ORIGINAL RESEARCH
published: 17 January 2022
doi: 10.3389/fbuil.2021.813399

Edited by:
Hadi Sarvari,
Islamic Azad University, Iran
Reviewed by:
Aynaz Lotfata,
Chicago State University,
United States
Hala Nassereddine,
University of Kentucky, United States
*Correspondence:
Fan Xue
xuef@hku.hk
†

ORCID:
Maosu Li
orcid.org/0000-0002-9970-4053
Yi Peng
orcid.org/0000-0002-7042-9254
Yijie Wu
orcid.org/0000-0003-1441-1583
Jinying Xu
orcid.org/0000-0001-9589-6396
Tan Tan
orcid.org/0000-0002-3374-8299
Hui Guo
orcid.org/0000-0003-2801-1652
Weisheng Lu
orcid.org/0000-0003-4674-0357
Anthony G. O. Yeh
orcid.org/0000-0002-0587-0588
Fan Xue
orcid.org/0000-0003-2217-3693
Specialty section:
This article was submitted to
Construction Management,
a section of the journal
Frontiers in Built Environment
Received: 11 November 2021
Accepted: 24 December 2021
Published: 17 January 2022
Citation:
Li M, Peng Y, Wu Y, Xu J, Tan T,
Guo H, Lu W, Yeh AGO and Xue F
(2022) Role of the Built Environment in
the Recovery From COVID-19:
Evidence From a GIS-Based Natural
Experiment on the City Blocks in
Wuhan, China.
Front. Built Environ. 7:813399.
doi: 10.3389/fbuil.2021.813399

Role of the Built Environment in the
Recovery From COVID-19: Evidence
From a GIS-Based Natural Experiment
on the City Blocks in Wuhan, China
Maosu Li 1†, Yi Peng 2†, Yijie Wu 3†, Jinying Xu 3†, Tan Tan 4†, Hui Guo 3†, Weisheng Lu 3†,
Anthony G. O. Yeh 1† and Fan Xue 3*†
1

Department of Urban Planning and Design, The University of Hong Kong, Pokfulam, Hong Kong, China, 2School of Public
Administration, Zhejiang University of Finance and Economics, Hangzhou, China, 3Department of Real Estate and Construction,
The University of Hong Kong, Pokfulam, Hong Kong, China, 4The Bartlett School of Sustainable Construction, University College
London, London, United Kingdom

The built environment closely relates to the development of COVID-19 and postdisaster recovery. Nevertheless, few studies examine its impacts on the recovery stage
and corresponding urban development strategies. This study examines the built
environment’s role in Wuhan’s recovery at the city block level through a natural
experiment. We ﬁrst aggregated eight built environmental characteristics (BECs) of
192 city blocks from the perspectives of density, infrastructure supply, and
socioeconomic environment; then, the BECs were associated with the recovery
rates at the same city blocks, based on the public “COVID-19-free” reports of
about 7,100 communities over the recovery stages. The results showed that three
BECs, i.e., “number of nearby designated hospitals,” “green ratio,” and “housing price”
had signiﬁcant associations with Wuhan’s recovery when the strict control measures
were implemented. At the ﬁrst time of reporting, more signiﬁcant associations were also
found with “average building age,” “neighborhood facility development level,” and
“facility management level.” In contrast, no associations were found for “controlled
residential land-use intensity” and “plot ratio” throughout the stages. The ﬁndings from
Wuhan’s recovery pinpointing evidence with implications in future smart and resilient
urban development are as follows: the accessibility of hospitals should be
comprehensive in general; and the average housing price of a city block can reﬂect
its post-disaster recoverability compared to that of the other blocks.
Keywords: built environment, post-disaster recovery, COVID-19-free community, natural experiment, GIS

HIGHLIGHTS
• Natural experiment for testing correlations between the built environment and Wuhan’s
recovery from COVID-19.
• Designated hospitals, neighborhood facility level, green ratio, housing price, building age, and
facility management level were associated with the early recovery (p < 0.01).
• Only designated hospitals, green ratio, and housing price were still associated with the recovery
in later stages (p < 0.05).

Frontiers in Built Environment | www.frontiersin.org

1

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

• No association found for controlled residential land-use
intensity and plot ratio throughout the whole stage.
• New evidence and implications for smarter and resilient
urban development.

et al., 2020). Although some cities have experienced recoveries
from the pandemic, most cities are still struggling with handling
COVID-19. Thus, the impacts of BECs on COVID-19 recovery
from different spatial scales are scarce for exploration. Compared
to the large-scale analysis, block-level BECs and the COVID-19
recovery can present more ﬁne-scale insights into a precious nonpharmaceutical intervention and inspire the future resilient urban
development (Li et al., 2021). However, its comprehensive
understanding is desired but overlooked.
Wuhan is a megacity with over 11 million inhabitants living in
about 7,100 residential communities. As the ﬁrst lockdown and
recovered city, the city was locked down on 23 January 2020 and
resumed in early April 2020. Strong interventions such as closed
community management were implemented to restrict the innercity transmission. The non-commutable urban environment
enables a natural experiment test bed to explore the role of
BECs in the COVID-19 recovery (Thomson 2020). This study
attempts to investigate the associations between the BECs and
their recovery from COVID-19 in the block level through
geospatial statistical analysis. Two questions are examined: 1)
What is the initial recovery distribution of Wuhan city blocks
with different built environments? 2) Under strong interventions
such as the intracity mobility restriction and community-based
lockdown, what is the role of the built environment in the later
urban recovery?

1 INTRODUCTION
Coronavirus disease 2019 (COVID-19) is one of the deadliest
pandemics in human history (Sohrabi et al., 2020).
Understanding its mechanism from biological, ecological, and
social perspectives is thus important, especially to avoid its wide
spread and infection and further facilitates urban recovery. Since
the transmission is a spatial-related process, substantial
geographical research has been made to monitor, analyze,
contain, and recover from the unprecedented pandemic
(Desjardins et al., 2020; Franch-Pardo et al., 2020).
The built environment is the collection of human-made
surroundings for human activity, ranging in scale from
buildings to neighborhoods and cities. Its direct physical
settings and indirect human behavior intervention have led to
its signiﬁcant inﬂuence on infectious disease dynamics, especially
in certain infections transmitted by contact, aerosols, or droplets
(Pinter-Wollman et al., 2018). Typical built environmental
characteristics (BECs) include population and land-use density
(Yu and Wen 2016; Ahmadian et al., 2019), infrastructure supply
(Loo and Lam 2012; Chen et al., 2019), and social-economic
environment (Aldrich 2012; Qu et al., 2018), the role of which has
been proved critical for the known infectious diseases such as
cholera (Pinter-Wollman et al., 2018), H1N1 (Lowcock et al.,
2012; Ponnambalam et al., 2012), and SARS (Fang et al., 2009).
Similarly, the relationship between these BECs and the COVID19 pandemic needs to be re-identiﬁed to inspire urban recovery
and future resilient urban development.
Numerous studies nowadays have been conducted on different
spatial scales to unfold the impacts of BECs on the COVID-19
development (Frank et al., 2019). For instance, there existed a
positive correlation between the county-level population density
and the conﬁrmed COVID-19 cases in Hubei Province, China,
(Xiong et al., 2020) and US (Klompmaker et al., 2021; Sy et al.,
2021). District-level land-use and population density as well as
infrastructure settings including green space ratio and hospital
density were signiﬁcantly correlated with the morbidity rate in
Wuhan of Hubei (You et al., 2020). At the block and community
level, commercial vitality and infrastructure supply were found to
have a signiﬁcant correlation with the number of conﬁrmed cases
in Wuhan (Li et al., 2020) and Huangzhou of Hubei (Li et al.,
2021) as well as Hong Kong (Kan et al., 2021).
Nevertheless, the role of the BECs in the COVID-19 recovery
may be different from that in urban development. For instance, it
is found that although the high land-use density may increase the
concentration of people and facilitate disease transmission, their
superior health and educational systems are also more prepared
to handle pandemics, leading to higher rates of recovery and
lower rates of mortality (Dye 2008; Hamidi et al., 2020). In
contrast, inadequate healthcare management would decrease
the rate of recovery among COVID-19 patients (Hopman

Frontiers in Built Environment | www.frontiersin.org

2 RESEARCH METHODS
2.1 Research Area
This study mainly focuses on the 192 city blocks in the central
urban area of Wuhan. They were from the eight districts,
including Caidian, Hanyang, Hongshan, Jiang’an, Jianghan,
Qiaokou, Qingshan, and Wuchang as shown in Figure 1. The
city block delineated by the Wuhan Natural Resources and
Planning Bureau is the study unit in this research (WNRPB
2015). A city block is a group of communities with the same
controlled residential land-use intensity.

2.2 Research Design
Figure 2 shows the ﬂowchart of the GIS-based data aggregation
and statistical analysis methods. First, eight typical BECs and
COVID-19-free rate (CFR) are deﬁned in Section 2.3. Four kinds
of datasets from different sources were collected and cleansed
using GIS spatial measurement, as described in Section 2.4.
Spearman’s rank correlation analysis between BECs and CFR
is explained in Section 2.5.

2.3 Deﬁnitions of BECs and CFR
Table 1 lists the deﬁnitions of eight BECs as independent
variables. The BECs represent three types of quality of the
built environment at the city block level, namely the density,
infrastructure supply, and socioeconomic environment. Besides
the eight BECs, the dependent variable CFR is also measured at
the city block level.
In this research, two BECs are deﬁned to represent the density
of the built environment. One is the controlled residential land-

2

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 1 | Location of 192 city blocks of eight districts in Wuhan, China.

TABLE 2 | Controlled residential land-use intensity (CR) deﬁned for Wuhan
(WNRPB 2015).
CR

Base plot ratio

Density of buildings

1
2
3
4

3.2
2.9
2.5
1.5

≤20%
≤25%

use intensity deﬁned by the Wuhan government for each city
block, as listed in Table 1. According to local regulations, the
controlled residential land-use intensity strictly limits the
intensity of residential land development. For example, a block
with the controlled residential land-use intensity  1 has the
guideline permission of developing not more than 20% density of
buildings and a base plot ratio at 3.2, as shown in Table 2.
Another BEC is the average of actual plot ratios of communities
within the city blocks reﬂecting the real land-use intensity, the

FIGURE 2 | Natural experiment research design of this study.

TABLE 1 | Eight BECs and CFR deﬁned for a city block in this study.

BECs

Type

Abbreviation.

Density

CR

Infrastructure supply

Socioeconomic
environment

Full name

Description

Range

Unit
—

R+

—

DH

Nearby designated hospitals

GR

Neighborhood facility
development level
Green ratio

0, 1,
2, . . .
[1, 4]

—

NF

[0, 1]

—

HP

Housing price

R+

BA
FM

Building age
Facility management level

Lower value indicates the higher planned base
residential plot ratio and building density
The actual average ratio of the gross ﬂoor area to the
total land area of communities
The number of designated hospitals and clinics in the
block or in an x-km buffer zone
The average level of matureness of the neighborhood
facilities. Higher is better
The average ratio of the green land area to total land area
of communities
The average unit price at which the estates were offered
for sale
The average building age
The average level of the facility management. Higher is
better
Ratio of communities won the COVID-19-free title

[1, 4]

PR

Controlled residential landuse intensity
Plot ratio

R+
[1, 4]

10,000 RMB/
m2
Year
—

[0, 1]

—

CFR

Frontiers in Built Environment | www.frontiersin.org

COVID-19-free rate

3

—

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 3 | Data collection and preprocessing for the eight BECs.

range of which belongs to R+ representing the positive real
numbers. In addition, it also indirectly reﬂects the population
density of the city blocks because of the blocked transportation
system.
Three BECs in Table 1 represent the infrastructure supply
for each city block. One is the number of designated hospitals
and clinics, which is measured within the block with a buffer of
3 km. The other two BECs are the average development level of
the neighborhood facility and the green ratio of the city blocks.
The neighborhood facility development level using a fourpoint scale is assessed as the matureness of the neighborhood
in terms of public facilities and commercial and transportation
services, whereas the green ratio was the average ratio of the
green land area to the total land area of the communities. For
both, the higher indices stand for the better infrastructure
settings.
Last, three BECs including housing price, building age,
and neighborhood facility management level are deﬁned
to depict the socioeconomic environment. Housing price refers
to the average unit price of residential communities in
the city blocks, where the unit is 10,000 RMB/m2. The
building age refers to the average age of residential buildings.
For the facility management level of blocks, we assessed them in a
four-point scale where 1 indicated “poor” and 4 stood for
“excellent.”
CFR is deﬁned as the “COVID-19-free ratio” of
communities in the city blocks. On 5th March 18:00,
42 days after the lockdown, the Centers for Disease Control
and Prevention of Wuhan consistently started to report a series
of “COVID-19-free community” honor lists. A community

Frontiers in Built Environment | www.frontiersin.org

was called “COVID-19-free” if both of the following
conditions were met (Xiao 2020):
(1) no new conﬁrmed, suspected, fever, or close contact cases in
the past 14 days;
(2) strictly social distancing, closed estate management, fully
screened inhabitants, and regular sterilization.

2.4 Data Sources

2.4.1 Built Environmental Data
Three datasets of the built environment were collected, as shown
in Figure 3. The printed plan of 192 city blocks was from the
Wuhan Natural Resources and Planning Bureau (2015). We georegistered and reconstructed the topographical boundaries of the
city blocks with ArcGIS 10.1 and saved the geographical
boundaries in the vector format of GeoJSON, as shown in
Figure 3. Meanwhile, we collected the controlled residential
land-use intensity data for the case area. The residential landuse intensities and their spatial distribution pattern are shown in
Figure 4A.
An online property dataset was collected using Python web
crawler for the proﬁle of the communities, as shown in Figure 3.
The data source was Fang.com (2020), a city-wide property
database listing residential estates in over 7,000 communities
of whole Wuhan City. After using a geospatial analysis, the
boundaries of the 192 city blocks selected 4,138 communities
for this study. For each community, there are six characteristics
related to the population density including 1) plot ratio,
infrastructure supply including 2) neighborhood facility
development level and 3) green ratio as well as socioeconomic

4

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 4 | Visualized patterns of the eight BECs of the city blocks in Wuhan. (A) Controlled residential land-use intensity (CR), (B) plot ratio (PR), (C) neighborhood
facility development level (NF), (D) green ratio (GR), (E) housing price (HP), (F) building age (BA), (G) facility management level (FM), and (H) nearby designated
hospitals (DH3).

Frontiers in Built Environment | www.frontiersin.org

5

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 5 | Accumulated COVID-19-free rates of communities in Wuhan. (A) Three sampling time points and (B) recovery rates in three stages.

sample was selected for an approximate average of CFR1 and
CFR3. We downloaded the reports of about 7,100 communities
from Changjiang Daily (CJD, 2020a; CJD, 2020b; CJD, 2020c),
parsed the name list from the report images. Then, we registered
and ﬁne-tuned the geo-locations of all the communities to the
WGS-84 globe using Amap.com (https://lbs.amap.com/api/
webservice/, ver. 3), which is one of the most popular digital
maps and offers accurate geocoding services in China. Finally,
about 4,970 residential communities were associated with the 192
city blocks in the case area.
Figure 5 visualizes the recovery rates of each block from the
three phases, including datum-to-CFR1 (early stage), CFR1-toCFR2 (middle stage), and CFR2-to-CFR3 (ﬁnal stage). A general
pattern is found that the western Wuhan relatively lagged in
terms of city block recovery. In this research, the basic recovery
rate, i.e., CFRP1  CFR1 was selected to present the initial recovery
situation of all blocks at the early age. Then, the percentages of
recovery for the rest two stages, i.e., CFRP2  (CFR2 − CFR1)/(1 −
CFR1) and CFRP3  (CFR3 – CFR2)/(1 − CFR1) were selected as
the ‘pure’ recovery rates of the city blocks.

environment including 4) housing price, 5) building age, and 6)
facility management level. We extracted all the residential
communities for the case area and associated each community
with the actual geolocation on WGS-84 globe (transformed from
the China’s local BD-09 coordinate system). Six related BECs
were extracted as an average by aggregating the communities into
city blocks. The spatial distribution patterns of the BECs are
shown in Figures 4B–G.
Furthermore, a list of 48 designated hospitals and 60
designated fever clinics were collected, as shown in Figure 3.
The name list of hospitals designated for COVID-19 treatment in
urban Wuhan was collected from the Wuhan Municipal Health
Commission (2020a), while the list of designated fever clinics was
collected from the Wuhan Municipal Health Commission
(2020b). By matching the two lists and removing the
duplicates, 72 hospitals and fever clinics were conﬁrmed, with
their full names, addresses, and geocoded locations on the WGS84 globe. We aggregated them into city blocks to examine the
spatial distribution pattern of DH3 as shown in Figure 4H.

2.4.2 Urban Recovery Data
As shown in Figure 5A, 20 lists of COVID-19-free communities
were authorized from 5th March to 14th May by the Wuhan
Municipal Headquarters. Before the reopening of the city in early
April 2020, as shown by the green line in Figure 5A, the CFR was
about 99% in urban Wuhan. On 30 June 2020, all 7,102
communities were branded COVID-19-free. From late January
to early March 2020, Wuhan increasingly implemented a series of
measures, such as bans on transportation, suspension of public
gatherings, closed management of villages, communities, and
units. Thus, we assume that the access to the built
environment was totally controlled when the ﬁrst report was
released.
We selected three data samples from the series of COVID-19free community reports, as shown in Figure 5A, based on the
overall CFR. The ﬁrst sample on 5 March 2020 represented CFR1
 29.2%, while the second was CFR2  68.6% released on 14th
March and the third was CFR3  99.0% on 1st April. The second

Frontiers in Built Environment | www.frontiersin.org

2.5 Statistical Analysis
We ﬁrst apply Spearman’s rank correlation to examine the
relations between the eight BECs and the three CFR variables.
Pearson’s linear correlation is inappropriate in this study due to
some non-normal variables. The Spearman’s correlation
coefﬁcient ρ is deﬁned as:

covx, y
ni1 (xi − x)yi − y
 
,
ρ  
2
n
n
2
σxσy

i1 (xi − x) i1 yi − y

(1)

where n is the total number of the city blocks, xi and yi stand for
the ranks of the variables, x and y are the mean ranks, cov (x, y)
represents the covariance, and σx and σy are the standard
deviations. The top 60 dense residential city blocks (number
of communities ≥20) were selected out of 192 to suppress errors
from insufﬁcient community samples. The greater the coefﬁcient
ρ is, the stronger the relationship is between two variables.

6

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 6 | Results of Spearman’s rank correlation analysis (n  60), with the histograms at diagonal subﬁgures (abbreviations referring to Figure 4), scatter plots in
the lower triangle, and the area of the circle in the upper triangle indicates the absolute value (cold color  positive; warm  negative) of the Spearman coefﬁcient ρ (twotailed signiﬁcance p: *** ≤0.001 < ** ≤0.01 < * ≤0.05).

3 RESULTS

particular, the neighborhood facility development level had a
weak positive association (ρ  0.34, p ≤0.05, n  60), while others
had moderate associations (0.41 ≤ ρ ≤ 0.49). In other words, on 5
March 2020, or the ﬁrst sample point, the city blocks with more
aged communities and nearby hospitals, a lower green ratio, more
mature public facilities, and a higher housing price and property
management level had higher recovery rates. The possible reasons
include 1) probably lower infection rates at the initial stage and 2)
earlier recoveries in such city blocks.
Next, the associations of the six BECs against CFRP1 had a sharp
decline against CFRP2, as shown in the red rectangle #2 in Figure 6. It
indicates that the strictly closed management, i.e., complete isolation
from the outdoor urban environment for most dwellers,
comprehensively weakened the associations. For example, the ρ

Figure 6 shows the results of analysis, where the data spreadsheet is
available in the Supplementary Material. The diagonal subﬁgures
show that some variables, such as CFRP1 and neighborhood facility
development level, were not in normal distributions. The upper
triangle in Figure 6 shows Spearman’s rank correlation tests, while
the lower triangle displays the scatter plots. The scales of two axes
target the two variables of the scatter plots.
As shown in the red rectangle #1 in Figure 6, CFRP1 was
signiﬁcantly associated with all six BECs from the infrastructure
supply and socioeconomic environment, i.e., nearby designated
hospitals (3 km), neighborhood facility development level, green
ratio, housing price, building age, and facility management. In

Frontiers in Built Environment | www.frontiersin.org

7

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

FIGURE 7 | Results of sensitivity analysis against the buffer zone size for nearby designated hospitals. (A) Spearman’s coefﬁcient and (B) signiﬁcance level (n  60).

60). It indicates that 3 km was a “sweet point” for Wuhan to
aggregate the number of designated hospitals and clinics in both
the early and later stages.

values were approximately halved except for housing price. Yet, three
BECs, nearby designated hospitals (3 km), green ratio, and housing
price, were still found signiﬁcant (p < 0.05, n  60). Besides, the signs
of the two ρ values for housing price were seen as consistent.
The correlations of the six BECs against CFRP3 in the red
rectangle #3 in Figure 6 were mirror reﬂections of those
regarding CFRP2. In contrast, the controlled residential landuse intensity and plot ratio representing the density of the built
environment, as shown in the red rectangle #4 of Figure 6, had no
considerable roles in resumptions in Wuhan.
Other signiﬁcant correlations between BECs mainly from the
infrastructure supply and socioeconomic environment were also
observed. For instance, aged communities in Wuhan were strongly
associated with lower green ratios (ρ  −0.64, p ≤0.001, n  60), as
shown in the red rectangle #5 in Figure 6. Such an association also
prevails in most of China’s megacities due to the fast urbanization.
Besides, the associations between the building age and other BECs
such as nearby designated hospitals (3 km), neighborhood facility
development level, housing price, facility management level, and
controlled residential land-use intensity were also in line with the
spatial distribution patterns shown in Figure 4. Large proportion of
aged city blocks was mainly located in the central part of Wuhan,
which owned a higher housing price, higher-level nearby facility
development, and more hospitals but less facility management.
In addition, a sensitivity analysis using eleven sample sizes of
buffer zones from 0 to 10 km was implemented to test the trends
of associations considering the modiﬁable areal unit problem
(MAUP) (Openshaw 1981). Figure 7 shows that the trends in the
ﬁrst two stages. The nearby designated hospitals within an
increasing buffer zone size had a consistent strengthening
correlation with CFRP1 and a declining p-value to 10–5 at
8 km. The correlation with CFRP2 had a different trend. When
the buffer zone size increased, the ﬁgure for the coefﬁcient in
Figure 7A rose at ﬁrst and reached the maximum at 3 km; then, it
declined gradually. Under the line of p  0.05 in Figure 7B, 3 km
was the only signiﬁcant point for CFRP2 (ρ  0.2810, p ≤0.05, n 

Frontiers in Built Environment | www.frontiersin.org

4 DISCUSSION
4.1 The Role of Built Environment in
Wuhan’s Recovery
The ﬁndings of this article suggest that both infrastructure supply
and socioeconomic environment factors had signiﬁcant associations
with city block–level COVID-19 recovery in Wuhan. Six examined
BECs, i.e., nearby designated hospitals (3 km), neighborhood facility
development level, green ratio, housing price, building age, and
facility management level were signiﬁcantly associated with Wuhan’s
initial recovery from the COVID-19 pandemic. The associations on
the ﬁrst sampling point may refer to the combination of BECs’
impacts on COVID-19 transmission and the different initial
conﬁrmed possibility of blocks in different areas, which should be
further explored.
Two BECs concerning the density of the built environment,
i.e., controlled residential land-use intensity and plot ratio, were
not signiﬁcant for the city block–level recovery throughout the
study period. In contrast, studies on COVID-19 transmission
have found a signiﬁcant positive association between the
population density and the possibilities of transmission
(Chhikara et al., 2020; You et al., 2020). The comparison of
the ﬁndings shows the macro-BECs have varied associations in
different stages such as disaster control (i.e., COVID-19
transmission) and post-disaster recovery.
Furthermore, these BECs were associated with each other.
Besides the ﬁndings of aged city blocks mentioned in Section 3,
the housing price was found to be the strongest correlation with the
number of nearby designated hospitals and clinics (medical
facilities). Surprisingly, the green ratio was negatively associated
with COVID-19 recovery in Wuhan’s case, which echoes some

8

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

studies in China, for example, high public green space density
increased the COVID-19 morbidity rate in Wuhan (You et al., 2020).
In the middle and ﬁnal recovery stages, the six BECs’
associations with the recovery rates were weakened in general.
It implies that strict measures against human mobility at the
neighborhood and city block level showed a statistical impact on
the BECs’ roles. However, three BECs, i.e., nearby designated
hospitals (3 km), green ratio, and housing price, still had
signiﬁcant associations in the later stages of recovery; in
addition, the three BECs may indicate the city blocks’ innate
urban resilience levels due to the full isolation of residents from
the environment. The effectiveness of human activity and
mobility control measures, as weakened associations with
BECs, also echoes the ﬁndings in the literature (Chhikara
et al., 2020; Hartley and Perencevich 2020; Hirschhorn et al.,
2020; Islam et al., 2020; Xiong et al., 2020).

recovery. In addition, the negative association between greenery
and housing price in Wuhan may infer a disparity in
accessibilities in Wuhan. That is, the high-income population
sacriﬁces their access to nature for public services and
resources—such as hospitals, subways, and education, which
echoes studies in property valuation (Huang & Yin 2015; Xu
et al., 2016). Thus, the housing price of a city block can reﬂect its
post-disaster recoverability to an extent. In the future, one way to
resolve the disparity of infrastructure supply is to encourage
integration and sharing of resources across city blocks from the
sharing economy perspective (Liu et al., 2016). Another possible
way is comprehensive and inclusive planning that equally meets
the needs of low-income and vulnerable groups in Wuhan and
other cities.

4.3 Limitations and Future Work
This study also has limitations. Our main data source was
government reports distributed by newspapers, which were
based on medical tests and self-reporting fever data. Due to
the incubation period and asymptomatic patients, the
aggregated CFR values may contain small errors. Furthermore,
the correlation analysis between aggregated COVID-19 recovery
and built environment in the city block level was reasonable, but
lost data details such as causes and causality data. Another
limitation lies in the source of the online property database,
which does not assess the streets and parks surrounding the
communities. The signiﬁcant associations in Wuhan may also be
limited by the urban plans, climate, ethnicity, and culture though
the methodology can be replicated in other cities.
In the future, we look forward to seeing studies on BECs
associations and impacts on different cities and countries with
different intervention conditions so that the ﬁndings can be
further veriﬁed thoroughly. More quantiﬁed urban data
sources, such as city information modeling (Xue, Wu et al.,
2021) and computable street view features (Xue, Li et al.,
2021), might reveal new evidence as well. Encouragements and
subsidies based on solid evidence, for example, to increase the
quantities of nearby hospitals, promote residential landscape
sanitary management levels, and take prioritized measures for
targeted economic city blocks should be further recommended
for sustainable and smart urban development.

4.2 Signiﬁcance
This study aims to ﬁll in the research gap of the BECs and COVID19 recovery associations at the city block level. The ﬁndings provide
new evidence at a ﬁne-spatial scale for researchers and practitioners
in the ﬁelds of urban development, urban health, post-disaster
recovery, and public policy. For example, Wuhan’s tight control
measures for COVID-19 have effectively weakened BECs’
associations with COVID-19 transmission. The ﬁndings also echo
classical urban development and post-disaster recovery theories. For
example, the infrastructure settings such as medical resources and
neighborhood greenery management show signiﬁcant but inverse
impacts on Wuhan’s recovery; some BECs of the socioeconomic
environment, such as housing price, can effectively reﬂect the
disaster and secondary disaster development (Aldrich 2012; Peng
et al., 2020; Rouhanizadeh et al., 2020).
The ﬁndings may inspire smarter and more resilient urban
development, in Wuhan and elsewhere. For Wuhan, a city that
just completed its 2010–2020 Town Plan (GovWH 2010), the
accessibility of medical infrastructures, nature, and public
services—besides the existing density and transport—should be
emphasized in the next plan. In addition, the BECs’ associations
may inspire the cities that are struggling with the pandemic
(Allam and Jones 2020; Parnell 2020). Under the strong mobility
intervention, preparedness should concern more on the
infrastructure supply and socioeconomic environment for
epidemic response and recovery.
The positive associations of nearby designated hospitals
suggest systematic review and reshaping of the density and
accessibility of public health and medical resources in urban
development. The trends of the buffer ranges to the medical
facilities seem to suggest a walkable distance, such as 3 km for
Wuhan, may be a “sweet point” for optimal infrastructure designs
for planning residential blocks. This ﬁnding may be more helpful
to the fast urbanization in China and other emerging economies,
where public health facilities in the marginal areas or new towns
cannot match the speed of real estate development.
The negative associations of green ratios revealed an unusual
case. It did not disprove the role of greenery on urban health in
megacities. Instead, it implies that public health facilities
outweighed a town’s greenery landscape in terms of epidemic

Frontiers in Built Environment | www.frontiersin.org

5 CONCLUSION
This study employed a randomized GIS-based natural
experiment study across the city blocks in Wuhan to
investigate
whether
different
built
environmental
characteristics (BECs) were associated with COVID-19
recovery rates. All the six BECs from the infrastructure supply
and socioeconomic environment were found signiﬁcantly
associated with the initial COVID-19 recovery in the megacity
of Wuhan. They were nearby designated hospitals (3 km),
neighborhood facility development level, green ratio, housing
price, building age, and facility management level. The strict
measures in the later recovery stages seemed to weaken all the six
associations in general. As a result, three BECs, i.e., nearby

9

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

designated hospitals (3 km), green ratio, and housing price, still
had signiﬁcant associations. In contrast, both BECs presenting
the density of the built environment had no association with the
recovery throughout the whole stage. The statistical results in the
article may be useful for the cities that are still ﬁghting COVID19. More importantly, the evidence can pinpoint several
implications for smart and resilient urban development, for
example, comprehensive and balanced accessibility of medical
facilities and public facility services.

AUTHOR CONTRIBUTIONS

DATA AVAILABILITY STATEMENT

SUPPLEMENTARY MATERIAL

The original contributions presented in the study are included in
the article/Supplementary Material, further inquiries can be
directed to the corresponding author.

The Supplementary Material for this article can be found online at:
https://www.frontiersin.org/articles/10.3389/fbuil.2021.813399/
full#supplementary-material

REFERENCES

GovWH (2010). Wuhan General Town Plan (2010-2020). Wuhan: The
Government of Wuhan City, China. Retrieved from http://gtghj.wuhan.gov.
cn/wu/pc-992-61109.html.
Hamidi, S., Ewing, R., and Sabouri, S. (2020). Longitudinal Analyses of the
Relationship between Development Density and the COVID-19 Morbidity
and Mortality Rates: Early Evidence from 1,165 Metropolitan Counties in the
United
States.
Health
&
Place
64,
102378.
doi:10.1016/
j.healthplace.2020.102378
Hartley, D. M., and Perencevich, E. N. (2020). Public Health Interventions for
COVID-19. Jama 323 (19), 1908–1909. doi:10.1001/jama.2020.5910
Hirschhorn, L., Smith, J. D., Frisch, M. F., and Binagwaho, A. (2020). Integrating
Implementation Science into COVID-19 Response and Recovery. BMJ 369,
m1888. doi:10.1136/bmj.m1888
Hopman, J., Allegranzi, B., and Mehtar, S. (2020). Managing COVID-19 in Low- and
Middle-Income Countries. Jama 323 (16), 1549–1550. doi:10.1001/jama.2020.4169
Huang, H., and Yin, L. (2015). Creating Sustainable Urban Built Environments: an
Application of Hedonic House price Models in Wuhan, China. J. Hous Built
Environ. 30 (2), 219–235. doi:10.1007/s10901-014-9403-8
Islam, N., Sharp, S. J., Chowell, G., Shabnam, S., Kawachi, I., Lacey, B., et al. (2020).
Physical Distancing Interventions and Incidence of Coronavirus Disease 2019:
Natural experiment in 149 Countries. BMJ 370, m2743. doi:10.1136/bmj.m2743
Kan, Z., Kwan, M.-P., Wong, M. S., Huang, J., and Liu, D. (2021). Identifying the SpaceTime Patterns of COVID-19 Risk and Their Associations with Different Built
Environment Features in Hong Kong. Sci. Total Environ. 772, 145379. doi:10.1016/
j.scitotenv.2021.145379
Klompmaker, J. O., Hart, J. E., Holland, I., Sabath, M. B., Wu, X., Laden, F., et al. (2021).
County-level Exposures to Greenness and Associations with COVID-19 Incidence
and Mortality in the United States. Environ. Res. 199, 111331. doi:10.1016/
j.envres.2021.111331
Li, B., Peng, Y., He, H., Wang, M., and Feng, T. (2021). Built Environment and Early
Infection of COVID-19 in Urban Districts: A Case Study of Huangzhou. Sustain.
Cities Soc. 66, 102685. doi:10.1016/j.scs.2020.102685
Li, X., Zhou, L., Jia, T., Peng, R., Fu, X., and Zou, Y. (2020). Associating COVID-19
Severity with Urban Factors: a Case Study of Wuhan. Ijerph 17 (18), 6712.
doi:10.3390/ijerph17186712
Liu, W., Liu, Y., Twum, P., and Li, S. (2016). National Equity of Health Resource
Allocation in China: Data from 2009 to 2013. Int. J. Equity Health 15 (1), 1–8.
doi:10.1186/s12939-016-0357-1
Loo, B. P., and Lam, W. W. (2012). Geographic Accessibility Around Health Care
Facilities for Elderly Residents in Hong Kong: A Microscale Walkability Assessment.
Environ. Plann. B 39 (4), 629–646. doi:10.1068/b36146
Lowcock, E. C., Rosella, L. C., Foisy, J., McGeer, A., and Crowcroft, N. (2012). The Social
Determinants of Health and Pandemic H1N1 2009 Inﬂuenza Severity. Am. J. Public
Health 102 (8), e51–e58. doi:10.2105/AJPH.2012.300814
Openshaw, S. (1981). The Modiﬁable Areal Unit Problem. Quantitative Geogr. A Br.
view, 60–69.

ML and FX contributed to the conception and design of the study.
ML and FX organized the database. ML, YW, and FX performed
the statistical analysis and charting. ML and FX wrote the ﬁrst
draft of the manuscript. YP, JX, TT, and HG wrote subsections of
the manuscript. All authors contributed to manuscript revision
and read and approved the submitted version.

Ahmadian, E., Sodagar, B., Mills, G., Byrd, H., Bingham, C., and Zolotas, A.
(2019). Sustainable Cities: The Relationships between Urban Built Forms
and Density Indicators. Cities 95, 102382. doi:10.1016/j.cities.2019.06.013
Aldrich, D. P. (2012). Building Resilience: Social Capital in post-disaster
Recovery. Chicago, USA: University of Chicago Press.
Allam, Z., and Jones, D. S. (2020). Pandemic Stricken Cities on Lockdown.
Where Are Our Planning and Design Professionals [now, Then and into the
Future]? Land use policy 97, 104805. doi:10.1016/j.landusepol.2020.104805
Chen, Y., Shen, L., Zhang, Y., Li, H., and Ren, Y. (2019). Sustainability Based
Perspective on the Utilization Efﬁciency of Urban Infrastructure --- A
China Study. Habitat Int. 93, 102050. doi:10.1016/j.habitatint.2019.102050
Chhikara, B. S., Rathi, B., Singh, J., and Poonam, F. N. (2020). Corona Virus
SARS-CoV-2 Disease COVID-19: Infection, Prevention and Clinical
Advances of the Prospective Chemical Drug Therapeutics. Chem. Biol.
Lett. 7 (1), 63–72.
CJD. (2020a, April 1). 99.0% of Communities Are COVID-19-free in Wuhan!
Don’t Relax, Insist on Not Going Out unless Necessary (in Chinese).
Changjiang Daily. Retrieved from https://mp.weixin.qq.com/s/qmJnAhOuhA4eBPR22QN_Q
CJD. (2020b, March 15). Wuhan Announced the Fourth-Batch Assessment of
COVID-19-free Communities. Changjiang Daily. Retrieved from http://
news.cjn.cn/wsqzgzb/jtd/202003/t3591318.htm
CJD. (2020c, March 6). Wuhan Announces the First Batch of COVID-19-free
Communities, See if Your home Is There (in Chinese). Changjiang Daily.
Retrieved from http://www.cjrbapp.cjn.cn/p/166086.html
Desjardins, M. R., Hohl, A., and Delmelle, E. M. (2020). Rapid Surveillance of
COVID-19 in the United States Using a Prospective Space-Time Scan
Statistic: Detecting and Evaluating Emerging Clusters. Appl. Geogr. 118,
102202. doi:10.1016/j.apgeog.2020.102202
Dye, C. (2008). Health and Urban Living. Science 319 (5864), 766–769.
doi:10.1126/science.1150198
Fang, L.-Q., de Vlas, S. J., Feng, D., Liang, S., Xu, Y.-F., Zhou, J.-P., et al. (2009).
Geographical Spread of SARS in mainland China. Trop. Med. Int. Health 14
(S1), 14–20. doi:10.1111/j.1365-3156.2008.02189.x
Fang.com (2020). Fangtianxia Wuhan. Beijing, China: Fang.com. Retrieved
from https://wuhan.fang.com/.
Franch-Pardo, I., Napoletano, B. M., Rosete-Verges, F., and Billa, L. (2020).
Spatial Analysis and GIS in the Study of COVID-19. A Review. Sci. Total
Environ. 739, 140033. doi:10.1016/j.scitotenv.2020.140033
Frank, L. D., Iroz-Elardo, N., MacLeod, K. E., and Hong, A. (2019). Pathways
from Built Environment to Health: a Conceptual Framework Linking
Behavior and Exposure-Based Impacts. J. Transport Health 12, 319–335.
doi:10.1016/j.jth.2018.11.008

Frontiers in Built Environment | www.frontiersin.org

10

January 2022 | Volume 7 | Article 813399

Li et al.

Role of the Built Environment

zrzyhgh.wuhan.gov.cn/zwgk_18/fdzdgk/jcgk/qtwjygk/202001/t20200107_
614962.shtml.
Xiao, J. (2020, March 6). The First List of COVID-19-free Communities in Wuhan,
Try to Find Your home (in Chinese). Changjiang News. Retrieved from http://
www.cjrbapp.cjn.cn/p/166086.html.
Xiong, Y., Wang, Y., Chen, F., and Zhu, M. (2020). Spatial Statistics and Inﬂuencing
Factors of the COVID-19 Epidemic at Both Prefecture and County Levels in
Hubei Province, China. Ijerph 17 (11), 3903. doi:10.3390/ijerph17113903
Xu, T., Zhang, M., and Aditjandra, P. T. (2016). The Impact of Urban Rail Transit
on Commercial Property Value: New Evidence from Wuhan, China.
Transportation Res. A: Pol. Pract. 91, 223–235. doi:10.1016/j.tra.2016.06.026
Xue, F., Li, X., Lu, W., Webster, C. J., Chen, Z., and Lin, L. (2021). Big Data-Driven
Pedestrian Analytics: Unsupervised Clustering and Relational Query Based on
Tencent Street View Photographs. Ijgi 10 (8), 561. doi:10.3390/ijgi10080561
Xue, F., Wu, L., and Lu, W. (2021b). Semantic Enrichment of Building and City
Information Models: A Ten-Year Review. Adv. Eng. Inform. 47, 101245.
doi:10.1016/j.aei.2020.101245
You, H., Wu, X., and Guo, X. (2020). Distribution of COVID-19 Morbidity
Rate in Association with Social and Economic Factors in Wuhan, China:
Implications for Urban Development. Ijerph 17 (10), 3417. doi:10.3390/
ijerph17103417
Yu, J., and Wen, J. (2016). Multi-criteria Satisfaction Assessment of the Spatial
Distribution of Urban Emergency Shelters Based on High-Precision
Population Estimation. Int. J. Disaster Risk Sci. 7 (4), 413–429.
doi:10.1007/s13753-016-0111-8

Parnell, S. (2020). The Enabling Conditions of post-pandemic City Government.
Environ. Plann. B: Urban Analytics City Sci. 47 (7), 1143–1145. doi:10.1177/
2399808320950041
Peng, Y., Gu, X., Zhu, X., Zhang, F., and Song, Y. (2020). Recovery Evaluation of
Villages Reconstructed with Concentrated Rural Settlement after the
Wenchuan Earthquake. Nat. Hazards 104, 139–166. doi:10.1007/s11069020-04241-z
Pinter-Wollman, N., Jelić, A., and Wells, N. M. (20182017). The Impact of the Built
Environment on Health Behaviours and Disease Transmission in Social
Systems. Phil. Trans. R. Soc. B 373 (1753), 20170245. doi:10.1098/
rstb.2017.0245
Ponnambalam, L., Samavedham, L., Lee, H. R., and Ho, C. S. (2012).
Understanding the Socioeconomic Heterogeneity in Healthcare in US
Counties: the Effect of Population Density, Education and Poverty on H1N1
Pandemic Mortality. Epidemiol. Infect. 140 (5), 803–813. doi:10.1017/
S0950268811001464
Qu, S., Hu, S., Li, Q., and Yang, S. (2018). Quantitative Evaluation of the Impacts of
Driving Factors on Urban Residential Land price and Analysis of Their SpatioTemporal Heterogeneity: a Case Study of Wuhan City (In Chinese with English
Abstract). Prog. Geogr. 37 (10), 1371–1380. doi:10.18306/dlkxjz.2018.10.007
Rouhanizadeh, B., Kermanshachi, S., and Nipa, T. J. (2020). Exploratory Analysis
of Barriers to Effective post-disaster Recovery. Int. J. Disaster Risk Reduction 50,
101735. doi:10.1016/j.ijdrr.2020.101735
Sohrabi, C., Alsaﬁ, Z., O’Neill, N., Khan, M., Kerwan, A., Al-Jabir, A., et al. (2020).
World Health Organization Declares Global Emergency: A Review of the 2019
Novel Coronavirus (COVID-19). Int. J. Surg. 76, 71–76. doi:10.1016/
j.ijsu.2020.02.034
Sy, K. T. L., White, L. F., and Nichols, B. E. (2021). Population Density and Basic
Reproductive Number of COVID-19 across United States Counties. PLoS One
16 (4), e0249271. doi:10.1371/journal.pone.0249271
Thomson, B. (2020). The COVID-19 Pandemic. Circulation 142 (1), 14–16.
doi:10.1161/CIRCULATIONAHA.120.047538
WMHC. (2020b, January 20). List of Fever Clinics and Designated Hospitals in
Wuhan. Wuhan, China: Wuhan Municipal Health Commission. Retrieved
from
http://wjw.wuhan.gov.cn/xwzx_28/gsgg/202004/t20200430_1199587.
shtml.
WMHC. (2020a, February 26). Use Statistics of Beds in Designated Hospitals in
Wuhan (February 25, 2020). Wuhan, China: Wuhan Municipal Health
Commission. Retrieved from http://wjw.wuhan.gov.cn/ztzl_28/fk/tzgg/
202004/t20200430_1197173.shtml.
WNRPB (2015). Announcement of the "Regulations on the Management of Land
Use Construction Intensity in the Urban Districts of Wuhan. Wuhan, China:
Wuhan Natural Resources and Planning Bureau. Retrieved from http://

Frontiers in Built Environment | www.frontiersin.org

Conﬂict of Interest: The authors declare that the research was conducted in the
absence of any commercial or ﬁnancial relationships that could be construed as a
potential conﬂict of interest.
Publisher’s Note: All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their afﬁliated organizations, or those of
the publisher, the editors, and the reviewers. Any product that may be evaluated in
this article, or claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
Copyright © 2022 Li, Peng, Wu, Xu, Tan, Guo, Lu, Yeh and Xue. This is an openaccess article distributed under the terms of the Creative Commons Attribution
License (CC BY). The use, distribution or reproduction in other forums is permitted,
provided the original author(s) and the copyright owner(s) are credited and that the
original publication in this journal is cited, in accordance with accepted academic
practice. No use, distribution or reproduction is permitted which does not comply
with these terms.

11

January 2022 | Volume 7 | Article 813399