Modelling the impact of control measures against the COVID-19 pandemic in Viet Nam ================================================================================== * Thu-Anh Nguyen * Quoc Nguyen Cuong * Anh Le Thi Kim * Huyen Nguyen Nguyen * Thao Nguyen Thi Huong ## Abstract **Objectives** Health care system of many countries are facing a surging burden of COVID-19. Although Vietnam has successfully controlled the COVID-19 pandemic to date, there is a sign of initial community transmission. An estimate of possible scenarios to prepare health resources in the future is needed. We used modelling methods to estimate impacts of mitigation measures on the COVID-19 pandemic in Vietnam. **Methods** SEIR model built in the COVIDSIM1.1 tool was adopted using available data for estimation. The herd immunization scenario was with no intervention implemented. Other scenarios consisted of isolation and social distancing at different levels (25%, 50%, 75% and 10%, 20%, 30%, respectively). Outcomes include epidemic apex, daily new and cumulative cases, deaths, hospitalized patients and ICU beds needed. **Results** By April 8, 2020, there would be 465 infected cases with COVID-19 in Viet Nam, of those 50% were detected. Cumulatively, there would be 1,400 cases and 30 deaths by end of 2020, if 75% of cases was detected and isolated, and 30% of social distancing could be maintained. The most effective intervention scenario is the detection and isolation of 75% infected cases and reduction of 10% social contacts. This will require an expansion of testing capacity at health facilities and in the community, posing a challenge to identify high-risk groups to prioritized testing. **Conclusions** In a localized epidemic setting, the expansion of testing should be the key measure to control the epidemic. Social distancing plays a significant role to prevent further transmission to the community. Key words * SARS-CoV-2 * modelling * measures * Vietnam ## Introduction The first SARS-COV-2 case was reported from Wuhan, China in December 2019, and has now spread out to 211 countries, territories, and 2 cruise ships with 1,353,361 infections and 79,235 deaths by 8 April 2020. [1] Since 11 March 2020, World Health Organization (WHO) has declared SARS-CoV-2 outbreak a global pandemic. According to WHO, R (R naught - the reproduction number), the average number of secondary cases attributable to infection by an index case after that case is introduced into a susceptible population, is estimated at 2.0 to 2.5. Viet Nam is considered successfully controlling the first stage of SARS-CoV-2 outbreak with 16 cases detected in connection with Wuhan. On 22 January 2020, the first COVID-19 case who returned from Wuhan was detected. Other cases afterwards either were related to returning from China or got infections from cases returning from China. By 13 February 2020, all these 16 cases were recovered and discharged from hospitals. However, from 2 March 2020, some passengers on the international flights to Viet Nam were found to be infected with SARS-CoV-2 and transmitted the virus to some of their close contacts. Some clusters were reported from Bach Mai hospital, Truong Sinh company in Hanoi, and Buddha bar in Ho Chi Minh city. By 8 April 2020, Viet Nam has reported 252 confirmed cases, of which 128 cases were recovered, 124 have been treated at health facilities, and no death reported. [2] Viet Nam has been implementing a number of rigorous interventions to control the COVID-19 pandemic, including school closure; contact tracing, isolation and testing of suspected and probable cases; isolation and treatment of confirmed cases; testing expansion; cancellation of flights from countries of epicenters; and social distancing. On one hand, these intervention strategies made Viet Nam reduce new cases. On the other hand, these rigorous interventions, if implemented in a long-term, could lead to negative consequences for society and economy in the future, such as, the increase of stress and anxiety, restriction of production, limited business and economic development. These consequences could become more serious for the poor and other vulnerable populations. In order to achieve comprehensive information in decision making for the next steps of pandemic control, we used modelling methods to estimate impacts of different mitigation measures on the pandemic. We also predicted resources of health system that needed for different pandemic scenarios. ## Methods The *COVIDSIM1*.*1*, a projection tool developed by ExploSYS GmbH, was used. [3] The tool adopted the SEIR (Susceptible - Exposed - Infectious - Recovered) model that has been widely used for simulating spread of communicable diseases. In this model, the population is categorized into 4 groups: susceptible, exposed, infected groups (symptomatic and asymptomatic), and a group with outcomes (recovery or death). This model was used to assess the effectiveness of the interventions for COVID-19 pandemic in New Zealand and one hypothetical European country. [4] [5] Details of the model and its algorithms are presented in the *Annex 1*. The *Annex 2* displays input data of the model. Key assumptions include a latency period of 4 days, a prodromal period of 1 day, a fully infective period of 10 days, infections leading to sickness of 82%, and R0 of 3.4 for worst case scenario and 2.5 for other intervention scenarios. Average duration of hospitalization was 15.4 days without critical care and 23.5 days with ICU. We assumed that 30% of sick patients are hospitalized and 10% require intensive care. These assumptions were developed using the data of Viet Nam that are accessible on the website of the Ministry of Health (MOH). Where local data is not available, we have used and adjusted other clinical parameters from other countries. We assumed that different measures, such as, contact tracing, isolation and testing of suspected and probable cases; isolation and treatment of confirmed cases would continue being implemented in a long-term. Therefore, we modelled two main intervention strategies (i) contact tracing and case isolation, and (ii) social distancing. Details are: * - Scenario 1- Current scenario: isolation of infected and probable cases was implemented at 75% and social distancing to reduce social contacts by 30% as of the second week of the 29th March 2020. [6] * - Scenario 2- Relaxed social distancing based on scenario 1: remain 75% isolation of infected and probable cases and 10% reduction of social contacts. * - Scenario 3- Herd immunization strategy: there was no intervention. In this scenario, we assumed that mass media could have effects on people’s behaviors when the number of cases and deaths increased. This could lead to a reduction of 5% of social contacts (or 95% social contacts would remain). * - Other scenarios: interventions of isolation and social distancing would be implemented at different levels, for instance, isolation of infected and probable cases at 25%, 50%, 75% and social distancing to reduce social contacts 10%, 20%, 30%. Assumptions to develop these intervention strategies considered the feasibility in terms of capacity and resources of health system and people in the community. The key outcomes of these models consisted of the numbers of new cases per day, cumulative cases, deaths, hospitals and ICU beds required. All these outcomes of different models were compared with each other to identify the best model for current capacity of health system. ## Results ### Application of the SEIR model to develop the current scenario By using a current scenario that R is 2.5, 75% of cases is detected and isolated right after disease confirmation, and 30% of social distancing is implemented, we estimated that by April 8, 2020, there would be 465 infected cases with SARS-CoV-2 in Viet Nam, of which 369 cases would be symptomatic (Figure 1). Number of cumulative cases by end of 2020 would be 1,400 with a total of 30 COVID-19 related death. There would be about 3 new cases in the community each day. The Figure 2 showed the reported daily number of new cases from 2 groups: cases imported from other countries and local transmission cases. This figure indicated that Viet Nam has controlled successfully the source of imported infection and the number of local transmission cases has been significantly decreased. The estimated number of new cases, patients and ICU beds would be low and not beyond the ability of healthcare system (Table 1). View this table: [Table 1:](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/T1) Table 1: Estimated new case, cumulative infection and death, hospitalization and ICU beds required of proposed scenarios ![Figure 1:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/04/29/2020.04.24.20078030/F1.medium.gif) [Figure 1:](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/F1) Figure 1: Number of estimated versus reported cases of COVID-19 in Viet Nam ![Figure 2:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/04/29/2020.04.24.20078030/F2.medium.gif) [Figure 2:](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/F2) Figure 2: Number of reported and confirmed cases* by day of immigration or day of testing (whichever earlier) ****The number of cases per day is calculated as the average of the 3 consecutive days to adjust for delay of reporting time (if any)*. As presented in the Figure 1, the number of detected cases would be 231, accounts for around 50% of the estimated cases. There are two possible reasons to explain the difference. First, there was a period of no isolation at quarantine centers for immigrants – especially from the US and Europe – to Viet Nam. The regulation of isolation for those from these areas had been released about mid of March while the first case was detected in late January in the US and at the early March 2020 in the Europe. Thus, there could be latent infected cases in the community. Second, there were some cases infected from index cases in the community but they had not been isolated immediately. Moreover, the implementation of contact tracing for confirmed and suspected cases could be incomplete, so their close contacts had been lost of follow up. All these problems could lead to uncontrolled latent cases in the community. In fact, Viet Nam has reported some cases with unknown source of infection. ### Modelling the impact of COVID-19 control measures Under the herd immunization scenario (scenario 3), we assumed that the Viet Nam Government did not implement any mitigation measures, the number of newly infected cases would peak at 2.9 million per day in June 2020. The number of cumulative cases by the end of 2020 would be 91.4 million, accounting for 95% of total population. The total number of deaths would be 3.2 million, about 2.4% of total population. Figure 3 shows the projection of newly daily cases under scenarios with different level of case isolation and social distancing. Figure 4 indicates the estimated ICU beds required under six scenarios. However, this scenario certainly will not occur since Viet Nam has implemented rigorous and definitive interventions from the outset of the epidemic. ![Figure 3:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/04/29/2020.04.24.20078030/F3.medium.gif) [Figure 3:](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/F3) Figure 3: Estimated number of newly daily infected cases by different scenarios ![Figure 4:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2020/04/29/2020.04.24.20078030/F4.medium.gif) [Figure 4:](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/F4) Figure 4: Estimated number of daily new ICU cases by the scenarios If 25% of infected cases would be isolated, and social contacts would decline by 10% or 20% (scenario 4a and 4b, respectively), the number of new cases per day will be very high, ranging from 1 million - 1.3 million at epidemic peak. The number of ICU beds needed at the peak will be 500,000 to 600,000. In this case, Viet Nam health system (e.g. human resources, ventilators, and infrastructure) will not be able to deal with this demand. If 50% of infected cases would be isolated, and social contacts would decline by 10% or 20% (scenario 5a and 5b), the pandemic would last and reach its peak by March and September 2022, respectively. Under the scenario 5b, vaccine and effective medication may be available by that time. With this scenario, 1,500 ICU beds and 10,000 patient beds should be prepared by the end of 2020. If we detected and isolated 75% of confirmed cases and reduced 10% of social contacts, there would be around 12 new cases per day. The cumulative number of deaths by the end of 2020 would be 90. In this scenario, the disease would continue to be endemic in the community with low prevalence. ## Discussion The results showed that if there was no intervention (scenario 3) or if only 25% of cases detected and quarantined (scenario 4), Viet Nam will experience huge numbers of infections, required IDU beds and deaths in the near future. In scenarios of isolation of 25% or 50% infected cases, in combination with social distancing of 10% or 20% social contacts, the total number of infected cases in two years (2020-2021) would range from 60%-80% of total population. The raw mortality rate would be 1.5%-1.8%. In other words, these scenarios would save 1.5-1.7 million of lives compared to the worst scenario. If there will be an effective vaccine or medication within this period, the number of infected cases and deaths will be much lower. The first scenario which is considered to be the current strategy of Viet Nam with rigorous intervention measures are being implemented and an assumption that these measures will last for at least one year would lead to low numbers of infections, required ICU beds and deaths. These intervention measures include detection and isolation of 75% infected cases, and social distancing strategies to reduce social contacts 30%. This scenario, in fact, will require the expansion of testing, not only at health facilities but also in the community to achieve the detection proportion of 75%. Korea, for instance, had done testing expansion, isolation and treatment of SARS-CoV-2 positive cases and the country has got success in reducing the number of new cases and deaths as well. However, the expansion of testing will be a challenge for health system. On average, Viet Nam has detected 1 case among 440 tests in the context of the existence of imported cases that have been under control. It will be a significant challenge to implement this testing strategy because of the required high cost, sufficient testing capacity, infrastructure, and procurement capacity, and ability to provide biological products and consumable supplies to large scale testing. However, if we cannot detect all cases, the probability of detection, and isolation as well, will not reach 75% cases, and the outbreak will then return in the community. Therefore, we might need to combine different approach such as assessing to identify high-risk groups to test them through strengthening the infectious disease surveillance system including influenza like illness surveillance at pharmacy stores, health clinics and hospitals. People presenting at health clinics and hospital with respiratory symptoms should receive a test for SARS-CoV-2. In addition to expansion of testing, contact tracing up to three generation, and case isolation should be implemented rigorously. People who have close contact with a confirmed case should be isolated for a minimum of 14 days regardless of their test result. The next generation of close contact case should be prepared for contact management procedure if any person in the chain becomes positive. [7] In addition, the scenario with strict and prolong social distancing to reduce a large number of social contacts would have negative effects on the economic development and social life. The second scenario though will bring the same challenge of an effective testing strategy to detect and isolate 75% of infected cases, it would have a more relaxed social distancing strategy which would reduce only 10% of social contacts. The number of cases and number of required ICU beds in this scenario would be still manageable. This scenario supports strategies to categorize the risk areas and businesses, gradually re-open the economy and social activities in Viet Nam in coming months. Our study has a number of limitations. So far, we still have limited knowledge of epidemiology and pathology of the SARS-CoV-2. Data of our models were based on multiple assumptions from data of countries where outbreaks have occurred. These models did not adjust for levels of virus transmission that could be affected by weather factors. In fact, we do not have evidences of changing levels of transmission by outdoor temperature. These models did not consider of unusual epicenters that their frequency and context of social contacts could be different from common contacts, such as the recent outbreak among men who have sex with men in Ho Chi Minh City. These models calculated mathematically the level of isolation and social distancing. In fact, defining what activities will reduce 10% or 20% social contacts, what actions will do to isolate infected cases is a matter. This requires consultation of researchers on social science. The mortality rate in our models only covers deaths of COVID-19, but other deaths due to overload of health system. Finally, these models did not consider impacts of the effective medication and vaccine. ## Data Availability Data used for the modeling is included in the Annex 2 of the manuscript. ## Funding source This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ## Conflicts of Interest The authors declared no conflict of interest. ## Authors’ Contributions TAN developed the modeling method and run the projection, QNC, ALTK, HNN, TNTH contributed to assumptions development and drafted the manuscript. ## Annex 1: Details of the model and its algorithms [3] Number of susceptible individuals: ![Formula][1] Number of individual s in the latent period: ![Formula][2] Number of individual s in the prodromal period: ![Formula][3] Number of individual s in the symptomatic period: ![Formula][4] Number of removed individuals: ![Formula][5] Number of dead individuals: ![Formula][6] Number of isolated cases at time *t*: ![Formula][7] Number of fully isolated cases at time *t*: ![Formula][8] Number of home isolated cases at time *t:* ![Formula][9] ### Initial values View this table: [Table2](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/T2) ### Parameters View this table: [Table3](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/T3) ### Derived variables View this table: [Table4](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/T4) ### Detection probability ![Formula][10] The probability that at least one test has been performed (and has returned a positive result) is then ![Formula][11] ## Annex 2: Input data and assumptions View this table: [Table5](http://medrxiv.org/content/early/2020/04/29/2020.04.24.20078030/T5) ## Acknowledgement We thanks Markus Schwhem and Ha Dao for their comments and contribution in the modeling process, and Duyen Duong for her assistance in preparation of reporting data. * Received April 24, 2020. * Revision received April 24, 2020. * Accepted April 29, 2020. * © 2020, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/) ## References 1. [1].W. H. Organization, “Situation report - 79. Coronavirus disease 2019 (COVID-19),” 08 April 2020. 2. [2].V. M. o. Health, “[https://ncov.moh.gov.vn/](https://ncov.moh.gov.vn/),” [Online]. Available: [https://ncov.moh.gov.vn/.content-type=](https://ncov.moh.gov.vn/.content-type=)[Accessed 08 April 2020]. 3. [3]. M. S. Martin Eichner, “CovidSIM version 1.1,” University of Tübingen and IMAAC NEXT Association, [Online]. Available: [http://covidsim.eu/?fbclid=IwAR1OiZpWVWDZHuIJ5EWA3LLz\_CFYBscSocAhy3fdCIKCM9F27LIrI46-08](https://covidsim.eu/?fbclid=IwAR1OiZpWVWDZHuIJ5EWA3LLz_CFYBscSocAhy3fdCIKCM9F27LIrI46-08). [Accessed 08 April 2020]. 4. [4].L. T. B. A. K. A. V. M. B. M. S. Nick Wilson, “Modelling the Potential Health Impact of the COVID-19 Pandemic on a Hypothetical European Country,” Preprint, p. [https://doi.org/10.1101/2020.03.20.20039776](https://doi.org/10.1101/2020.03.20.20039776), 2020. 5. [5].L. T. B. A. K. M. B. Nick Wilson, “Potential Health Impacts from the COVID-19 Pandemic for New Zealand if Eradication Fails:,” University of Otago Wellington, 23/03/2020. 6. [6].Google, “Vietnam COVID-19 Community Mobility Report ([https://www.google.com/covid19/mobility/](https://www.google.com/covid19/mobility/)),” 29 March 2020. 7. [7].Q. N. C. A. L. T. K. T. N. H. H. N. N. G. J. F. G. B. M. T-A. Nguyen, “Adapting a Tuberculosis contact investigation strategy for COVID-19,” IJTLD, no. [https://www.theunion.org/news-centre/news/adapting-a-tb-contact-investigation-strategy-for-covid-19](https://www.theunion.org/news-centre/news/adapting-a-tb-contact-investigation-strategy-for-covid-19), 2020. 8. [8].GSO, “Results of the population and housing census 2019 ([https://vietnam.unfpa.org/en/news/results-population-and-housing-census-2019](https://vietnam.unfpa.org/en/news/results-population-and-housing-census-2019)),” 2019. 9. [9].P. S. C. B. S. Y. Z. T. Y. W. e. a. Li R, “Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV2).,” Science, 2020. 10. [10].C. S. C. n. w. g. F. S. E. R. M. Quilty Billy J, “Effectiveness of airport screening at detecting travellers infected with novel coronavirus (2019-nCoV),” Euro Surveill., vol. 25, no. 5 ([https://doi.org/10.2807/1560-7917.ES.2020.25.5.2000080](https://doi.org/10.2807/1560-7917.ES.2020.25.5.2000080)), 2020. 11. [11].[https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College\---|COVID-19\---|Relative-Sensitivity-International-Cases.pdf](https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College\---|COVID-19\---|Relative-Sensitivity-International-Cases.pdf). 12. [12].V. N. E. R. D. F. Ashleigh R. Tuite, “Estimation of COVID-19 outbreak size in Italy based on international case exportations,” Preprint: [https://doi.org/10.1101/2020.03.02.20030049](https://doi.org/10.1101/2020.03.02.20030049)., 2020. 13. [13]. W.-C. J. Mission., “Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19).,” 2020 (16-24 February). 14. [14].A. B. C. A. M. G. G. Z. Alessia Lai, “Early phylogenetic estimate of the effective reproduction number of SARS-CoV-2,” Journal of Medical Viology, no. [https://doi.org/10.1002/jmv.25723](https://doi.org/10.1002/jmv.25723), 25 February 2020. 15. [15].Y. D. Z. X. e. a. Penghui Yang, “Epidemiological and clinical features of COVID-19 patients with and without pneumonia in Beijing, China,” [https://doi.org/10.1101/2020.02.28.20028068](https://doi.org/10.1101/2020.02.28.20028068)., 2020. 16. [16].Y. W. S. M. e. a. Qifang Bi, “Epidemiology and Transmission of COVID-19 in Shenzhen China: Analysis of 391 cases and 1,286 of their close contacts,” [https://doi.org/10.1101/2020.03.03.20028423](https://doi.org/10.1101/2020.03.03.20028423)., 2020. 17. [17].J. Y. W. W. e. a. Xiaowei Deng, “Case fatality risk of novel coronavirus diseases 2019 in China,” [https://doi.org/10.1101/2020.03.04.20031005](https://doi.org/10.1101/2020.03.04.20031005)., 2020. 18. [18]. K. Z. Davide Manca, “Analysis of COVID-19 data on numbers in intensive care from Italy,” European Society of Anaesthesiology, 2020. 19. [19].K. K. A. Z. G. C. Kenji Mizumoto, “Estimating the asymptomatic proportion of coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship, Yokohama, Japan,” Eurosurveillance, vol. 25, no. 10 ([https://doi.org/10.2807/1560-7917.ES.2020.25.10.2000180](https://doi.org/10.2807/1560-7917.ES.2020.25.10.2000180)), 2020. 20. [20].E. C. f. D. P. a. Control, “Disease background of COVID-19,” 2020. 21. [21].Z.-y. N. Y. H. e. a. Wei-jie Guan, “Clinical Characteristics of Coronavirus Disease 2019 in China,” New England Journal of Medicine, no. doi:10.1056/NEJMoa2002032, 2020. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1056/NEJMoa2002032&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=32109013&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2020%2F04%2F29%2F2020.04.24.20078030.atom) 22. [22].S. Q. Z. L. J. R. J. X. Pengfei Sun, “Clinical characteristics of 50466 patients with 2019-nCoV infection,” no. DOI: 10.1101/2020.02.18.20024539. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoibWVkcnhpdiI7czo1OiJyZXNpZCI7czoyMToiMjAyMC4wMi4xOC4yMDAyNDUzOXYyIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjAvMDQvMjkvMjAyMC4wNC4yNC4yMDA3ODAzMC5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 23. [23].S. Z. Q. L. Zian Zhuang, “Preliminary estimating the reproduction number of the,” no. [https://doi.org/10.1101/2020.03.02.20030312](https://doi.org/10.1101/2020.03.02.20030312), 2020. [1]: /embed/graphic-6.gif [2]: /embed/graphic-7.gif [3]: /embed/graphic-8.gif [4]: /embed/graphic-9.gif [5]: /embed/graphic-10.gif [6]: /embed/graphic-11.gif [7]: /embed/graphic-12.gif [8]: /embed/graphic-13.gif [9]: /embed/graphic-14.gif [10]: /embed/graphic-20.gif [11]: /embed/graphic-21.gif