Free access
Original Research
10 March 2020

The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and ApplicationFREE

Publication: Annals of Internal Medicine
Volume 172, Number 9
Visual Abstract. The Incubation Period of COVID-19 From Publicly Reported Confirmed Cases Using news reports and press releases from provinces, regions, and countries outside Wuhan, Hubei province, China, this analysis estimates the length of the incubation period of COVID-19 and its public health implications.
Visual Abstract. The Incubation Period of COVID-19 From Publicly Reported Confirmed Cases
Using news reports and press releases from provinces, regions, and countries outside Wuhan, Hubei province, China, this analysis estimates the length of the incubation period of COVID-19 and its public health implications.

Abstract

Background:

A novel human coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was identified in China in December 2019. There is limited support for many of its key epidemiologic features, including the incubation period for clinical disease (coronavirus disease 2019 [COVID-19]), which has important implications for surveillance and control activities.

Objective:

To estimate the length of the incubation period of COVID-19 and describe its public health implications.

Design:

Pooled analysis of confirmed COVID-19 cases reported between 4 January 2020 and 24 February 2020.

Setting:

News reports and press releases from 50 provinces, regions, and countries outside Wuhan, Hubei province, China.

Participants:

Persons with confirmed SARS-CoV-2 infection outside Hubei province, China.

Measurements:

Patient demographic characteristics and dates and times of possible exposure, symptom onset, fever onset, and hospitalization.

Results:

There were 181 confirmed cases with identifiable exposure and symptom onset windows to estimate the incubation period of COVID-19. The median incubation period was estimated to be 5.1 days (95% CI, 4.5 to 5.8 days), and 97.5% of those who develop symptoms will do so within 11.5 days (CI, 8.2 to 15.6 days) of infection. These estimates imply that, under conservative assumptions, 101 out of every 10 000 cases (99th percentile, 482) will develop symptoms after 14 days of active monitoring or quarantine.

Limitation:

Publicly reported cases may overrepresent severe cases, the incubation period for which may differ from that of mild cases.

Conclusion:

This work provides additional evidence for a median incubation period for COVID-19 of approximately 5 days, similar to SARS. Our results support current proposals for the length of quarantine or active monitoring of persons potentially exposed to SARS-CoV-2, although longer monitoring periods might be justified in extreme cases.

Primary Funding Source:

U.S. Centers for Disease Control and Prevention, National Institute of Allergy and Infectious Diseases, National Institute of General Medical Sciences, and Alexander von Humboldt Foundation.
In December 2019, a cluster of severe pneumonia cases of unknown cause was reported in Wuhan, Hubei province, China. The initial cluster was epidemiologically linked to a seafood wholesale market in Wuhan, although many of the initial 41 cases were later reported to have no known exposure to the market (1). A novel strain of coronavirus belonging to the same family of viruses that cause severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS), as well as the 4 human coronaviruses associated with the common cold, was subsequently isolated from lower respiratory tract samples of 4 cases on 7 January 2020 (2). Infection with the virus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), can be asymptomatic or can result in mild to severe symptomatic disease (coronavirus disease 2019 [COVID-19]) (3). On 30 January 2020, the World Health Organization declared that the SARS-CoV-2 outbreak constituted a Public Health Emergency of International Concern, and more than 80 000 confirmed cases had been reported worldwide as of 28 February 2020 (4, 5). On 31 January 2020, the U.S. Centers for Disease Control and Prevention announced that all citizens returning from Hubei province, China, would be subject to mandatory quarantine for up to 14 days (6).
Our current understanding of the incubation period for COVID-19 is limited. An early analysis based on 88 confirmed cases in Chinese provinces outside Wuhan, using data on known travel to and from Wuhan to estimate the exposure interval, indicated a mean incubation period of 6.4 days (95% CI, 5.6 to 7.7 days), with a range of 2.1 to 11.1 days (7). Another analysis based on 158 confirmed cases outside Wuhan estimated a median incubation period of 5.0 days (CI, 4.4 to 5.6 days), with a range of 2 to 14 days (8). These estimates are generally consistent with estimates from 10 confirmed cases in China (mean incubation period, 5.2 days [CI, 4.1 to 7.0 days] [9]) and from clinical reports of a familial cluster of COVID-19 in which symptom onset occurred 3 to 6 days after assumed exposure in Wuhan (1). These estimates of the incubation period of SARS-CoV-2 are also in line with those of other known human coronaviruses, including SARS (mean, 5 days; range, 2 to 14 days [10]), MERS (mean, 5 to 7 days; range, 2 to 14 days [11]), and non-SARS human coronavirus (mean, 3 days; range, 2 to 5 days [12]).
The incubation period can inform several important public health activities for infectious diseases, including active monitoring, surveillance, control, and modeling. Active monitoring requires potentially exposed persons to contact local health authorities to report their health status every day. Understanding the length of active monitoring needed to limit the risk for missing SARS-CoV-2 infections is necessary for health departments to effectively use limited resources. In this article, we provide estimates of the incubation period of COVID-19 and the number of symptomatic infections missed under different active monitoring scenarios.

Methods

Data Collection

We searched for news and public health reports of confirmed COVID-19 cases in areas with no known community transmission, including provinces, regions, and countries outside Hubei. We searched for reports in both English and Chinese and abstracted the data necessary to estimate the incubation period of COVID-19. Two authors independently reviewed the full text of each case report. Discrepancies were resolved by discussion and consensus.
For each case, we recorded the time of possible exposure to SARS-CoV-2, any symptom onset, fever onset, and case detection. The exact time of events was used when possible; otherwise, we defined conservative upper and lower bounds for the possible interval of each event. For most cases, the interval of possible SARS-CoV-2 exposure was defined as the time between the earliest possible arrival to and latest possible departure from Wuhan. For cases without history of travel to Wuhan but with assumed exposure to an infectious person, the interval of possible SARS-CoV-2 exposure was defined as the maximum possible interval of exposure to the infectious person, including time before the infectious person was symptomatic. We allowed for the possibility of continued exposure within known clusters (for example, families traveling together) when the ordering of transmission was unclear. We assumed that exposure always preceded symptom onset. If we were unable to determine the latest exposure time from the available case report, we defined the upper bound of the exposure interval to be the latest possible time of symptom onset. When the earliest possible time of exposure could not be determined, we defined it as 1 December 2019, the date of symptom onset in the first known case (1); we performed a sensitivity analysis for the selection of this universal lower bound. When the earliest possible time of symptom onset could not be determined, we assumed it to be the earliest time of possible exposure. When the latest time of possible symptom onset could not be determined, we assumed it to be the latest time of possible case detection. Data on age, sex, country of residence, and possible exposure route were also collected.

Statistical Analysis

Cases were included in the analysis if we had information on the interval of exposure to SARS-CoV-2 and symptom onset. We estimated the incubation time using a previously described parametric accelerated failure time model (13). For our primary analysis, we assumed that the incubation time follows a log-normal distribution, as seen in other acute respiratory viral infections (12). We fit the model to all observations, as well as to only cases where the patient had fever and only those detected inside or outside mainland China in subset analyses. Finally, we also fit 3 other commonly used incubation period distributions (gamma, Weibull, and Erlang). We estimated median incubation time and important quantiles (2.5th, 25th, 75th, and 97.5th percentiles) along with their bootstrapped CIs for each model.
Using these estimates of the incubation period, we quantified the expected number of undetected symptomatic cases in an active monitoring program, adapting a method detailed by Reich and colleagues (14). We accounted for varying durations of the active monitoring program (1 to 28 days) and individual risk for symptomatic infection (low risk: 1-in-10 000 chance of infection; medium risk: 1-in-1000 chance; high risk: 1-in-100 chance; infected: 1-in-1 chance). For each bootstrapped set of parameter estimates from the log-normal model, we calculated the probability of a symptomatic infection developing after an active monitoring program of a given length for a given risk level. This model conservatively assumes that persons are exposed to SARS-CoV-2 immediately before the active monitoring program and assumes perfect ascertainment of symptomatic cases that develop under active monitoring. We report the mean and 99th percentile of the expected number of undetected symptomatic cases for each active monitoring scenario.
All estimates are based on persons who developed symptoms, and this work makes no inferences about asymptomatic infection with SARS-CoV-2. The analyses were conducted using the coarseDataTools and activemonitr packages in the R statistical programming language, version 3.6.2 (R Foundation for Statistical Computing). All code and data are available at https://github.com/HopkinsIDD/ncov_incubation (release at time of submission at https://zenodo.org/record/3692048) (15).

Role of the Funding Source

The findings and conclusions in this manuscript are those of the authors and do not necessarily represent the views of the U.S. Centers for Disease Control and Prevention, the National Institute of Allergy and Infectious Diseases, the National Institute of General Medical Sciences, and the Alexander von Humboldt Foundation. The funders had no role in study design, data collection and analysis, preparation of the manuscript, or the decision to submit the manuscript for publication.

Results

We collected data from 181 cases with confirmed SARS-CoV-2 infection detected outside Hubei province before 24 February 2020 (Table 1). Of these, 69 (38%) were female, 108 were male (60%), and 4 (2%) were of unknown sex. The median age was 44.5 years (interquartile range, 34.0 to 55.5 years). Cases were collected from 24 countries and regions outside mainland China (n = 108) and 25 provinces within mainland China (n = 73). Most cases (n = 161) had a known recent history of travel to or residence in Wuhan; others had evidence of contact with travelers from Hubei or persons with known infection. Among those who developed symptoms in the community, the median time from symptom onset to hospitalization was 1.2 days (range, 0.2 to 29.9 days) (Figure 1).
Table 1. Characteristics of Patients With Confirmed COVID-19 Included in This Analysis (n = 181)*
Table 1. Characteristics of Patients With Confirmed COVID-19 Included in This Analysis (n = 181)*
Figure 1. SARS-CoV-2 exposure (blue), symptom onset (red), and case detection (green) times for 181 confirmed cases. Shaded regions represent the full possible time intervals for exposure, symptom onset, and case detection; points represent the midpoints of these intervals. SARS-CoV-2 = severe acute respiratory syndrome coronavirus 2.
Figure 1. SARS-CoV-2 exposure (blue), symptom onset (red), and case detection (green) times for 181 confirmed cases.
Shaded regions represent the full possible time intervals for exposure, symptom onset, and case detection; points represent the midpoints of these intervals. SARS-CoV-2 = severe acute respiratory syndrome coronavirus 2.
Fitting the log-normal model to all cases, we estimated the median incubation period of COVID-19 to be 5.1 days (CI, 4.5 to 5.8 days) (Figure 2). We estimated that fewer than 2.5% of infected persons will show symptoms within 2.2 days (CI, 1.8 to 2.9 days) of exposure, and symptom onset will occur within 11.5 days (CI, 8.2 to 15.6 days) for 97.5% of infected persons. The estimate of the dispersion parameter was 1.52 (CI, 1.32 to 1.72), and the estimated mean incubation period was 5.5 days.
Figure 2. Cumulative distribution function of the COVID-19 incubation period estimate from the log-normal model. The estimated median incubation period of COVID-19 was 5.1 days (CI, 4.5 to 5.8 days). We estimated that fewer than 2.5% of infected persons will display symptoms within 2.2 days (CI, 1.8 to 2.9 days) of exposure, whereas symptom onset will occur within 11.5 days (CI, 8.2 to 15.6 days) for 97.5% of infected persons. Horizontal bars represent the 95% CIs of the 2.5th, 50th, and 97.5th percentiles of the incubation period distribution. The estimate of the dispersion parameter is 1.52 (CI, 1.32 to 1.72). COVID-19 = coronavirus disease 2019.
Figure 2. Cumulative distribution function of the COVID-19 incubation period estimate from the log-normal model.
The estimated median incubation period of COVID-19 was 5.1 days (CI, 4.5 to 5.8 days). We estimated that fewer than 2.5% of infected persons will display symptoms within 2.2 days (CI, 1.8 to 2.9 days) of exposure, whereas symptom onset will occur within 11.5 days (CI, 8.2 to 15.6 days) for 97.5% of infected persons. Horizontal bars represent the 95% CIs of the 2.5th, 50th, and 97.5th percentiles of the incubation period distribution. The estimate of the dispersion parameter is 1.52 (CI, 1.32 to 1.72). COVID-19 = coronavirus disease 2019.
To control for possible bias from symptoms of cough or sore throat, which could have been caused by other more common pathogens, we performed the same analysis on the subset of cases with known time of fever onset (n = 99), using the time from exposure to onset of fever as the incubation time. We estimated the median incubation period to fever onset to be 5.7 days (CI, 4.9 to 6.8 days), with 2.5% of persons experiencing fever within 2.6 days (CI, 2.1 to 3.7 days) and 97.5% having fever within 12.5 days (CI, 8.2 to 17.7 days) of exposure.
Because assumptions about the occurrence of local transmission and therefore the period of possible exposure may be less firm within mainland China, we also analyzed only cases detected outside mainland China (n = 108). The median incubation period for these cases was 5.5 days (CI, 4.4 to 7.0 days), with the 95% range spanning from 2.1 (CI, 1.5 to 3.2) to 14.7 (CI, 7.4 to 22.6) days. Alternatively, persons who left mainland China may represent a subset of persons with longer incubation periods, persons who were able to travel internationally before symptom onset within China, or persons who may have chosen to delay reporting symptoms until they left China. Based on cases detected inside mainland China (n = 73), the median incubation period is 4.8 days (CI, 4.2 to 5.6 days), with a 95% range of 2.5 (CI, 1.9 to 3.5) to 9.2 (CI, 6.4 to 12.5) days. Full results of these sensitivity analyses are presented in Appendix Table 1.
Appendix Table 1. Percentiles of SARS-CoV-2 Incubation Period From Selected Sensitivity Analyses*
Appendix Table 1. Percentiles of SARS-CoV-2 Incubation Period From Selected Sensitivity Analyses*
We fit other commonly used parameterizations of the incubation period (gamma, Weibull, and Erlang distributions). The incubation period estimates for these alternate parameterizations were similar to those from the log-normal model (Appendix Table 2).
Appendix Table 2. Parameter Estimates for Various Parametric Distributions of the Incubation Period of SARS-CoV-2 Using 181 Confirmed Cases*
Appendix Table 2. Parameter Estimates for Various Parametric Distributions of the Incubation Period of SARS-CoV-2 Using 181 Confirmed Cases*
Given these estimates of the incubation period, we predicted the number of symptomatic infections we would expect to miss over the course of an active monitoring program. We classified persons as being at high risk if they have a 1-in-100 chance of developing a symptomatic infection after exposure. For an active monitoring program lasting 7 days, the expected number of symptomatic infections missed for every 10 000 high-risk persons monitored is 21.2 (99th percentile, 36.5) (Table 2 and Figure 3). After 14 days, it is highly unlikely that further symptomatic infections would be undetected among high-risk persons (mean, 1.0 undetected infections per 10 000 persons [99th percentile, 4.8]). However, substantial uncertainty remains in the classification of persons as being at “high,” “medium,” or “low” risk for being symptomatic, and this method does not consider the role of asymptomatic infection. We have created an application to estimate the proportion of missed COVID-19 cases across any active monitoring duration up to 100 days and various population risk levels (16).
Table 2. Expected Number of Symptomatic SARS-CoV-2 Infections That Would Be Undetected During Active Monitoring, Given Varying Monitoring Durations and Risks for Symptomatic Infection After Exposure*
Table 2. Expected Number of Symptomatic SARS-CoV-2 Infections That Would Be Undetected During Active Monitoring, Given Varying Monitoring Durations and Risks for Symptomatic Infection After Exposure*
Figure 3. Proportion of known symptomatic SARS-CoV-2 infections that have yet to develop symptoms, by number of days since infection, using bootstrapped estimates from a log-normal accelerated failure time model. The solid line represents the mean estimate, the dashed line represents the 99th percentile estimate, and the dotted line represents the first percentile estimate. See Table 2 for exact estimates at various time points and at different levels of population risk for symptomatic infection. SARS-CoV-2 = severe acute respiratory syndrome coronavirus 2.
Figure 3. Proportion of known symptomatic SARS-CoV-2 infections that have yet to develop symptoms, by number of days since infection, using bootstrapped estimates from a log-normal accelerated failure time model.
The solid line represents the mean estimate, the dashed line represents the 99th percentile estimate, and the dotted line represents the first percentile estimate. See Table 2 for exact estimates at various time points and at different levels of population risk for symptomatic infection. SARS-CoV-2 = severe acute respiratory syndrome coronavirus 2.

Discussion

We present estimates of the incubation period for the novel coronavirus disease (COVID-19) that emerged in Wuhan, Hubei province, China, in 2019. We estimated the median incubation period of COVID-19 to be 5.1 days and expect that nearly all infected persons who have symptoms will do so within 12 days of infection. We found that the current period of active monitoring recommended by the U.S. Centers for Disease Control and Prevention (14 days) is well supported by the evidence (6). Symptomatic disease is frequently associated with transmissibility of a pathogen. However, given recent evidence of SARS-CoV-2 transmission by mildly symptomatic and asymptomatic persons (17, 18), we note that time from exposure to onset of infectiousness (latent period) may be shorter than the incubation period estimated here, with important implications for transmission dynamics.
Our results are broadly consistent with other estimates of the incubation period (1, 7–9). Our analysis, which was based on 181 confirmed COVID-19 cases, made more conservative assumptions about the possible window of symptom onset and the potential for continued exposure through transmission clusters outside Wuhan. Of note, the use of fixed times of symptom onset, as used in 3 of the 4 prior analyses, will truncate the incubation period distribution by either decreasing the maximum possible incubation period (if the earliest possible time of symptom onset is used) or increasing the minimum possible incubation period (if the midpoint or latest possible time of symptom onset is used). Therefore, using a symptom onset window more accurately accounts for the full distribution of possible incubation periods.
Although our results support current proposals for the length of quarantine or active monitoring of persons potentially exposed to SARS-CoV-2, longer monitoring periods might be justified in extreme cases. Among those who are infected and will develop symptoms, we expect 101 in 10 000 (99th percentile, 482) will do so after the end of a 14-day monitoring period (Table 2 and Figure 3), and our analyses do not preclude this estimate from being higher. Although it is essential to weigh the costs of extending active monitoring or quarantine against the potential or perceived costs of failing to identify a symptomatic case, there may be high-risk scenarios (for example, a health care worker who cared for a COVID-19 patient while not wearing personal protective equipment) where it could be prudent to extend the period of active monitoring.
This analysis has several important limitations. Our data include early case reports, with associated uncertainty in the intervals of exposure and symptom onset. We have used conservative bounds of possible exposure and symptom onset where exact times were not known, but there may be further inaccuracy in these data that we have not considered. We have exclusively considered reported, confirmed cases of COVID-19, which may overrepresent hospitalized persons and others with severe symptoms, although we note that the proportion of mild cases detected has increased as surveillance and monitoring systems have been strengthened. The incubation period for these severe cases may differ from that of less severe or subclinical infections and is not typically an applicable measure for those with asymptomatic infections.
Our model assumes a constant risk for SARS-CoV-2 infection in Wuhan from 1 December 2019 to 30 January 2020, based on the date of symptom onset of the first known case and the last known possible exposure within Wuhan in our data set. This is a simplification of infection risk, given that the outbreak has shifted from a likely common-source outbreak associated with a seafood market to human-to-human transmission. Moreover, phylogenetic analysis of 38 SARS-CoV-2 genomes suggests that the virus may have been circulating before December 2019 (19). To test the sensitivity of our estimates to that assumption, we performed an analysis where cases with unknown lower bounds on exposure were set to 1 December 2018, a full year earlier than in our primary analysis. Changing this assumption had little effect on the estimates of the median (0.2 day longer than for the overall estimate) and the 97.5th quantile (0.1 day longer) of the incubation period. In data sets such as ours, where we have adequate observations with well-defined minimum and maximum possible incubation periods for many cases, extending the universal lower bound has little bearing on the overall estimates.
This work provides additional evidence for a median incubation period for COVID-19 of approximately 5 days, similar to SARS. Assuming infection occurs at the initiation of monitoring, our estimates suggest that 101 out of every 10 000 cases will develop symptoms after 14 days of active monitoring or quarantine. Whether this rate is acceptable depends on the expected risk for infection in the population being monitored and considered judgment about the cost of missing cases (14). Combining these judgments with the estimates presented here can help public health officials to set rational and evidence-based COVID-19 control policies.

References

1.
Huang CWang YLi Xet al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020;395:497-506. [PMID: 31986264]  doi: 10.1016/S0140-6736(20)30183-5
2.
Zhu NZhang DWang Wet alChina Novel Coronavirus Investigating and Research Team, A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med. 2020;382:727-733. [PMID: 31978945]  doi: 10.1056/NEJMoa2001017
3.
The Novel Coronavirus Pneumonia Emergency Response Epidemiology Team. The Epidemiological Characteristics of an Outbreak of 2019 Novel Coronavirus Diseases (COVID-19)—China, 2020. China CDC Weekly. 2020;2:113-22.
4.
World Health Organization. Coronavirus disease 2019 (COVID-19): Situation Report – 38. 27 February 2020. Accessed at www.who.int/docs/default-source/coronaviruse/situation-reports/20200227-sitrep-38-covid-19.pdf?sfvrsn=9f98940c_2 on 28 February 2020.
5.
World Health Organization. Statement on the second meeting of the International Health Regulations (2005) Emergency Committee regarding the outbreak of novel coronavirus (2019-nCoV). 30 January 2020. Accessed at www.who.int/news-room/detail/30-01-2020-statement-on-the-second-meeting-of-the-international-health-regulations-(2005)-emergency-committee-regarding-the-outbreak-of-novel-coronavirus-(2019-ncov) on 31 January 2020.
6.
The White House. Press Briefing by Members of the President's Coronavirus Task Force. 31 January 2020. Accessed at www.whitehouse.gov/briefings-statements/press-briefing-members-presidents-coronavirus-task-force on 1 February 2020.
7.
Backer JAKlinkenberg DWallinga J. Incubation period of 2019 novel coronavirus (2019-nCoV) infections among travellers from Wuhan, China, 20–28 January 2020. Euro Surveill. 2020;25. [PMID: 32046819]  doi: 10.2807/1560-7917.ES.2020.25.5.2000062
8.
Linton NMKobayashi TYang Yet al. Incubation period and other epidemiological characteristics of 2019 novel coronavirus infections with right truncation: a statistical analysis of publicly available case data. J Clin Med. 2020;9. [PMID: 32079150]  doi: 10.3390/jcm9020538
9.
Li QGuan XWu Pet al. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med. 2020. [PMID: 31995857]  doi: 10.1056/NEJMoa2001316
10.
Varia MWilson SSarwal Set alHospital Outbreak Investigation Team. Investigation of a nosocomial outbreak of severe acute respiratory syndrome (SARS) in Toronto, Canada. CMAJ. 2003;169:285-92. [PMID: 12925421]
11.
Virlogeux VFang VJPark Met al. Comparison of incubation period distribution of human infections with MERS-CoV in South Korea and Saudi Arabia. Sci Rep. 2016;6:35839. [PMID: 27775012]  doi: 10.1038/srep35839
12.
Lessler JReich NGBrookmeyer Ret al. Incubation periods of acute respiratory viral infections: a systematic review. Lancet Infect Dis. 2009;9:291-300. [PMID: 19393959]  doi: 10.1016/S1473-3099(09)70069-6
13.
Reich NGLessler JCummings DAet al. Estimating incubation period distributions with coarse data. Stat Med. 2009;28:2769-84. [PMID: 19598148]  doi: 10.1002/sim.3659
14.
Reich NGLessler JVarma JKet al. Quantifying the risk and cost of active monitoring for infectious diseases. Sci Rep. 2018;8:1093. [PMID: 29348656]  doi: 10.1038/s41598-018-19406-x
15.
Lauer SA, Grantz KH, Bi Q, et al. Estimating the incubation time of the novel coronavirus (COVID-19) based on publicly reported cases using coarse data tools. 2020. Accessed at https://github.com/HopkinsIDD/ncov_incubation on 3 March 2020.
16.
Determining Durations for Active Monitoring. Accessed at https://iddynamics.jhsph.edu/apps/shiny/activemonitr on 28 February 2020.
17.
Chan JFYuan SKok KHet al. A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster. Lancet. 2020;395:514-523. [PMID: 31986261]  doi: 10.1016/S0140-6736(20)30154-9
18.
Rothe CSchunk MSothmann Pet al. Transmission of 2019-nCoV infection from an asymptomatic contact in Germany [Letter]. N Engl J Med. 2020. [PMID: 32003551]  doi: 10.1056/NEJMc2001468
19.
Genomic epidemiology of novel coronavirus (HCoV-19). 2020. Accessed at https://nextstrain.org/ncov on 29 January 2020.

Comments

0 Comments
Sign In to Submit A Comment
Qingyuan Zhao, PhD, Nianqiao Ju, Sergio Bacallado, PhD, Rajen Shah, PhD4 May 2020
Great caution needed to use the Lauer study in the guidance of health policies
(This is a request to replace our earlier comment by the following, thank you.)

In a paper published by the Annals of Internal Medicine, Lauer et al. [3] (“Lauer study” below) estimated the incubation period of the coronavirus disease (COVID-19) using confirmed cases outside Wuhan, China. Their results are being used by the Centers for Disease Control and Prevention (CDC) in guidelines for management of confirmed COVID-19 patients [2].

When analyzing emerging epidemic outbreaks using limited observation data, there are many forms of biases in the estimation of basic epidemiological parameters [1]. Although the Lauer study acknowledged some potential limitations

in their discussion, they failed to give adequate warnings about the potential magnitude of selection bias in their analysis.

We believe there are at least three major sources of biases in this study:

1. Ignoring the epidemic growth. Among the 181 cases used by the Lauer study, the majority (161) were residents of the Hubei province or had known travel to Wuhan. A crucial assumption, as acknowledged by the authors, is a “constant risk for SARS-CoV-2 infection in Wuhan from 1 December 2019 to 30 January 2020”. However, the epidemic was almost certainly growing rapidly in Wuhan before its lock-down.

2. Right-truncation. Although in the abstract the authors stated that the COVID-19 cases they used were confirmed between 4 January 2020 and 24 February 2020, a closer examination of their dataset reveals that only 4 cases were confirmed in February. Therefore, most of the cases who left Wuhan days before its lock-down on 23 January and had longer incubation periods were not confirmed by the end of January and were not included in the dataset.

3. Non-random sample selection. The Lauer study did not give a clear description of how they decided which cases were included in their dataset and which were excluded. In their discussion they mentioned that cases in their dataset may under-represent mild symptoms, but they did not comment on whether their dataset may even adequately represent the more severe cases (or any other well-defined population).

Any of the above issues could incur severe bias. By analyzing a carefully constructed dataset with a generative statistical model, we found that these issues indeed lead to substantial biases, and 5% of the symptomatic cases may develop symptoms after 14 days since infection. Our study is available as a preprint on arXiv [4], has its own limitations, and has yet to undergo rigorous peer review. Nevertheless, the unaccounted and under-acknowledged biases in the Lauer study suggest that their results should be used with great caution in the guidance of health policies.

[1] T. Britton and G. Scalia Tomba. Estimation in emerging epidemics: Biases and remedies. Journal of the Royal Society Interface, 16(150):20180670, 2019.
[2] Centers for Disease Control and Prevention. Interim clinical guidance for management of patients with confirmed coronavirus dis-
ease (COVID-19). https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html. Retrieved: April 15, 2020.
[3] S. A. Lauer, K. H. Grantz, Q. Bi, F. K. Jones, Q. Zheng, H. R. Meredith, A. S. Azman, N. G. Reich, and J. Lessler. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Annals of Internal Medicine, 2020.
[4] Q. Zhao, N. Ju, and S. Bacallado. BETS: The dangers of selection bias in early analyses of the coronavirus disease (COVID-19) pandemic, 2020. arXiv: 2004.07743
Ari B Cuperfain9 April 2020
14-Day Quarantine, Incubation Period, and Asymptomatic Transmission of COVID-19
Lauer, Grantz and colleagues’ study on the incubation period for COVID-19 tracked early cases originating in travelers from Wuhan, China and found that 101 out of 10 000 cases would develop symptoms of infection after 14 days of exposure (1). They concluded that their results support the 14-day quarantine, noting that this period may be extended in “extreme cases”. Much has changed about our understanding of COVID-19, including the increasingly appreciated role of asymptomatic infection in disease transmission (2). However, the recommended 14-day length of quarantine instituted in many jurisdictions for individuals with actual or potential exposure to the virus has remained virtually constant.

Little is known about asymptomatic cases, or the corresponding dynamics of transmission, but some studies provide at least minimal guidance. The proportion of asymptomatic infections is estimated at 15-20% based on observations from passengers on the Diamond Princess cruise ship (3). The period of communicability in asymptomatic cases is challenging to predict. In one cohort (of symptomatic patient) the median length of illness from onset of symptoms until a negative viral test was 10.5 days, with 2.5 days occurring after resolution of symptoms (4). Thus, assuming a period of communicability of 3-10 days, the proportion of potentially transmissible asymptomatic individuals at the end of 14 days in quarantine is 0.4% on the lower end, and up to 15% on the higher end, by considering a 75% chance of developing COVID-19 on or after day 4 of quarantine, with 20% of those potentially unknowingly communicable ten days later. Moreover, the risk of transmission from asymptomatic cases would be additive to those who have not yet shown symptoms as described by Lauer, Grantz et al.

The authors’ study design did not intend to capture the risk of asymptomatic transmission, and less was known about this risk at the time the study was conducted. Indeed, at the time this study was published online, the World Health Organization had not yet declared COVID-19 a pandemic. But given the emergence of new data, the study’s conclusion is misleading. Perhaps a more accurate take-away is that, after 14 days, the number of individuals who are neither showing symptoms, nor asymptomatically infected, is low. This nuanced distinction may hold significant implications in determining length of quarantine and/or the utility of testing those at high-risk of contracting the virus even in the absence of symptoms.

1. Lauer SA, Grantz KH, Bi Q, Jones FK, Zheng Q, Meredith HR, et al. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Annals of internal medicine. 2020.
2. Bai Y, Yao L, Wei T, Tian F, Jin D-Y, Chen L, et al. Presumed asymptomatic carrier transmission of COVID-19. Jama. 2020.
3. Mizumoto K, Kagaya K, Zarebski A, Chowell G. Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship, Yokohama, Japan, 2020. Eurosurveillance. 2020;25(10).
4. Chang D, Mo G, Yuan X, Tao Y, Peng X, Wang F, et al. Time Kinetics of Viral Clearance and Resolution of Symptoms in Novel Coronavirus Infection. American Journal of Respiratory and Critical Care Medicine. 2020(ja).
Cynthia L. Gong, PharmD, PhD,1 Nadine K. Zawadzki, MPH,2 Roy S. Zawadzki,3 Sang K. Cho, PharmD, MPH, PhD,2 Joel W. Hay, PhD223 March 2020
Optimizing Policy in Response to COVID-19
Lauer et al. report the infection rates of COVID-19, with a median incubation time of 5.1 days (95% CI: 4.5-5.8). Though COVID-19 has been declared a global pandemic, social distancing measures to reduce infections will likely have a far greater cost in economic productivity and welfare than this virus will. Extreme, hasty lockdown measures in several countries (e.g. United States, Italy, and China) have effectively halted global, regional, and local economies. Officials have attributed the slowing epidemic in China and elsewhere to such lockdown measures; yet in South Korea, where widespread testing has been most successfully implemented, cases have slowed through traditional case-contact tracing rather than social distancing, which is not effective unless implemented immediately and aggressively.

We estimated country-specific logistic growth curves as functions of time from Day Zero (when the cumulative number of reported cases surpasses 100)2 using linear models and ridge regression with leave-one-out cross-validation, estimating Hubei Province separately from the rest of China. Cumulative case curves are similarly shaped among all countries with growth rates lying between those for Hubei and South Korea. While case and fatality reporting in different countries may be subject to differential reporting biases, they are unlikely time-varying. All countries reporting cases 25+ days after Day Zero have demonstrated rapid declines in new cases by Day 25, with growth rates in all countries declining by Day 15. (https://rpubs.com/nzawadzki/covid19-by-country, https://drive.google.com/file/d/1LX2IamvOebtg7Xan9nIvKJ-zVAqbVQlr/view?usp=sharing).

Each 1% increase in unemployment corresponds to ~4000 additional deaths per year. Based on these curves, global cumulative cases are unlikely to exceed 500,000 or deaths 50,000, regardless of measures taken. Financial markets have lost >$15 trillion, resulting in more lives lost from panic and economic disruption than gained from social distancing and economy shutdown, at a cost of ~$300 million per life.

References
Lauer SA, Grantz KH, Bi Q, Jones FK, Zheng Q, Meredith HR, et al. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Ann Intern Med. 2020.

Kuhn A. South Korea's Drive-Through Testing For Coronavirus Is Fast — And Free. Online: NPR; 2020. Available: https://www.npr.org/sections/goatsandsoda/2020/03/13/815441078/south-koreas-drive-through-testing-for-coronavirus-is-fast-and-free

Maharaj S, Kleczkowski A. Controlling epidemic spread by social distancing: do it well or not at all. BMC Public Health. 2012;12:679.

Hastie T. cv.glmnet - Cross-Validation for Glmnet. RDocumentation. Online: R. Available: https://www.rdocumentation.org/packages/glmnet/versions/3.0-2/topics/cv.glmnet

Roelfs DJ, Shor E, Davidson KW, Schwartz JE. Losing life and livelihood: a systematic review and meta-analysis of unemployment and all-cause mortality. Soc Sci Med. 2011;72(6):840-54.
R Matthew Chico MPH PhD, Nguyen Tien Huy MD PhD17 March 2020
COVID-19 symptoms and viral shedding: Implications for testing and self-Isolation
In 10 weeks of time, COVID-19 has spread from the Hubei Province in Central China to 110 countries, prompting the World Health Organization to declare the outbreak a global pandemic in an acknowledgment that the novel coronavirus threatens to impact all sectors of every country in the world.
Lauer and colleagues contributed importantly to our nascent understanding of COVID-19 in their recent Annals publication regarding the incubation of symptomatic infections. The mean period is 5.1 days (95% CI, 4.5 to 5.8 days), an estimate based on analyses of 181 confirmed cases spanning 24 countries and regions outside mainland China, and 25 provinces within mainland China (1). Thus, under the best of circumstances, the average person will seek diagnosis in five days of exposure, with results available two days later.
Wölfel and colleagues have added an equally valuable piece to the COVID-19 puzzle in a non-peer-reviewed report (2). Nine patients in Germany, for whom the time of exposure to an index case of SARS-CoV-2 was known, provided biological samples for virology testing. Viral loads in the upper respiratory tract were detected among the first samples collected 48 hours from the onset of symptoms and peaked before day five. Viral concentrations were 1000 times higher than those observed in Hong Kong during the 2004 outbreak of SARS, a related coronavirus (3). Compared to SARS, the viral concentrations of COVID-19 have made contact tracing difficult, particularly in Western countries with highly itinerant populations. Regardless, diagnostic testing remains vital to estimating COVID-19 case-fatality rates, identifying and responding to emerging hotspots, and tailoring medical care at the individual level.
Within this context there is now a clear rationale for prioritizing self-isolation of the elderly and immunocompromised for an extended period of time, and promoting social distancing behavior for everyone else. Whether or not lower-risk population groups exposed to SARS-CoV-2 infection will end up developing related IgG, IgM, IgA antibodies and reduce the risk of exposure to groups most at risk of mortality, individuals who acquire immunological defenses will be better able to provide social support to the masses in self-isolation. These are among the most warranted measures now as we bide our time for a deployable COVID-19 vaccine 12 to 18 months from now.

References
1. Lauer SA, Grantz KH, Bi Q, Jones FK, Zheng Q, Meredith HR, et al. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Annals of Internal Medicine. 2020.
2. Woelfel R, Corman VM, Guggemos W, Seilmaier M, Zange S, Mueller MA, et al. Clinical presentation and virological assessment of hospitalized cases of coronavirus disease 2019 in a travel-associated transmission cluster. medRxiv. 2020:2020.03.05.20030502.
3. Poon LL, Chan KH, Wong OK, Cheung TK, Ng I, Zheng B, et al. Detection of SARS coronavirus in patients with severe acute respiratory syndrome by conventional and real-time quantitative reverse transcription-PCR assays. Clinical chemistry. 2004;50(1):67-72.

Disclosures: None

Richard M Fleming, PhD, MD, JD; Matthew R Fleming, BS, NRP; Tapan K Chaudhuri, MD10 March 2020
The greatest thing we have to fear from CoVid-19.
This coronavirus is behaving exactly like a virus. By that we mean it is transmitted in specific ways: it attaches to certain tissues within the body and its goal is not to kill its host but to survive within the host.

If we knew nothing more, that should be enough to calm our fears and promote cooperation among all of us: that should be enough to halt the run on store products and stop the price gouging of tissue, hand cleaner and other items; and while calmly promoting restoration of the stock market, the world economy, and our countries.

Why?

Because this is not some unknown enemy attacking us that we don’t know how to deal with.

This virus is transmitted by people sneezing or coughing on you. Masks are for those people who are coughing and sneezing—for them to wear to reduce their coughing or sneezing their virus upon you—not for you to wear when you’re not the one coughing or sneezing. This behavior of everyone wearing masks doesn’t stop the spread; in fact it may increase the potential for warm moist areas for the virus to survive and it promotes unnecessary fear.

(1) A major method of spreading the virus includes touching your face with your hands and then spreading the virus by touching others, as well as increasing the likelihood of further infecting yourself with more of the virus. As Ignaz Semmelweis demonstrated more than 150 years ago, hand washing (hand soap) dramatically reduces the transmission of pathogens from person to person.

(2) The virus attaches itself to the lungs and GI track, where IgA is primarily responsible for addressing immunologic responses. This means we know what to look for and what to treat, allowing those who are sickest to be best-taken care of. This is why we see the elderly immune-compromised and those with heart and lung problems most susceptible.

(3) Viruses don’t try to kill their host. If they were successful at that, it would prevent them from reproducing themselves and surviving.

CoVid-19 also presents us with the opportunity to learn and potentially develop new treatments for IgA disorders, CAD, and cancer.

This is not some unknown invader which we need to fear. This is a virus with all the limitations of a virus—not a zombie apocalypse.

Disclosures: No COI to declare.

Alessandro Testori22 May 2020
QUARANTINE AND VIRUS CLEARANCE IN COVID-19

I am writing in response to an article which appeared on the May 5, 2020 issue of Annals of Internal Medicine, reporting that 101 out of every 10,000 cases of Covid-19 will develop symptoms after 14 days of quarantine (1). These data suggest that a 14-day quarantine is 99% effective, but 1% of cases would be missed and could still spread the infection after completing the quarantine. Complicating things is the fact that 15% to 20% of infections are asymptomatic (2). I believe the way to solve this dilemma is to test the patient for coronavirus before the quarantine is lifted. After successful recovery from the illness it is vital to confirm clearance of the virus.

Patients should be tested with two separate nasal swabs performed at least 24 hours apart. A positive test should trigger repeat testing at 7 days interval, until the test becomes negative. This procedure should be also followed for patient that are completing the quarantine. The 1% of patients that remains positive for coronavirus would continue quarantine for 7 days, at which point they will be tested again until they clear the virus and the quarantine will be stopped only if they clear the virus.

1. Lauer S, Grantz K, Bi Q, Jones F, et al. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Ann. Intern. Med. 2020;172(9):577-582. doi: 10.7326/M20-0504.

2. Mizumoto K, Kagaya K, Zarebski A, Chowell G. Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship, Yokohama, Japan, 2020. Eurosurveillance. 2020;25(10).

Information & Authors

Information

Published In

cover image Annals of Internal Medicine
Annals of Internal Medicine
Volume 172Number 95 May 2020
Pages: 577 - 582

History

Published online: 10 March 2020
Published in issue: 5 May 2020

Keywords

Authors

Affiliations

Stephen A. Lauer, MS, PhD
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Kyra H. Grantz, BA
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Qifang Bi, MHS
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Forrest K. Jones, MPH
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Qulu Zheng, MHS
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Hannah R. Meredith, PhD
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Andrew S. Azman, PhD
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Nicholas G. Reich, PhD
School of Public Health and Health Sciences, University of Massachusetts, Amherst, Massachusetts, and Ludwig-Maximilians-Universität, Munich, Germany (N.G.R.)
Justin Lessler, PhD
Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (S.A.L., K.H.G., Q.B., F.K.J., Q.Z., H.R.M., A.S.A., J.L.)
Acknowledgment: The authors thank all who have collected, prepared, and shared data throughout this outbreak. They are particularly grateful to Dr. Kaiyuan Sun, Ms. Jenny Chen, and Dr. Cecile Viboud from the Division of International Epidemiology and Population Studies, Fogarty International Center, National Institutes of Health; Dr. Moritz Kraemer and the open COVID-19 data working group; and the Johns Hopkins Center for Systems Science and Engineering.
Grant Support: By the U.S. Centers for Disease Control and Prevention (NU2GGH002000), the National Institute of Allergy and Infectious Diseases (R01 AI135115), the National Institute of General Medical Sciences (R35 GM119582), and the Alexander von Humboldt Foundation.
Disclosures: Dr. Lauer reports grants from the National Institute of Allergy and Infectious Diseases and the U.S. Centers for Disease Control and Prevention during the conduct of the study. Ms. Grantz reports a grant from the U.S. Centers for Disease Control and Prevention during the conduct of the study. Dr. Reich reports grants from the National Institute of General Medical Sciences and the Alexander von Humboldt Foundation during the conduct of the study. Dr. Lessler reports a grant from the U.S. Centers for Disease Control and Prevention during the conduct of the study. Authors not named here have disclosed no conflicts of interest. Disclosures can also be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M20-0504.
Editors' Disclosures: Christine Laine, MD, MPH, Editor in Chief, reports that her spouse has stock options/holdings with Targeted Diagnostics and Therapeutics. Darren B. Taichman, MD, PhD, Executive Editor, reports that he has no financial relationships or interests to disclose. Cynthia D. Mulrow, MD, MSc, Senior Deputy Editor, reports that she has no relationships or interests to disclose. Eliseo Guallar, MD, MPH, DrPH, Deputy Editor, Statistics, reports that he has no financial relationships or interests to disclose. Jaya K. Rao, MD, MHS, Deputy Editor, reports that she has stock holdings/options in Eli Lilly and Pfizer. Christina C. Wee, MD, MPH, Deputy Editor, reports employment with Beth Israel Deaconess Medical Center. Sankey V. Williams, MD, Deputy Editor, reports that he has no financial relationships or interests to disclose. Yu-Xiao Yang, MD, MSCE, Deputy Editor, reports that he has no financial relationships or interest to disclose.
Reproducible Research Statement: Study protocol: Not applicable. Statistical code and data set: Available at https://github.com/HopkinsIDD/ncov_incubation.
Corresponding Author: Justin Lessler, PhD, Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, 615 North Wolfe Street, Baltimore, MD 21205; e-mail, [email protected].
Previous Posting: This manuscript was posted as a preprint on medRxiv on 4 February 2020. doi:10.1101/2020.02.02.20020016
Current Author Addresses: Drs. Lauer, Meredith, and Lessler; Ms. Grantz; Ms. Bi; Mr. Jones; and Ms. Zheng: Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, 615 North Wolfe Street, Baltimore, MD 21205.
Dr. Azman: Médecins Sans Frontières, Rue de Lausanne 72, 1202 Genève, Switzerland.
Dr. Reich: Department of Biostatistics and Epidemiology, Amherst School of Public Health and Health Sciences, University of Massachusetts, 715 North Pleasant Street, Amherst, MA 01003-9304.
Author Contributions: Conception and design: S.A. Lauer, K.H. Grantz, F.K. Jones, N.G. Reich, J. Lessler.
Analysis and interpretation of the data: S.A. Lauer, K.H. Grantz, Q. Bi, F.K. Jones, N.G. Reich, J. Lessler.
Drafting of the article: S.A. Lauer, K.H. Grantz, Q. Bi, F.K. Jones, A.S. Azman, N.G. Reich.
Critical revision of the article for important intellectual content: Q. Bi, F.K. Jones, A.S. Azman, N.G. Reich, J. Lessler.
Final approval of the article: S.A. Lauer, K.H. Grantz, Q. Bi, F.K. Jones, Q. Zheng, H.R. Meredith, A.S. Azman, N.G. Reich, J. Lessler.
Statistical expertise: Q. Bi, N.G. Reich, J. Lessler.
Collection and assembly of data: S.A. Lauer, K.H. Grantz, Q. Bi, F.K. Jones, Q. Zheng, H.R. Meredith.
This article was published at Annals.org on 10 March 2020.
* Dr. Lauer and Ms. Grantz share first authorship.

Metrics & Citations

Metrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Format





Download article citation data for:
Stephen A. Lauer, Kyra H. Grantz, Qifang Bi, et al. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Ann Intern Med.2020;172:577-582. [Epub 10 March 2020]. doi:10.7326/M20-0504

View Options

View options

PDF/ePub

View PDF/ePub

Get Access

Login Options:
Purchase

You will be redirected to acponline.org to sign-in to Annals to complete your purchase.

Create your Free Account

You will be redirected to acponline.org to create an account that will provide access to Annals.

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media