Connecting clusters of COVID-19: an epidemiological and serological investigation

Abstract

Background
Elucidation of the chain of disease transmission and identification of the source of coronavirus disease 2019 (COVID-19) infections are crucial for effective disease containment. We describe an epidemiological investigation that, with use of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) serological assays, established links between three clusters of COVID-19.

Methods
In Singapore, active case-finding and contact tracing were undertaken for all COVID-19 cases. Diagnosis for acute disease was confirmed with RT-PCR testing. When epidemiological information suggested that people might have been nodes of disease transmission but had recovered from illness, SARS-CoV-2 IgG serology testing was used to establish past infection.

Findings
Three clusters of COVID-19, comprising 28 locally transmitted cases, were identified in Singapore; these clusters were from two churches (Church A and Church B) and a family gathering. The clusters in Church A and Church B were linked by an individual from Church A (A2), who transmitted SARS-CoV-2 infection to the primary case from Church B (F1) at a family gathering they both attended on Jan 25, 2020. All cases were confirmed by RT-PCR testing because they had active disease, except for A2, who at the time of testing had recovered from their illness and tested negative. This individual was eventually diagnosed with past infection by serological testing. ELISA assays showed an optical density of more than 1·4 for SARS-CoV-2 nucleoprotein and receptor binding domain antigens in titres up to 1/400, and viral neutralisation was noted in titres up to 1/320.

Interpretation
Development and application of a serological assay has helped to establish connections between COVID-19 clusters in Singapore. Serological testing can have a crucial role in identifying convalescent cases or people with milder disease who might have been missed by other surveillance methods.

Funding
National Research Foundation (Singapore), National Natural Science Foundation (China), and National Medical Research Council (Singapore).

Full text

Since January 2020 Elsevier has created a COVID-19 resource centre with
free information in English and Mandarin on the novel coronavirus COVID-
19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.
Elsevier hereby grants permission to make all its COVID-19-related
research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.

www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
1
Articles
Connecting clusters of COVID-19: an epidemiological and
serological investigation
Sarah Ee Fang Yong*, Danielle Elizabeth Anderson*, Wycliffe E Wei, Junxiong Pang, Wan Ni Chia, Chee Wah Tan, Yee Leong Teoh,
Priyanka Rajendram, Matthias Paul Han Sim Toh, Cuiqin Poh, Valerie T J Koh, Joshua Lum, Nur-Afidah Md Suhaimi, Po Ying Chia,
Mark I-Cheng Chen, Shawn Vasoo, Benjamin Ong, Yee Sin Leo, Linfa Wang*, Vernon J M Lee*
Summary
Background Elucidation of the chain of disease transmission and identification of the source of coronavirus disease
2019 (COVID-19) infections are crucial for eective disease containment. We describe an epidemiological investigation
that, with use of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) serological assays, established links
between three clusters of COVID-19.
Methods In Singapore, active case-finding and contact tracing were undertaken for all COVID-19 cases. Diagnosis for
acute disease was confirmed with RT-PCR testing. When epidemiological information suggested that people might
have been nodes of disease transmission but had recovered from illness, SARS-CoV-2 IgG serology testing was used
to establish past infection.
Findings Three clusters of COVID-19, comprising 28 locally transmitted cases, were identified in Singapore; these
clusters were from two churches (Church A and Church B) and a family gathering. The clusters in Church A and
Church B were linked by an individual from Church A (A2), who transmitted SARS-CoV-2 infection to the primary
case from Church B (F1) at a family gathering they both attended on Jan 25, 2020. All cases were confirmed by
RT-PCR testing because they had active disease, except for A2, who at the time of testing had recovered from their
illness and tested negative. This individual was eventually diagnosed with past infection by serological testing. ELISA
assays showed an optical density of more than 1·4 for SARS-CoV-2 nucleoprotein and receptor binding domain
antigens in titres up to 1/400, and viral neutralisation was noted in titres up to 1/320.
Interpretation Development and application of a serological assay has helped to establish connections between
COVID-19 clusters in Singapore. Serological testing can have a crucial role in identifying convalescent cases or people
with milder disease who might have been missed by other surveillance methods.
Funding National Research Foundation (Singapore), National Natural Science Foundation (China), and National
Medical Research Council (Singapore).
Copyright © 2020 Elsevier Ltd. All rights reserved.
Introduction
As of April 15, 2020, more than 1·9 million cases of
coronavirus disease 2019 (COVID-19), and more than
120 000 deaths from the disease, have been recorded
worldwide.1 Many initial cases reported outside of China
were imported or were linked to travellers from China.2,3
However, as community transmission has become
widespread, the source of cases of COVID-19 in several
countries has not been established.
In Singapore, a globally connected city-state in southeast
Asia, health ocials have attempted to contain the spread
of COVID-19 through intensive epidemiological investi-
gations coupled with isolation of cases and quarantine of
close contacts. However, establishing the source of infec-
tion to ascertain the possible extent of spread can be
dicult, because scant epidemiological data might be
available. Even when possible nodes of transmission are
retrospectively identified through epidemiological investi-
gations, nucleic acid-based tests would not be diag-
nostically useful if these infected individuals have
recovered and no longer shed the virus. Hence, serological
tests are needed to identify convalescent cases and aid
investigations and containment eorts.
We present findings of investigations from Jan 29 to
Feb 24, 2020, that linked two people with COVID-19 from
Wuhan, China, to three clusters of COVID-19 cases in
Singapore. Serological testing had a crucial role in estab-
lishing a link between clusters, showing its use in
identifying convalescent COVID-19 cases and supporting
epidemiological investigations.
Methods
Surveillance methods and identification of cases
In Singapore, several surveillance methods are used to
identify people with COVID-19. On Jan 2, 2020, a suspect
case-definition of COVID-19 was circulated to all doctors
in Singapore;4 doctors are legally required to notify
the Ministry of Health of cases of COVID-19.5 From
Jan 31, 2020, Singapore began testing all patients
with pneumonia in hospital for severe acute respiratory
Lancet Infect Dis 2020
Published Online
April 21, 2020
https://doi.org/10.1016/
S1473-3099(20)30273-5
See Online/Comment
https://doi.org/10.1016/
S1473-3099(20)30322-4
*Contributed equally
Ministry of Health, Singapore
(S E F Yong MMed, W E Wei MPH,
P Rajendram MPH, C Poh MPH,
V T J Koh MPH, J Lum BA,
N-A M Suhaimi PhD, B Ong FRCP,
V J M Lee PhD); National
University Health System,
Singapore (S E F Yong, B Ong);
Programme in Emerging
Infectious Diseases,
Duke-National University
of Singapore Medical School,
Singapore (D E Anderson PhD,
W N Chia PhD, C W Tan PhD,
Prof L Wang PhD); National
Centre for Infectious Diseases,
Singapore (W E Wei,
M P H S Toh MMed,
P Y Chia MRCP, M I-C Chen PhD,
S Vasoo MRCP,
Prof Y S Leo FRCP); Centre for
Infectious Disease
Epidemiology and Research
(J Pang PhD), and
Saw Swee Hock School of
Public Health (J Pang,
Y L Teoh MMed, M P H S Toh,
M I-C Chen, V J M Lee), National
University of Singapore,
Singapore (B Ong); Singapore
Clinical Research Institute,
Singapore (Y L Teoh); Ministry
of Health Office for Healthcare
Transformation, Singapore
(Y L Teoh); and Tan Tock Seng
Hospital, Singapore (P Y Chia,
S Vasoo, Prof Y S Leo)
Correspondence to:
Dr Vernon J M Lee,
Communicable Diseases
Division, Ministry of Health,
Singapore 169854
vernon_lee@moh.gov.sg

Articles
2
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
syndrome coronavirus 2 (SARS-CoV-2); this testing was
later expanded to include people with pneumonia in
primary care. The diagnosis of COVID-19 is confirmed
either by a respiratory sample testing positive for
SARS-CoV-2 using a laboratory-based RT-PCR6 or by a
serum sample testing positive for SARS-CoV-2 on sero-
logical analysis.7
Once individuals with COVID-19 were identified, their
activities from 14 days before symptom onset until they
were isolated were mapped and their close contacts traced.
Contact tracing before symptom onset was done to identify
the source of exposure that led to the case being infected,
allowing for further active case-finding around the source.
Contact tracing after symptom onset until isolation was
done to identify exposed individuals for quarantine to
break the transmission chain. Both these approaches were
part of the containment strategy. A close contact was
defined as anyone who had prolonged contact within 2 m
of the case. All close contacts with active or recent
symptoms were tested, whereas those who were asymp-
tomatic and exposed while the case was symptomatic were
quarantined. Activity maps were reviewed and cross-
checked to establish potential expo sures and identify
possible epidemiological links between cases and clusters.
All epidemiological investigations and outbreak con-
tain ment measures were implemented under the
Infectious Diseases Act,5 which allows use of data for
analysis to control outbreaks.
Laboratory techniques
For laboratory confirmation of COVID-19, we did RT-PCR
testing for SARS-CoV-2, using previously published
methods.6 Two serological platforms were developed for
confirmation of specific antibody responses to SARS-CoV-2
in people with suspected infection or individuals with
PCR-confirmed disease. A virus neutralisation test (VNT)
was established at the Duke-National University of
Singapore Medical School ABSL3 facility using a
SARS-CoV-2 virus isolate (BetaCoV/Singapore/2/2020;
GISAID accession number EPI_ISL_407987) cultured
from a patient in Singapore; VNT was done using protocols
previously published for severe acute respiratory syndrome
coronavirus (SARS-CoV).7 For ELISA assays, we used
recombinant nucleocapsid protein from SARS-CoV and
SARS-CoV-2 expressed in mammalian cell culture using
the pcDNA3.1 vector (ThermoFisher Scientific, Carlsbad,
CA, USA), according to previously published methods,8
and a recombinant receptor binding domain (RBD) of
the SARS-CoV-2 spike protein custom-produced by a
commercial provider (GenScript, Piscataway, NJ, USA).
ELISA wells were coated with 100 ng of the respective
protein per well and serum samples were used at dilutions
from 1/50 to 1/400, followed by horseradish peroxidase-
conjugated goat anti-human IgG (Santa Cruz, Dallas, TX,
USA) used at a dilution of 1/2000.
Role of the funding source
The funders had no role in study design, data collection,
data analysis, data interpretation, or writing of the report.
The corresponding author had full access to all data
in the study and had final responsibility for the decision
to submit for publication.
Results
As of April 6, 2020, Singapore had recorded 1375 cases of
COVID-19, of which 554 were imported and 821 locally
transmitted. Three clusters were identified that involved
two churches (Church A and Church B) and one family
gathering. These clusters comprised 28 locally transmitted
cases. The clusters were linked to two travellers (W1 and
W2) from Wuhan, China, who attended a church service
at Church A on Jan 19, 2020 (figure 1). The clusters at
Research in context
Evidence before this study
We searched PubMed on March 3, 2020, for reports on
serological testing in individuals with coronavirus disease 2019
(COVID-19). We used the keywords (“COVID-19”, OR
“2019-nCoV”, OR “SARS-CoV-2”) AND (“serology” OR
“serologic testing”). Our search did not identify any reports
of the epidemiological application of serological tests in
COVID-19. In one report, researchers described serological
characteristics of COVID-19, and in other publications,
researchers have commented on the potential importance
of COVID-19 serological tests. In another study, epidemiological
investigations were reported of the epidemic in Singapore,
but serological methods had not been used.
Added value of this study
In our epidemiological investigation, we used RT-PCR and
serological testing to diagnose cases of COVID-19 and
establish links between clusters. RT-PCR testing alone
is limited by its ability to detect convalescent cases of
COVID-19, because RT-PCR can only detect severe acute
respiratory syndrome coronavirus 2 during the period of viral
shedding, which is the acute phase of infection. Serological
testing can be useful in detecting previous infection in people
with suspected infection who have recovered, assisting in
epidemiological investigation and containment efforts.
Implications of all the available evidence
COVID-19 laboratory testing is focused on use of quantitative
RT-PCR for diagnosis, and serological testing can be
overlooked. We have highlighted the importance of serological
testing for epidemiological investigation of COVID-19 cases,
and we urge further development of serological testing
capabilities.

Articles
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
3
Church A and Church B were linked by one individual
from Church A (A2), who probably transmitted the
infection to the primary case of the Church B cluster (F1)
at a family gathering on Jan 25, 2020. All cases were
confirmed by RT-PCR testing, except for A2, who was
diagnosed by serological testing.
The clusters at Church A and Church B were detected
in early February and mid-February, respectively.
Although W1 and W2 were diagnosed with COVID-19 at
the end of January, their possible link to Church A was
only discovered after the Church A cluster was identified,
through investigation and repeat interviews.5 By that
time, A2 had recovered from COVID-19 and was not
immediately linked to either cluster. Family members
who had been infected at the family gathering on
Jan 25, 2020, were first linked to F1 and were initially
regarded as part of the cluster at Church B. However,
subsequent investigations into case histories indicated
that the family cluster was a distinct cluster and that A2
was most probably the missing link between the two
church clusters; this idea was substantiated when A2’s
serological results were confirmed to be positive.
Five locally transmitted cases of COVID-19 (A1–A5)
were linked to Church A. These people attended a church
service on Jan 19, 2020, the same day W1 and W2 visited
the church. Although all five people had developed
symptoms by Feb 2, 2020 (figure 2), only A1, A4, and A5
were diagnosed (between Feb 6 and Feb 8, 2020), because
they had been hospitalised for pneumonia and tested for
SARS-CoV-2 as part of enhanced surveillance measures
to test all patients admitted to hospital with pneumonia.
A2 and A3 were not diagnosed when symptomatic in late
January because their symptoms were mild and they did
not meet the suspect case-definition at that time. A2 and
A3 were tested only after mapping of activities and
movements of other cases suggested that A2 could be the
missing link between the clusters at Church A and
Church B. An RT-PCR test of a nasopharyngeal specimen
taken from A3 on Feb 18, 2020, was positive, although
this individual had clinically recovered from the illness,
which persisted from Jan 28 to Feb 10, 2020. Serological
analysis of a serum sample obtained on the same day as
the nasopharyngeal specimen was also positive. Although
A2 had two negative RT-PCR tests, the serological result
was positive, indicating past infection.
A2 and A3 attended a Chinese New Year family
gathering on Jan 25, 2020, at the home of F1. Nine cases
(A2, A3, and F1–F7) were linked to this family gathering.
A2, whose symptoms started on Jan 23, 2020, was
unwell at this event and most probably was the primary
Figure 1: Transmission map of COVID-19
Map shows how COVID-19 was linked to two travellers from Wuhan, China, and two church clusters and a family gathering in Singapore. COVID-19=coronavirus
disease 2019.
Case
Primary case of a cluster
Church A service Jan 19, 2020
Wuhan travellers
Church A
Family gathering
Church B
Transmission
Possible transmission
Indirect link
Attended Church A service
W1
A1
A2
A3
A4
A5
A2
F1
F2
F3
F4
F5
F6
F7
F1
B2 B3
B6
B11
B13
B15
B5
B7
B10
B12
B8
B9
B16
B4
B14
W2
19
January February
Date
20 21 22 23 24 25 26 27 28 29 30 31 1234567891011121314151617
B1

Articles
4
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
Figure 2: Incubation period,
duration of symptoms, and
length of admission, from
Jan 14 to Feb 26, 2020
Data for 30 people with
coronavirus disease 2019 are
shown. Individuals are
labelled with cluster letter and
number, age (years) and sex
(M=male; F=female). Median
age of affected individuals
was 50·5 (IQR 25–79) years
and half the cohort (15 of 30)
were female.
Incubation period
Possible incubation period
Symptom onset until admission
Duration of admission
Length of current admission
Symptom onset until laboratory confirmation (not admitted)
Isolation date
Laboratory confirmation date
W1, 56F
Travellers from Wuhan
Church A cluster
Family gathering cluster
Church B cluster
W2, 56M
A1, 53M
A2, 58F
A3, 54M
A4, 39F
A5, 52F
F1, 28M
F2, 79F
F3, 27F
F4, 35F
F5, 38F
F6, 25M
F7, 30F
B1, 34M
B2, 46M
B3, 48M
B4, 55M
B5, 57M
B6, 51F
B7, 44F
B8, 56F
B9, 43M
B10, 54M
B11, 55F
B12, 54F
B13, 26M
B14, 57F
B15, 29M
B16 50M
14 19 24 29 3813 18 23 28
January February
Date

Articles
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
5
source of transmission. A3 developed symptoms later,
on Jan 28, 2020, and was, therefore, unlikely to be the
source of infection at this gathering.
17 locally transmitted cases were linked to Church B
(B1–B16 and F1). Thorough review of activity maps
ascertained that F1, who had developed symptoms on
Jan 29, 2020, and continued to work at Church B while
ill, was the primary case of the Church B cluster.
In mid-February, epidemiological and clinical evidence
strongly suggested that A2 was the missing link between
the clusters at Church A and Church B through attendance
at the family gathering on Jan 25, 2020, while symptomatic.
However, by the time this link was ascertained, more than
3 weeks had passed from symptom onset on Jan 23, 2020,
and symptoms had resolved completely a week previously,
on Feb 8–10, 2020. A2 had two negative RT-PCR test
results from samples taken from the nasopharynx, with
testing done 1 day apart on Feb 18 and Feb 19, 2020. The
sample for serological testing was taken on Feb 18, 2020,
and a positive result was confirmed on Feb 22, 2020.
ELISA results for A2 and A3 on Feb 20, 2020 (figure 3A)
showed a strong antibody response to the SARS-CoV-2
RBD protein, which was not seen in serum samples
from SARS-CoV patients. The results were further
confirmed by VNT on Feb 22, 2020 (figure 3B).
Discussion
This investigation shows how SARS-CoV-2 serological
analysis (ELISA detecting IgG and VNT detecting
neutralising antibodies), in addition to use of traditional
epidemiological methods, was important in establishing
links among locally transmitted COVID-19 cases and
tracing the transmission chain to an imported source.
Detection of COVID-19 can be dicult because of the non-
specific mild respiratory symptoms in many aected
individuals and because some people might recover
without being diagnosed.9 Although PCR tests oer a rapid
diagnostic solution, they can only detect SARS-CoV-2
during the period of viral shedding, which is the acute
phase of infection. The duration of viral shedding for
COVID-19 is not certain,10 but SARS-CoV data indicate that
21 days after symptom onset, 53% of cases achieved viral
clearance in nasopharyngeal aspirate samples.11 As such,
PCR testing alone is limited by its ability to detect
convalescent cases.
Serological testing can be especially useful in detecting
a previous suspected infection in people who have
recovered. For seroconversion kinetics, past coronavirus
studies indicate that all patients with Middle East
respiratory syndrome (MERS) seroconverted 3 weeks
after symptoms started,12 and 93% of patients with severe
acute respiratory syndrome (SARS) seroconverted at
an average of 20 days from symptom onset.11 The first
preliminary analysis of SARS-CoV-2 IgM and IgG
indicated that the antibody response in COVID-19
patients is similar to, if not earlier than, these times.13
Cross-reactivity of immunoglobulins to closely related
viruses such as SARS-CoV is a potential issue,14 but our
Figure 3: Serological testing of two patients
(A) ELISA testing using SARS-CoV-2 NP and RBD antigens. In addition to serum samples from the two suspected cases (A2 and A3), known positive samples for
COVID-19 and SARS, and negative samples, were included in the testing, and serial dilutions were done. (B) VNT results at the highest dilution that could effectively
neutralise the SARS-CoV-2 infection. COVID-19=coronavirus disease 2019. NP=nucleocapsid protein. RBD=receptor binding domain. SARS=severe acute respiratory
syndrome. SARS-CoV-2=severe acute respiratory syndrome coronavirus 2. VNT=virus neutralisation test.
Optical density
0
0·5
1
1·5
2
AB
SARS-CoV-2 NP
SARS-CoV-2 RBD
Optical density
0
1/50 1/100 1/200 1/400
SARSCOVID-19 Control A2 A3
0·5
1
1·5
2
Replicate
123
1/80 1/160 1/80
Negative
Negative
Negative
Negative
Negative
Negative
1/160 1/320 1/320
1/160 1/160 1/80
COVID-19
SARS
Control
A2
A3

Articles
6
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
RBD-based ELISA showed sucient dierentiation
power, a finding further supported by VNT results.
IgM serological testing might hold promise as
a diagnostic method, although for SARS and MERS, limi-
tations for its use have been noted.12,13 For MERS, IgM was
not detected earlier than IgG, and IgM against prevalent
human coronaviruses showed cross-reactivity.12 Pre-
liminary data for SARS-CoV-2 IgM are promising,13 but
more work is needed to assess the feasibility of IgM
serological analysis as a rapid diagnostic method to
enhance COVID-19 detection capabilities.
For most people in the three clusters we report here,
transmission of infection was accounted for by close
contact with a symptomatic case. Our findings suggest
that COVID-19 is largely transmitted by close contact,
particularly when contact occurs over a prolonged period
and in close congregation. Two churches were the
setting for COVID-19 transmission in our report.
A church cluster has also been reported in South Korea.15
Churches host prolonged repeated activities, during
which close contact occurs, thus providing the oppor-
tunity for disease spread through droplets or fomites.
Singing (a common practice in churches) can generate
droplets in a similar quantity to coughing.16,17 Repeated
social interactions of church groups has also facilitated
discovery of transmission, compared with other settings
in which people might not know each other. The inter-
actions are similar in nature to large family gatherings,
which other cases in our report were linked to. Other
similar settings include school and workplaces, in which
respiratory disease transmission is not uncommon and
should be the focus for prepared ness, surveillance, and
containment measures.
Risk of transmission could be reduced if symptomatic
people do not attend events in which prolonged social
interactions take place (eg, at the family gathering when
A2 was unwell, and F1 who continued to work at
Church B while unwell). Other risk reduction measures
could include having smaller group activities and
prevention of interactions across these groups. Moreover,
there may be common touchpoints within each setting
that could result in contact transmission. To prevent
transmission, people should practice increased personal
hygiene and reduce physical contact to minimise indirect
transmission risks.
Linking disease transmission to an imported source
and contact tracing for each identified case has facilitated
a high capture of cases in Singapore. This successful
linking of a large proportion of cases to imported sources
provides encouraging evidence that the intense contain-
ment measures undertaken in Singapore have been
eective. The three clusters we report here occurred
relatively early in the emergence of COVID-19 in
Singapore, within 4 weeks of the first imported case.
As the epidemic continues, it might be progressively
dicult to establish linkages by relying on traditional
epidemiological methods alone. Challenges include
diculty in obtaining information from cases and
contacts, which could be inaccurate because of recall and
other biases. In our investigation, information for
symptom onset of W1 and W2 was based on case history
alone and could not be independently corroborated.5
In such instances, determining the transmission chain
would have to account for possible inconsistencies in
case history and rely on triangulation with other methods.
The development and adoption of additional laboratory
techniques, such as serological tests and phylogenetic
analysis by whole-genome sequencing, could help
identify possible links between cases.
Serological testing is a key method in the response to
the COVID-19 epidemic. As shown in our study, it enabled
detection of a convalescent case, which could be key in
initial containment eorts to discover transmission links
to support containment eorts. Serological testing also
detects people with mild or asymptomatic disease who
have recovered, allowing for more accurate deter mination
of the number of people probably infected in a cluster or
the population. Identifying people who were probably
infected in household or school clusters could help
ascertain attack rates by age, particularly among children
who mostly manifest less severe disease.18 Calculating
population-level attack rates is also important to estimate
disease incidence and the case-fatality rate (CFR). Thus
far, the CFR for COVID-19 has been based on PCR-
diagnosed symptomatic cases. Serological surveys would
be important to estimate CFR more accurately and would
better inform calibrated responses to COVID-19. As the
pandemic progresses, monitoring seroprevalence would
enable countries to track transmission dynamics and
population immunity levels and to inform disease
control policies. Such monitoring would necessitate the
development of ready serological testing solutions that are
cost-eective and the establishment of population-level
surveillance prog rammes to obtain blood samples.
Development and application of serological assays
has helped to unravel connections between three
clusters of COVID-19 in Singapore, linking the disease
to two travellers from China. Serological assays should
be considered to identify mild or subclinical infections
in the community.
Contributors
VJML and YLT had the idea for the study and contributed to study design.
CP, VTJK, JL, N-AMS, and PYC contributed to the epidemiological
investigation and data collection. VJML, YLT, SEFY, PR, MPHST, JP, and
WEW contributed to analysis and interpretation of epidemiological data.
LW, DEA, WNC, and CWT developed the serological methods and
contributed to the analysis and interpretation. SEFY, WEW, and JP wrote
the report, and VJML, MI-CC, SV, BO, YSL, LW, and DEA contributed to
critical revision.
Declaration of interests
We declare no competing interests.
Acknowledgments
The development of serological tests was funded in part by a joint grant
under the National Research Foundation (Singapore) and the National
Natural Science Foundation of China (NRF2016NRFNSFC002-013).
The study was partly funded by the National Medical Research Council

Articles
www.thelancet.com/infection Published online April 21, 2020 https://doi.org/10.1016/S1473-3099(20)30273-5
7
(Singapore) COVID19 research fund (COVID19RF-001). We thank our
colleagues who contributed to the investigations, including the
COVID-19 epidemiology workgroup, contact tracing team, and hospital
epidemiology teams; and Viji Vijayan, Benson Ng, and Velraj Sivalingam,
(Duke-National University of Singapore ABSL3 facility) for logistics
management and assistance.
References
1 WHO. Coronavirus disease 2019 (COVID-19): situation report—86.
April 15, 2020. https://www.who.int/docs/default-source/
coronaviruse/situation-reports/20200415-sitrep-86-covid-19.
pdf?sfvrsn=c615ea20_6 (accessed April 16, 2020).
2 Hoehl S, Rabenau H, Berger A, et al. Evidence of SARS-CoV-2
infection in returning travelers from Wuhan, China.
N Engl J Med 2020; 382: 1278–80.
3 Rothe C, Schunk M, Sothmann P, et al. Transmission of 2019-nCoV
infection from an asymptomatic contact in Germany.
N Engl J Med 2020; 382: 970–71.
4 Ministry of Health Singapore. Circular no 04/2020: viral pneumonia
of unknown cause in Wuhan city, China. Singapore: Ministry of
Health, 2020.
5 Singapore Statutes Online. Infectious Diseases Act 2003.
July 31, 2003. https://sso.agc.gov.sg/Act/IDA1976 (accessed
April 1, 2020).
6 Pung R, Chiew CJ, Young BE, et al. Investigation of three clusters
of COVID-19 in Singapore: implications for surveillance and
response measures. Lancet 2020; 395: 1039–46.
7 Li W, Shi Z, Yu M, et al. Bats are natural reservoirs of SARS-like
coronaviruses. Science 2005; 310: 676–79.
8 Ge X-Y, Li J-L, Yang X-L, et al. Isolation and characterization
of a bat SARS-like coronavirus that uses the ACE2 receptor.
Nature 2013; 503: 535–38.
9 Novel Coronavirus Pneumonia Emergency Response Epidemiology
Team. The epidemiological characteristics of an outbreak of 2019
novel coronavirus diseases (COVID-19) in China.
Zhonghua Liu Xing Bing Xue Za Zhi 2020; 41: 145–51 (in Chinese).
10 Vetter P, Eckerle I, Kaiser L. Covid-19: a puzzle with many missing
pieces. BMJ 2020; 368: m627.
11 Peiris JSM, Chu CM, Cheng VCC, et al. Clinical progression and
viral load in a community outbreak of coronavirus-associated SARS
pneumonia: a prospective study. Lancet 2003; 361: 1767–72.
12 Corman VM, Albarrak AM, Omrani AS, et al. Viral shedding and
antibody response in 37 patients with Middle East respiratory
syndrome coronavirus infection. Clin Infect Dis 2015; 62: 477–83.
13 Zhang W, Du R-H, Li B, et al. Molecular and serological
investigation of 2019-nCoV infected patients: implication of
multiple shedding routes. Emerg Microbes Infect 2020; 9: 386–89.
14 Meyer B, Drosten C, Muller MA. Serological assays for emerging
coronaviruses: challenges and pitfalls. Virus Res 2014; 194: 175–83.
15 Korea Centers for Disease Control and Prevention. The updates
of COVID-19 in Korea. Feb 29, 2020. https://www.cdc.go.kr/board/
board.es?mid=a30402000000&bid=0030&act=view&list_
no=366406&tag=&nPage=7 (accessed April 1, 2020).
16 Loudon RG, Roberts RM. Singing and the dissemination of
tuberculosis. Am Rev Respir Dis 1968; 98: 297–300.
17 Asadi S, Wexler AS, Cappa CD, Barreda S, Bouvier NM,
Ristenpart WD. Aerosol emission and superemission during
human speech increase with voice loudness. Sci Rep 2019; 9: 2348.
18 Qiu H, Wu J, Hong L, Luo Y, Song Q, Chen D. Clinical and
epidemiological features of 36 children with coronavirus disease
2019 (COVID-19) in Zhejiang, China: an observational cohort study.
Lancet Infect Dis 2020; published online March 25.
https://doi.org/10.1016/S1473-3099(20)30198-5.