If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Quantifying the improvement in confirmation efficiency of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) during the early phase of the outbreak in Hong Kong in 2020
1 JR and SZ contribute equally, and thus they are joint first authors.
Affiliations
JC School of Public Health and Primary Care, Chinese University of Hong Kong, Hong Kong, ChinaShenzhen Research Institute of Chinese University of Hong Kong, Shenzhen, China
JC School of Public Health and Primary Care, Chinese University of Hong Kong, Hong Kong, ChinaShenzhen Research Institute of Chinese University of Hong Kong, Shenzhen, China
JC School of Public Health and Primary Care, Chinese University of Hong Kong, Hong Kong, ChinaShenzhen Research Institute of Chinese University of Hong Kong, Shenzhen, China
Since the first imported case was confirmed in Hong Kong on January 23, 2020, local public interventions for SARS-CoV-2 control and prevention have been enhanced.
•
A regression model is adopted to quantify the association between the confirmation delay and the calendar time, validated by a Cox proportional hazard model.
•
We estimate that the SARS-CoV-2 confirmation efficiency improved 5.40% (95%CI: 2.54% − 8.33%) per day in Hong Kong.
•
The sustaining enforcement in timely confirmation and other effective control measures are recommended to prevent the spreading of the SARS-CoV-2 infection.
Abstract
Backgrounds
The emerging virus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), caused a large outbreak of coronavirus disease, COVID-19, in Wuhan, China, since December 2019. COVID-19 soon spread to other regions of China and overseas. In Hong Kong, local mitigation measures were implemented since the first imported case was confirmed on January 23, 2020. Here we evaluated the temporal variation of detection delay from symptoms onset to laboratory confirmation of SARS-CoV-2 in Hong Kong.
Methods
A regression model is adopted to quantify the association between the SARS-CoV-2 detection delay and calendar time. The association is tested and further validated by a Cox proportional hazard model.
Findings
The estimated median detection delay was 9.5 days (95%CI: 6.5 − 11.5) in the second half of January, reduced to 6.0 days (95%CI: 5.5 − 9.5) in the first half of February 2020. We estimate that SARS-CoV-2 detection efficiency improved at a daily rate of 5.40% (95%CI: 2.54 − 8.33) in Hong Kong.
Conclusions
The detection efficiency of SARS-CoV-2 was likely being improved substantially in Hong Kong since the first imported case was detected. Sustaining enforcement in timely detection and other effective control measures are recommended to prevent the SARS-CoV-2 infection.
The deadly coronavirus disease (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2, formerly 2019-nCoV), has emerged in Wuhan, China, in December 2019 (
’Pneumonia of unknown cause – China,’ Emergencies preparedness, response, Disease outbreak news, World Health Organization (WHO) [https://www.who.int/csr/don/05-january-2020-pneumonia-of-unkown-cause-china/en/]
). The first imported cases in Hong Kong were confirmed and reported on January 23, 2020, see Fig. 1(a) . Since then, the government of Hong Kong has implemented a series of control and prevention measures for COVID-19, including enhanced border screening and traffic restrictions (
Summary of data and outbreak situation of the Severe Respiratory Disease associated with a Novel Infectious Agent, Centre for Health Protection, the government of Hong Kong. [https://www.chp.gov.hk/en/features/102465.html]
). The official news related to the COVID-19 outbreak was first released by the CHP on December 31, 2019. Outbreak-related news reports have been released daily since January 2 and updated at least twice per day since January 23, 2020 (
). After the first week of February, COVID-19 was controlled at 1 incidence onset from February 9 to the present (as of February 15), 2020.
Fig. 1The daily SARS-CoV-2 cases time series in (a), and count of tested cases (b) and the association between detection delay (τ) and time in (c). Panel (a) shows the time series of the daily number of SARS-CoV-2 cases aggregated by the date of onset (grey horizontal bars) and by the laboratory-confirmed date (black vertical bars). Panel (b) shows the time series of the daily number of tested cases. Panel (c) shows the detection delays (τ) of the cases who are Hong Kong residence (blue dots) and of the cases who are not Hong Kong residence (red triangles) vary over the calendar date. The bold curves are the fitting results of the regression in Eqn (1), and the dashed curves are the 95% confidence interval (95%CI).
Summary of data and outbreak situation of the Severe Respiratory Disease associated with a Novel Infectious Agent, Centre for Health Protection, the government of Hong Kong. [https://www.chp.gov.hk/en/features/102465.html]
Laboratory testing for 2019 novel coronavirus (2019-nCoV) in suspected human cases, World Health Organization (WHO) [https://www.who.int/health-topics/coronavirus/laboratory-diagnostics-for-novel-coronavirus]
The case detection efficiency is the reciprocal of detection delay, denoted by τ. The detection delay (τ) is measured by the time interval between the date of COVID-19 symptoms onset and the date of laboratory confirmation. For the i-th patient, the detection delay is denoted by τi,t, where t denotes symptoms onset on the t-th day. Thus, the time evolution of the detection delay is formulated by a regression model in Eqn (1).
(1)
where E[·] represents the expectation. The ‘isHKi’ is an indicator that is 1 if the i-th patient is a Hong Kong resident but 0 otherwise. The denotes the average number of cases being tested between the t-th and (t + τ)-th days, and thus it accounts for the effect of time-varying working load. The β is the regression coefficient for the onset time variable t. The β quantifies the rate of change in the detection delay. Since a shorter detection delay means a higher detection efficiency, we expect a negative estimate of β for an improving detection efficiency over time. The [exp(− β) − 1] × 100% quantifies the percentage improvement in the detection efficiency per day in Hong Kong. We fit the model by a Gamma-distributed likelihood framework. The 95% confidence interval (95%CI) of β is calculated by using its mean estimate plus and minus its standard deviation (SD) multiplied by the Student's t quantile.
Validation analysis is conducted by using the Cox proportional hazard (PH) model defined as , which has a similar form as Eqn (1). The S(·) is the survival function, and the S0(·) is the baseline survival function. Similarly, the α is the regression coefficient for t in the hazard function, and thus exp(α) is the hazard ratio (HR). In contrast to β, we expect a positive estimate of α for an improving detection efficiency over time. For validation, we exam the signs of the estimated α and β, and their statistical significance levels.
3. Results and discussion
Fig. 1(a) shows the epidemic curve of all 56 COVID-19 cases as of February 15, 2020, in Hong Kong. We reported that the onset cases peaked on January 30, and the confirmed cases peaked on February 9, 2020. The daily number of tests started increasing rapidly from January 23 until the end of January, see Fig. 1(b). The estimated median detection delay was 9.5 days (95%CI: 6.5 − 11.5) in the second half of January, and this number was reduced to 6.0 days (95%CI: 5.5 − 9.5) in the first half of February 2020. These estimates indicate the enhancement of infection screening and public interventions for the prevention and control of SARS-CoV-2.
The fitting results of Eqn (1) are shown in Fig. 1(c); and the Nagelkerke's pseudo-R-squared is 0.38. The non-resident cases are likely to be detected earlier than the resident cases even though there were only 5 non-Hong Kong resident cases. We estimate that SARS-CoV-2 detection efficiency improves 5.40% per day (95%CI: 2.54 − 8.33) in Hong Kong. The estimates of α and β are consistent in signs, i.e., α is positive, and β is negative, and statistical significance levels, i.e., both are significant with p-values < 0.05, which validate our estimates. Therefore, we report that the SARS-CoV-2 infection detection efficiency was likely being improved substantially and significantly since January 23, 2020, when the first imported case was detected in Hong Kong.
Due to the unavailability of data, our findings are restricted to the early phase of the COVID-19 outbreak in Hong Kong. However, our analytical approach can be extended to a more complex context. In Table 1, the events and control measures during the early outbreak are summarized. Most of the control measures aimed at strengthening the testing efforts, i.e., number of tests and testing efficiency, and reducing the size of the outbreak. Though the control measures would partially affect the testing efforts, we have accounted for these effects by adjusting the term in the models. Hence, our estimates should quantify the temporal improvement in detection efficiency.
Table 1Summary of the events and control measures during the early phase of the COVID-19 outbreak in Hong Kong (HK).
Date or period
Event type
Details
January 17 -
control measure
Temperature check of all inbound travellers at borders
January 19
Fact
First COVID-19 case in HK symptom onset
January 21 -
control measure
COVID-19 reporting system in operation
January 24 -
control measure
Testing for all suspected, confirmed cases and close contacts Isolation and quarantine
January 25 -
control measure
Activate the ‘Emergency’ level of response plan against the outbreak
January 23 -
fact and control measure
Local schools’ holiday, and school closure
January 25 - January 28
Fact
Chinese lunar new year holiday
January 26 -
control measure
Cancellation of major gathering events, and closure of parks
January 27 -
control measure
Flight suspension between HK and Wuhan Boarder restriction for non-HK residents with travel history to Hubei
January 30 -
control measure
Close 6 border-control points
February 4 -
control measure
Close 4 additional border-control points
February 9 -
control measure
Quarantine for all individuals coming from mainland China
We note that the sustaining enforcement in timely detection and other effective control measures are recommended in response to reduce the risk associated with SARS-CoV-2.
Declarations
Ethics approval and consent to participate
The follow-up data of individual patients were collected via the public domain (
Summary of data and outbreak situation of the Severe Respiratory Disease associated with a Novel Infectious Agent, Centre for Health Protection, the government of Hong Kong. [https://www.chp.gov.hk/en/features/102465.html]
Summary of data and outbreak situation of the Severe Respiratory Disease associated with a Novel Infectious Agent, Centre for Health Protection, the government of Hong Kong. [https://www.chp.gov.hk/en/features/102465.html]
DH was supported by the General Research Fund (Grant Number 15205119) of the Research Grants Council (RGC) of Hong Kong, China, and an Alibaba (China) - Hong Kong Polytechnic University Collaborative Research project (0031768). WW was supported by the National Natural Science Foundation of China (Grant Number 61672013) and Huaian Key Laboratory for Infectious Diseases Control and Prevention (Grant Number HAP201704), Huaian, Jiangsu, China.
Competing Interests
DH was supported by an Alibaba-Hong Kong Polytechnic University Collaborative Research project. Other authors declared no competing interests.
Disclaimer
The funding agencies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.
Conflict of Interests
The authors declared no competing interests.
Authors’ Contributions
SZ conceived the study, carried out the analysis. JR and SZ drafted the first manuscript. All authors discussed the results, critically read and revised the manuscript, and gave final approval for publication.
Acknowledgments
None.
References
’Pneumonia of unknown cause – China,’ Emergencies preparedness, response, Disease outbreak news, World Health Organization (WHO) [https://www.who.int/csr/don/05-january-2020-pneumonia-of-unkown-cause-china/en/]
Summary of data and outbreak situation of the Severe Respiratory Disease associated with a Novel Infectious Agent, Centre for Health Protection, the government of Hong Kong. [https://www.chp.gov.hk/en/features/102465.html]
Laboratory testing for 2019 novel coronavirus (2019-nCoV) in suspected human cases, World Health Organization (WHO) [https://www.who.int/health-topics/coronavirus/laboratory-diagnostics-for-novel-coronavirus]
30321-0&id=gr1.jpg){kind=link}