Original Article

A problem of self-isolation in Japan: The relationship between self-isolation and COVID-19 community case

Introduction

The global pandemic of coronavirus disease-2019 (COVID-19) began from an outbreak at China in 2019, it spread to worldwide later. During pandemic, multiple countries have been dealing with many new cases implicated as waves of infection. This is partly due to the early relaxation of control measures and restrictions. The appearance of variant strains of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)^1,2 have also crippled effectors to abolish the disease. Indeed, the delta variant has been reported as a dangerous factor given its high transmissibility and global spread. This variant has significantly increased the mean adequate reproduction number by 97 percent.³ Countries have employed a multitude of non-pharmaceutical interventions worldwide to ascertain control over the pandemic. Movement restrictions have been put in place in various degrees across 186 countries. Full or partial lockdown was constituted in 82 countries in the past year. In the absence of definitive treatment for COVID-19, many countries started developing vaccines against COVID-19. As of April 16th, 2021, there are 88 vaccines in clinical development and 184 vaccines in pre-clinical development. A few vaccines have already obtained emergency use authorization in multiple countries.^4,5 However, the over-dependence on vaccines alone can be detrimental to areas stricken with poverty.

A recent study suggests that over a fifth of the global population (low- and middle-income countries) would not access the existing vaccines until 2022.⁵ Hence, controlling the source of infection by breaking the chains of human-to-human transmission remains the best way to manage the outbreak. An essential method of controlling the source of infection is the immediate isolation of COVID-19 patients. In some countries such as Vietnam, the government has started to control the declaration of health status and history of the movement by using a BlueTooth-based mobile application that serves to caution its users with close contact with a COVID-19 patient.⁶ Either home-based or institution-based isolation accomplishes this. Institution-based isolation is better in controlling community transmission of COVID-19,^7,8 though home-based isolation is understandably the preferred method. Another method is the early diagnosis of COVID-19 cases. The early diagnosis allows to immediately isolate and prevent the further spread of disease. However, it has been found that lack of timely diagnosis can significantly affect the primary reproduction number, increase the risk of transmission to the community leading to an outbreak of cases in the community.⁹

Despite having an effective healthcare system alongside technological advances and a robust universal health insurance system for its citizens, Japan has been no exception to the pandemic. They are currently facing the fourth wave of COVID-19 infections^10,11 and have declared a third state of emergency from April 25th to May 11th, 2021, ahead of the Olympic Game.^12,13 This action appeared to express limitation on their strategy against COVID-19 regardless the availability of vaccine. Besides, the Japanese government has encouraged the citizens to avoid the “Three Cs” (3Cs) in daily life, including closed spaces with poor ventilation, crowded places with many people nearby, and close-contact settings.¹⁴ This policy is considered as one of the critical measures to prevent the occurrence of the new clusters. Along with 3Cs, the authority advised people having mild COVID-19 symptoms to self-isolate at home, and avoid work or public spaces.^15-17 In contrast, more severe individuals were treated at health posts.¹⁷ Home isolation with mild cases was similar to nations at Europe and North America.^18,19 However, the compliance to self-isolation measures is questionable as the findings of a recent study on Japanese workers showed that very few people practiced self-isolation.²⁰ Therefore, it is likely that mild or asymptomatic cases could be an infectious source of new outbreak if they do not obey the government’s regulation. This study aimed to analyze the correlation between the raised community cases and home-based cases with mild or limited symptoms in Japan.

Materials and Methods

Data sources

This was a secondary data analysis study with data from three different databases during March 1st, 2020 to March 31st, 2021:

COVID-19 database of Center for Systems Science and Engineering (CSSE) at Johns Hopkins University is an interactive web-based dashboard to visualize and track COVID-19 cases in real-time. It reports cases at province level in China, county level in the USA, state level in Australia, Canada, Japan, and at the country level otherwise. The location, number of confirmed cases, deaths, and recoveries was illustrated. The data source is collected from international organizations and local governments.^21,22

• The COVID-19 dataset in Japan is an open nationwide data from Japanese Ministry of Health, Labour and Welfare. It includes demographic data based on the government website, daily information regarding the testing and medical care provision published by prefectures, and information regarding cluster events, obtained from the media or other sources. In this study, we analysed patient cases, number of tests and new community cases.²³

• Community Mobility Reports is an anonymized Google database of movement trends over time by regions and place categories. It aims to gauge how people change in response to government policies during COVID-19.²⁴ This data collected through Google Maps, and consisted of six different categories: residential, grocery and pharmacy, parks, retail and recreation, workplaces, transit stations in the form of changes comparing to the baseline period between January and February 2020. The daily data is affected by activities during the week, so we have converted the original data into the rolling seventh-day average data to evaluate the impact of policies.

Procedure

Indicators in the study are defined in Table 1.

Statistical analysis

As of March 31st, 2021, the proportion of the population vaccinated in Japan has been at 0.69%, and SARS-CoV-2 is human-to-human, so we assumed that the community cases would be directly affected by Actual Active Cases, Government’s policy (Adjusted Stringency index), people’s compliance based on Community Mobility Reports, and Adjusted Number of Tests.²⁶ Although we all knew that each infected case was in interaction, for this study, we also assumed that the data was independent and only affected by Actual Active Cases, Stringency Index, Number of Tests, and Community Mobility Reports. Therefore, we utilized an independent variable for a given date to predict the future number of community cases using a linear regression model. Based on the incubation period of SARS-CoV-2, a mean of 5.2 days and a 95% confidence interval of 4.1 to 7.0 days, plus a late reporting time of 2 days (corresponding to test result waiting time and update data delay in Japan).²⁷ Our study assumed that the number of cases (inpatient cases, outpatient cases), the adjusted number of tests, the adjusted the stringency index, and Google Mobility Trend (Residential index; Grocery and pharmacy index; Retail and recreation index; Parks index; Workplaces index; Transit stations index) were related to the number of new community cases at 6^th, 7^th, 8^th, and 9^th day later.

We used Bayesian model averaging - an application of Bayesian inference to determine the optimal model based on R² value, BIC index, and Post-Prob index to select variables in the aforementioned linear model. The analysis was implemented on Bayesian Model Averaging (BMA) package by R software version 4.0.3 (CRAN main site at https://CRAN.R-project.org).

Results

shows the time series of the number of new community cases, inpatient cases, and outpatient cases. The peak of COVID-19 waves with the highest community cases corresponded with the top of outpatient cases. While outpatients had stationary, the community cases decreased rapidly. This chart also presents that the gap between inpatients and outpatients widened during outbreaks and narrowed during stable community cases.

Close

Figure 1. Time series of the number of new community cases, inpatient cases, and outpatient cases. New community cases, inpatient cases, and outpatient cases were represented in light blue line (secondary y-axis), blue line, and red line (primary y-axis), respectively. The peak of COVID-19 outbreaks with the highest community cases corresponded with the top of outpatient cases

The timeline of an association of Community Mobility Reports and new community cases was described in . Eight days before the peaks of waves, retail and recreation index, transit stations, workplaces started to decline, and the lowest levels are strikingly similar to the second and third waves. Meanwhile, the grocery and pharmacy index and parks index increased in the first and second waves, but decreased slightly in the third wave. In contrast, the residential index tended to grow over time and rapidly increased during waves.

Close

Figure 2. The time series of Community Mobility Reports and the number of new community cases. The blue line represents the new community cases (primary y-axis), while the secondary y-axis depicts six indicators of Community Mobility Reports with different colors

Table 2 shows the predictive model community cases corresponding to the incubation periods from 4 to 7 days. The model predicted the number of community cases on 8^th day (matching the incubation period of 6 days and 2 days waiting to report results) was the best-fitting model. It achieved the highest value of R2 (R2 = 0.89). emphasized the predictive values ​​of this model in the forecast of community instances. In this model, an increase of 100 outpatient cases related to the apparition of approximately 24 community cases. Furthermore, it was revealed that the increment of 1% of Community Mobility Reports index in residential, grocery and pharmacy, retail and recreation, the workplace would escalate the new community cases by 278, 112, 73, and 57 cases, respectively. For the parks index, the elevation of one unit of adjusted the stringency index, the increase of every 100 inpatient cases, and the increment of 1% of Community Mobility Reports would significantly reduced community cases (14, 3, and 33 points, respectively). with the number of new community cases on the eighth day and the forecast from the best-fitting model showed that the predicted value was relatively same as the actual value, even when the number of new community cases broke out and the periods of this indicator was stable.

Close

Figure 3. Predictive values ​​of the model on the eighth day and the number of community cases in real time. The predicted value was relatively same as the actual value

Discussion

We focused on analyzing the number of COVID-19 outpatients having mild or moderate symptoms who were self-isolated and treated at home. Our result indicated that the outpatients could become a major factor causing the new outbreaks in Japan. The highest outpatient cases were respective to top of COVID-19 waves from a predictive model of community cases in the past outbreaks (an incubation period of 4 to 7 days^27-29). The model showed that outpatients contributed significantly to the number of community cases. Among the predictive models, the analysis on the 8^th day was the most appropriate when it showed a strong association between government reports and an incubation period of 5.2 days plus 2 days to wait the announcement. Proposed models on the 7^th, 8^th, and 9^th day did not comprise the Adjusted Number of Tests variable because this variable was not statistically significant, even reduced the quality of the model.

Our hypothesis found that a tremendous gap has been happening in the self-isolation of Japan since the COVID-19 pandemic emerged. The Japanese government could overlook the potential role of outpatients as an important infection source. The regulation only requires hospitalization for severe cases or high-risk patients. Other cases could be managed at home or accommodation facility.¹⁷ Although the public health centers took responsibility for outpatients, it might be ineffective because the rule of isolation could be broken anytime. The reason could be several factors. We inclined that the poor attitude and low knowledge were key elements. The government needed to consider a better strategy of health communication to promote knowledge and attitude of COVID-19 prevention together with responsibility. On the other hand, supervising outpatients with strict measurements was another strategy, for example, quarantining COVID-19 mild cases or outpatients in isolation facilities like Vietnam or control through follow-up application in South Korea.^30-32

In terms of movement, it was necessary to manage residential and retail strictly and recreation activities; however, opening the parks could be considered as long as ensuring the social distance. Studies showed that SARS-CoV-2 was at higher risk of infection in closed spaces, high population densities where exposure distances greater than 2 meters can hardly be guaranteed. During social distancing due to COVID-19, activities would be increased at residential areas or parks. In a residential area, activities among neighbours can make it challenging to maintain the six-feet distance, but it can be easier reached if activities are in the park. Therefore, visiting parks may be a good alternative for those seeking entertainment.^33,34

We also offered that the government measure policies against outbreaks by designing or applying a specific index like the Stringency index. In our study showed that Government Stringency Index should be adjusted according to the number of active cases to avoid the possibility of too tight or too loose application that negatively affects social performance and epidemic control.

Propose solutions

Japan employed a variety of strategies in an attempt to control over the past COVID-19 waves. However, the potential disadvantage in their strategy could pose new outbreaks. One of them was effective management of COVID-19 outpatients. Here, we suggest several supplement measurements to prevent community transmission. Our solutions are not absolutely appropriate but they deserve to become valuable references for the governments. We hope that they will help Japanese avoiding a state of emergency in the future.

First of all, improving knowledge, attitude of COVID-19 prevention and better self-isolation management.

Secondly, the intensive tracing could be a critical solution to pass the new outbreaks without an emergency state although this is not induced by our results. In more details, the tracing should not only focus on closed contacts but also expand their contacts as well.^6,35 The closed contacts should test via PCR and stay home until the result is available ( and ). The undeniable efficacy of the Japan cluster-based approach will surge, paying more attention to F2 (contact of closed contact with confirmed cases), who were ignored. This might pose the silent transmission in case the tracing was not performed. The efficacy of F2 monitoring appeared in Vietnam, where the classification procedure of COVID-19 defines patients and their contacts from F0 to F5 (first generation of infection, second to fifth generation of infection).^6,35 During the period of March to April 2020, a successful containment of a localized outbreak of COVID-19 in Bach Mai Hospital, Vietnam involved identification of 27893 F1 and F2 contacts who were traced, tested and quarantined at centralised centres; and about 24346 F3 and F4 cases were traced and quarantined at home.³⁶ As of April 18th, 2021 with 2784 COVID-19 positive cases, the Government of Vietnam has already quarantined more than 15.7 million people.³⁷

Close

Figure 4. Proposal of COVID-19 tracing strategy for Japan

Close

Figure 5. Subjects of COVID-19 tracing strategy

Finally, early detection can be achieved by modifying the current medical declaration used by Japan called COCOA (Contact-Confirming Application) which is a smartphone application employing the principle of bluetooth and needs user approval to receive notification about potential contact within 1 meter for more than 15 minutes with a person who tested positive for COVID-19 whilst maintaining anonymity.³⁸ Now, it has been applied to positive polymerase chain reaction (PCR) cases. We thought the application should cover all individuals with fever or flu-like syndrome (cough, shortness of breath, and runny nose). Besides, we can apply social network analysis (SNA) for analysing reported locations from like-flu symptomatic individuals to detect suspicious transmission. We can use SNA software packages (UCINET or Gephi) to explore the centrality of connections, and location detection can be simply completed.³⁹ As a result, the early detection of COVID-19 clusters, as well as influenza, could be achieved.

Conclusion

Outpatient cases and Community Mobility Reports including residential index, grocery and pharmacy index, retail and recreation index, workplaces index were positively associated with COVID-19 community cases. It implied that isolation policy and compliance were closely related to the increase of community cases. It recommended that governments should manage effectively outpatients with mild or moderate symptoms and asymptomatic case with positive test, and evaluate the compliance through the Google Mobility Trend index to make appropriate adjustments.

Acknowledgments

We would like to appreciate all the teams of Center for Systems Science and Engineering (CSSE) (Johns Hopkins University), Japanese Ministry of Health, Labour and Welfare, and Google for public data of COVID-19. We also thank Editors and Reviewers of the Journal Health Promotion Perspectivesfor their valuable comments.

Nam Xuan Ha^1,2#, Truong Le-Van^2,3#, Nguyen Hai Nam^2,4,5, Akshay Raut^2,6, Joseph Varney^2,7, Nguyen Tien Huy^2,8*

Authors’ contributions

NXH, TLV, and NTH made substantially contributions to the study’s conception and design. TLV analysed the data, while NXH and TLV interpreted the results. NXH reviewed the manuscript after all authors participated in the writing. NTH critically revised all works. All authors read and approved the final manuscript.

Funding

No funding was available in this study.

Ethical approval

Not applicable.

Competing interests

All authors declare that no competing interests exist

References

1. Beware of the second wave of COVID-19. Lancet 2020; 395(10233):1321-2. doi: 10.1016/s0140-6736(20)30845-x [Crossref]
2. WHO. SARS-CoV-2 Variants. Available from: https://www.who.int/csr/don/31-december-2020-sars-cov2-variants/en/. Accessed May 2, 2021.
3. Increased transmissibility and global spread of SARS-CoV-2 variants of concern as at June 2021. Euro Surveill 2021; 26(24). doi: 10.2807/1560-7917.es.2021.26.24.2100509 [Crossref]
4. WHO. Draft landscape and tracker of COVID-19 candidate vaccines. Available from: https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines. Accessed May 2, 2021.
5. Reserving coronavirus disease 2019 vaccines for global access: cross sectional analysis. BMJ 2020; 371:m4750. doi: 10.1136/bmj.m4750 [Crossref]
6. Community-based measures to against the COVID-19: an experience from Vietnam with consecutive 3 months of no new infection in the community during the first wave of pandemic. Front Public Health 2021; 9:583655. doi: 10.3389/fpubh.2021.583655 [Crossref]
7. Institutional, not home-based, isolation could contain the COVID-19 outbreak. Lancet 2020; 395(10236):1541-2. doi: 10.1016/s0140-6736(20)31016-3 [Crossref]
8. Institutional versus home isolation to curb the COVID-19 outbreak - Authors’ reply. Lancet 2020; 396(10263):1632-3. doi: 10.1016/s0140-6736(20)32171-1 [Crossref]
9. Effect of delay in diagnosis on transmission of COVID-19. Math Biosci Eng 2020; 17(3):2725-40. doi: 10.3934/mbe.2020149 [Crossref]
10. The Japan Times. Top Japan virus expert acknowledges fourth wave as variants take hold. Available from: https://www.japantimes.co.jp/news/2021/04/15/national/fourth-wave-omi-coronavirus/. Accessed May 3, 2021.
11. NHK. Omi Calls the Virus Surge a Fourth Wave. Available from: https://www3.nhk.or.jp/nhkworld/en/news/20210414_20. Accessed May 3, 2021.
12. Reuters. Three months before Olympics, Japan declares ‘short’ emergency in Tokyo. Available from: https://www.reuters.com/business/healthcare-pharmaceuticals/japan-says-seek-short-powerful-state-emergency-tokyo-osaka-elsewhere-2021-04-22/. Accessed May 2, 2021.
13. Office for Novel Coronavirus Disease Control CS, Government of Japan. COVID-19 Declaration of State of Emergency. Available from: https://corona.go.jp/en/. Accessed May 2, 2021.
14. Ministry of Health, Labour and Welfare, Japan. Preventing COVID-19 (novel coronavirus) and stopping its spread. Available from: https://www.c19.mhlw.go.jp/covid-19-en.html. Accessed April 20, 2021.
15. Ministry of Health, Labour and Welfare, Japan. Japan’s cluster-based approach. Available from: https://www.mhlw.go.jp/content/10900000/000639224.pdf. Accessed May 2, 2021.
16. Cluster-based approach to coronavirus disease 2019 (COVID-19) response in Japan, from February to April 2020. Jpn J Infect Dis 2020; 73(6):491-3. doi: 10.7883/yoken.JJID.2020.363 [Crossref]
17. Ministry of Health, Labour and Welfare, Japan. Clinical Management of Patients with COVID-19: A guide for front-line healthcare workers (Version 2.1). Available from: https://www.mhlw.go.jp/content/000646531.pdf. Accessed May 2, 2021.
18. Società Italiana di Medicina Generale e delle Cure Primarie. COVID-19: La gestione del paziente con sintomi compatibili. Available from: https://www.simg.it/covid-19-la-gestione-del-paziente-con-sintomi-compatibili/. Accessed May 2, 2021.
19. Centers for Disease Control and Prevention (CDC). Interim Clinical Guidance for Management of Patients with Confirmed Coronavirus Disease (COVID-19). Available from: https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html. Accessed 5 May 5, 2021.
20. The actual implementation status of self-isolation among Japanese workers during the COVID-19 outbreak. Trop Med Health 2020; 48:63. doi: 10.1186/s41182-020-00250-7 [Crossref]
21. Johns Hopkins University. COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. Available from: https://github.com/CSSEGISandData/COVID-19. Accessed April 20, 2021.
22. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect Dis 2020; 20(5):533-4. doi: 10.1016/s1473-3099(20)30120-1 [Crossref]
23. Ministry of Health, Labour and Welfare, Japan. Open data. Available from: https://www.mhlw.go.jp/stf/covid-19/open-data_english.html. Accessed April 20, 2021.
24. Google. COVID-19 Community Mobility Reports. Available from: https://www.google.com/covid19/mobility/. Accessed April 20, 2021.
25. A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nat Hum Behav 2021; 5(4):529-38. doi: 10.1038/s41562-021-01079-8 [Crossref]
26. Our World in Data. Coronavirus (COVID-19) Vaccinations. Available from: https://ourworldindata.org/covid-vaccinations. Accessed April 20, 2021.
27. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med 2020; 382(13):1199-207. doi: 10.1056/NEJMoa2001316 [Crossref]
28. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med 2020; 382(18):1708-20. doi: 10.1056/NEJMoa2002032 [Crossref]
29. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Ann Intern Med 2020; 172(9):577-82. doi: 10.7326/m20-0504 [Crossref]
30. Policy responses to the COVID-19 pandemic in Vietnam. Int J Environ Res Public Health 2021; 18(2):559. doi: 10.3390/ijerph18020559 [Crossref]
31. Kim MS. South Korea is watching quarantined citizens with a smartphone app. Available from: https://www.technologyreview.com/2020/03/06/905459/coronavirus-south-korea-smartphone-app-quarantine. Accessed March 6, 2021.
32. Facilities for centralized isolation and quarantine for the observation and treatment of patients with COVID-19. Engineering (Beijing) 2021; 7(7):908-13. doi: 10.1016/j.eng.2021.03.010 [Crossref]
33. Superspreading events suggest aerosol transmission of SARS-CoV-2 by accumulation in enclosed spaces. Phys Rev E 2021; 103(3-1):033109. doi: 10.1103/PhysRevE.103.033109 [Crossref]
34. Mechanistic transmission modeling of COVID-19 on the Diamond Princess cruise ship demonstrates the importance of aerosol transmission. Proc Natl Acad Sci U S A 2021; 118(8):e2015482118. doi: 10.1073/pnas.2015482118 [Crossref]
35. Vietnam Times. Prevention of Covid 19: Vietnamese Customs continues giving directives about exporting face masks. Available from: https://vietnamtimes.org.vn/prevention-of-covid-19-vietnamese-customs-continues-giving-directives-about-exporting-face-masks-18450.html. Accessed April 20, 2021.
36. Nosocomial coronavirus disease outbreak containment, Hanoi, Vietnam, March-April 2020. Emerg Infect Dis 2021; 27(1):10-7. doi: 10.3201/eid2701.202656 [Crossref]
37. World Health Organization (WHO). COVID-19 in Viet Nam Situation Report 38. Available from: https://www.who.int/vietnam/internal-publications-detail/covid-19-in-viet-nam-situation-report-38. Accessed April 20, 2021.
38. Ministry of Health, Labour and Welfare, Japan. Install the contact-confirming application (COCOA) to protect yourself, your loved ones, your community and society as a whole. Available from: https://www.mhlw.go.jp/content/10900000/000647649.pdf. Accessed April 20, 2021.
39. Wasserman S, Faust K. Social Network Analysis: Methods and Applications. Cambridge: Cambridge University Press; 1994. p. 196-8.