Pre-Symptomatic Transmission of Sars-Cov-2 Infection: A Secondary Analysis Using Published Data

medRxiv preprint doi: https://doi.org/10.1101/2020.05.08.20094870.this version posted June 11, 2020.
The copyright holder for this preprint

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-ND 4.0 International license .
Pre-symptomatic transmission of SARS-CoV-2

infection: a secondary analysis using published data
Miriam Casey1, John Griffin1, Conor G. McAloon2, Andrew W. Byrne3, Jamie M Madden1,
David Mc Evoy4, Áine B. Collins1,5, Kevin Hunt6, Ann Barber1, Francis Butler6, Elizabeth A.
Lane1,5, Kirsty O’Brien7, Patrick Wall4, Kieran A. Walsh7, Simon J. More1
1. Centre for Veterinary Epidemiology and Risk Analysis, UCD School of Veterinary
Medicine, University College Dublin, Belfield, Dublin D04 W6F6, Ireland
2. Section of Herd Health and Animal Husbandry, UCD School of Veterinary Medicine,
University College Dublin, Dublin D04 W6F6, Ireland
3. One-Health Scientific Support Unit, Department of Agriculture, Food and the Marine
(DAFM), Kildare Street, Dublin 2, Ireland.
4. School of Public Health, Physiotherapy and Sports Science, Woodview House
University College Dublin, Belfield, Dublin, Ireland
5. Department of Agriculture, Food and the Marine, Kildare St., Dublin 2, Ireland
6. Centre for Food Safety, UCD School of Biosystems and Food Engineering,
University College Dublin, Belfield, Dublin D04 W6F6, Ireland
7. Health Information and Quality Authority (HIQA), Unit 1301, City Gate, Cork,
Ireland.
Correspondence to: Miriam Casey; miriam.casey@ucd.ie, Centre for Veterinary

Epidemiology and Risk Analysis, UCD School of Veterinary Medicine, University College
Dublin, Belfield, Dublin D04 W6F6, Ireland
Key words: “COVID-19”; ‘SARS-CoV-2”, “Pre-symptomatic”; “Transmission”
1
medRxiv preprint doi: https://doi.org/10.1101/2020.05.08.20094870.this version posted June 11, 2020. The copyright holder for this preprint
Abstract
Objective: To estimate the proportion of pre-symptomatic transmission of SARS-CoV-2
infection that can occur and timing of transmission relative to symptom onset.
Setting/design: Secondary analysis of international published data.
Data sources: Meta-analysis of COVID-19 incubation period and a rapid systematic review
of serial interval and generation time, which are published separately.
Participants: Studies were selected for analysis if they had transparent methods and data
sources and they provided enough information to simulate full distributions of serial interval
or generation time. Twenty-three estimates of serial interval and five of generation time from
17 publications were included.
Methods: Simulations were generated of incubation period and of serial interval or
generation time. From these, transmission times relative to symptom onset were calculated
and the proportion of pre-symptomatic transmission was estimated.
Outcome measures: Transmission time of SARS-CoV-2 relative to symptom onset and
proportion of pre-symptomatic transmission.
Results: Transmission time ranged from a mean of 2.91 (95% CI: 3.18-2.64) days before
symptom onset to 1.20 (0.86-1.55) days after symptom onset. Unweighted pooling of
estimates of transmission time based on serial interval resulted in a mean of 0.60 days before
symptom onset (3.01 days before to 1.81 days after). Proportion of pre-symptomatic
transmission ranged from 42.8% (39.8%-45.9%) to 80.6% (78.1%-83.0%). The proportion of
pre-symptomatic transmission from pooled estimates was 56.4% (34.9%-78.0%).
Conclusions: Whilst contact rates between symptomatic infectious and susceptible people
are likely to influence the proportion of pre-symptomatic transmission, there is substantial
potential for pre-symptomatic transmission of SARS-CoV-2 in a range of different contexts.
Our work suggests that transmission is most likely in the day before symptom onset whereas
estimates suggesting most pre-symptomatic transmission highlighted mean transmission
times almost three days before symptom onset. This highlights the need for rapid case
detection, contact tracing and quarantine.
2
Strengths and weaknesses of this study

• We estimate the extent and variation of pre-symptomatic transmission of SARS-CoV-
2 infection across a range of contexts. This provides important information for
development and targeting of control policies and for the parameterisation of
transmission models.
• This is a secondary analysis using simulations based on published data, some of
which is in pre-print form and not yet peer-reviewed. There is overlap in the contact
tracing data that informed some of our source publications. We partially address this
by summarising data at source location level as well as at study level.
• Populations where symptomatic people are rapidly isolated are likely have relatively
more pre-symptomatic transmission. This should be borne in mind whilst interpreting
our results, but does not affect our finding that there is substantial potential for pre-
symptomatic transmission of SARS-CoV-2 infection.
• A strength of our approach is that it builds an understanding of pre-symptomatic
transmission from a range of estimates in the literature, facilitates discussion for the
drivers of variation between them, and highlights the consistent message that
consideration of pre-symptomatic transmission is critical for COVID-19 control
policy.
3
Introduction
There is currently a pandemic of coronavirus disease (COVID-19), a recently emerged and
rapidly spreading infectious disease that is caused by the novel coronavirus, severe acute
respiratory syndrome coronavirus 2 (SARS-CoV-2). There are large direct impacts of
COVID-19 amongst known cases. As of 2nd of June 2020, the World Health Organization has
reported 6, 194, 533 confirmed cases and 376, 320 deaths due to COVID-19 [1]. In China,
14% and 5% of cases were classified as severe and critical, respectively [2], and a report
from Italy showed that 18% of cases required intensive care [3]. There are also major
indirect impacts of COVID-19 and its control measures on other aspects of health care [4–6]
and on the economy [7,8].
As there is currently no COVID-19 vaccine ready for widespread use, primary control
measures entail reducing transmission from infectious individuals. These include case
isolation, contact tracing and quarantine, physical distancing and hygiene measures [9].
Infectious people are predominantly identified by reported symptoms of COVID-19.
In absence of active surveillance, infectious people without symptoms may not be

quarantined, and therefore may have more contacts with susceptible people resulting in
increased COVID-19 transmission. Therefore, quantifying the transmission potential of
COVID-19 before or in the absence of symptoms will inform disease control measures and
predictions of epidemic progression.
Characteristics of pre-symptomatic and asymptomatic transmission are potentially different,

and separate approaches may be required to understand them. We aimed to capitalise upon
the considerable information about pre-symptomatic transmission that can be inferred from
contact tracing studies. Therefore, we focus here on transmission from people before they
develop symptoms rather than that from people who never develop symptoms. This addresses
the urgent need for more data on extent of pre-symptomatic transmission which has been
highlighted by those developing models to inform policies [10].
The pre-symptomatic transmission potential of COVID-19 has been highlighted by case

reports [11–20]. The potential for pre-symptomatic transmission was also suggested by
detection of viral genome in upper respiratory samples prior to symptoms [21–23].
4
Live virus has been isolated very soon after symptom onset [24]. These findings are
supported by quantitative studies based on contact tracing, reporting serial intervals or
generation times similar in duration or shorter than incubation periods in some situations [25–
28], and even evidence of symptoms manifesting in the infectee prior to the infector in some
cases [28–31]. Several studies have quantified the proportion [25,27–29,32] and timing
[25,28,29,32] of pre-symptomatic transmission, using a variety of datasets and
methodologies.
Here, using secondary analysis of data collated in meta-analysis [33] and a rapid systematic
review [34] that are published separately, we apply a standardised methodology to estimate
the proportion and timing of pre-symptomatic transmission of COVID-19 in a range of
different contexts.
Methods
Principles of methodology
If generation time, the duration in days between time of infection of a secondary case
(infectee) and that of its primary case (infector), is longer than incubation period, the time
between infection and symptom onset in the infector, transmission will have occurred after
symptom onset (Scenario A in Figure 1). If generation time is shorter than incubation period,
pre-symptomatic transmission will have occurred (Scenarios B and C in Figure 1). If an
infector and infectee incubation periods are taken to be independent and identically
distributed, serial interval, the time between infector and infectee symptom onset, can be
taken as an approximation of generation time [35,36], although serial interval will have more
variation [26]. Table 1 contains definitions relevant to our analysis.
Meta-analysis and rapid systematic review

Data about incubation period, serial interval and generation time were sourced through on our
separately published metanalysis of incubation period [33] and rapid systematic review of
serial interval and generation time [34]. As described fully elsewhere [33,34], literature
searches covered publication dates between the 1st of December 2019 and 8th of April 2020
for incubation period, extending to the 27th of April 2020 for generation time / serial interval.
A dedicated team searched for publications on the electronic databases PubMed [37], Google
5
Scholar [38], MedRxiv [39] and BioRxiv [40] with the following keywords: “Novel
coronavirus” OR “SARSCoV2” OR “2019-nCoV” OR “COVID-19” AND “serial
interval” OR “latent period” OR “incubation period” OR “generation time” OR
“infectiousness” OR “pre-symptomatic” OR “asymptomatic”). The dynamic curated PubMed
database “LitCovid” [41,42] was also monitored. In addition, publicly available reports from
the World Health Organization [1,9,43], European Centre for Disease Prevention and Control
[44] and Centres for Disease Control and Prevention Morbidity and Mortality Weekly
Reports [45] were monitored, as well as curated summaries on relevant topics from the
American Association for the Advancement of Science [46] and the Nature Journal [47].
Both our meta-analysis [33] and rapid systematic review [34] completed checklists to show
fulfilment of Preferred Reporting Items for Systematic reviews and Meta-Analyses -
extension for Scoping Reviews (PRISMA-ScR) [48].
Based on the estimates reported by our meta-analysis [33] and rapid systematic review [34],
we simulated data for incubation period, serial interval and generation time. We subtracted
incubation period from serial interval or generation time to infer transmission time relative to
onset of symptoms and to estimate the proportion of pre-symptomatic transmission.
Data extraction
All analyses were conducted in the R statistical environment [49]. For each publication
included in our analyses, parameters describing the distributions of generation time or serial
interval, and the location and dates of collection of the contact tracing data from which they
were generated were collated.
If not reported directly, gamma shape and rate parameters were calculated from mean and
standard deviation, either directly or using the “epitrix” [50] package as follows.

ଶ

ଶ

ଶ
6
If lognormal distribution parameters (Meanlog and SDlog) were not directly reported, they
were estimated from the reported mean and standard deviation (SD) as follows:

ln 1

ଶ
ଶ

ln

2
Weibull parameters were estimated from the reported mean and standard deviation using the
“ “mixdist” [51] R package.
The incubation period that we used from the meta-analysis [33] had a lognormal distribution
(Mean = 5.8 (95% CI (5.01, 6.69 days). It, Median = 5.1 (4.5, 5.8) days, , SD = 3 (,2.68-
3.34), Meanlog = 1.63, (1.51, 1.75) and SDlog = 0.5 (0.45, 0.55) respectively.).
As serial interval has more variation than generation time [26], we considered them
separately for plotting and summary purposes. For the two studies [26,27] that estimated
generation time, we also generated serial interval simulations to allow a more direct
comparison with the other estimates based on serial interval. One of these, [26], related serial
interval to generation time with the following approach: Serial interval of an infectee can be
expressed as generation time of the infectee plus the difference between the incubation
periods of the infectee and the infector (Figure 1). That is, the incubation period used for the
generation time estimation was simulated twice to generate two samples (“inc 1” and “inc2”).
The extra variation in serial interval compared to generation time was then simulated by:
! 1 2
We repeated this estimation to simulate the serial interval for study [26] and cross-checked
the simulation against the summary statistics were reported in that publication. We then
estimated a serial interval from the generation time of study [27], using the same
methodology and the same incubation period as was used to infer generation time in study
[27].
7
One further study [52] did not directly supply enough information to simulate a distribution,
but supplied their code and data from which they estimated the serial interval. For this study,
we fitted distributions to their data by maximum likelihood estimation with the “fitdistrplus”
package [53] and chose one based on AIC values and visual cross-checking by plotting.
Simulation A
Simulation A was conducted to take into account the uncertainty around the reported
parameter estimates for incubation period, generation time and serial interval. Where we
could extract enough information from the publications, we sampled from distributions
capturing the 95% confidence intervals around the mean and standard deviation for each
reported distribution (n=1000). We converted these samples to the relevant parameters for the
distribution (e.g. shape and rate for a gamma distribution) and simulated distributions using
these parameters (n= 1000). The incubation period sample was subtracted from each
generation time or serial interval sample to give a resultant distribution indicating
transmission time relative to onset of symptoms. The resultant 1,000,000 samples were
resampled with replacement (n=1000 samples from each of 10,000 repeats) and 95%
confidence intervals from bootstrapping were calculated.
As there are known drivers of variation of generation time and serial interval [34], and
therefore of transmission time relative to symptom onset estimated based upon them,
estimates were plotted and summarised individually to show this variation. Estimates were
also grouped, plotted and summarised at the level of source location for the contact tracing
data that they were inferred from. Simple unweighted pooled estimates of transmission time
relative to symptom onset based on serial interval and generation time were presented for
interpretation in the context of the variation between the individual results.
Simulation B
Simulation B incorporated data from every study that reported enough information to
simulate a full distribution for serial interval or generation time but did not take uncertainty
around the parameter estimates into account (as this was not reported in all of the studies).
We simulated data (n = 100,000 samples) for incubation period and generation time or serial
interval from each study. The distributions were plotted, summarised and cross-checked
against the summary statistics and plots reported in the papers. The incubation period sample
8
was subtracted from the generation time or serial interval sample to give a resultant
distribution indicating transmission time relative to onset of symptoms. This was plotted and
summarised and the proportion of samples of transmission times relative to symptom onset
that were negative (indicating transmission before symptoms was reported) was calculated.
This simulation was repeated 20,000 times to explore the uncertainty from within the
simulation. As with Simulation A, summaries at estimate, source location and pooled level
were reported. Unlike Simulation A, confidence intervals were based only on the variation
from within the simulation.
4. Results
Included studies
Of the 19 studies reporting serial interval or generation time included in our rapid systematic
review [34], 17 were included in this study. We excluded the study [54] as it defined the start
of the exposure window for the infectee as the time of symptom onset in the infector,
excluding the possibility of transmission before symptom onset. We excluded one further
study, [20], pending clarification from the authors, as we could not replicate the distribution
described for serial interval. Table 2 lists the estimates considered for inclusion. Figure 2
describes the data available for 28 estimates for serial interval or generation time from the 17
publications that were included in this study.
Simulation A results
Studies that reported 95% confidence intervals for both the mean and standard deviation of
the generation time or serial interval could be incorporated into Simulation A (Table 2).
These included four generation time estimates and nine serial interval estimates from eight
different studies. Four of these estimates came from mixed locations, two each came from
Hong Kong, Mainland China excluding Hubei, Singapore and Tianjin, and one came from
Shenzhen. The uncertainty captured by the simulation was slightly less than that reported in
the source publications for serial interval and generation time. (Supplementary Table 1).
Table 3 summarises transmission time relative to symptom onset for each of the 13 estimates
included in Simulation A. These ranged from a mean of 2.91 (95% CI: 3.18, 2.64) days
before symptom onset to 1.20 (95% CI: 0.86, 1.55) days after symptom onset. Simple
9
unweighted pooling of estimates of transmission time relative to symptom onset based on

serial interval estimates resulted in a mean of 0.60 days before symptom onset (95% CI: 3.01
days before, 1.81 days after). Pooling based on generation time estimates gave a mean of
1.83 (95% CI: 3.48, 0.17) days before symptom onset.
Proportion of pre-symptomatic transmission calculated from the 13 individual estimates

ranged from 42.8% (95% CI:39.8%, 45.9%) to 80.6% (78.1%, 83.0%) (Table 3).
From unweighted pooling of serial interval and generation time estimates respectively, the
proportion of pre-symptomatic transmission was estimated to be 56.4% (95% CI: 34.9%,
78.0%) and 68.3% (95% CI: 47.9%, 88.6%) (Table 3).
Simple unweighted pooling at source location level resulted in mean transmission time
relative to symptoms ranging from 2.38 (95% CI: 3.45, 1.30) days before symptom onset in
Tianjin to a mean of 0.51 (95% CI: 0.18, 0.84) days after symptom onset in Shenzhen.
Proportion of pre-symptomatic transmission ranged from 48.6% (95% CI: 45.5%, 51.7%) in
Shenzhen to 74.9% (95% CI: 63.5%, 86.3%) in Tianjin (Table 4).
Further details of standard deviation, 2.5th, 25th, 50th, 75th and 97.5th quantiles of the
distributions described above, and their 95% confidence intervals are shown in Tables 3 and
4.
Simulation B results
A total of 28 estimates from 17 studies were included for Simulation B. Transmission time
relative to symptom onset estimates were based on five estimates of generation times and 23
estimates of serial interval. Several studies generated more than one estimate. This was due
to separate estimates for different locations [25,26], different models used to infer generation
time [26], sub-setting of data depending on confidence in transmission pair identification and
exposure windows [52,55], and estimation of both generation times and serial intervals from
the same papers [26,27]. Of the two models used in [26] , one only allowed positive serial
intervals to be inferred for missing data whereas a second model allowed negative serial
intervals for missing data.
10
Table 5 shows the counts of estimates and studies that came from specific locations or mixed
sources under nine different data source categories (Mixed sources, Tianjin, Singapore,
Mainland China excluding Hubei, Hong Kong, northern Italy, pooled data from Hong Kong
and Shenzhen, and Wuhan).
Results for Simulation B, incorporating a relatively broader range of studies, were similar to
those for Simulation A. Figure 3 and Table 6 show the variation in transmission time relative
to symptom onset amongst the 28 estimates ranging from a mean (median) of 2.92 (2.91)
days before symptom onset to 1.72 (1.69) days after symptom onset. Proportion of pre-
symptomatic transmission associated with the 28 different estimates ranged from 33.5% to
80.7%.
Table 6 also shows the summary statistics for a simple unweighted pooling of the estimates
based on serial interval and the five estimates based on generation time. The mean (median)
time of transmission for the pooled samples based on serial interval was 0.66 (0.62) days
before symptoms and pre-symptomatic transmission was estimated to be 56.0%. The
pooling of five generation time based estimates for transmission time relative to symptom
onset gave a mean (median) estimate of 1.60 (1.31) days before symptom onset and 65.5%
pre-symptomatic transmission.
Figure 4 and Table 7 show the variation in transmission relative to symptom onset amongst
estimates based on the nine different source categories ranging from a mean (median) of 2.05
(1.90) days before symptom onset for Hong Kong to 1.72 (1.68) days after symptom onset
for Wuhan. Proportion of pre-symptomatic transmission amongst the different data source
categories ranged from 33.5% in Wuhan to 72.7% in Hong Kong.
For simulation B, the uncertainty captured by 20,000 repeat simulations was small (Table 6),
as would be expected for large samples in simulations from defined distributions.
5. Discussion
Our results show that transmission time ranged from mean of 2.92 days before symptom
onset to 1.72 days after symptom onset. Simple unweighted pooling of the 23 estimates based
11
on serial intervals resulted in a mean time of transmission of 0.66 days before symptoms.
From this, it can be inferred that transmission of SARS-CoV-2 is most likely in the day
before symptom onset. The estimates suggesting most pre-symptomatic transmission
highlighted a mean transmission times almost 3 days before symptom onset. This is
consistent with other estimates in the literature, ranging from 2.65 days before symptoms to
2.3 days after symptoms [25–29,32]. A study focussed on inferring infectiousness profile
from 77 transmission pairs, reported that infectiousness started from 2.3 days before
symptom onset and peaked at 0.7 days before symptom onset [28]. Our study, analysing all
estimates from a variety of locations has produced a similar estimate (mean from pooled
estimate of 0.67 days before symptom onset).
We estimated that the proportion of pre-symptomatic transmission ranged from 33.5% to

80.7% depending on the study analysed. Pooled serial interval based estimates suggested
56.0% pre-symptomatic transmission. Our source level analyses describe a trend in pre-
symptomatic transmission ranging from the greatest proportion in Hong Kong (72.7%) to the
smallest proportion in Wuhan (33.5%). The range in the proportion of pre-symptomatic
transmission that we report is consistent with other studies [25,26,28,29,32] where reports of
pre-symptomatic transmission range from 37% to >80%.
The wide variation in estimates of the proportion of pre-symptomatic transmission and

transmission time relative to symptom onset is expected, as the observed transmission events
depend on contact rates between infectious and susceptible people as well as the natural
course of infectiousness. The variation in observed transmission events is manifested by
variations in serial interval and generation time, and therefore in transmission relative to
symptom onset. Griffin et al. [34] discuss drivers of this variation and the limitations in
different approaches to estimating generation time and serial interval from transmission pairs.
If people are quarantined once their symptoms become apparent, a greater proportion
transmission will be pre-symptomatic. Whilst data relating to the early stages of the COVID-
19 outbreak in Wuhan amount to only six pairs, we see a trend towards lower proportions of
pre-symptomatic transmission in this context, as well as in the early stages of the outbreaks in
Italy, possibly corresponding to relatively more transmission from symptomatic people in the
early stages of disease incursion. Three other studies also highlight this contrast in the
proportion of pre-symptomatic transmission in a different context. Zhao et al. [52] show that
serial interval became shorter in Hong Kong and Shenzhen as time elapsed from initial cases,
12
and suggest that this is due to increasing effectiveness of quarantining people with symptoms
over time. Zhang [32] contrasts a mean transmission time of 2.3 days after symptoms in the
early stages of the Wuhan outbreak (with relatively fewer quarantine measures) to a mean
transmission time of 2.4 days before symptom onset amongst imported cases outside Wuhan.
A study in Shenzhen [64] highlighted the valuable information that can be inferred from
knowledge of when transmission occurred and when the infector was isolated. Serial interval
was much shorter when the infectors were isolated within two days of symptom onset
compared to those isolated 3-5 days after symptom onset. However, lack of isolation beyond
five days made little difference, suggesting that there may be relatively more potential for
transmission before compared to after five days of symptom onset. A further analysis [56]
highlights the difference in pre-symptomatic transmission in Shenzhen between a subset with
accelerated case isolation (46% pre-symptomatic transmission) one without (23%).
Despite this understandable variation in estimates from different contexts, our work
consistently suggests the potential for pre-symptomatic transmission. A person presenting
with COVID-19 symptoms has potentially been infectious to others for several days.
Therefore, in absence of severe social distancing measures, extremely effective and rapid
contact tracing and quarantine will be required to control the spread of COVID-19.
This potential of the COVID-19 spread to be too fast to be controlled by conventional contact
tracing been highlighted with Ferretti et al.’s model [27]. This model suggests that pre-
symptomatic transmission alone can account for a basic reproductive number of 0.9 (47% of
the overall reproductive basic number), almost enough to sustain an epidemic on its own.
However, this estimate may be influenced by the low level of asymptomatic infectiousness
(10% relative to a symptomatic case) assumed by that model. This uncertainty highlights the
need for transmission from asymptomatically infected people to be more fully understood,
and to be considered as having potentially distinct characteristics compared to the pre-
symptomatic transmission that we report on in this paper.
We used a straightforward approach of simulating large numbers of samples from both the
incubation period distribution from our meta-analysis [33] and the distributions of 28 serial
interval or generation time estimates from our rapid systematic review [34], and subtracting
the incubation period samples from the serial interval or generation time samples to give a
resultant distribution of transmission time relative to symptom onset. This methodology is
13
similar to that applied by Tindale et al. and Ganyani et al. [25,26] to data from Singapore and
Tianjin. An alternative method, using maximum likelihood estimation, was used by He et al.
[28] and Zhang [32], and a Bayesian approach to estimate the probability of pre-symptomatic
transmission was used by Ferretti et al. [27]. Ma et al. [29] estimated transmission time
directly from manual interrogation of the data. Despite the variety of methodologies used to
estimate transmission time, results of our analysis gave similar ranges to the six other studies
that investigated pre-symptomatic transmission in different contexts [25–29,32].
Case reports [11–20] and virological studies [21–23] support the occurrence of pre-
symptomatic transmission suggested by the quantitative approaches reported here and
elsewhere [25–29,32]. Whilst samples testing positive by polymerase chain reaction (PCR)
do not always fully correlate with infectiousness [24,57,58], relatively lower cycle threshold
(CT) values suggest higher virus loads. Two studies [21,23] that reported pre-symptomatic
PCR CT values included some relatively low values. A report of pre-symptomatic PCR
positive samples in ten nursing home residents reported a mean time of 3 days from sampling
to onset of symptoms. In addition, the isolation of live virus from upper respiratory samples
very soon after patient presentation with symptoms has been reported [24].
Limitations and strengths

Table 4 shows the potential overlap in contact tracing data that the studies we analysed are
based on. Therefore, a limitation of our study is that all our data sources cannot be considered
completely independent. We partially address this by grouping estimates by the source
location of the contact-tracing data upon which they were based. Another challenge was to
fully capture the uncertainty in estimates with a simulation study. Where enough information
was available from the source publications, we incorporated measures of uncertainty into
Simulation A, and reported a comparison of the uncertainty suggested in our simulation to
that in the original estimates. A strength of our approach is that it builds a picture of pre-
symptomatic transmission from a range of estimates in the literature, facilitates discussion for
the drivers of variation between them, and highlights the consistent message that
consideration of pre-symptomatic transmission is critical for COVID-19 control policy. The
important insights into COVID-19 transmission gleaned from the studies that contributed to
our analyses, that used publicly available transmission pair data, highlights the immense
value of allowing public access to anonymised transmission pair data.
14
Conclusion
Although contact rates between symptomatic infectious and susceptible people are likely to
influence the proportion of pre-symptomatic transmission, our study highlights substantial
potential for pre-symptomatic transmission of COVID-19 in a range of different contexts.
Our work suggests that transmission of SARS-CoV-2 is most likely in the day before
symptom onset, whereas estimates suggesting most pre-symptomatic transmission
highlighted a mean transmission times almost 3 days before symptom onset. These findings
highlight the urgent need for extremely rapid and effective case detection, contact tracing and
quarantine measures if strict social distancing measures are to be eased.
Funding
All investigators are full-time employees (or retired former employees) of University College
Dublin, the Irish Department of Agriculture, Food and the Marine (DAFM), or the Irish
Health information and Quality Authority (HIQA). No additional funding was obtained for
this research.
Competing interests
All authors have completed the ICMJE 508 uniform disclosure form at
www.icmje.org/coi_disclosure.pdf and declare: no support from any organisation for the
submitted work; no financial relationships with any organisations that might have an interest
in the submitted work in the previous three years; no other relationships or activities that
could appear to have influenced the submitted work
Patient and public involvement statement

It was not appropriate or possible to involve patients or the public in the design, or conduct,
or reporting, or dissemination plans of our research
Author contributions
MC conceptualized the study, extracted parameter definitions from the literature, performed
the analyses and drafted the manuscript. JG led the rapid systematic review upon which the
generation time and serial interval simulations are based. CM led the meta-analysis upon
which the incubation period simulations are based upon. AC, KH, KW and KOB performed
systematic literature searches upon which the incubation period, generation time, serial
interval and pre-symptomatic transmission information reported here are based upon. SM
15
conceptualized, initiated and managed the overall project. All authors supplemented the
literature review and assessed the literature. All authors contributed to the manuscript and
reviewed it.
Data sharing statement

The data and code used for the analyses described in this paper are available in the Github
repository: https://github.com/miriamcasey/covid-19_presymptomatic_project
Word count
3888 words
Acknowledgements:
Dr. Jamie Tratalos and Mr. Guy McGrath of the Centre for Veterinary Epidemiology and
Risk Analysis gave valuable feedback on this manuscript.
References:
1 WHO. COVID-19 situation reports.
2020.https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-
reports (accessed 2 Jun 2020).
2 Wu Z, McGoogan JM. Characteristics of and Important Lessons From the Coronavirus
Disease 2019 (COVID-19) Outbreak in China. JAMA Published Online First: 24
February 2020. doi:10.1001/jama.2020.2648
3 Cereda D, Tirani M, Rovida F, et al. The early phase of the COVID-19 outbreak in
Lombardy, Italy. Published Online First: 2020.http://arxiv.org/abs/2003.09320
4 Roberts L. Polio, measles, other diseases set to surge as COVID-19 forces suspension
of vaccination campaigns | Science | AAAS. Sci. Mag. Newsl.
2020.https://www.sciencemag.org/news/2020/04/polio-measles-other-diseases-set-
surge-covid-19-forces-suspension-vaccination-campaigns (accessed 4 May 2020).
5 Saini KS, de Las Heras B, de Castro J, et al. Effect of the COVID-19 pandemic on
cancer treatment and research. Lancet Haematol 2020;0. doi:10.1016/S2352-
3026(20)30123-X
6 Kawohl W, Nordt C. COVID-19, unemployment, and suicide. The Lancet Psychiatry
2020;7:389–90. doi:10.1016/S2215-0366(20)30141-3
7 Fernandes N. Economic effects of coronavirus outbreak ( COVID-19 ) on the world
16
economy. SSRN Electron J 2020;:0–

29.https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3557504
8 McKibbin WJ, Fernando R. The Global Macroeconomic Impacts of COVID-19: Seven
Scenarios. SSRN Electron J Published Online First: 2020. doi:10.2139/ssrn.3547729
9 WHO. COVID-19 Strategy Update. 2020. https://www.who.int/publications-
detail/covid-19-strategy-update---14-april-2020
10 Peak CM, Kahn R, Grad YH, et al. Individual quarantine versus active monitoring of
contacts for the mitigation of COVID-19: a modelling study. Lancet Infect Dis 2020;0.
doi:10.1016/S1473-3099(20)30361-3
11 Huang R, Xia J, Chen Y, et al. A family cluster of SARS-CoV-2 infection involving
11 patients in Nanjing, China. Lancet Infect Dis 2020;0. doi:10.1016/S1473-
3099(20)30147-X
12 Wei WE, Li Z, Chiew CJ, et al. Presymptomatic Transmission of SARS-CoV-2 —
Singapore, January 23–March 16, 2020. Cent. Dis. Control Morb. Mortal. Wkly. Rep.
2020;69:411–5. doi:10.15585/mmwr.mm6914e1
13 Tong Z-D, Tang A, Li K-F, et al. Potential Presymptomatic Transmission of SARS-
CoV-2, Zhejiang Province, China, 2020. Emerg Infect Dis 2020;26:1052–4.
doi:10.3201/eid2605.200198
14 Qian G, Yang N, Ma AHY, et al. COVID-19 Transmission Within a Family Cluster by
Presymptomatic Carriers in China. Clin Infect Dis Published Online First: 23 March
2020. doi:10.1093/cid/ciaa316
15 Liu Y-C, Liao C-H, Chang C-F, et al. A Locally Transmitted Case of SARS-CoV-2
Infection in Taiwan. N Engl J Med 2020;382:1070–2. doi:10.1056/NEJMc2001573
16 Yu P, Zhu J, Zhang Z, et al. A familial cluster of infection associated with the 2019
novel coronavirus indicating potential person-to-person transmission during the
incubation period. J Infect Dis 2020;:1–5. doi:10.1093/infdis/jiaa077
17 Rothe C, Schunk M, Sothmann P, et al. Transmission of 2019-nCoV Infection from an
Asymptomatic Contact in Germany. N Engl J Med 2020;382:2019–20.
doi:10.1056/nejmc2001468
18 Zhang S, Na J, Zhou W, et al. A familial cluster of COVID-19 indicating virus can be
transmitted by asymptomatic carriers. Bull World Heal Organ Published Online First:
2020. doi:10.2471/BLT.20.251561
19 Lavezzo E, Franchin E, Ciavarella C, et al. Suppression of COVID-19 outbreak in the
municipality of Vo, Italy. medRxiv 2020;:2020.04.17.20053157.
17
doi:10.1101/2020.04.17.20053157
20 Liao J, Fan S, Chen J, et al. Epidemiological and clinical characteristics of COVID-19
in adolescents and young adults. medRxiv 2020;:2020.03.10.20032136.
doi:10.1101/2020.03.10.20032136
21 Kimball A, Hatfield KM, Arons M, et al. Asymptomatic and Presymptomatic SARS-
CoV-2 Infections in Residents of a Long-Term Care Skilled Nursing Facility — King
County, Washington, March 2020. MMWR Morb Mortal Wkly Rep 2020;69:377–81.
doi:10.15585/mmwr.mm6913e1
22 Hu Z, Song C, Xu C, et al. Clinical characteristics of 24 asymptomatic infections with
COVID-19 screened among close contacts in Nanjing, China. Sci China Life Sci
Published Online First: 2020. doi:10.1007/s11427-020-1661-4
23 Kam K, Yung CF, Cui L, et al. A Well Infant With Coronavirus Disease 2019 With
High Viral Load. Clin Infect Dis Published Online First: 28 February 2020.
doi:10.1093/cid/ciaa201
24 Wölfel R, Corman VM, Guggemos W, et al. Virological assessment of hospitalized
patients with COVID-2019. Nature 2020;:1–10. doi:10.1038/s41586-020-2196-x
25 Tindale L, Coombe M, Stockdale JE, et al. Transmission interval estimates suggest
pre-symptomatic spread of COVID-19. medRxiv 2020;:2020.03.03.20029983.
doi:10.1101/2020.03.03.20029983
26 Ganyani T, Kremer C, Chen D, et al. Estimating the generation interval for
coronavirus disease (COVID-19) based on symptom onset data, March 2020.
Eurosurveillance 2020;25:2000257. doi:10.2807/1560-7917.ES.2020.25.17.2000257
27 Ferretti L, Wymant C, Kendall M, et al. Quantifying SARS-CoV-2 transmission
suggests epidemic control with digital contact tracing. Science Published Online First:
31 March 2020. doi:10.1126/science.abb6936
28 He X, Lau EHY, Wu P, et al. Temporal dynamics in viral shedding and
transmissibility of COVID-19. Nat Med 2020;26:672–5. doi:10.1038/s41591-020-
0869-5
29 Ma S, Zhang J, Zeng M, et al. Epidemiological parameters of coronavirus disease
2019: a pooled analysis of publicly reported individual data of 1155 cases from seven
countries. medRxiv 2020;:1–21.
30 Du Z, Xu X, Wu Y, et al. Serial Interval of COVID-19 among Publicly Reported
Confirmed Cases. Emerg Infect Dis 2020;26. doi:10.3201/eid2606.200357
31 Du Z, Xu xiaoke, Wu Y, et al. COVID-19 serial interval estimates based on
18
confirmed cases in public reports from 86 Chinese cities. medRxiv

2020;:2020.04.23.20075796. doi:10.1101/2020.04.23.20075796
32 Zhang W. Estimating the presymptomatic transmission of COVID19 using incubation
period and serial interval data. medRxiv 2020;:2020.04.02.20051318.
doi:10.1101/2020.04.02.20051318
33 McAloon CG, Collins A, Hunt K, et al. The incubation period of COVID-19: A rapid
systematic review and meta-analysis of observational research. medRxiv
2020;:2020.04.24.20073957. doi:10.1101/2020.04.24.20073957
34 Griffin JM, Collins A, Hunt K, et al. A rapid review of available evidence on the serial
interval and generation time of COVID-19. medRxiv 2020.
35 Vink MA, Bootsma MCJ, Wallinga J. Serial Intervals of Respiratory Infectious
Diseases: A Systematic Review and Analysis. Am J Epidemiol 2014;180:865–75.
doi:10.1093/aje/kwu209
36 Svensson Å. A note on generation times in epidemic models. Math Biosci
2007;208:300–11. doi:10.1016/j.mbs.2006.10.010
37 NCBI. National Center for Biotechnology Information.
2020.https://www.ncbi.nlm.nih.gov/ (accessed 26 Apr 2020).
38 Google. Google Scholar. 2020.https://scholar.google.com/ (accessed 26 Apr 2020).
39 medRxiv. medRxiv.org - the preprint server for Health Sciences.
2020.https://www.medrxiv.org/ (accessed 26 Apr 2020).
40 bioRxiv. bioRxiv.org - the preprint server for Biology. 2020.https://www.biorxiv.org/
(accessed 26 Apr 2020).
41 NIH. LitCovid - NCBI - NLM - NIH.
2020.https://www.ncbi.nlm.nih.gov/research/coronavirus/ (accessed 26 Apr 2020).
42 Chen Q, Allot A, Lu Z. Keep up with the latest coronavirus research. Nature
2020;579:193–193. doi:10.1038/d41586-020-00694-1
43 WHO. WHO | Bulletin of the World Health Organization. WHO Published Online
First: 2020.https://www.who.int/bulletin/en/ (accessed 26 Apr 2020).
44 ECDC. European Centre for Disease Prevention and Control Rapid Risk Assessments.
2020.https://www.ecdc.europa.eu/en/threats-and-outbreaks/reports-and-data/risk-
assessments (accessed 26 Apr 2020).
45 CDC. Novel Coronavirus Reports | Morbidity and Mortality Weekly Report.
2020.https://www.cdc.gov/mmwr/Novel_Coronavirus_Reports.html (accessed 26 Apr
2020).
19
46 AAAS. Coronavirus: Research, Commentary, and News.

2020.https://www.sciencemag.org/collections/coronavirus?intcmp=ghd_cov%3Futm_
medium%3Demail&utm_source=membership&utm_campaign=tbc-2020-acq-fos-
global-lead-list-
2&utm_content=email1&dmc=P2BSO1EMAQL2&et_rid=35079646&et_cid=328571
8 (accessed 26 Apr 2020).
47 Nature. Nature Coronavirus updates. 2020.
48 Tricco AC, Lillie E, Zarin W, et al. PRISMA Extension for Scoping Reviews
(PRISMA-ScR): Checklist and Explanation. Ann Intern Med 2018;169:467.
doi:10.7326/M18-0850
49 R Core Team. R: A language and environment for statistical computing. R Foundation
for Statistical Computing. 2019.http://www.r-project.org.
50 Jombart T. Package 'epitrix" Small Helpers and Tricks for Epidemics Analysis.
2019.https://github.com/reconhub/epitrix/issues (accessed 26 Apr 2020).
51 Macdonald P. Package ’mixdist’: Finite Mixture Distribution Models.
2018.https://cran.r-project.org/web/packages/mixdist/mixdist.pdf (accessed 26 Apr
2020).
52 Zhao S, Cao P, Chong MKC, et al. The time-varying serial interval of the coronavirus
disease (COVID-19) and its gender-specific difference: A data-driven analysis using
public surveillance data in Hong Kong and Shenzhen, China from January 10 to
February 15, 2020. Infect Control Hosp Epidemiol 2020;:1–2. doi:10.1017/ice.2020.64
53 Delignette-Muller ML, Dutang C. fitdistrplus: An R Package for Fitting Distributions.
J Stat Softw 2015;64:1–34. doi:10.18637/jss.v064.i04
54 Cheng H-Y, Jian S-W, Liu D-P, et al. High transmissibility of COVID-19 near
symptom onset. medRxiv 2020;:2020.03.18.20034561.
doi:10.1101/2020.03.18.20034561
55 Nishiura H, Linton NM, Akhmetzhanov AR. Serial interval of novel coronavirus
(COVID-19) infections. Int J Infect Dis Published Online First: 2020.
doi:10.1016/j.ijid.2020.02.060
56 Yang Liu, CMMID nCov working group, Sebastian Funk, et al. The Contribution of
Pre-symptomatic Transmission to the COVID-19 Outbreak | CMMID Repository. C.
Repos. 2020.https://cmmid.github.io/topics/covid19/pre-symptomatic-
transmission.html (accessed 10 May 2020).
57 Hu F, Chen F, Wang Y, et al. Failed detection of the full-length genome of SARS-
20
CoV-2 by ultra-deep sequencing from the recovered and discharged patients retested
viral PCR positive Authors. medRxiv Published Online First: 2020.
doi:10.1101/2020.03.27.20043299
58 Byrne AW, McEvoy D, Collins A, et al. Inferred duration of infectious period of
SARS-CoV-2: rapid scoping review and analysis of available evidence for
asymptomatic and symptomatic COVID-19 cases. medRxiv
2020;:2020.04.25.20079889. doi:10.1101/2020.04.25.20079889
59 ECDC. Systematic review on the incubation and infectiousness/ shedding period of
communicable diseases in children. 2016. doi:10.2900/337026
60 Anderson RM, May RM (Robert M. Infectious diseases of humans: dynamics and
control. Oxford University Press 1991.
https://global.oup.com/academic/product/infectious-diseases-of-humans-
9780198540403?cc=ie&lang=en& (accessed 19 Mar 2020).
61 Wu JT, Leung K, Bushman M, et al. Estimating clinical severity of COVID-19 from
the transmission dynamics in Wuhan, China. Nat Med 2020;26:506–10.
doi:10.1038/s41591-020-0822-7
62 Wang Y, Teunis PFM. Strongly heterogeneous transmission of COVID-19 in
mainland China: local and regional variation. medRxiv 2020;:2020.03.10.20033852.
doi:10.1101/2020.03.10.20033852
63 Zhang J, Litvinova M, Wang W, et al. Evolving epidemiology and transmission
dynamics of coronavirus disease 2019 outside Hubei province, China: a descriptive
and modelling study. Lancet Infect Dis 2020;0. doi:10.1016/S1473-3099(20)30230-9
64 Zhao S, Gao D, Zhuang Z, et al. Estimating the serial interval of the novel coronavirus
disease (COVID-19): A statistical analysis using the public data in Hong Kong from
January 16 to February 15, 2020. medRxiv 2020;:2020.02.21.20026559.
doi:10.1101/2020.02.21.20026559
65 Bi Q, Wu YY, Mei S, et al. Epidemiology and transmission of COVID-19 in 391 cases
and 1286 of their close contacts in Shenzhen, China: a retrospective cohort study.
Lancet Infect Dis 2020;0. doi:10.1016/S1473-3099(20)30287-5
66 Li Q, Guan X, Wu P, et al. Early Transmission Dynamics in Wuhan, China, of Novel
Coronavirus–Infected Pneumonia. N Engl J Med 2020;382:1199–207.
doi:10.1056/NEJMoa2001316
21
Table 1: Definitions referred to in this review.

Asymptomatic An infected person who never develops symptoms of the disease
Pre-symptomatic An infected person before they develop symptoms of the disease.
Duration of The time interval in days during which an infectious agent may be transferred directly or
infectiousness indirectly from an infected person to another person [59].
Incubation period The time interval in days between invasion by an infectious agent and appearance of the first
signs or symptoms of the disease in question [59].
Serial interval The duration in days between symptom onset of a secondary case and that of its primary case.
Generation time or The duration in days between time of infection of a secondary case (infectee) and that of its
generation interval primary case (infector).
Transmission pair An infected person (infector) and a person who they transmit the pathogen to (infectee).
Latent period The period from the point of infection to the beginning of the state of infectiousness [60]. This
period corresponds to the “exposed” (E) compartment of a susceptible-exposed-infectious-
recovered/removed (SEIR) model.
Transmission time The time of transmission of an infectious agent from an infector to an infectee in days relative to
relative to symptom the onset of symptoms in the infector.
onset
Proportion of pre- The proportion of all transmission events that occur before the onset of symptoms in the
symptomatic infector.
transmission
22
Table 2: A summary of all generation time and serial interval estimates included in our rapid
systematic review [34] considered for inclusion in this study, those included in both
Simulations A and B, those included in Simulation A only, and those excluded. SI = Serial
interval. GT = Generation time. Ref = reference.
Included in Simulation A and B
Ref Sub-set/Location N Distribution SI or GT
[64] Shenzhen 48 pairs Gamma SI
[26a] Singapore Model 1 91 cases (54 with known contacts) Gamma GT
Singapore Model 2 (allowing
inference of negative values for
[26c] unknown SI’s) 91 cases (54 with known contacts) Gamma GT
[26e] Tianjin Model 1 135 cases (114 with known contacts) Gamma GT
[26g] Tianjin Model 2 135 cases (114 with known contacts) Gamma GT
[60] Mixed 43 pairs Gamma SI
Hong Kong: 12 more certain
[63a] pairs 12 case pairs Gamma SI
Hong Kong: 21 (all available
[63b] pairs) 21 transmission chains Gamma SI
Mixed: 28 pairs (all available
[55b] pairs) 28 pairs Lognormal SI
[55a] Mixed: 18 more certain pairs 18 pairs Weibull SI
[30] MCNH 468 pairs Normal SI
[31] MCNH 339 pairs Normal SI
[29] Mixed 689 pairs Normal SI
Included in Simulation B only: Confidence intervals for specific distribution parameters could not be incorporated into Simulation
A using asymptotic approach
Sub-set/Location N Distribution SI or GT
[3] Northern Italy 90 pairs Gamma SI
[27a] Mixed: GT from paper 40 pairs Weibull GT
[52] Hong Kong and Shenzhen 48 pairs Gamma SI
Included in Simulation B only: The serial interval was simulated from the generation time and incubation period from the same
paper. A distribution for serial interval was not defined. For simulation B we repeated this simulation to generate the serial interval.
Sub-set / Location N Distribution SI or GT
[26b] Singapore Model 1 SI 91 cases (54 with known contacts) Not reported SI
[26d] Singapore Model 2 SI 91 cases (54 with known contacts) Not reported SI
[26f] Tianjin Model 1 SI 135 cases (114 with known contacts) Not reported SI
[26h] Tianjin Model 2 SI 135 cases (114 with known contacts) Not reported SI
[27a] Mixed: SI estimated from GT 40 pairs Not reported SI
Included in Simulation B only: Confidence intervals for standard deviation not reported
Sub-set/ Location N Distribution SI or GT
[19] Northern Italy 166 cases, 120 pairs Gamma SI
[65] Wuhan 6 pairs Gamma SI
[62] MCNH 35 pairs Gamma SI
[25a] Singapore 93 pairs Normal SI
[25b] Tianjin 125 pairs Normal SI
[28] Mixed 77 pairs Shifted Gamma 7.6% negative SIs SI
Included in Simulation B only: Confidence intervals for mean and standard deviation not reported
[61] Tianjin 112 cases Gamma SI
Excluded
[54] Taiwan 12 pairs Gamma SI
[20] Chongqing, China 12 pairs Gamma SI
23
Table 3: Summary statistics for transmission time relative to symptom onset and proportion pre-symptomatic transmission estimated from
Simulation A. The studies are ordered first by whether they are based on generation time (GT) or serial interval (SI), and then by median
transmission time relative to symptoms. A negative number means days before symptom onset. MCNH = Mainland China excluding Hubei
province. Mixed = pooled data from multiple countries. SD = Standard deviation. Ref = reference number corresponding to Table 2 and main
text. Confidence intervals are based incorporating measures of uncertainty from the source publications into the simulation.

Ref Location SI Percentage pre- Mean (95% CI) SD (95% CI) 2.5th percentile 25th percentile Median (95% CI) 75th percentile 97.5th percentile
or symptomatic (95% CI) (95% CI) (95% CI) (95% CI)
GT transmission 95%
CI)
[26a] Singapore GT 52.1 (49.0, 55.2) -0.57 (-0.80, -0.34) 3.71 (3.44, 3.98) -9.03 (-10.17, -7.89) -2.48 (-2.80, -2.16) -0.16 (-0.40, 0.08) 1.79 (1.55, 2.02) 5.55 (4.96, 6.15)
[26e] Tianjin GT 69.3 (66.4, 72.1) -1.84 (-2.06, -1.63) 3.54 (3.27, 3.81) -10.13 (-11.27, -8.99) -3.60 (-3.91, -3.28) -1.38 (-1.61, -1.15) 0.43 (0.22, 0.64) 3.76 (3.25, 4.27)
[26c] Singapore GT 71.1 (68.3, 73.9) -1.98 (-2.25, -1.71) 4.35 (3.99, 4.71) -10.87 (-12.00, -9.73) -4.31 (-4.63, -3.98) -1.91 (-2.17, -1.64) 0.41 (0.11, 0.71) 6.44 (5.28, 7.61)
[26g] Tianjin GT 80.6 (78.1, 83.0) -2.91 (-3.18, -2.64) 4.39 (4.01, 4.77) -11.70 (-12.82,-10.58) -5.19 (-5.50, -4.87) -2.91 (-3.15, -2.66) -0.71 (-1.01, -0.40) 6.03 (4.70, 7.35)
[29] Mixed SI 42.8 (39.8, 45.9) 0.90 (0.52, 1.28) 6.11 (5.81, 6.40) -11.59 (-12.81,-10.37) -3.01 (-3.54, -2.48) 1.08 (0.62, 1.53) 5.03 (4.54, 5.52) 12.31 (11.38, 13.23)
[60] Mixed SI 43.4 (40.4, 46.5) 1.20 (0.86, 1.55) 5.54 (5.19, 5.88) -8.86 (-9.96, -7.76) -2.20 (-2.56, -1.83) 0.76 (0.40, 1.12) 4.24 (3.76, 4.72) 13.38 (11.92, 14.84)
[64] Shenzhen SI 48.6 (45.5, 51.7) 0.51 (0.18, 0.84) 5.28 (4.95, 5.62) -9.31 (-10.46, -8.17) -2.67 (-3.02, -2.32) 0.15 (-0.19, 0.50) 3.41 (2.95, 3.86) 11.97 (10.59, 13.35)
[31] MCNH SI 51.9 (48.8, 55.0) -0.47 (-0.86, -0.08) 6.28 (5.97, 6.58) -13.24 (-14.50,-11.98) -4.51 (-5.04, -3.97) -0.30 (-0.77, 0.18) 3.77 (3.26, 4.28) 11.29 (10.34, 12.24)
[55a] Mixed SI 56.2 (53.1, 59.3) -0.78 (-1.06, -0.50) 4.46 (3.45, 5.46) -10.10 (-11.22, -8.97) -3.44 (-3.80, -3.08) -0.66 (-0.99, -0.33) 2.18 (1.82, 2.53) 7.26 (6.63, 7.90)
[55b] Mixed SI 58.8 (55.7, 61.9) -0.81 (-1.11, -0.50) 4.81 (4.29, 5.33) -10.23 (-11.37, -9.08) -3.59 (-3.94, -3.24) -0.92 (-1.24, -0.61) 1.96 (1.58, 2.35) 8.68 (7.58, 9.78)
[30] MCNH SI 61.7 (58.6, 64.7) -1.84 (-2.19, -1.48) 5.75 (5.47, 6.04) -13.69 (-14.89,-12.48) -5.47 (-5.96, -4.99) -1.64 (-2.07, -1.21) 2.05 (1.59, 2.51) 8.81 (7.97, 9.64)
[63b] Hong Kong SI 65.1 (62.1, 68.1) -1.42 (-1.70, -1.13) 4.54 (4.20, 4.89) -10.51 (-11.64, -9.37) -3.93 (-4.26, -3.60) -1.43 (-1.72, -1.15) 1.10 (0.77, 1.43) 7.67 (6.46, 8.88)
[63a] Hong Kong SI 79.3 (76.8, 81.8) -2.69 (-2.93, -2.45) 3.83 (3.46, 4.20) -11.16 (-12.32, -9.99) -4.62 (-4.94, -4.31) -2.36 (-2.60, -2.13) -0.40 (-0.64, -0.16) 3.85 (3.08, 4.62)
Unweighted pooling GT 68.3 (47.9,88.6) -1.83 (-3.48, -0.17) 4.00 ( 3.19, -10.43 (-12.66, -8.20) -3.89 (-5.86, -1.92) -1.59 ( -3.54, 0.37) 0.48 ( -1.27, 2.23) 5.45 ( 3.22, 7.67)
4.80)
SI 56.4 (34.9,78.0) -0.60 ( -3.01, 1.81) 5.18 (3.58, -10.96 (-14.22, -7.70) -3.72 (-5.67, -1.76) -0.59 ( -2.72, 1.54) 2.59 ( -0.55, 5.74) 9.47 ( 3.76, 15.18)
6.77)
24
Table 4: Summary statistics for transmission time relative to symptom onset and proportion pre-symptomatic transmission estimated from
Simulation A. MCNH = Mainland China excluding Hubei province. N Italy = Northern Italy. Mixed = data came from multiple countries. SD =
Standard deviation. Confidence intervals in Simulation A are based on incorporating measures of uncertainty from the source publications into
the simulation.

Percentage pre-
2.5th percentile (95% 25th percentile (95% 75th percentile (95% 97.5th percentile (95%
symptomatic Mean (95% CI) SD (95% CI) Median (95% CI)
CI) CI) CI) CI)
Location transmission
Shenzhen 48.6 (45.5, 51.7) 0.51 (0.18, 0.84) 5.28 (4.95, 5.62) -9.31 (-10.46, -8.17) -2.67 (-3.02, -2.32) 0.15 (-0.19, 0.50) 3.41 (2.95, 3.86) 11.97 (10.59, 13.35)
Mixed 50.3 (35.8, 64.9) 0.13 (-1.72, 1.98) 5.23 (3.84, 6.62) -10.19 (-12.41, -7.98) -3.06 (-4.19, -1.92) 0.06 (-1.68, 1.80) 3.35 (0.74, 5.97) 10.41 (5.36, 15.46)
MCNH 56.8 (46.8, 66.8) -1.15(-2.55, 0.24) 6.01 (5.42, 6.61) -13.46 (-14.77,-12.15) -4.99 (-6.07, -3.91) -0.97 (-2.36, 0.43) 2.91 (1.16, 4.67) 10.05 (7.45, 12.64)
Singapore 61.6 (42.7, 80.5) -1.2 (-2.68, 0.13) 4.03 (3.33, 4.73) -9.95 (-12.08, -7.82) -3.39 (-5.21, -1.57) -1.04 (-2.77, 0.70) 1.10 (-0.27, 2.47) 6.00 (4.72, 7.27)
Hong Kong 72.2 (58.0, 86.3) -2.05(-3.33,-0.78) 4.19 (3.40, 4.97) -10.83 (-12.15, -9.52) -4.28 (-5.03, -3.52) -1.90 (-2.84, -0.95) 0.35 (-1.15, 1.85) 5.76 (1.88, 9.64)
Tianjin 74.9 (63.5, 86.3) -2.38(-3.45,-1.30) 3.97 (3.07, 4.86) -10.91 (-12.83, -9.00) -4.39 (-5.98, -2.80) -2.14 (-3.66, -0.63) -0.14 (-1.29, 1.01) 4.89 (2.46, 7.33)
25
Table 5: A summary of the counts (N) of studies and estimates used in Simulation B coming from specific locations, or mixed sources.
Location / data source Count of estimates Count of studies References
Mixed 7 5 [27–29,55,61]
Tianjin 6 3 [25,26,62]
Singapore 5 2 [25,26]
Mainland China excluding 3 3 [30,31,63]

Hubei
Hong Kong 2 1 [64]
Northern Italy 2 2 [3,19]
Hong Kong and Shenzhen 1 1 [52]
Shenzhen 1 1 [65]
Wuhan 1 1 [66]
Total 28 19
26
Table 6: Summary statistics based on Simulation B for transmission time relative to symptom onset and proportion pre-symptomatic
transmission (PPS) for each of 5 estimates based on generation time (GT) and 23 estimates based on serial interval (SI). The studies are ordered
first by whether they are based on generation time or serial interval, and then by median transmission time relative to symptoms, to correspond
to Figure 3. A negative number means days before symptom onset. MCNH = Mainland China excluding Hubei province. Mixed = SI or GT was
calculated from pooled data from multiple countries. HKSZ = pooled data from Hong Kong and Shenzhen. SD = Standard deviation. Figure 2

ref = Reference number corresponding to Table 2, Figure 3 and main text. Confidence intervals are based on the variation between 20,000 repeat
simulations.
Reference Location N for study Based on % pre- Mean Standard 2.5th 25th Median 75th 97.5th
(description) symptomatic deviation percentile percentile percentile percentile
transmission
[26a] Singapore 91 GT 52.14 -0.59 3.53 -8.87 -2.38 -0.16 1.71 5.13
(54) cases (51.83, 52.46) (-0.61, -0.57) (3.50, 3.55) (-8.99, -8.76) (-2.41, -2.35) (-0.18, -0.13) (1.69, 1.73) (5.09, 5.18)
[27a] Mixed 40 pairs GT 54.14 -0.74 3.64 -9.19 -2.66 -0.33 1.70 5.16
(53.83, 54.45) (-0.76, -0.71) (3.61, 3.66) (-9.30, -9.07) (-2.69, -2.63) (-0.35, -0.30) (1.68, 1.73) (5.12, 5.20)
[26e] Tianjin 135 GT 69.42 -1.83 3.43 -10.01 -3.53 -1.35 0.42 3.58
(114) cases (69.14, 69.71) (-1.85, -1.81) (3.41, 3.46) (-10.12, -9.89) (-3.56, -3.50) (-1.38, -1.33) (0.39, 0.44) (3.53, 3.62)
[26c] Singapore 93 GT 70.89 -1.93 4.06 -10.63 -4.13 -1.81 0.43 5.99
Model 2 (54) cases (70.61, 71.18) (-1.95, -1.90) (4.04, 4.09) (-10.74, - (-4.16, -4.09) (-1.83, -1.78) (0.40, 0.46) (5.91, 6.08)
10.52)
[26g] Tianjin Model 137 GT 80.70 -2.92 4.18 -11.61 -5.13 -2.88 -0.73 5.74
2 (114) cases (80.45, 80.94) (-2.95, -2.90) (4.15, 4.21) (-11.72,-11.50) (-5.16, -5.10) (-2.90, -2.85) (-0.76, -0.70) (5.63, 5.84)
[65] Wuhan 6 pairs SI 33.47 1.72 4.59 -7.63 -1.00 1.69 4.50 10.89
(33.18, 33.76) (1.69, 1.75) (4.56, 4.62) (-7.74, -7.52) (-1.04, -0.97) (1.66, 1.73) (4.46, 4.54) (10.80, 10.98)
[19] N Italy 120 pairs SI 36.20 1.15 4.20 -7.81 -1.23 1.29 3.76 9.11
(35.90, 36.49) (1.13, 1.18) (4.17, 4.22) (-7.92, -7.70) (-1.26, -1.20) (1.26, 1.32) (3.73, 3.79) (9.03, 9.19)
[29] Mixed 689 pairs SI 42.69 0.92 6.04 -11.48 -2.96 1.09 5.01 12.27
(42.39, 43.00) (0.88, 0.95) (6.02, 6.07) (-11.60,-11.35) (-3.02, -2.91) (1.05, 1.14) (4.96, 5.06) (12.18, 12.36)
[60] Mixed 43 pairs SI 43.22 1.22 5.45 -8.80 -2.14 0.77 4.23 13.29
(42.91, 43.53) (1.18, 1.25) (5.42, 5.49) (-8.91, -8.69) (-2.18, -2.11) (0.73, 0.81) (4.18, 4.28) (13.15, 13.43)
[3] N Italy 90 pairs SI 47.77 0.90 5.78 -9.30 -2.66 0.26 3.93 14.12
(47.46, 48.09) (0.86, 0.93) (5.74, 5.81) (-9.41, -9.19) (-2.70, -2.63) (0.22, 0.30) (3.88, 3.98) (13.96, 14.28)
[64] Shenzhen 48 pairs SI 48.68 0.52 5.21 -9.24 -2.63 0.14 3.39 11.95
(48.37, 49.00) (0.48, 0.55) (5.18, 5.24) (-9.35, -9.13) (-2.66, -2.59) (0.11, 0.18) (3.34, 3.43) (11.82, 12.08)
[28] MCNH 339 pairs SI 51.54 -0.02 5.44 -10.49 -3.49 -0.19 3.34 11.18
(51.23, 51.85) (-0.06, 0.01) (5.41, 5.47) (-10.60, 10.38) (-3.53, -3.45) (-0.23, -0.16) (3.29, 3.38) (11.07, 11.29)
27
Reference Location N for study Based on % pre- Mean Standard 2.5th 25th Median 75th 97.5th
(description) symptomatic deviation percentile percentile percentile percentile
transmission
[31] Mixed 77 pairs SI 52.15 -0.49 6.15 -13.07 -4.45 -0.32 3.67 11.08
(51.84, 52.46) (-0.53, -0.46) (6.12, 6.18) (-13.20,-12.95) (-4.50, -4.40) (-0.37, -0.28) (3.62, 3.72) (10.98, 11.17)
[26b] Singapore 92 SI 53.02 -0.59 5.30 -11.78 -3.80 -0.37 2.87 9.35
(54) cases (52.71, 53.33) (-0.62, -0.56) (5.28, 5.33) (-11.90,-11.66) (-3.85, -3.76) (-0.41, -0.33) (2.83, 2.91) (9.26, 9.44)
[27b] Mixed 40 pairs SI 54.30 -0.74 4.67 -10.82 -3.50 -0.46 2.36 7.70
(53.99, 54.61) (-0.77, -0.71) (4.65, 4.70) (-10.94,-10.71) (-3.54, -3.46) (-0.49, -0.43) (2.32, 2.39) (7.63, 7.77)

[62] MCNH 35 pairs SI 56.19 -0.67 4.09 -9.46 -2.93 -0.53 1.81 7.11
(55.88, 56.50) (-0.70, -0.65) (4.06, 4.11) (-9.57, -9.35) (-2.96, -2.90) (-0.56, -0.50) (1.78, 1.84) (7.04, 7.19)
[25a] Singapore 93 pairs SI 59.57 -1.22 3.23 -9.20 -2.74 -0.62 0.98 3.25
(59.26, 59.87) (-1.24, -1.20) (3.20, 3.25) (-9.31, -9.09) (-2.77, -2.71) (-0.64, -0.60) (0.96, 1.00) (3.22, 3.27)
[55a] Mixed 18 pairs SI 58.22 -0.98 3.85 -9.62 -3.09 -0.69 1.53 5.70
(n=18) (57.91, 58.53) (-1.01, -0.96) (3.82, 3.87) (-9.74, -9.51) (-3.12, -3.05) (-0.72, -0.66) (1.50, 1.56) (5.64, 5.75)
[61] Tianjin 112 cases SI 59.96 -0.98 4.10 -9.77 -3.24 -0.85 1.48 6.90
(59.66, 60.27) (-1.01, -0.96) (4.07, 4.12) (-9.88, -9.66) (-3.27, -3.21) (-0.88, -0.83) (1.45, 1.51) (6.81, 6.98)
[25b] Tianjin 125 pairs SI 64.88 -1.56 3.11 -9.41 -2.96 -0.90 0.59 2.42
(64.59, 65.18) (-1.58, -1.54) (3.08, 3.13) (-9.52, -9.29) (-2.99, -2.93) (-0.92, -0.88) (0.57, 0.60) (2.41, 2.44)
[52] HKSZ 48 pairs SI 59.34 -0.61 5.05 -10.12 -3.56 -0.98 2.07 10.60
(59.04, 59.65) (-0.64, -0.58) (5.02, 5.09) (-10.23, - (-3.59, -3.53) (-1.01, -0.94) (2.02, 2.11) (10.46, 10.74)
10.01)
[55b] Mixed 28 pairs SI 62.63 -1.08 4.23 -9.84 -3.34 -1.03 1.19 7.50
(n=28) (62.33, 62.93) (-1.11, -1.06) (4.20, 4.26) (-9.95, -9.73) (-3.37, -3.31) (-1.05, -1.00) (1.16, 1.22) (7.39, 7.61)
[63b] Hong Kong 21 pairs SI 65.24 -1.38 4.30 -10.30 -3.78 -1.38 1.08 7.32
(n=21) (64.94, 65.53) (-1.41, -1.36) (4.27, 4.33) (-10.41,-10.19) (-3.81, -3.74) (-1.40, -1.35) (1.04, 1.11) (7.22, 7.41)
[30] MCNH 468 pairs SI 63.17 -1.83 5.24 -12.93 -4.99 -1.60 1.59 7.96
(62.87, 63.47) (-1.86, -1.80) (5.21, 5.27) (-13.05,-12.81) (-5.03, -4.94) (-1.63, -1.56) (1.55, 1.63) (7.87, 8.05)
[26f] Tianjin 136 SI 61.74 -1.82 5.66 -13.53 -5.42 -1.63 2.01 8.72
(114) cases (61.45, 62.04) (-1.86, -1.79) (5.63, 5.69) (-13.65,-13.41) (-5.47, -5.37) (-1.67, -1.59) (1.97, 2.06) (8.63, 8.80)
[26d] Singapore 94 SI 63.91 -1.93 5.67 -13.51 -5.43 -1.85 1.68 9.18
Model 2 (54) cases (63.61, 64.20) (-1.96, -1.89) (5.64, 5.70) (-13.63,-13.39) (-5.48, -5.39) (-1.89, -1.81) (1.64, 1.72) (9.07, 9.28)
[63a] Hong Kong 12 pairs SI 79.90 -2.68 3.57 -10.97 -4.49 -2.29 -0.44 3.37
(n=12) (79.65, 80.15) (-2.71, -2.66) (3.54, 3.59) (-11.08,-10.85) (-4.52, -4.46) (-2.31, -2.27) (-0.47, -0.42) (3.31, 3.42)
[26h] Tianjin Model 138 SI 71.10 -2.92 5.76 -14.51 -6.47 -2.91 0.64 8.61
2 (114) cases (70.82, 71.38) (-2.96, -2.89) (5.72, 5.79) (-14.63,-14.39) (-6.51, -6.42) (-2.95, -2.87) (0.59, 0.68) (8.49, 8.73)
Simple pooled estimate of GT based transmission time relative to 65.46 -1.60 3.88 -10.25 -3.71 -1.31 0.88 5.12
symptom onset (65.27, 65.65) (-1.62, -1.58) (3.85, 3.90) (-10.34,-10.15) (-3.73, -3.68) (-1.33, -1.29) (0.87, 0.90) (5.09, 5.14)
Simple pooled estimate of SI based transmission time relative to 56.04 -0.66 5.05 -11.10 -3.51 -0.62 2.21 9.60
symptom onset (55.89, 56.19) (-0.68, -0.64) (5.03, 5.06) (-11.17,-11.03) (-3.54, -3.49) (-0.64, -0.60) (2.20, 2.23) (9.58, 9.63)
28
Table 7: Summary statistics from Simulation B for transmission time relative to symptom onset and proportion pre-symptomatic transmission
(PPS) for simulated samples, based on 23 estimates of serial interval, pooled according to the source location of the original data upon which
estimates were based. MCNH = Mainland China excluding Hubei province. N Italy = Northern Italy. Mixed = data came from multiple
countries. The studies are ordered by median transmission time relative to symptoms, to correspond to Figure 4. SD = Standard deviation.

Location Count of Proportion pre- Mean SD 2.5th percentile 25th percentile Median 75th percentile 97.5th
estimates symptomatic percentile
transmission
Wuhan 1 33.5% 1.72 4.62 -7.63 -1.02 1.68 4.53 10.96
N Italy 2 42.1% 1.01 5.05 -8.66 -2.00 0.83 3.81 11.76
Shenzhen 1 48.9% 0.50 5.22 -9.26 -2.65 0.12 3.40 11.90
Mixed 6 52.2% -0.13 5.09 -10.29 -3.13 -0.23 2.81 10.49
MCNH 3 56.8% -1.01 5.41 -12.48 -4.19 -0.80 2.40 9.35
Singapore 3 58.8% -1.25 4.89 -11.95 -3.97 -0.87 1.64 8.27
HKSZ 1 59.3% -0.62 5.06 -10.16 -3.59 -0.96 2.07 10.64
Tianjin 4 64.9% -1.84 4.72 -12.31 -4.40 -1.40 0.99 7.09
Hong Kong 2 72.7% -2.05 4.00 -10.67 -4.19 -1.90 0.24 5.81
29
Figure 1: Schematic illustration of incubation period, generation time and serial interval at
infector-infectee pair level. Scenario A: if serial interval/generation time is longer than
incubation period, transmission occurs after symptom onset. Scenario B: if serial
interval/generation time is shorter than incubation period, transmission occurs prior to
symptom onset. Scenario C: A negative serial interval is possible if symptoms manifest in the
infectee before the infector. If incubation period is assumed to be independent and identically
distributed, mean serial interval will approximate mean generation time.
Figure 2: Generation time (GT, upper panels) and serial interval (SI, lower panels) estimates
from the publications included in this study. Mean (left panels) and Standard Deviation (SD,
right panels) are shown. Circles represent mean estimates and bars represent 95% confidence
intervals.
Figure 3: A boxplot showing time of transmission relative to onset of symptoms inferred

from simulations of incubation period and generation time (GT) (n = 5 estimates) or serial
interval (SI) (n = 23 estimates). The purple triangles represent the mean of the simulated
samples. The vertical red line represents onset of symptoms. The numbers on the left axis are
the reference numbers for each estimate (described in Table 2).
Figure 4: A boxplot showing time of transmission relative to onset of symptoms inferred

from simulations from incubation period and serial interval (n = 23 estimates). The purple
triangles represent the mean of the simulated samples. The vertical red line represents onset
of symptoms. The simulated samples were pooled according to the source location of the
original data upon which estimates were based. MCNH = Mainland China excluding Hubei
province. N Italy = Northern Italy. HKSZ = pooled data from Hong Kong and Shenzhen.
Mixed = data came from multiple countries.
30
A: Sym ptomatic transmission: Incubation period ≤ [Serial interval or Generation time]
infection symptoms
IncP
Infector
infection symptoms
IncP
Infectee
Serial interval
Generation time
B: Pre-symptomatic transmission: Incubation period > [Serial interval or generation time]

and Serial interval > 0
infection symptoms
IncP
Infector
symptoms
infection IncP
Infectee
Serial interval
Generation time
C: Pre-symptomatic transm ission: Incubation period > [Serial interval or Generation time]
and Serial interval < 0
infection symptoms
IncP
Infector Legend
IncP = incubation
infection symptoms
period
Infectee
Serial interval
Generation time
Time
Mean SD
[26a] ● ●
[27a] ● ●
GT
[26e] ● ●
[26c] ● ●
[26g] ● ●
[65] ● ●
[60] ● ●
[19] ● ●
[29] ● ●
[3] ● ●
[64] ● ●
[28] ● ●
Reference
[31] ● ●
[26b]
● ●
[52] ● ●
[62] ● ●
SI
[61] ● ●
[55a] ● ●
[55b] ● ●
[25a] ● ●
[63b] ● ●
[25b] ● ●
[30] ● ●
[26f] ● ●
[26d] ● ●
[63a] ● ●
[26h] ● ●
0 10 20 2 4 6
Days
[26a]
[27a]
GT
[26e]
[26c]
[26g]
[65]
[19]
[29]
[60]
[3]
[64]
[28]
[31]
Study
[26b]
[27b]
[62]
SI
[25a]
[55a]
[61]
[25b]
[52]
[55b]
[63b]
[26f]
[30]
[26d]
[63a]
[26h]
−20 −10 0 10 20
Transmission time in days relative to symptom onset
Wuhan
N Italy
Shenzen
Mixed
Location
MCNH
Singapore
HKSZ
Tianjin
Hong Kong
−20 −10 0 10 20
Transmission time in days relative to symptom onset

Pre-Symptomatic Transmission of Sars-Cov-2 Infection: A Secondary Analysis Using Published Data

Uploaded by

Copyright:

Available Formats

Pre-Symptomatic Transmission of Sars-Cov-2 Infection: A Secondary Analysis Using Published Data

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pre-Symptomatic Transmission of Sars-Cov-2 Infection: A Secondary Analysis Using Published Data

Uploaded by

Copyright:

Available Formats

medRxiv preprint doi: https://doi.org/10.1101/2020.05.08.20094870.this version posted June 11, 2020.

The copyright holder for this preprint

Pre-symptomatic transmission of SARS-CoV-2

Correspondence to: Miriam Casey; miriam.casey@ucd.ie, Centre for Veterinary

Key words: “COVID-19”; ‘SARS-CoV-2”, “Pre-symptomatic”; “Transmission”

Strengths and weaknesses of this study

In absence of active surveillance, infectious people without symptoms may not be

Characteristics of pre-symptomatic and asymptomatic transmission are potentially different,

The pre-symptomatic transmission potential of COVID-19 has been highlighted by case

Meta-analysis and rapid systematic review

  !      1   2

unweighted pooling of estimates of transmission time relative to symptom onset based on

Proportion of pre-symptomatic transmission calculated from the 13 individual estimates

We estimated that the proportion of pre-symptomatic transmission ranged from 33.5% to

The wide variation in estimates of the proportion of pre-symptomatic transmission and

Limitations and strengths

Patient and public involvement statement

Data sharing statement

economy. SSRN Electron J 2020;:0–

confirmed cases in public reports from 86 Chinese cities. medRxiv

46 AAAS. Coronavirus: Research, Commentary, and News.

Table 1: Definitions referred to in this review.

It is made available under a CC-BY-ND 4.0 International license .

It is made available under a CC-BY-ND 4.0 International license .

It is made available under a CC-BY-ND 4.0 International license .

It is made available under a CC-BY-ND 4.0 International license .

It is made available under a CC-BY-ND 4.0 International license .

It is made available under a CC-BY-ND 4.0 International license .

N Italy 2 42.1% 1.01 5.05 -8.66 -2.00 0.83 3.81 11.76

Shenzhen 1 48.9% 0.50 5.22 -9.26 -2.65 0.12 3.40 11.90

Mixed 6 52.2% -0.13 5.09 -10.29 -3.13 -0.23 2.81 10.49

MCNH 3 56.8% -1.01 5.41 -12.48 -4.19 -0.80 2.40 9.35

Singapore 3 58.8% -1.25 4.89 -11.95 -3.97 -0.87 1.64 8.27

HKSZ 1 59.3% -0.62 5.06 -10.16 -3.59 -0.96 2.07 10.64

Tianjin 4 64.9% -1.84 4.72 -12.31 -4.40 -1.40 0.99 7.09

Figure 3: A boxplot showing time of transmission relative to onset of symptoms inferred

Figure 4: A boxplot showing time of transmission relative to onset of symptoms inferred

B: Pre-symptomatic transmission: Incubation period > [Serial interval or generation time]

You might also like

! 1 2