometimes, so health
problems affect work or life, and are defined as ‘0’ in the unhealthy category, whereas the
others are defined as ‘1’ in the healthy category.
In terms of independent variables, socioeconomic variables included annual income,
age, marriage, class, education, gender, household size, and housing area per capita;
location variables included urban–rural and east–west sub-regions; activity preference
variables included the frequency of physical exercise and travel mode; environmental
pollution variables included air pollution, water pollution, noise pollution, food pollution,
lack of green space and degradation of arable land quality, as shown in Table 1.
Int. J. Environ. Res. Public Health 2022,19, 14225 4 of 17
Table 1. Variable assignment.
Variable Symbol 1 Symbol 2 Description Relevance
Health level Health a16a 1
Annual income Income a8a Individual total annual income of last year (2012) +
Urban–rural Urban s5a Urban = 1 and rural = 0 +
Region Province s41 Survey area (province)
Age Age a3aa Age
Marriage Marriage a69
Unmarried = 1; cohabiting = 2; first married with a
spouse = 3; remarried with a spouse = 4; separated
and not divorced = 5; divorced = 6; widowed = 7
Stratum Stratum a43a
The highest ‘10 points’ represents the top stratum,
and the lowest ‘1 point’ represents the bottom
stratum.
+
Education Education a7a
1 = no education; 2 = private school and literacy
classes; 3 = primary school; 4 = junior high school;
5 = vocational high school; 6 = general high school;
7 = secondary school; 8 = technical school;
9 = university specialist (adult higher education);
10 = university specialist (formal higher education);
11 = university undergraduate (adult higher
education); 12 = university undergraduate (formal
higher education); 13 = graduate; 14 = doctoral
students and above.
+
Gender Sex a2 1 = male and 2 = female
Household size Family size a63 How many people usually live in your household at
the moment (including yourself)
Housing area per
capita
Housing area per
capita a11 Housing area per capita +
Frequency of
physical exercise
Frequency of
physical exercise a3009 1 = daily; 2 = several times a week; 3 = several times
a month; 4 = several times a year or less; 5 = never
Travel mode Travel mode b1105
I always take a taxi or private car when I go out
1 = very much so; 2 = more so; 3 = not very much so;
and 4 = very little so
Air pollution Air pollution b21b01 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Water pollution Water pollution b21b02 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Noise pollution Noise pollution b21b03 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Food pollution
Food contamination
b21b10 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Lack of green space Insufficient green
space b21b06 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Degradation of
arable land quality
Degradation of
cultivated land
quality
b21b08 1 = very serious; 2 = more serious; 3 = less serious;
4 = not serious; 5 = average; and 6 = no such problem
+
Note: “+” is positive correlation, and is negative correlation.
2.3. Statistical Description
Descriptive statistics of the data are shown in Table 2. In terms of socioeconomic
variables, the annual income level of all individuals interviewed was approximately CNY
8365; the mean age was approximately 46 years; and the level of self-assessed social class
was low. Furthermore, family size, educational level, and housing area per capita were
small. In terms of urban–rural distribution, rural samples were predominant. In terms
of activity preferences, physical exercise was infrequent, and the mode of travel was not
in line with ‘always take a taxi or private car when going out’. Regarding environmental
pollution, the quality of the air environment varied greatly due to the vast size of China.
Int. J. Environ. Res. Public Health 2022,19, 14225 5 of 17
Table 2. Statistical description of variables.
Variable Symbol Maximum Minimum Average
Health level a16a 1 0 0.8
Annual income a8a 1,000,000 0 28,365.04
Urban–rural s5a 5 1 2.691315
Age a3aa 96 17 46.45288
Marriage a69 7 1 3.092723
Stratum a43a 10 1 4.459759
Education a7a 14 1 5.661469
Gender a2 2 1 1.456908
Household size a63 12 1 3.094903
Housing area per capita a11a 700 1.5 41.57517
Frequency of physical exercise a3009 5 1 3.719651
Travel mode b1105 4 1 3.307344
Air pollution b21b01 6 1 3.348759
Water pollution b21b02 6 1 3.473587
Food contamination
(food safety) b21b10 6 1 3.682428
Lack of green space b21b06 6 1 4.129297
2.4. Model Setting
Objective air pollution was represented by the annual average concentration of PM2.5
(particulate matter 2.5
µ
m or less) in 2012, using air quality monitoring data. Meanwhile,
subjective air pollution was evaluated using the survey question a16, ‘How often did
health problems affect your work or other daily activities in the past four weeks’ from
the CGSS database. Then, the ratings were assessed. The correlation coefficient was used
to explore the correlation between the subjective and objective air pollution variables.
Then, subjective air pollution and objective air pollution were introduced separately into
the logistic regression model to assess the effect of both on the health levels using the
significance of the variables.
3. Empirical Analysis
3.1. Coupled Relationships between Social Inequality, Air Pollution, Lack of Green Space,
and Health
First, basic statistical analysis charts were used to explore the coupling relationships
between the explanatory variables and health. In the results of the statistical analysis, the
health level varied with different individual social characteristics, economic level, travel
preferences, and environmental pollution with the corresponding linear trends, as shown
in Figure 1.
The response variable in this subsection of the statistical analysis is the mean level
of health. The explanatory variables are the values of each statistical survey variable.
Frequency distributions, scatter plots, and linear trend line distributions in the statistical
analysis adequately express the degree of clustering of the sample in different characteristic
groups and the direction of correlation. Furthermore, the slope of the trend line expresses
the magnitude of the change in mean health level caused by the explanatory variables.
Int. J. Environ. Res. Public Health 2022,19, 14225 6 of 17
Int. J. Environ. Res. Public Health 2022, 19, x 6 of 17
Figure 1. Coupling of variables with the health levels.
The response variable in this subsection of the statistical analysis is the mean level of
health. The explanatory variables are the values of each statistical survey variable. Fre-
quency distributions, scatter plots, and linear trend line distributions in the statistical anal-
ysis adequately express the degree of clustering of the sample in different characteristic
groups and the direction of correlation. Furthermore, the slope of the trend line expresses
the magnitude of the change in mean health level caused by the explanatory variables.
Figure 1. Coupling of variables with the health levels.
3.1.1. Age
Sample frequencies with different age distributions were mainly concentrated in the
young and middle-aged groups. The degree of health was negatively correlated with age
(i.e., the younger the person, the higher the degree of health). Figure 1a demonstrates
differences in the magnitude of change in health levels for different age groups, with a
smaller decline in health levels between 17 and 36 years and a larger decline in the average
health level above 37 years. Specifically, the decline in health levels with age in the young
adult group was much smaller than in the elderly group. The histogram in Figure 1a shows
Int. J. Environ. Res. Public Health 2022,19, 14225 7 of 17
that the population age distribution had a clear trend toward a normal distribution, with
the largest number of people in the 37–56 age group and the smallest number of people
aged 77 years and older.
3.1.2. Annual Income
Annual income was directly proportional to self-rated health, with higher income
associated with higher health (average health level). In Figure 1b, the number of people
with an annual income between CNY 0 and CNY 4999 was higher, with the next highest
number of people in the CNY 20,000–CNY 24,999 range. The slope of the trend line was
0.016 with an increase in annual income, which was less than the slope of the age–health
trend line, indicating that the average level of health decreased less than the age–health
range with a decrease in annual income.
3.1.3. Number of Family Members
The number of household members was positively related to self-rated health, with
higher income being associated with higher health. The number of family members was
related to self-rated health in an ‘inverted U-shaped’ curve, as verified by a polynomial. In
Figure 1c, the frequency distribution of family members and health levels in this survey
sample showed a normal distribution trend, with the highest proportion of ‘three-member
families’ and ’two-member families’ up to 50%.
3.1.4. Marriage
Marriage was inversely related to self-rated health, and married people had a higher
level of health than unmarried individuals. In Figure 1d, the highest number of people
were married to a spouse, followed by the second-highest number of single people. The
slope below the marital status health trend line indicates that the decline in health levels
with marital separation was higher than the marital health margin.
3.1.5. Education
The level of education was positively related to self-rated health, with higher levels of
education associated with higher levels of health. The frequency of the sample distribu-
tion gradually increased from ‘no education’ to ‘junior high school’ and then decreased
with increasing level of education, which follows the general trend of the distribution of
education in China. The increase in the average level of health in these two stages was as
follows: before junior high school, the health level increased rapidly with the increase in
education level (from 0.5 to 0.77 through three levels of education), whereas after junior
high school, the health level slowly increased with the increase in educational level (from
0.77 to 0.97 through nine levels of education) from vocational high school to postgraduate
level and above.
3.1.6. Mode of Travel
When travelling, the level of compliance with always taking a taxi or private car
decreased from 1 to 4, with 1 being ‘very compliant’ and 4 being ‘very non-compliant’.
Figure 1f shows that the mode of travel was inversely related to health, with those who con-
stantly travelled by car having the highest level of health. Thus, exposure to environmental
pollution and travel time by car must be reduced.
3.1.7. Exercise Frequency
The frequency of exercise decreased from 1 to 5, with 1 being ‘every day’ and 5 being
‘never’. Figure 1g shows that the frequency of exercise was negatively correlated with
the level of health (i.e., the more frequent exercise, the higher the level of health, and
vice versa).
Int. J. Environ. Res. Public Health 2022,19, 14225 8 of 17
3.1.8. Air Quality (Air Pollution)
The air environment quality (air pollution) variable gradually decreased from 1 to 7,
as shown by the air quality–health level trend line. The less polluted the air, the higher the
level of health.
3.1.9. Housing Area Per Capita
Housing area per capita health statistics showed that the housing area per capita
variable was positively correlated with health.
4. Results
4.1. Modeling Results
Logistic regression models differ from other general linear regression models in terms
of simulation accuracy. Linear regression typically uses R-squared values to assess the
accuracy of regression results. Meanwhile, logistic regression models use sample prediction
accuracy, receiver operating characteristic (ROC) curves, and Akaike information criterion
values. The ROC curve is a comprehensive assessment of the degree of correct prediction
for different samples and is mainly expressed in terms of the area under the curve (AUC).
The larger the AUC, the higher the accuracy of the model regression results and vice versa.
Based on the comparison of the model results in this study, the competent full-sample air
pollution model had the highest accuracy, whereas the full-sample objective air pollution
model had the lowest accuracy.
According to a series of logistic and logit regression model simulation accuracy com-
parison studies, the perceptibility curve (ROC curve) is based on a comprehensive analysis
of a series of indices such as the accuracy rate, true positive rate, false positive rate, and re-
call rate to assess the equation simulation accuracy, but it still needs to be combined with the
Akaike information criterion (AIC) to further verify the model fit goodness. Examination
of the AIC value can explore the causal relationships and mechanisms of action between
the independent and dependent variables with a minimum number of free parameters
to complete. In many regression analyses, increasing the number of free parameters can
improve the accuracy of the model but can also easily result in overfitting. AIC is another
method to validate the optimal fit of a model proposed by statisticians to address this
problem. The AIC can be expressed as
2log(L) + 2p, where k is the number of parameters,
L is the log-likelihood, n is the number of observations, and p is the number of variables in
the model. It is assumed that the model errors follow an independent normal distribution.
The red pool AIC is inversely correlated with the goodness of fit of the model (i.e., the
smaller the AIC, the better the model simulation accuracy). A comparison of the model
results showed that the simulation accuracy of the full-sample objective air pollution model
was better than the full-sample competent air pollution model.
An overview of previous studies revealed that the multiple independent variables in
the modeled prediction equations were mostly related to each other and were not entirely
independent of each other. Spatial regression measures the strength of this linkage using
multicollinearity, which is measured using the variance inflation factor (VIF) or tolerance,
with an inverse correlation between VIF and tolerance. A higher VIF or lower tolerance in-
dicates substantial multicollinearity between the variables. According to relevant statistical
studies, a VIF <0.5 or tolerance >0.2 proves that there is no multicollinearity between the
variables in the equation (i.e., the structure of the variables in the equation is reasonable).
The results of the model simulations in this work, which used stepwise regression, showed
that the VIF of each variable was <0.5 after several iterations, indicating minimal co-linearity
between the explanatory variables in the health impact factor exploration equation; that is,
the results were scientifically reasonable.
4.2. Interpretation of Model Variables
The results of logistic regression model 1 show that the 10 variables—urban rural,
class, household size, education, income, area per capita, lack of green space, food pollution
Int. J. Environ. Res. Public Health 2022,19, 14225 9 of 17
(food safety), east–west, and subjective air pollution—have a significant positive effect
on health levels in Table 3. The four variables of gender, age, frequency of physical
activity, and mode of travel showed a significant reduction effect on the level of health.
The regression parameters of model 1 showed that the probability of women’s health
improving relative to that of men decreased by 14.18%, whereas the probability of health
deteriorating increased by 4.50% if the age increased by one year. As all of the respondents
in this study were a sample of people aged 17 years or older, they avoided the need to
consider requirements such as breakpoint regression pre-processing for underage and adult
populations. The results showed that the linear negative effect of aging on the probability
of improved health is scientifically justified. The mechanism of gender and age further
indicates that the negative health effects of disadvantaged groups (women and elderly)
are more pronounced. Therefore, in the process of residential planning, environmental
improvement, and the layout of facilities, the sensitivity of the living environment and
the health improvement needs of disadvantaged groups need to be further considered,
which is an inevitable requirement for the development of an aging society. The significant
p-value for the variable “insufficient green space” further highlights the health-enhancing
effect of green space.
Table 3. Comparison of the full-sample multi-model regression results.
Variable Code Model 1 Model 2 Model 3
Coe. b Sig. Coe. b Sig. Coe. b Sig.
Gender a2 0.153 0.034 ** 0.154 0.033 ** 0.056 0.000 ***
Age a3aa 0.046 0 *** 0.046 0.000 *** 0.017 0.019 **
Urban-rural s5aa 0.189 0.034 ** 0.204 0.020 ** 0.096 0.000 ***
Stratum a43a 0.16 0 *** 0.159 0.000 *** 0.058 0.001 ***
Household size a63 0.046 0.098 * 0.045 0.109 0.020 0.000 ***
Education a7a 0.06 0 *** 0.060 0.000 *** 0.017 0.036 **
Income a8aaa 0.294 0.08 * 0.296 0.078 * 0.080 0.000 ***
Area per capita a11aa 0.051 0.016 ** 0.049 0.019 ** 0.015 0.011 **
Frequency of
physical exercise a3009 0.093 0 *** 0.094 0.000 *** 0.062 0.034 **
Travel mode b1105 0.125 0.013 ** 0.130 0.009 *** 0.039 0.000 ***
Lack of green space b21b06 0.052 0.012 ** 0.045 0.024 ** 0.015 0.009 ***
Food contamination
(food safety) b21b08 0.02 0.173 ** 0.018 0.224 0.003 0.013 **
East and West s41 East
and West 0.318 0 *** 0.264 0.006 *** 0.040 0.384
Air pollution
(subjective) S6kq 0.024 0.025 ** - - 0.000 0.165
PM2.5 (objective) PM2.5 - 0.003 0.216 0.002 0.978
Constants e 2.577 0 *** 2.678 0.000 *** 4.729 0.000 ***
ROC AUC 0.854 0.806 0.837
Predicted correct rate 88.261 80.345 85.543
AIC 5270 4988 -
Note: ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Model 1 is the logistic model
with subjective air pollution. Model 2 is the logistic model with the objective of air pollution. Model 3 is a linear
model. Abbreviations: PM2.5, particulate matter 2.5
µ
m or less; ROC AUC, Receiver operating characteristic area
under curve; AIC, Akaike information criterion.
The subjective and objective air pollution variables were added to the logistic regres-
sion models as input variables, respectively (Table 3—models 1, 2, and 3). The regression
results showed that the p-value for the subjective air pollution variables was less than
0.01, which had a significant effect; the p-value for the objective air pollution variables
was greater than 0.1, which meant that they were excluded from the model variables. The
evidence shows that the relationship between the objective air pollution levels and subjec-
tive air pollution levels can be quantified from the perspective of objective environmental
Int. J. Environ. Res. Public Health 2022,19, 14225 10 of 17
quality monitoring and subjective human perception, and that there is no ‘one-to-one’
relationship between the two, but there is significant heterogeneity because subjective
perceptions of air pollution are influenced by differences in individual characteristics. Past
empirical studies have shown that at the same level of objective air quality, vulnerable
groups such as women and the elderly have higher subjective perceptions of air pollution,
that higher income groups are more sensitive to air pollution, and sensitive people are
more likely to be affected by air pollution at the perception level.
The regression results from model 4 (urban sample) and model 5 (rural sample) in
Table 4showed that the significance of ‘air pollution and lack of green space’ held for
the urban sample, whereas the significance of ‘air pollution and lack of green space’ for
the rural sample was greater than. This means that ‘air quality and green space’ have a
significant impact on the health of urban residents. In terms of ‘food safety’, the coefficient
for the urban sample was smaller than the coefficient for the rural sample, indicating
that rural residents are more likely to experience negative health effects when exposed
to food contamination, thus suggesting that rural residents are at greater risk of food
contamination exposure.
Table 4. A comparison of the regression results for urban–rural differences in the whole sample.
Variables Code Model 1 Model 4 Model 5
Coe. b Sig. Coe. b Sig. Coe. b Sig.
Gender a2 0.153 0.034 ** 0.196 0.031 ** 0.041 0.740
Age a3aa 0.046 0.000 *** 0.045 0.000 *** 0.042 0.000 ***
Urban–rural s5aa 0.189 0.034 **
Stratum a43a 0.16 0 *** 0.178 0.000 *** 0.120 0.001 ***
Household size a63 0.046 0.098 * 0.042 0.259 0.044 0.300
Education a7a 0.06 0 *** 0.050 0.005 *** 0.114 0.003 ***
Income a8aaa 0.294 0.08 * 0.053 0.074 * 2.161 0.000 ***
Area per capita a11aa 0.051 0.016 ** 0.029 0.341 0.064 0.032 **
Frequency of physical
exercise a3009 0.093 0 *** 0.103 0.001 *** 0.051 0.358
Travel mode b1105 0.125 0.013 ** 0.165 0.006 *** 0.016 0.858
Lack of green space b21b06 0.052 0.012 ** 0.058 0.052 * 0.051 0.101
Food contamination
(food safety) b21b08 0.02 0.173 ** 0.017 0.036 ** 0.027 0.330
East and West s41 East
and West 0.318 0 *** 0.326 0.004 *** 0.266 0.027 **
Air pollution S6kq 0.024 0.025 ** 0.026 0.032 ** 0.025 0.141
Constants e 2.577 0 *** 3.072 0.000 *** 1.251 0.071 *
ROC AUC 0.854 0.813 0.727
Predicted correct rate 88.261 82.534 75.457
AIC 5270 3337 1942
Note: ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. Model 1 is the full-sample.
Model 4 is the urban sample. Model 5 is the rural sample. Abbreviations: PM2.5, particulate matter of 2.5
µ
m or
less; ROC AUC, receiver operating characteristic area under curve; AIC, Akaike information criterion.
5. Discussion
5.1. Analysis of the Mechanism of the Impact of Environmental Pollution on the Health Level of
the Population
5.1.1. Subjective and Objective Air Pollution
For ambient air pollution, many studies in the past have stayed at the level of objective
environmental pollution air quality, thus neglecting subjective environmental pollution [
25
].
Objective air pollution is a direct expression of air quality monitoring data, often proven
to be directly related to human health and is influenced by certain factors such as vehicle
emissions and urban morphology [
26
28
]. This series of studies is based on objective
environmental pollution and ignores subjective environmental pollution. Several studies
Int. J. Environ. Res. Public Health 2022,19, 14225 11 of 17
have gradually revealed the endogeneity of subjective and objective air pollution and
well-being, finding that subjective air pollution affects well-being, life satisfaction, and
psychological well-being to a more significant degree.
Individual differences in environmental sensitivity lead to differences in subjective
evaluations and objective measures of environmental pollution, so subjective air pollution
levels can show significant differences between individuals. Objective air pollution is
indicated by the actual pollutant composition and pollution indices monitored. From
these actual measurements, it is possible to infer to what extent objective air pollution
threatens physical health at a physical level such as an increased probability of respiratory
and cardiovascular diseases, but it is impossible to accurately characterize the reduction
in subjective mental health levels associated with objective air pollution. Subjective air
pollution is an evaluation of the level of pollution from the residents’ perceptions of the
objective air environment. It is a comprehensive expression of the objective measured
data and subjective perceptions of experience, and is a comprehensive evaluation of the
extent to which air pollution affects physical health and mental health, and is a more
scientific assessment of the extent of health effects for the surveyed individuals [
29
]. The
more concerned and sensitive people are about air pollution, the greater their subjective
perception of the impact of air pollution levels on their daily lives, and the greater their
perception of the health risks. Therefore, subjective air pollution indicators are of greater
research significance in the study of the assessment of individual health level and its
influence mechanism.
Air pollution affects the health of the population at both the physical and mental
levels. The impact of air pollution on physical health is mainly reflected in the triggering of
respiratory diseases such as respiratory infections, asthma, chronic obstructive pulmonary
disease, and lung cancer, and even long-term exposure to heavy air pollution can raise
the mortality rates. Haze can have a significant impact on urban economies, the tourism
industry, and residential travel. Subjective air pollution poses risks such as mobility
restrictions, safety threats, and health threats, and the subjective risk perceptions that
residents generate from the haze can create feelings of stress and anxiety. This range of
negative emotions will cause a decrease in mental health.
5.1.2. Insufficient Green Space
Green space has a significant health-enhancing effect. Green space has been a key
focus of health research in the past, and studies on the health effects of green space have
developed rapidly in the country and abroad in recent years, with a large body of literature
focusing on the effects of land use type and green space on physical and mental health.
Green spaces can improve the frequency, duration, and willingness of residents to be
physically active through walkability, and improve air and habitat quality, which improves
physical fitness [
30
,
31
]. Furthermore, green views and open green spaces are conducive to
reducing stress and tension and improving psychological well-being. This study empirically
emphasizes the negative health effects of insufficient green spaces.
Although green space has a significant impact on the health levels of residents, a
variety of quantitative approaches and green space variables quantify how to measure the
accessibility, scale, quality, and type of green space. In the early literature, the only variable
chosen to measure the explanatory variables of green space was ‘accessibility of green
space’ [
32
34
]. In later literature, the heterogeneity of ‘size and quality’ of green spaces
on resident activity preferences, activity types, and health was gradually considered [
35
].
The variable ‘insufficient green space’ selected in this study was a subjective evaluation of
the adequacy of green space by urban and rural residents from a ‘humanistic’ perspective
and a comprehensive evaluation of the ‘accessibility, scale, quality, and type of green space
of the residence. It is also a comprehensive evaluation of the accessibility, scale, quality,
and type of green space, which is more conducive to reflecting the degree of influence and
mechanism of green space on the residents’ health, and provides a new direction for future
research on the health effects of green spaces.
Int. J. Environ. Res. Public Health 2022,19, 14225 12 of 17
5.1.3. Food Contamination (Food Safety)
The ‘food environment’ element is neglected in urban planning, geography, and
health geography studies. This study fills the gap between these fields and their related
intersections: ‘urban residents’ subjective perceptions of food contamination and how it
affects health levels. In the current era of health, there is an urgent need to improve the
quality of the food environment. This study measures the “quality of the food environment”
through the variable “food safety”, which provides a side-effect on the severity of “food
desertification”. Food safety factors affect human health through accessibility and food
availability and have a clear urban–rural divide.
5.2. Analysis of the Mechanisms of Socioeconomic Inequalities and Urban–Rural Differences on the
Health Level of the Population
Socioeconomic inequalities lead to significant mechanisms of differentiation in their
individual health. Urban rural, income, class, household size, education, and housing space
per capita reflect the social class differences, wealth disparities, and urban–rural disparities
from the side as well as feedback on the socioeconomic conditions and mechanisms acting
on health from a social inequality perspective. The urban–rural divide had a significant
impact on health, with cities having a 20.8% higher probability of improving the health of
the population than rural areas, an increase that was higher than other influencing factors;
highlighting that cities outperform rural areas in terms of health improvement. Increasing
incomes, increasing classes, and educational attainment bring better access to health care,
housing, and infrastructure, which are beneficial to health improvement, with probabilities
of 34.18%, 17.35%, and 6.18%, respectively. This finding shows that improvements in
urban/rural income and class levels can cause qualitative changes in health.
Differences in air quality and green space between urban and rural areas led to
prominent differences in the health levels of urban and rural residents. The overall health
level of urban residents was higher. The overall health level of urban residents was higher
than residents in urban–rural areas and rural areas. There were significant differences in
the characteristics of urban and rural habitats, with urban areas densely populated and
industrially active whereas rural areas were sparsely populated and mainly agricultural,
resulting in lower concentrations of air pollutants and lower air mobility in urban areas than
in rural areas, increasing the concentration and duration of exposure to urban air pollution
for people, which led to the health effects of air quality being more pronounced in cities.
The high density, spatial compactness, and fast pace of urban life increase the demand for
air quality and green space by urban residents compared to the slow-paced and low-density
rural areas, further increasing the extent to which both environmental factors affect the
health of urban populations. Thus, there is a clear mechanism of urban–rural locational
differentiation in the unequal health effects of air pollution and insufficient green space,
with both having a more significant negative health effect on urban populations. The health
level of urban residents was more significantly influenced by two types of environmental
factors, namely air pollution and insufficient green space. Thus, urban residents are more
concerned about their health from the perspective of environmental quality, and urban
residents are more at risk of being affected by air pollution and insufficient green space
factors. Furthermore, the evidence shows that urban residents spend more on health in
response to environmental pollution than rural residents.
For each unit increase in income, the health improvement of rural residents is much
greater than that of urban residents, which shows that the impact of the urban–rural income
disparity on health is significant. In recent years, the gap between the rich and the poor
in China’s urban and rural areas has gradually narrowed but still exists, and the overall
income level of rural residents is lower than urban residents; however, this study showed
that the health improvement of the low-income level group with increased income was
much higher than the high-income level group, further proving the practical significance
of rural revitalization and precise poverty alleviation in China. In contrast, the effects of
Int. J. Environ. Res. Public Health 2022,19, 14225 13 of 17
gender, household size, travel, and physical activity on the health of the rural population
were not significant under the perspective of socioeconomic inequality.
5.3. Analysis of the Mechanism of Activity Preference and Travel Behavior on the Health Level of
the Population Activity Preferences and Travel Behavior Directly Affect Health
Daily physical activity significantly improves health, whereas never travelling me-
chanically increases the time and level of exposure to environmental pollution and reduces
self-rated health. Human activity and behavior are the expressions and feedback of human
mobility and experience in the spatial and temporal dimensions. The frequency of physical
activity and travel mode variables used in this study are a quantitative spatial and tempo-
ral portrayal of the mobility of the human subject compared with the effects of objective
physical space such as human testing indicators and the living environment on health from
a ‘behavioral perspective’. This notion complements the ‘social-economic-environmental’
framework of health impacts. Exploring the coupling between human behavior and health
is also challenging. Although the continuous development of GIS and spatial positioning
technology has made it possible to monitor microscopic human mobility trajectories and
their spatial and temporal characteristics, and to improve the accuracy of behavioral trajec-
tory monitoring, it is still based on small samples and small areas. Differences in the privacy
and security requirements of different groups of people can also result in limitations in the
types of samples to be monitored and tracked. Reflective scales are mainly an individual’s
overall evaluation of behavioral activities and the health assessment, real-time feedback, or
recall assessment of objective facts, with the advantage of simplicity and speed, wide use,
and a true reflection of objective concepts to some extent. Accordingly, the reflective scale
questionnaire in this study still has the advantage of being used in a large sample of large
regions on a national scale and further validates the accuracy and scientific validity of the
result that ‘activity preference and travel behavior’ have a significant effect on health. The
negative health effect of the travel mode variable further suggests that motorized travel
helps to avoid the risk of exposure to ambient environmental pollution, thus improving
subjective perceived health.
5.4. An Important Strategy to Improve the Health of the Population
5.4.1. Positive Realization of the Healthy City Plan
In the rapid urbanization process, urban planning to further strengthen environmental
protection and environmental pollution control and improve the quality of the residents’
living environment is significantly important to improve the health of the residents. At
present, environmental pollution such as air and water pollution, caused by urban ex-
pansion, is a serious threat to the physical and psychological health of residents. Many
Chinese cities have gradually emphasized that “along with China’s rapid industrialization,
pollution of the air, water, soil, and other ecological environments, as well as food and drug
safety issues, constitute a major health hazard for the nation”. The Healthy China strategy
is an important element of China’s basic national development strategy, and healthy cities
are a major goal that Chinese cities are striving to achieve. In 2013, 2018, and 2021, cities
have continued to emphasize the major strategy of “fighting the battle against pollution”,
promoting urban planning to strengthen “environmental protection and governance”. At
present, urban planning has achieved good results in the planning, management, and
remediation of environmental pollution issues such as air and water pollution. Food safety
is one of the most important urban issues for urban management and is a key point for
the health of residents. To further improve the health of the population, it is therefore of
great theoretical and practical importance to study and implement strategies for healthy
city planning that address a range of issues such as air pollution, water pollution, and
food safety.
Int. J. Environ. Res. Public Health 2022,19, 14225 14 of 17
5.4.2. Enhancing the Scale and Accessibility of Blue-Green Spaces
Enhancing the scale of urban green space improves the accessibility of green space
for residents, so urban and rural residents can have access to sufficient green space. The
urban blue-green space can provide residents with places to exercise, which is conducive
to improving their physical health; at the same time, the blue-green space also has a signifi-
cant psychological health-enhancing effect. The urban blue-green space has a significant
PM2.5 reduction effect. Green spaces mitigate PM2.5 pollution through their open space
characteristics and biological purification functions, whereas blue spaces reduce PM2.5
concentrations due to their good ventilation and diffusion functions. In urban landscape
planning and green space system planning, the scale of green space and green space cov-
erage can be further enhanced. More blue and green spaces as dots and strips can be
distributed in high-density urban spaces, which can improve their spatial utilization effi-
ciency and health effects. The diversity of land use types can be improved, and a reasonable
scale of blue-green space can be allocated in residential and working areas.
5.4.3. Scientific Mechanisms for Physical Exercise
1
Further increase the frequency of physical activity where conditions allow. Whereas
spatial diversity and green space accessibility can enhance a healthy quality of life, this
depends on the individual’s walkability choices and frequency of walking. Therefore,
increasing the frequency of physical activity can further improve people’s health, satisfac-
tion, and well-being.
2
The choice of open spaces or parks as places of activity is a good
indicator of the pollution-reducing effect of blue-green urban spaces, which allow residents
to breathe fresh air when exercising. This choice avoids areas with high air pollution such
as near city roads and along the road during peak commuting hours. The area near city
roads is a notable area for the accumulation of vehicle emissions, which makes it easier to
inhale harmful substances when doing physical exercise, which is not good for one’s health.
3
Make scientific travel and exercise plans, use environmental pollution monitoring data,
focus on the level of air pollution, water pollution, and other environmental problems
in activity in advance, and avoid long-term exposure to areas with serious air pollution.
4
Monitor the physical exercise process for any discomfort, shortness of breath, coughing,
and other problems.
5.4.4. Bridging the Gap between Urban and Rural Areas and Increasing Residents’ Incomes
1
Actively promote urban renewal, improve urban governance, renovate dilapidated
and decaying facilities in the city, and build a safer and more comfortable public space
environment. This will help improve the health of residents by attracting more participation
in public activities.
2
Retrofit communities with age-appropriate facilities, select safe and
suitable locations for elderly people’s activities and encourage them to participate in
physical exercise.
3
Based on the characteristic resources of the countryside, promote
the development of rural industries, optimize the industrial layout, increase the income
level of farmers, and improve their income level. Scientifically lay out the ecological space
for production and living in the countryside, preserve as much of the original landforms
and natural ecology as possible, systematically protect the natural scenery and idyllic
landscapes of the countryside, and improve its appearance.
4
Improve and improve the
living environment in economically disadvantaged areas, narrow the gap between urban
and rural areas, and compensate for the shortcomings in infrastructure and public service
facilities in economically disadvantaged areas.
5
Insist on the integration of urban and
rural areas, reflecting the distinction between urban and rural areas, and design different
development strategies for different types of villages according to local conditions. Focus
on the common construction and sharing of infrastructure and public service facilities
between villages, and strengthen the farmers’ service function of villages.
Int. J. Environ. Res. Public Health 2022,19, 14225 15 of 17
6. Conclusions
This study used big data from dynamic air quality monitoring and social survey
data (e.g., the China CGSS National Survey) to assess the spatial differences and coupling
relationships between subjective and objective air pollution. Furthermore, this study
explored the mechanisms of social inequality, air pollution, food security, and insufficient
green space for health, based on a logistic model. The study found that:
(1) Significant heterogeneity was observed between subjective and objective air pollution.
The correlation coefficient between subjective and objective air pollution was small
and the internal association between the two was insignificant. The significance of
the subjective and objective air pollution variables in the logistic model was differ-
ent, with the subjective variables being significant, whereas the objective variables
were excluded from the model variables. Thus, subjective air pollution has a more
significant influence on the residents’ health.
(2)
Based on a health study of a complete sample of urban and rural residents, income
inequality, air pollution, food pollution, and travel behavior can significantly affect
the health level of residents, and the negative health effects of environmental pol-
lution from air pollution, food pollution, and insufficient green space are evident.
Furthermore, urban–rural health inequalities from the perspective of socioeconomic
inequalities are also particularly evident, with gender, household size, travel, and
physical activity having insignificant effects on the health of the rural population.
Health improvement from increased income is much higher for groups with lower
income levels than for those with higher income levels. The health-enhancing benefits
per unit of income are much higher for rural residents than for urban residents.
(3)
This study found a significant urban–rural differentiation mechanism for environ-
mental health effects from a health perspective. Logistic regressions were conducted
on urban and rural samples to determine the degree of influence and significance of
the variables between them based on the coefficients. The results indicate that urban
residents are more concerned about health from an environmental quality perspective
and are more at risk from air pollution and insufficient green space elements. In this
study, food pollution variables were included in the independent variable system of
the regression equation to explore the health impact factors, and a negative effect of
food pollution on health was found. Furthermore, rural residents were more likely to
have negative health effects when affected by food pollution due to the urban–rural
divide in healthy food desertification, and their risk of exposure to food pollution
exposure was greater. In this study, food contamination variables were included in
the independent variable system of the exploratory regression equation for factors
that influence health, and negative health effects of food contamination were found.
Furthermore, an urban–rural health inequality differentiation mechanism caused by
the urban–rural divide of healthy food desertification was found.
In summary, ensuring social parity, strict control of environmental pollution, healthy
exercise, and travel can help narrow the gap between rich and poor, promote urban–rural
health equity, and improve human health in China.
Author Contributions:
Conceptualization, P.Z. and T.C.; Methodology, P.Z. and Y.P.; Software, Y.P.
and W.X.; Writing—original draft preparation, P.Z., T.C. and S.S.; Supervision, T.C., W.X. and H.Z. All
authors have read and agreed to the published version of the manuscript.
Funding:
This research was supported by the National Natural Science Foundation of China
(52278076), Hubei Provincial Social Science Fund General Project (later funded project; 2020158), the
Construction Science and Technology Plan Project of Hubei Provincial Department of Housing and
Urban Rural Development (research on the evaluation system of urban human settlement environ-
ment quality based on multivariate big data, 2021018), and the Hubei University Student Innovation
and Entrepreneurship Project (S202010490027).
Institutional Review Board Statement: Not applicable.
Int. J. Environ. Res. Public Health 2022,19, 14225 16 of 17
Informed Consent Statement: Not applicable.
Data Availability Statement:
Restrictions apply to the availability of data. Data were obtained from
the China Comprehensive Social Survey (CGSS).
Acknowledgments:
The authors thank the anonymous reviewers for their valuable comments and
constructive suggestions.
Conflicts of Interest:
The authors declare no conflict of interest. This company had no role in
the design of the study; in the collection, analyses, or interpretation of data; in the writing of the
manuscript, or in the decision to publish the results.
References
1.
Liu, W.; Xu, Z.; Yang, T. Health Effects of Air Pollution in China. Int. J. Environ. Res. Public Health
2018
,15, 1471. [CrossRef]
[PubMed]
2.
Li, Z.; Folmer, H.; Xue, J. To what extent does air pollution affect happiness? The case of the Jinchuan mining area, China. Ecol.
Econ. 2014,99, 88–99. [CrossRef]
3.
Song, Y.; Li, M.; Zhang, M.; Sun, X. Study on the impact of air pollution control on urban residents’ happiness from microscopic
perspective. J. Clean. Prod. 2019,229, 1307–1318. [CrossRef]
4. Levinson, A. Valuing public goods using happiness data: The case of air quality. J. Public Econ. 2012,96, 869–880. [CrossRef]
5.
Ferreira, S.; Akay, A.; Brereton, F.; Cuñado, J.; Martinsson, P.; Moro, M.; Ningal, T.F. Life satisfaction and air quality in Europe.
Ecol. Econ. 2013,88, 1–10. [CrossRef]
6. MacKerron, G.; Mourato, S. Life satisfaction and air quality in London. Ecol. Econ. 2009,68, 1441–1453. [CrossRef]
7.
Kimic, K.; Ostrysz, K. Assessment of Blue and Green Infrastructure Solutions in Shaping Urban Public Spaces—Spatial and
Functional, Environmental, and Social Aspects. Sustainability 2021,13, 11041. [CrossRef]
8.
Chen, C.; Guo, P. Can Indoor Residents Perceive Green and Blue Spaces in Communities as Posted Sentiments? A Verification in
Nanchang. Forests 2022,13, 1421. [CrossRef]
9.
Wolch, J.R.; Byrne, J.; Newell, J.P. Urban green space, public health, and environmental justice: The challenge of making cities
‘just green enough’. Landsc. Urban Plan. 2014,125, 234–244. [CrossRef]
10.
James, A.J.; Asal, M.J. Urban-Rural Differences in Childhood and Adolescent Obesity in the United States: A Systematic Review
and Meta-Analysis. Child. Obes. 2015,11, 233–241. [CrossRef]
11.
Abera, A.; Friberg, J.; Isaxon, C.; Jerrett, M.; Malmqvist, E.; Sjostrom, C.; Taj, T.; Vargas, A.M. Air Quality in Africa: Public Health
Implications. Public Health 2021,42, 193–210. [CrossRef] [PubMed]
12.
Sacks, J.D.; Lloyd, J.M.; Zhu, Y.; Anderton, J.; Jang, C.J.; Hubbell, B.; Fann, N. The Environmental Benefits Mapping and Analysis
Program—Community Edition (BenMAP-CE): A tool to estimate the health and economic benefits of reducing air pollution.
Environ. Modell. Softw. 2018,104, 118–129. [CrossRef]
13.
Li, Z.; Liu, Q.; Zhang, Y.; Yan, K.; Yan, Y.; Xu, P. Characteristics of Urban Parks in Chengdu and Their Relation to Public Behaviour
and Preferences. Sustainability 2022,14, 6761. [CrossRef]
14.
Feng, T.; Du, H.B.; Lin, Z.G.; Zuo, J. Spatial spillover effects of environmental regulations on air pollution: Evidence from urban
agglomerations in China. J. Environ. Manage. 2020,272, 10. [CrossRef] [PubMed]
15.
Qu, L.L.; Li, Y.R.; Feng, W.L. Spatial-temporal differentiation of ecologically-sustainable land across selected settlements in China:
An urban-rural perspective. Ecol. Indic. 2020,112, 10. [CrossRef]
16.
Ma, J.; Li, C.J.; Kwan, M.P.; Chai, Y.W. A Multilevel Analysis of Perceived Noise Pollution, Geographic Contexts and Mental
Health in Beijing. Int. J. Environ. Res. Public Health 2018,15, 18. [CrossRef] [PubMed]
17.
Zhou, Q.; Zhang, X.L.; Shao, Q.L.; Wang, X.L. The non-linear effect of environmental regulation on haze pollution: Empirical
evidence for 277 Chinese cities during 2002–2010. J. Environ. Manage. 2019,248, 12. [CrossRef]
18.
Zhu, Y.; Wang, J.; Meng, B.; Ji, H.; Wang, S.; Zhi, G.; Liu, J.; Shi, C. Quantifying Spatiotemporal Heterogeneities in PM2.5-Related
Health and Associated Determinants Using Geospatial Big Data: A Case Study in Beijing. Remote Sens.
2022
,14, 4012. [CrossRef]
19.
Muttoo, S.; Jeena, P.M.; Röösli, M.; de Hoogh, K.; Naidoo, R.N. A Review of Ambient Air Pollution Exposure Assessment Methods
in Determining Childhood Respiratory Health Effects in Children under Five. Environments 2022,9, 107. [CrossRef]
20.
HE, S.; CHAI, Y.w.; GUO, W.b.; TAN, Y.m.; LIU, B.c. City women’ leisure satisfaction and its influencing factors on a daily scale:a
case study of Xining city. Econ. Geogr. 2019,39, 224–231. [CrossRef]
21. Liu, C.; Li, X. Air pollution and urban-rural income gap -Based on health perspective. Stat. Decis. 2021,37, 100–103. [CrossRef]
22.
Ma, X.j.; Wang, C.x.; Zhang, Z.y. Quantitative Study of Residents’ Wellbeing from the Perspective of a “Two-dimensional”
Environment: New Evidence from China CGSS Data. Stat. Res. 2019,36, 56–67. [CrossRef]
23.
Shi, X.; Li, X.; Chen, X.; Zhang, L. Objective air quality index versus subjective perception: Which has a greater impact on life
satisfaction? Environ. Dev. Sustain. 2022,24, 6860–6877. [CrossRef]
24.
Yan, D.; Chen, G.; Lei, Y.; Zhou, Q.; Liu, C.; Su, F. Spatiotemporal Regularity and Socioeconomic Drivers of the AQI in the Yangtze
River Delta of China. Int. J. Environ. Res. Public Health 2022,19, 9017. [CrossRef]
Int. J. Environ. Res. Public Health 2022,19, 14225 17 of 17
25.
Liang, W.; Yang, M. Urbanization, economic growth and environmental pollution: Evidence from China. Sust. Comput.
2019
,21,
1–9. [CrossRef]
26.
Chen, L.L.; Shen, W. Spatiotemporal differentiation of urban-rural income disparity and its driving force in the Yangtze River
Economic Belt during 2000–2017. PLoS ONE 2021,16, e0245961. [CrossRef]
27.
Gong, P.; Liang, S.; Carlton, E.J.; Jiang, Q.W.; Wu, J.Y.; Wang, L.; Remais, J.V. Urbanisation and health in China. Lancet
2012
,379,
843–852. [CrossRef]
28.
Chen, J.M.; Li, C.L.; Ristovski, Z.; Milic, A.; Gu, Y.T.; Islam, M.S.; Wang, S.X.; Hao, J.M.; Zhang, H.F.; He, C.R.; et al. A review of
biomass burning: Emissions and impacts on air quality, health and climate in China. Sci. Total Environ.
2017
,579, 1000–1034.
[CrossRef]
29.
Lu, X.C.; Lin, C.Q.; Li, W.K.; Chen, Y.; Huang, Y.Q.; Fung, J.C.H.; Lau, A.K.H. Analysis of the adverse health effects of PM2.5
from 2001 to 2017 in China and the role of urbanization in aggravating the health burden. Sci. Total Environ.
2019
,652, 683–695.
[CrossRef]
30.
Chen, Y.; Ke, X.; Min, M.; Zhang, Y.; Dai, Y.; Tang, L. Do We Need More Urban Green Space to Alleviate PM2.5 Pollution? A Case
Study in Wuhan, China. Land 2022,11, 776. [CrossRef]
31.
Lee, A.C.; Maheswaran, R. The health benefits of urban green spaces: A review of the evidence. J. Public Health
2011
,33, 212–222.
[CrossRef] [PubMed]
32.
Costache, R.; Arabameri, A.; Moayedi, H.; Pham, Q.B.; Santosh, M.; Nguyen, H.; Pandey, M.; Pham, B.T.J.G.I. Flash-flood potential
index estimation using fuzzy logic combined with deep learning neural network, naïve Bayes, XGBoost and classification and
regression tree. Geocarto Int. 2022,37, 6780–6807. [CrossRef]
33.
Ekkel, E.D.; de Vries, S. Nearby green space and human health: Evaluating accessibility metrics. Landsc. Urban Plan.
2017
,157,
214–220. [CrossRef]
34.
Browning, M.; Rigolon, A.; McAnirlin, O.; Yoon, H. Where greenspace matters most: A systematic review of urbanicity, greenspace,
and physical health. Landsc. Urban Plan. 2022,217, 13. [CrossRef]
35.
Markevych, I.; Schoierer, J.; Hartig, T.; Chudnovsky, A.; Hystad, P.; Dzhambov, A.M.; de Vries, S.; Triguero-Mas, M.; Brauer, M.;
Nieuwenhuijsen, M.J.; et al. Exploring pathways linking greenspace to health: Theoretical and methodological guidance. Environ.
Res. 2017,158, 301–317. [CrossRef]
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Ecological infrastructures (EIs), such as public and urban green and blue spaces (GBSs), have been well demonstrated to benefit visitors’ mental well-being. Experiences in community GBSs may also evoke positive emotions for their residents. In this study, 54 communities in Nanchang were chosen as objective sites, where landscape metrics of GBSs were remotely evaluated. A total of 2105 local residents’ facial expressions (with happy, sad, and neutral emotions) were obtained from Sina Weibo. Inhabitants showed more net positive emotions (happy minus sad) in cold seasons, and females smiled more frequently than males. Newly constructed communities with houses for sale had larger areas of normalized difference vegetation index (NDVI) and built-up index compared to communities with no houses for sale. Neither the availability of houses for sale nor housing price had any effect on facial expression scores. Poisson regression revealed significant coefficients (β) of positive emotions with largeness of green space (GS) and blue space (BS). Overall, BS had a stronger contribution (β, 0.6–1.1) to residents showing positive emotions relative to GS (β, −2.45–0.89), whose area ratio of NDVI increased the frequency of showing happiness. We recommend constructing GBSs with over 0.68 km2 of GS and over 2000 m2 of BS per community, where the area ratio of GS should be more than 70% of the total if the goal is to evoke more happiness in residents.
Article
Full-text available
Various epidemiological studies have reported on air pollution exposure-related lung function decline and respiratory health effects in children. Children have increased susceptibility to ambient air pollutants as physiological and structural changes of the lung are still occurring within the first five years of life after birth. This review examines applications in air pollution exposure assessment methods when evaluating lung function and respiratory health concentration–response effects in young children, while considering the effects of critical windows of exposure. We identified 13 studies that used various methods of exposure assessment in assessing respiratory health outcomes (presence of lower respiratory tract infections, respiratory symptoms, wheezing and asthma) in children under five. The methods applied included personal monitoring (n = 1), proximity-based methods (n = 3), inverse distance weighting (n = 2), geographic weighted regression (n = 1), dispersion modeling (n = 1), satellite-based methods (n = 2) and land use regression modeling (n = 5). These studies assessed exposure and outcomes at different “windows of susceptibility”: antenatally/specific trimesters (n = 8), infancy (n = 5) and early childhood (n = 6). In most studies, the reported measures of air pollutants were noted to be below the prescribed limits, though for some, a cause–effect association was observed. It was also noted that there was very little variation in estimates between time points or trimesters of exposure, likely attributed to limitations in the selected exposure assessment method. Moderate to high correlations between trimesters were reported for most studies.
Article
Full-text available
Air pollution has brought about serious challenges to public health. With the limitations of available data, previous studies overlooked spatiotemporal heterogeneities in PM2.5-related health (PM2.5-RH) and multiple associated factors at the subdistrict scale. In this research, social media Weibo data was employed to extract PM2.5-RH based on the Bidirectional Encoder Representations from Transformers (BERT) model, in Beijing, China. Then, the relationship between PM2.5-RH and eight associated factors was qualified based on multi-source geospatial big data using Geographically Weighted Regression (GWR) models. The results indicate that the PM2.5-RH in the study area showed a spatial pattern of agglomeration to the city center and seasonal variation in the spatially non-stationary effects. The impacts of varied factors on PM2.5-RH were also spatiotemporally heterogeneous. Specifically, nighttime light (NTL), population density (PD) and the normalized difference built-up index (NDBI) had outstanding effects on PM2.5-RH in the four seasons, but with spatial disparities. The impact of the normalized difference vegetation index (NDVI) on PM2.5-RH was significant in summer, especially in the central urban areas, while in winter, the contribution of the air quality index (AQI) was increased. This research further demonstrates the feasibility of using social media data to indicate the effect of air pollution on public health and provides new insights into the seasonal impacts of associated driving factors on the health effects of air pollution.
Article
Full-text available
Air pollution has caused adverse effects on the climate, the ecological environment and human health, and it has become a major challenge facing the world today. The Yangtze River Delta (YRD) is the region with the most developed economy and the most concentrated population in China. Identifying and quantifying the spatiotemporal characteristics and impact mechanism of air quality in this region would help in formulating effective mitigation policies. Using annual data on the air quality index (AQI) of 39 cities in the YRD from 2015 to 2018, the spatiotemporal regularity of the AQI is meticulously uncovered. Furthermore, a geographically weighted regression (GWR) model is used to qualify the geographical heterogeneity of the effect of different socioeconomic variables on the AQI level. The empirical results show that (1) the urban agglomeration in the YRD presents an air pollution pattern of being low in the northwest and high in the southeast. The spatial correlation of the distribution of the AQI level is verified. The spatiotemporal regularity of the “high clustering club” and the “low clustering club” is obvious. (2) Different socioeconomic factors show obvious geographically heterogeneous effects on the AQI level. Among them, the impact intensity of transportation infrastructure is the largest, and the impact intensity of the openness level is the smallest. (3) The upgrading of the industrial structure improves the air quality status in the northwest more than it does in the southeast. The impact of transportation infrastructure on the air pollution of cities in Zhejiang Province is significantly higher than the impact on the air pollution of other cities. The air quality improvement brought by technological innovation decreases from north to south. With the expansion of urban size, there is a law according to which air quality first deteriorates and then improves. Finally, the government should promote the upgrading of key industries, reasonably control the scale of new construction land, and increase the cultivation of local green innovative enterprises.
Article
Full-text available
Urban parks can offer a variety of ecosystem services such as beautifying the landscape, ecological regulation, leisure and recreation, and maintaining biodiversity. Understanding how urban parks affect people’s lives can help guide the construction and planning of parks in a direction that is more beneficial to the public. Therefore, it is worth studying the extent to which different urban parks with different characteristics affect public behaviour and preferences. This paper takes five typical urban parks in Chengdu and analyses the relationship between characteristics of the park ecosystem and public behaviour and preferences. The characteristics include the park scale, blue-green space ratio, plant diversity, and degree of re-wilding. Visit frequency, stay time, and park preference characterise public behaviour and preferences. The results show: (1) There are obvious differences in the ecosystem characteristics of the five parks: Qinglong Lake Wetland Park is the largest; the proportion of blue-green space in Jiangjiayiyuan Garden is the highest; the degree of re-wilding in Bailuwan Wetland Park is the highest; the proportion of green space and plant diversity in Guixi Ecological Park is the highest; and the proportion of blue space in Jincheng Lake Wetland Park is the highest. (2) There are differences in public behaviour and preferences for different parks. Tourists visit Qinglong Lake Wetland Park the most in spring and autumn and they choose Guixi Ecological Park instead in summer and winter. The public stays longer in Qinglong Lake Wetland Park and shorter in Jincheng Lake Wetland Park. (3) The scale of urban parks, the proportion of blue-green space, and the degree of re-wilding, especially the proportion of blue space, have a positive impact on the public’s evaluation and promote public visits. The results of the study could help improve public awareness of the relationship between park characteristics and ecological services and well-being.
Article
Full-text available
Urban green space can help to reduce PM2.5 concentration by absorption and deposition processes. However, few studies have focused on the historical influence of green space on PM2.5 at a fine grid scale. Taking the central city of Wuhan as an example, this study has analyzed the spatiotemporal trend and the relationship between green space and PM2.5 in the last two decades. The results have shown that: (1) PM2.5 concentration reached a maximum value (139 μg/m3) in 2010 and decreased thereafter. Moran’s I index values of PM2.5 were in a downward trend, which indicates a sparser distribution; (2) from 2000 to 2019, the total area of green space decreased by 25.83%. The reduction in larger patches, increment in land cover diversity, and less connectivity led to fragmented spatial patterns of green space; and (3) the regression results showed that large patches of green space significantly correlated with PM2.5 concentration. The land use/cover diversity negatively correlated with the PM2.5 concentration in the ordinary linear regression. In conclusion, preserving large native natural habitats can be a supplemental measure to enlarge the air purification function of the green space. For cities in the process of PM2.5 reduction, enhancing the landscape patterns of green space provides a win-win solution to handle air pollution and raise human well-being.
Article
Full-text available
Blue and Green Infrastructure (BGI) provide one of the key Nature Based Solution (NBS) approaches for sustainable stormwater management in cities, in conjunction with extending the scope of Ecosystem Services (ES). In both the process of planning and designing highly urbanized areas, the implementation of BGI is important for the improvement of living conditions and counteracting the negative effects of climate change. Based on the literature review, 19 BGI solutions were identified and then valorized in relation to the following three key aspects: spatial and functional, environmental, and social. The results of the assessment were derived using the scoring method and allowed for the identification of BGI solutions with a high, medium or low value for shaping sustainable urban public spaces. Using the potential of analyzed BGI solutions to improve the functioning and attractiveness of urban areas requires a comprehensive approach. Conscious planning and designing should use the knowledge presented to make the implementation of BGI solutions as effective as possible in relation to the above-mentioned aspects of shaping urban public spaces.
Article
Full-text available
Greenspace in urban areas may have greater protective health effects than elsewhere. Urban dwellers experience more environmental harmful exposures, attentional demands, and stressors than their suburban/rural counterparts. In this systematic review, we synthesize the results of studies that examined how the greenspace and health relationship varies by urbanicity. We searched for articles in April 2019 that found positive associations between greenspace and physical health. Included articles tested for effect modification by urbanicity among one or more (of eight total) outcomes relevant to health and environmental equity. We coded results as 1 = stronger association in more urban areas, −1 = stronger association in less urban areas, or 0 = no difference. We found 57 analyses in 37 articles that met our inclusion criteria. Among these analyses, 50.9% showed no difference, 38.6% showed a stronger association for more urban areas, and 10.5% showed a stronger association for less urban areas. More urban areas had stronger associations for cardiovascular-related, birth, and mortality outcomes and for greenspace measured within 500 m. Stronger greenspace-health associations in more urban areas might be explained in part by the mechanistic pathways underlying these associations. Greenspace can reduce harms from environmental exposures (i.e., air pollution, noise, heat, and artificial light at night) in addition to alleviating attentional demands, reducing chronic stressors, and promoting healthy behavior - factors which might be more necessary, prevalent, or stronger in urban areas. These potential explanations warrant further investigation. The findings of this review inform public health policy and planning professionals who are attempting to make cities livable for all residents.
Article
Full-text available
Although the PM2.5 concept was put forward in 1997, it came into Chinese people's sight in 2011, and became a household name in 2013 in China. Previous studies did not involve this indicator, so we need to re-examine and compare the effects of objective air quality data and subjective perception on Life satisfaction (LS). The data were collected from Xi'an, a megacity of Northwest China. Particles smaller than 2.5 μm in diameter (PM2.5), nitrogen dioxide (NO2), sulfur dioxide (SO2), and Air quality index (AQI) acquired from air quality stations were used to calculate pollutant concentrations near the vicinity of subjects’ home by Geographical information system (GIS) software. In this paper, the ordinal logistic regression models are used to explain ratings of LS in terms of the objective air pollution data, social and economic variables. The results indicate that the objective air pollution data has significant and negative effects on LS of the surveyed respondents. However, when the subjective perception of air pollution is considered in the regression models, it shows that the subjective perception of air pollution has a significant negative impact on LS, while the effects of PM2.5, NO2, SO2, and AQI on the LS become insignificant. Furthermore, the research significances and policy suggestions are discussed in this paper. Graphic abstract
Article
Flash floods pose a major challenge in various regions of the world, causing serious damage to life and property. Here we investigated the Izvorul Dorului river basin from Romania, to identify slope surfaces with a high potential for flash-flood employing a combination of Fuzzy Logic algorithm with the following four machine learning models: Classification and Regression Tree, Deep Learning Neural Network, XGBoost, and Naïve Bayes. Ten flash-flood predictors were used as independent variables to determine the flash-flood potential index. As a dependent variable, we used areas with ttorrential phenomena divided into training (70%) and validating data set (30%). Predictive ability and the degree of Correlation between factors were assessed through the Correlation-based Feature Selection (CFS) method and through the Confusion Matrix, respectively. In the training phase, all ensemble models yielded good and very good accuracies of over 84%. The spatialization of Flash-Flood Potential Index (FFPI) over the study area showed that high and very high values of flash-flood potential occur in the northern half of the region and occupy the following weights within the study area: 53.11% (FFPI Fuzzy-CART), 45.09% (Fuzzy-DLNN), 45.58% (Fuzzy-NB) and 44.85% (Fuzzy-XGBoost). The validation of the results was done through the ROC Curve method. Thus, according to Success Rate, Fuzzy-XGBoost (AUC =0.886) is the best model, while in terms of Prediction Rate, the ideal one is Fuzzy-DLNN (AUC =0.84). The novelty of this work is the application of the four ensemble models in evaluating this natural hazard.