key: cord-0300627-xb73t8j2
authors: Luo, W.; Liu, Z.; Zhou, Y.; Zhao, Y.; Li, Y. E.; Masrur, A.; Yu, M.
title: Spatiotemporal patterns and progression of the Delta variant of COVID-19 and their health intervention linkages in Southeast Asia
date: 2021-12-21
journal: nan
DOI: 10.1101/2021.12.20.21268140
sha: cdb89977a96826e971cd8d939ec9e4cff3447d4c
doc_id: 300627
cord_uid: xb73t8j2

The global pandemic of COVID-19 presented an unprecedented challenge to all countries in the world, among which Southeast Asia (SEA) countries managed to maintain and mitigate the first wave of COVID-19 in 2020. However, these countries were caught in the crisis after the Delta variant was introduced to SEA, though many countries had immediately implemented non-pharmaceutical intervention (NPI) measures along with vaccination in order to contain the disease spread. To investigate the potential linkages between epidemic dynamics and public health interventions, we adopted a prospective space-time scan method to conduct spatiotemporal analysis at the district level in the seven selected countries in SEA from June 2021 to October 2021. Results reveal the spatial and temporal propagation and progression of COVID-19 risks relative to public health measures implemented by different countries. Our research benefits continuous improvements of public health strategies in preventing and containing this pandemic.

circumstance, monitoring outbreaks and identifying the space-time clusters of infection have become significant for a coordinated response to the epidemic in SEA.

Spatio-temporal analysis has been widely used in research of COVID-19 propagation to illustrate the characteristics and mechanism of COVID-19 spatial propagation, which provided public health authorities with important information to help mitigate the situation [9] [10] [11] . Among diverse spatio-temporal methods, space-time scan is one of the most popular methods adopted to explore spatiotemporal clusters in a particular region around the world, such as Mainland China [12] , the United States [13] , Mexico [14] , Spain [15] , Malaysia [16] , Bangladesh [17] etc. In SEA, previous studies have applied this analysis to investigate the first wave of COVID-19 cases in SEA [18, 19] . However, those studies mainly focused on each individual country without exploring propagation patterns and progression characteristics with collective public health interventions at the regional scale. Previous research has shown that regional-wide coordination could interrupt transmission of COVID-19 in an effective way [10, 17, [20] [21] [22] . In order to contain the emerging epidemic of COVID-19 and minimize its risk, countries in SEA have deployed various preventive and containment measures such as lockdowns, distancing restrictions, and mandatory tracking and trace method [23, 24] .

Delta variant transmission had shown significant spatial heterogeneity in SEA because different countries adopted different interventions along with the spread. Hence, this paper aims to identify the space-time clusters of outbreaks of COVID-19 caused by SARS-CoV-2 delta variant in SEA. We utilized district-scale daily confirmed cases of seven SEA countries from June 2021 to October 2021 to identify the active and emerging clusters of the disease outbreaks and summarized the relative policies to investigate the potential linkage between government interventions and the transmission. Our work contributes to regional surveillance of COVID- 19 progression in SEA and provides essential information of COVID-19 propagation from space and . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint time perspective to public health authorities, which is beneficial to timely policy making according to the dynamic COVID-19 situation.

Our study focused on seven countries in SEA, namely Indonesia, Malaysia, the Philippines, Singapore, Thailand, Vietnam and Brunei, because they disclose relevant data at a district level. During the second COVID-19 outbreak caused by the Delta variant, the dynamic of interventions implemented by different countries along the way may cause the fluctuation of transmission. For example, Thailand, Singapore, Malaysia, Philippines, and Vietnam began to relax their restrictions around August, which may cause significant changes in the epidemic patterns after then ( Table 1) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021.

We obtained or extracted data of COVID-19 confirmed cases from the official websites of public health authorities in the seven countries and Johns Hopkins University's Center for Systems Science and Engineering GIS dashboard ( Table 2) . We adopted data from 1 June 2021 to 31 October 2021, which is the approximate date of the second COVID-19 outbreak of these seven countries in SEA ( Figure 1 ). We aggregated the data at the first administrative level except for those in Singapore and Brunei which were aggregated at country level considering the similar magnitude of area and population in each analytic unit. We obtained or extracted population data from statistical reports and yearbooks from those countries ( Table 2 ). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

To explore emerging and active space-time clusters of COVID-19 cases in SEA, we is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint reaching a specific maximum spatial and temporal upper bound which is set to 10% of the population risk and 50% of the study period respectively in this study. Besides, we defined the minimum duration of each cluster to two days for surveillance of the continuously existing clusters. Moreover, the minimum number of cases in each cluster is set to 3 in order to ensure that there must be at least three cases in each cluster.

We assumed that the COVID-19 cases follow a Poisson distribution according to the population of study units in SEA. The null hypothesis H0 indicates that the model reflects infection of COVID-19 having a constant intensity μ within or outside the cylinder, which is proportional to at-risk population. Alternative hypothesis HA indicates that the observed cases are more than expected cases, which reflects an increased risk within a cylinder. Expected cases are calculated by Equation (1) [13, 29] :

where p represents the population within a study unit, C represents the total COVID-19 cases in our study area (i.e., seven countries in SEA), and P represents the total estimated population within our study area.

A maximum likelihood ratio test is used to evaluate the null and alternative hypotheses. It can identify scanning windows with an elevated risk for COVID-19, which is defined by equation (2) [30, 31] :

where represents the likelihood function for cylinder Z; represents the likelihood function for H0; represents the number of COVID-19 cases in a cylinder; represents the number of expected cases in cylinder Z; N represents the total number of observed cases for the seven countries in SEA across all time periods. When the likelihood ratio . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ;  is greater than 1, there is an elevated risk in the cylinder, and the cylinder with the maximum likelihood ratio should be the most likely cluster.

The relative risk of COVID-19 is assumed homogeneous throughout different districts within the same cluster. To make the results more reasonable, we calculated relative risk (RR) for each study unit within a cluster to explore the spatial heterogeneity of relative risk of COVID-19 (equation (3)) [32] :

where c represents the total number of COVID-19 in a study unit; e represents the total number of expected cases in a study unit; C represents the total number of observed cases in the seven countries of SEA. The formula indicates that RR represents the estimated risk in a study unit divided by the risk outside that unit. Specifically, if a location (cluster or study unit) has a RR of 3, the population within the location are 3 times more likely to be exposed to COVID-19 infection than its outside.

In the following sections, the results reveal significant emerging clusters of COVID-19

cases in seven countries of SEA from 1 June to 31 October. Considering that some SEA countries (e.g., Malaysia, Philippines, Singapore, Thailand) started loosening their restrictions from August along with strengthening their vaccination plan, we divided the timeline into two parts (i.e., June 1 to August 31 and June 1 to October 31) in order to identify the dynamics of clusters. Additionally, we explored the variation of RR across each district in each half-month, which is approximately equal to the incubation period of an infection [33] , using a cumulative half a month prospective scanning approach. Then we compared the interventions with the discovered space-time characteristics to identify the potential linkage between political intervention and progression of the Delta variant of COVID-19.

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

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint

As shown in table 3, 14 significant space-time clusters were identified during June 1 and August 31 in SEA, including 7 high-risk clusters (RR > 1) and 7 low-risk clusters (RR < 1). Most of the high-risk clusters emerged between mid-July and late-August, which means the situation of COVID-19 in SEA became severe during this period. Specifically, Philippines also emerged as high-risk clusters from July 22 to August 31, and from August 11 to August 31 respectively. There were also 2 clusters emerging in Indonesia and they contained only 1 district, which was Jakarta, with an RR of 2.38, and Daerah Istimewa Yogyakarta (DIY), with an RR of 3.88. Three high-risk clusters in Indonesia reveal that population within a number of regions in Indonesia were more likely to be exposed to COVID-19 compared to other regions in SEA during this period.

Additionally, there were 7 low-risk clusters distributing across other regions of SEA (e.g., north of Vietnam, south of Philippines, some other districts of Indonesia), which means population within these clusters were less likely to be exposed to COVID-19.

Note that, cluster 10, with an RR of 0.60, contained 2 high-risk districts in Philippines (i.e., Region VII, with an RR of 1.07, and Region X, with an RR of 1.08). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; 

Eleven significant clusters were detected between June 1 and October 31, which was three less than the early phase. Among the clusters, there were only 4 high-risk clusters (RR > 1), decreasing from 7 in the previous period ( Table 4 ). The most likely cluster was the same as one of the previous periods, which covered partial districts of Thailand and Malaysia. The RR of cluster 1, however, decreased from 5.45 to 3.91 in this period, while the number of high-risk districts increased from 39 to 45, which implies that more districts were influenced by COVID-19. Cluster 2 was also the same as one of the previous periods, containing Binh Duong and Ho Chi Minh city. Besides, cluster 7 evolved from cluster 4 in the previous period, with the exclusion of Kalimantan Timur. The RR of this cluster increased from 3.50 to 4.62, which means that people in this cluster were more likely to be infected in this period. Similarly, Cordillera Administrative Region, Region II, and Region I merged with one high-risk cluster in the north of Philippines and formed a larger cluster. The rest of the clusters were low-risk clusters with RR less than 1. Figure 3 visualizes the distribution of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint the capital of Indonesia. Note that Jakarta was one of the major emerging risk districts early in the second outbreak ( Figure 5 ) and was still faced with a relatively high RR (2.79) until the end of our study period. Nevertheless, characterized as the largest city nationwide with a high-density population, Jakarta's success in preventing exacerbation of risk impact of COVID-19 is most likely to be related to its persistent is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint . Also, Ho Chi Minh City has a larger population base and more developed economy than other districts which resulted in more infections via contact with a relatively larger number of crowds. It has been suggested that population density and contact intensity are the main drivers for the propagation and amplification of this virus [36] . As for Malaysia, 10 out of 16 districts showed an increase in RR and took up a major proportion of the country. Although Labuan and Kuala Lumpur observed the most obvious decrease among SEA districts, it should be noted that they were also the most severely threatened districts by COVID-19 risks early in the study period (RR>10) and remain the relatively highest RR amidst SEA until the end of the study. This is probably because the Malaysian government gradually adopted more loose measures than the other SEA countries with its phase development of the National Recovery Plan since the beginning of our study period. Additionally, Singapore saw an increased RR of 1.43, which is probably related to its intermittent relaxation of social restrictions. RR in Brunei also increased, but it still remained less than 1. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint relative risk of some districts slightly declined (e.g., Bangkok, Samut Sakhon, Jakarta, Riau, Sarawak, etc.). Note that, most of the countries tightened the restriction before this time except for Malaysia, which loosened their restrictions. This phenomenon also reflected the strong infection and danger of the Delta variant. In the next one and a half month (i.e., August 15 to September 30), the situation of Indonesia had been improving, which was inseparable with continuous extension of activity restrictions in Indonesia (e.g., extended PPKM for the ninth time). The north of Philippines, however, showed increased relative risk in mid-September, which may be due to the relaxing restriction of Manila capital region on August 19 and mid-September ( Figure 6 ). In the next month, relative risk of Singapore increased from 0.44 to 1.53, which was probably due to gradually loosened restrictions. Besides, the situation across north of Malaysia and south of Thailand was getting worse as well. Up to October 31, the situation of Thailand, Malaysia, Singapore, capital areas in Vietnam, and capital areas in Philippines was still alarming, which should be noticed with further surveillance by obtaining daily case reports in the future. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

In this study, we utilized prospective space-time scan statistics to detect the emerging and existing space-time clusters of COVID-19 in SEA. We found that most districts in is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint restrictions were beneficial on epidemic control, especially for those developing regions with weak public health systems and relatively low vaccination rates. To our knowledge, this is the first attempt to explore the space-time progression of delta variant outbreak in SEA, as well as summarize the potential linkage between the epidemic dynamics and diverse public health policies. Spatio-temporal statistics is an effective method to identify high-risk clusters and regions [37] , which can help explore spatial temporal patterns of epidemic progression and better understand the dynamics and characteristics of the Delta variant in SEA. Furthermore, by combining the results with relevant measures that reflect the appropriateness of public health interventions, governments can also timely improve the interventions. In the meantime, health authorities can timely adjust their interventions by comparing the RR with current measures, which may reflect the appropriateness of these interventions. In short, the prospective space-time scan can continuously be used to monitor the dynamics of COVID-19 with the most updated data, and timely adjust the potential gaps in domestic public health intervention to prevent further deterioration of the pandemic.

Public health interventions played an important role in epidemic containment, in which social restriction policies effectively mitigate the propagation of COVID-19 [29] .

Restrictions of mass gathering and travel, keeping social distance, and reducing human mobility are beneficial to control COVID-19 because these measures could reduce the probability of exposure to virus infection [38, 39] . Our study discovered the potential is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ; https://doi.org/10.1101/2021.12.20.21268140 doi: medRxiv preprint restriction would be less valuable than domestic measures. In this case, authorities should emphasize more on domestic intervention in order to reduce community transmission [40] . Nevertheless, it is well known that those intervention measures against COVID-19 come with economic costs from other perspectives (i.e., labor market etc.). When making intervention regulations, economic and social justifications for public health policies are one of the priorities that governments would consider [41] . Hence, it is one of the challenges for all countries to live with COVID-19 and weigh the balance between epidemic development and social economic loss [42] . Given the social and economic status, most countries in SEA have decided to change their strategies from the elimination of cases to living with COVID-19 since August 2021 despite the new outbreak of Delta variant [8] . A concurrent trend observed in SEA is that all countries except for Indonesia have been gradually loosening social restrictions, allowing international communications whilst boosting vaccination attempting to achieve group immunity among citizens. For instance, the Singapore Ministry of Health believes that with the assistance of a high vaccination rate, it is assured that the growth of COVID-19 cases, though cannot be eliminated, will be controlled at a certain level. By doing this, economy and social norms could be restored without causing another disease outbreak or a breakdown in the hospital system [43] . Vaccination is increasingly essential to protect the crowd from the exacerbating threat of morbidity and mortality of this Delta variant and future variant [44] . Effective vaccination is highly beneficial to avert infection, while wide-scale vaccination is proven to be able to successfully control transmission of COVID-19 so far [45] [46] [47] . However, it should be noted that vaccinations cannot be depended on entirely. The previous experiment found that effectiveness of available vaccines against the Delta variant (B.1.617.2) showed a reduction compared to previous virus variants [48] , implying that current vaccination is highly likely to become ineffective in defending against many mutations in the future [49] . On the other hand, considering the fragile health systems in SEA, contact tracing, quick isolation and strict restrictions are still essential to avoid any miscalculation and prevent another outbreak [50, 51] . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021. ;  International coordination including vaccine allocation, food, finance, and equitable access is essential for the world to respond to the pandemic [52] . The pandemic did not only cause the breakdown of developing countries, but also developed countries, which indicates that countries should enhance transparency, share information, and strengthen cooperation to protect international health and security [53] . As for Southeast Asia, due to strong socioeconomic connectivity among all countries, it is vital for countries in SEA to adopt collaborative policies to overcome the challenges of COVID-19. Sustainable development in SEA during the pandemic era requires multisectoral cooperation including tourism, economy, health etc., which can help effectively coordinate border restrictions, international trade, or allocation of medical supplies. To ensure cooperation among countries in SEA, regional surveillance and information sharing should be indispensable for each country, which can support more precise control and prevention of COVID-19. Although some countries (Singapore, Vietnam) have made efforts to donate medical equipment and provide aid to their neighbors [3] , we hope that there will be more interactive assistance among countries in SEA in the future, especially along the continuous emergence of COVID-19 variants. Furthermore, multilateral collaboration with countries or regions (e.g., European Union) outside SEA is considered as an effective and valuable way to help respond to the pandemic, although there will still be a long way to go in the future [3] .

Overall, SEA should focus on regional coordination with long-term plans to solve a series of threats caused by the epidemic, considering the long-term challenges caused by the COVID-19 pandemic.

Despite the insights from our study, there are notable limitations in the COVID-19 data.

There are only seven out of twelve countries provided with data at the primary administrative district level, so we were not able to explore the complete propagation process in SEA. In addition, if high spatial resolution data become available (i.e., city, county, even block or subzone), more specific and detailed patterns could be revealed. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted December 21, 2021.

As a matter of fact, many previous studies were also faced with a shortage or loss of available data (i.e., insufficient pediatrics data) [54] [55] [56] . Moreover, insufficient knowledge of the data or dynamics would lead to invalidity and unreliability of responses to COVID-19 [57] . Therefore, we strongly suggest that public health authorities should disclose more representative and reliable data [58, 59] . Apart from this, COVID-19 transmission and its impacts have shown environmental inequality in terms of household income, education level, age, gender etc. [60, 61] . The potential correlation between environmental inequality and COVID-19 should be further studied using diverse data, which can provide significant insights to resource allocation and regional prevention.

Transmission of SARS-CoV-2: implications for infection prevention precautions: scientific brief

An interactive web-based dashboard to track COVID-19 in real time

Responses to COVID-19 in Southeast Asia: Diverse paths and ongoing challenges

COVID-19 in Southeast Asia: Implications for workers and unions

ASEAN Economies Hit Hard by Escalating Covid Delta Waves. Ihs Markit

A Comparative Study on Covid-19 Corona Virus Variants: Alpha

Southeast Asia is an emerging hotspot for COVID-19

Southeast Asia Nations Plan To 'Live With Virus' As Delta Surges. Thediplomat

Spatial analysis and GIS in the study of COVID-19. A review

A data driven agent-based model that recommends non-pharmaceutical interventions to suppress Coronavirus disease 2019 resurgence in megacities

Effect of nonpharmaceutical interventions to contain COVID-19 in China

Fine-scale space-time cluster detection of COVID-19 in Mainland China using retrospective analysis

Rapid surveillance of COVID-19 in the United States using a prospective space-time scan statistic: Detecting and evaluating emerging clusters

The main factors influencing COVID-19 spread and deaths in Mexico: A comparison between phases I and II

Real time surveillance of COVID-19 space and time clusters during the summer 2020 in Spain

Spatial cluster analysis of COVID-19 in Malaysia

Space-time patterns, change, and propagation of COVID-19 risk relative to the intervention scenarios in Bangladesh

Early Spatiotemporal Patterns and Population Characteristics of the COVID-19 Pandemic in Southeast Asia. Healthcare

Scenario analysis of COVID-19 transmission dynamics in Malaysia with the possibility of reinfection and limited medical resources scenarios

A hybrid multi-scale model of COVID-19 transmission dynamics to assess the potential of non-pharmaceutical interventions

COVID-19 Propagation Model Based on Economic Development and Interventions

Assessing the impact of coordinated COVID-19 exit strategies across

Lessons learned from three Southeast Asian countries during the COVID-19 pandemic

Response of ASEAN member states to the spread of COVID-19 in Southeast Asia

Southeast Asia Covid-19 Tracker

Daily surveillance of COVID-19 using the prospective space-time scan statistic in the United States

Space-time clusters and cooccurrence of chikungunya and dengue fever in Colombia from 2015 to 2016

Detecting space-time clusters of COVID-19 in Brazil: mortality, inequality, socioeconomic vulnerability, and the relative risk of the disease in Brazilian municipalities

Identifying the space-time patterns of COVID-19 risk and their associations with different built environment features in Hong Kong

Estimation of incubation period and serial interval of COVID-19: analysis of 178 cases and 131 transmission chains in Hubei province

Cabinet Secretariat of the Republic of Indonesia

High population densities catalyse the spread of COVID-19

Using geographic information systems and spatial and space-time scan statistics for a population-based risk analysis of the 2002 equine West Nile epidemic in six contiguous regions of Texas

Association between population density and infection rate suggests the importance of social distancing and travel restriction in reducing the COVID-19 pandemic

China provides important lessons for global response

Delta variant makes COVID-19 border closures unsustainable in Southeast Asia. CNA. 2021. Commentary

The political economy of the COVID-19 pandemic

The effects of different travel modes and travel destinations on COVID-19 transmission in global cities

Ministry of Health Singapore

A strategic approach to COVID-19 vaccine R&D

Effectiveness of COVID-19 vaccines in preventing SARS-CoV-2 infection among frontline workers before and during B. 1.617. 2 (Delta) variant predominance-eight US locations

COVID-19 vaccine impact in Israel and a way out of the pandemic

Epidemiological differences in the impact of COVID-19 vaccination in the United States and China

Progress of the Delta variant and erosion of vaccine effectiveness, a warning from Utah

Viral Variants and Vaccinations: If We Can Change the COVID-19

Should We? medRxiv

Impact of national containment measures on decelerating the increase in daily new cases of COVID-19 in 54 countries and 4 epicenters of the pandemic: comparative observational study

COVID-19: where do we go from here? The Lancet

How an outbreak became a pandemic: a chronological analysis of crucial junctures and international obligations in the early months of the COVID-19 pandemic. The Lancet

Obesity and immunodeficiencies are the main pre-existing conditions associated with mild to moderate COVID-19 in children

Asymptomatic infection and atypical manifestations of COVID-19: comparison of viral shedding duration

Excess deaths from COVID-19 and other causes in the US

Impact of the COVID-19 pandemic on the surveillance of antimicrobial resistance

The prevention of COVID-19 and the need for reliable data

Clinical Trial Data Sharing for COVID-19-Related Research

GIS-based spatial modeling of COVID-19 incidence rate in the continental United States