Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

ORIGINAL ARTICLE

Spatial and Temporal Analysis of Covid-19 in Sibu, Sarawak

Dora Jeta1,2, Alia Azmi1
1
Centre of Environmental Health & Safety, Faculty of Health Sciences, Universiti Teknologi MARA Puncak Alam Campus, 42300
Puncak Alam, Selangor, Malaysia
2
Pejabat Kesihatan Bahagian Sibu, Wisma Persekutuan Blok 3, Jalan Teng Chin Hua, 96000 Sibu, Sarawak.

ABSTRACT

Introduction: The high incidence of COVID-19 highlights the need for better understanding of the factors
that contribute to disease transmission and its geographic distribution, which could help predict global
outbreaks and enhance public health initiatives. The aim of this study was to analyse the spatio-temporal
pattern based on the weekly COVID-19 surveillance data in Sibu, Sarawak. Methods: This is a retrospective
study involving secondary COVID-19 data obtained from Sarawak State Health Department report of cases
in Sibu division from weeks 1-26 in 2021. Data were analyzed using IBM SPSS and ArgGIS (Kernel Density
Analysis and Average Nearest Neighbour Analysis). Results: According to the summary report of Average
Nearest Neighbour Analysis, the nearest neighbour ratio is equal to 0.020554, with a z-score of -229.845638
and a p-value of 0.00000. The pattern displays clustering because the average nearest neighbour ratio, or
index, is smaller than 1. With a z-score of -229.845638, the chances that this clustered pattern is the result
of random chance is less than 1%. Analysis also revealed that all districts had a clustered pattern.
Conclusion: Spatial and temporal analysis can also offer an evidence-based approach for prevention and
control activities, not just for COVID-19 but also for other infectious diseases. In this study, district level maps
of the COVID-19 distribution were created to provide detailed information to the local public health
authorities so they may create appropriate SOPs or interventions to control the COVID-19 outbreak.

Keywords: Spatial analysis; Temporal analysis; COVID-19; Public health

Corresponding Author: Although data on COVID-19 is now widely accessible,

Alia Azmi, PhD there is still much to be learned. Additionally, given
Email: [email protected] that COVID-19 is now in its endemic phase, the global
Tel: +603-32584576 health community needs to have the understanding
and resources necessary for better monitoring of this
INTRODUCTION disease, particularly for public health workers. To
that end, numerous studies on spatial and temporal
As the most recent pandemic in human memory, analysis have been conducted. For instance, a
it stands to reason that COVID-19 will remain fresh COVID-19 study was conducted in Nepal, and the
in the minds of the global population for the next findings of the spatial analysis were beneficial to the
several years. Most would still remember how it relevant authorities in helping them make appropriate
started, with only several cases initially found in decisions and carry out a comprehensive assessment
Wuhan, China. In December 2019, China notified and management that included the implementation
the outbreak of the COVID-19 to the World Health of efficient public health measures in the various
Organization, prompting officials to take actions provinces of Nepal (2). Meanwhile, in the United States,
by closing down the market that was believed to be a study employing spatial transmission showed the
the source of the outbreak. However, cases spread main regions that will be the focus of public health
rapidly, first throughout China, before finally entering efforts to reduce COVID-19 outbreaks (3). A case study
the international community. Globally, the number in Iraq used GIS spatial analysis to illustrate the spread
of infected people with COVID-19 is 17 times that of risk of COVID-19 across the entire country, through
he SARS-CoV outbreak in Guandong, China, in 2003 a comparison of maps between the distribution of
and 50 times that of the MERS-CoV outbreak in 2012, actual and predicted infected individuals across all of
which had also become a global public health the provinces (4). These maps allow decision-makers
concern (1). in the health and government sectors to anticipate

Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023 85

Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

events in order to develop proactive plans to combat Sibu (12). Sibu, a district in Sarawak, had reported
the disease. In addition, a study of GIS-based review 37 COVID-19 positive cases on 9th January, 2021,
for monitoring spatial distribution of COVID-19 in which kickstarts the emergence of the first cluster in
Haryana has been carried out to map the real-time Sibu, known as Cluster Passai (13).
mapping of infected area at a variety of zones since
it was so beneficial in making cooperation among For the record, no spatial temporal analysis of
other supporting agencies easier (5). COVID-19 had been done in Sarawak nor Sibu,
and this study can pave the way for further study on
The examples given make it evident that GIS is a epidemiology towards precision public health. This
great tool since it aids in better planning, decision- study may be useful for the management of other
making, and mobilization, particularly by offering infectious diseases as well as the early case detection
more tools for the dynamics of spatial transmission, and of future outbreaks and pandemics. Understanding
it helps to improve infection count estimations. The the geographic distribution of the infections is crucial
usage of this technique is also possible for identifying for forecasting local epidemics and developing public
potential new outbreaks, enhancing the location of health interventions. The spatial distribution and
health care facilities and controls, comprehending and clustering pattern of COVID-19 cases from January to
mapping people mobility, and other related tasks (6). June 2021 were mapped using temporal and spatial
In fact, the use of temporal and spatial analysis will analysis, which will assist in identifying potential
increase precision public health through possibilities high-risk regions. Undoubtedly, more precise disease
for targeted surveillance and intervention, thus surveillance can also reveal disease causes for
allowing for better pandemic management (7). With improved prevention methods. Understanding the
the aim of providing the right intervention at the spatial and temporal dynamics of COVID-19 is
right time, precision public health makes use of a essential for mitigation as it facilitates planning, decision-
combination of already accessible technology to making, and community action.
enable a more precise description and analysis of
people and their surroundings over the course of a MATERIALS AND METHODS
person’s life. This improves population health overall
and enables customization of preventative actions for Data collection
at-risk populations. This is a retrospective study using COVID-19 data
obtained from reports of cases in Sibu from weeks
In Malaysia, COVID-19 was confirmed to have 1-26 in 2021. The cases were from Sibu Division,
reached the country in January 2020, following the which consist of Sibu, Kanowit and Selangau. The
first outbreak in Hubei Province, China (8). Within data used is the cumulative cases of COVID-19
a few weeks, Malaysia had recorded the largest that are registered in the Sarawak State Health
cumulative number of confirmed COVID-19 infections Department, from the Ministry of Health. All
in Southeast Asia, breaching the number of 2000 confirmed cases were downloaded from e-notification
active cases by the end of March 2020 (9). Since (CDCIS) system and imported into Microsoft Excel
then, various study has been conducted to better 2007. The information from the downloaded data
understand the spatial and temporal analysis of only included basic information of patients, such
COVID-19 in this country; and this particular study as name, age, occupation, and case’s location
is one of the effort made towards better management coordinates.
of COVID-19. To note, this study concentrated
primarily on a central division in Sarawak, Malaysia, Statistical Analysis
rather than the whole country. Unlike other regions Data were analyzed using IBM SPSS Statistical
of Malaysia, Sarawak’s population is distributed software version 21 (IBM SPSS Statistics) and ArgGIS
uniquely, which is why is it used as a case study software version 10.8.2 (Kernel Density Analysis and
for this research. Even though the government had Average Nearest Neighbor Analysis). The following
offered a variety of resources to help the state formulas define how the kernel density for points
manage the rising number of cases, the high rise of is calculated and how the default search radius is
COVID-19 cases in 2021 put a pressure on Sarawak’s determined within the kernel density formula.
public health system (10). As most of the COVID-19
cases in Sibu were from longhouse, this made the Predicting the density for points;
study location more unique. In longhouse’s structure, The predicted density at a new (x,y) location is
the veranda area, also referred to as the “ruai,” was determined by the following formula:
connected to each other, and made it difficult to
control how people moved inside the longhouse (11).
Before Malaysia’s ban on interstate travel on January
13, 2021, COVID-19 clusters had already started in

86 Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023

 Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

where:
• i = 1,…,n are the input points. Only include points
in the sum if they are within the radius distance of
the (x,y) location.
• opi is the population field value of point i, which is
an optional parameter.
• disti is the distance between point i and the (x,y)
location.

The calculated density is multiplied by the number

of points or the total population field. With this
adjustment, the spatial integral will no longer
always equal 1, but will instead equal the number
of points (or sum, or population field). This
implementation makes use of a Quartic kernel (14). Fig. 1 : Clusters of Covid-19 in Sibu Division.
Every location’s density has been calculated using
the formula. The calculations are applied to the centre
of each cell in the output raster since a raster is being
generated. For distance analysis, the Average Nearest
Neighbour (ANN) method is used. The results of the
ANN analysis give the p-value and z-scores of the
COVID-19 cases. As a result, the distribution’s
pattern can be divided into three categories: clustered,
dispersed, and random.

RESULTS

Fig. 1 displays the COVID-19 cluster’s within the Sibu

division. From the figure, more than 15 different clusters
emerged in the first half of 2021 with the biggest being
the Passai cluster located in Sibu district. Meanwhile,
Fig. 2 : Epidemic Curve of Sibu Division (Half year of
the epidemic curve of COVID-19 cases in the Sibu
2021).
division for the first half of 2021 is depicted in Fig. 2.
Epidemic curves, or epid curves, are helpful because
it can reveal details about the pattern of spread,
magnitude, outliers, time trend, and the duration of
exposure to the disease or its incubation period (15).
Although the graph only displayed the overall number
of COVID-19 cases for the first half of 2021, it is
obvious that since the start of Passai cluster near Sibu/
Selangau border (Fig. 1) in second week of January,
there were daily records of more than 5 cases. The
most cases ever reported were on May 7, 2021, at 245
cases. Table I displays the incidence rate, infection
rate, and mortality rates for Sibu from January to
June 2021. Sibu district had a total incidence rate
of 5029.56, followed by Kanowit with 5443.12 and
Selangau with 5775.89, with Sibu district having a Fig. 3 : Trend of COVID-19 Cases according to Districts
rate of 4912.60. Selangau district had the highest (Half year of 2021).
Infected Rate (IR) (57.75888514%), Kanowit district
came in second (54.43116668%), and Sibu district
came in third (49.12599044%). In the first half of starting from January until June 2021, while Fig. 4
2021, the COVID-19 death rate was over 1.0% in presents visualization of the same data. As seen in
both the Sibu and Selangau districts. All population Fig. 3, Sibu district had the highest overall number
data were obtained from the Sarawak Government’s of cases with 12,207 cases reported for the first half
official website (16). of 2021. Since the Passai cluster, more cases were
reported with EW7 recording the highest number at
Fig. 3 displayed the COVID-19’s overall trend broken 718 cases. In contrast, the overall number of cases
down by district and epidemiological week (EW) in the Kanowit area increased in EW3 (134 cases)

Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023

87
Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

Table I : Summary of Descriptive Statistic of COVID-19 in Sibu (Jan-June, 2021)

Districts Total Cases Population Incidence Rate Infected Rate (IR) Death Rate (DR)
per 100,000 per 100 per 100

Sibu 12,183 247,995 4912.60 49.12599044 1.05064434

Kanowit 1,576 28,954 5443.12 54.43116668 0.697969543
Selangau 1,318 22,819 5775.89 57.75888514 1.062215478

Total for Sibu

15,077 299,768 5029.56 50.2955619 1.014790741
division

and EW4 (126 cases). Following a somewhat stable

number of cases, EW17 recorded a jump in the
number of cases (196 cases) in Kanowit. Overall, in
Kanowit with 1,604 number of cases were reported
during the first half of 2021. Unlike Sibu and Kanowit,
however, the trend in the Selangau were flat initially.
Cases increased starting from EW15 with Selangau
recording 1,318 cases overall for the first half
of 2021.

Fig. 5 displays the hotspots that was calculated using

Kernel Density Estimation (KDE). The map displayed
the spatial distribution of COVID-19 cases over a
six-month period. Red indicated high, yellow indicated
low, and green indicated no cases. The figures are
organized according to the months, starting with
January (A), February (B), March (C), April (D), May
(E), and ultimately June (F). Based on these figures, it
is evident that between January and March 2021,
Sibu district exhibited the majority of the cases. For Fig. 4 : Spatial Distribution of COVID-19 in Sibu
Kanowit and Selangau, however, a lot of changes at Division in 2021 (Jan-June).
the division level were observed after April, although
still lower than Sibu district. Even though clusters
had developed in all of Sibu’s divisions, only Sibu
district presented bigger clusters (as evidenced based
on the colour red) compared to the other two districts.
Within the six months, the pattern began to gradually
spread from Sibu districts to other districts, affecting
not only Sibu division but the entire state of Sarawak.
The clusters in the Sibu division mainly emerged
after the Passai cluster, which is the largest cluster.

Fig. 6 depicts the COVID-19 Average Nearest

Neighbour (ANN) Summary for the Sibu division.
Almost similar to the KDE result in Fig. 5, the ANN test
is appropriate for infectious diseases like COVID-19
because it not only graphically displays the hotspot
pattern but also displays the significance level
of the data under test. ANN provides a detailed
explanation of what had occurred in the Sibu division
between January and June of 2021. The Sibu division
map polygon in shapefile was also used to execute
this test, along with the data that was evaluated Fig. 5 : Hotspot Identification of COVID-19 in Sibu
using the coordinates of each COVID-19 case. Fig. 6 Division.

88 Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023

 Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

ethnic in Sarawak) is mostly to blame for the high

number of occurrences, as travelers were required to
take tests prior to their journey, even when moving
between districts. Despite the mass testing and the
application of other Standard Operating Procedures
(SOP) for the Gawai and Aidilfitri festivals, all failed
in flattening the epid curve. Not only that, The Health
Ministry of Malaysia also suggested keeping the
prohibition on interstate travel throughout the Eid
holiday season, yet the number of COVID-19 cases
was still increasing (17). Following the increase of
the number of cases during Eid season, claims were
made that numerous Muslims disregarded the
COVID-19 safety regulations that forbade them from
seeing one another, bringing the Malaysia’s overall
cases to 549,514 with 2,552 fatalities. Additionally,
about 40% of COVID-19-related deaths in Malaysia
had taken place during that festive month (18). Not
only that, although measures were in place to reduce
movement during the following Gawai Festival, the
epid curve still did not flatten (19). The epid curve
highlights that despite measures, it is difficult to
Fig. 6 : Average Nearest Neighbour Summary of restrict the number of cases at times of high
COVID-19 in Sibu Division. population density, such as Cluster Passai or any
cluster that occurred after Aidilfitri and Gawai.
indicates the summary report. According to the Regardless, the epid curve highlights the potential
figure, the nearest neighbour ratio is equal to to anticipate a high number of new clusters following
0.020554, with a z-score of -229.845638 and a a specific time or period, such as a huge gathering or
p-value of 0.00000. The pattern displays clustering any celebrations. Therefore, this can be used as
because the average nearest neighbour ratio, or index, a monitoring tool to increase public health vigilance.
is smaller than 1. This is expected, as COVID-19 is
an infectious disease, therefore easily transmitted. In Additionally, in contrast to Sibu district, the other two
fact, with a z-score of -229.845638, the chances that districts do not have nearly as many cases, despite
this clustered pattern is the result of random chance having a few clusters. The significant number of cases
is less than 1%. The result of the ANN highlights the that were registered in Sibu districts was also largely
need to strengthen prevention measures as well as to due to the highly convenient facility that Sibu offered
better understand the risk behaviours associated with for handling COVID-19 cases. As COVID-19 cases
the spread of COVID-19. Since COVID-19 infections were reported under the district where the patients
can spread from one person to another via airborne were tested, this undoubtedly had an impact on
or airdrops, clusters emerged rapidly. This puts not Sibu’s COVID-19 statistics and indicates the need to
only additional burden on public health officers to organize the COVID-19 outbreak response better.
maintain stringent prevention measures, but also to
the public, as close contact carries high risk of Not only that, Sibu district also serves as the
transmission. geographic centre of Sarawak state as well as the
centre of Sibu division. People who drive or relocate
DISCUSSION to another division for work tend to stay in the Sibu
district before moving to another districts. Due to
The study’s findings revealed a significant variation in its localized position between other divisions such
the spatial distribution of COVID-19 incidence in Sibu as Bintulu, Sarikei, Kapit and Mukah, Sibu is a
division, as well as the fact that the geographic range convenient address to reside in. In fact, during the
of confirmed cases has expanded over time in other first two years of the pandemic, anyone entering the
districts: Kanowit and Selangau. There were certain Bintulu division was required to submit to a
reasons why there were high number of cases in COVID-19 test, and since Sibu division was where
certain period. For example, the Hari Raya and Gawai most individuals were tested, it is thus where the cases
festival that falls on 13 and 14 May 2021 (EW19) were registered. All of this contributed to the high
and 1 and 2 June 2021 (EW22) meant that high rate number of cases in the district.
of samples collected before Sarawakians celebrated
Aidilfitri (Muslim holiday celebrated after Ramadhan) The findings of this study, which provide details on
and Gawai Festival (holiday celebrated by Dayak the epidemiological distribution of COVID-19 in Sibu

Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023 89

Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

division, can be used to help develop more precise COVID-19 using geographic information system
public health measures that will lessen the effects of (March and April, 2020). Available from: https://
viral transmission. Overall, KDE was used to portray doi.org/10.1007/s40201-020-00565-x
the overall condition that existed in the Sibu division 2. Basnet BB, Pant RR, Bishwakarma K, Paudel S,
and to visualise the COVID-19 cases, while ANN Pandey N, Adhikari SK, et al. A Year Trend Analysis
was crucial in identifying whether the pattern of and Spatial Distribution of COVID-19 Cases in
distribution was clustered, dispersed, or random. It Nepal. Asia Pac J Public Health. 2021;33(5):641–
also lays the foundation for future studies on socio- 4.
demographic factors that affect changing disease 3. Andersen LM, Harden SR, Sugg MM, Runkle JD,
patterns. The results can also be used to identify Lundquist TE. Analyzing the spatial determinants
hotspots, or areas that require more organized of local Covid-19 transmission in the United States.
surveillance and public health measures. Vol. 754, Science of the Total Environment. 2021.
4. Moneer Yahya B, Samier Yahya F, Ghazi Thannoun
CONCLUSION R. COVID-19 prediction analysis using artificial
intelligence procedures and GIS spatial analyst: a
By utilising GIS technologies, this study was able case study for Iraq. 2021; Available from: https://
to visualize the COVID-19 distribution. The study’s doi.org/10.1007/s12518-021-00365-4
findings revealed the spread of COVID-19 in the Sibu 5. Kumar S, Singh R, Veer V, Jambheshwar Sr Sec
division for six consecutive months in the first half School G, Professor A. GIS-Based Review for
of 2021 by using KDE and ANN. By illustrating the Monitoring the Spatial Distribution of Covid-19: A
spatial spread of the disease, health authorities can Case Study of Haryana. 2022;10.
develop a more precise surveillance system that is 6. Franch-Pardo I, Napoletano BM, Rosete-Verges
more capable in predicting COVID-19 clusters. As F, Billa L. Spatial analysis and GIS in the study
a result, control measures can be carried out more of COVID-19. A review. Science of the Total
effectively, without wasting unnecessary resources. Environment. 2020;739.
Spatial and temporal analysis can also offer an 7. Curtis AJ, Ajayakumar J, Curtis J, Brown S. Spatial
evidence-based approach for prevention and control Syndromic Surveillance and COVID-19 in the U.S.:
activities, not just for COVID-19 but also for other Local Cluster Mapping for Pandemic Preparedness.
infectious diseases. In this research, district level Int J Environ Res Public Health. 2022 Aug 1;19(15).
maps of the COVID-19 distribution were created. 8. Edinur HA, Safuan S. Sri petaling COVID-19
In contrast to other studies that use the mapping cluster in malaysia: Challenges and the mitigation
method at the national level, the focus of this study strategies. Acta Biomedica. 2020;91(4):1–4.
is more on the division level to provide detailed 9. The Straits Times. Malaysia reports new daily
information to the local public health authorities high of 15,573 Covid-19 cases. The Straits Times
so they may create appropriate SOPs or interventions [Internet]. 2021 Jul 24; Available from: https://
to control the COVID-19 outbreak. In light of this, the www.straitstimes.com/asia/se-asia/record-high-
researcher believes that the Ministry of Health should of-15573-new-covid-19-cases-in-malaysia-more-
prioritize spatial and temporal analysis in order to variants-of-concern-identified
manage infectious diseases especially COVID-19. 10. Md Hamzah N, Yu MM, Fong See K. Assessing the
Lessons learnt from managing COVID-19 will also efficiency of Malaysia health system in COVID-19
be essential for future references largely due to the prevention and treatment response. Available
threat of future pandemics as the result of climate from: https://doi.org/10.1007/s10729-020-09539-
change. 9
11. Mahayuddin SA, Harun SN, Mahat N.
ACKNOWLEDGMENT Sustainability Of Building Elements in Bidayuh
Traditional Longhouse Construction Behavioural
Authors would like to extend our gratitude to on Built Environment Sustainability (Volume 2)
Universiti Teknologi MARA for the opportunity to View Project Documenting Building Typology,
carry out this research. Ethical approval for Construction Method And Environmental
this research was obtained from Malaysian Performance Of Traditional Longh [Internet].
Ministry of Health Medical Research Ethics 2015. Available from: https://www.researchgate.
Committtee (MREC) with Ref: NMRR ID-22-01757 -SXS net/publication/313629690
(IIR). 12. Hazlin Hassan. Malaysia bans interstate, inter-
district travel from May 10 to June 6 to curb
REFERENCES Covid-19. New Straits Times. 2021 May 10;
13. Phillip Wong. Major Covid-19 outbreak in Pasai
1. Shariati M, Mesgari T, Kasraee M, Jahangiri-Rad M. Siong, Sibu; 37 cases confirmed, eight longhouses
Spatiotemporal analysis and hotspots detection of under lockdown. Borneo Post Online. 2021 Jan 9;

90 Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023

 Malaysian Journal of Medicine and Health Sciences (eISSN 2636-9346)

14. Gesler W. The uses of spatial analysis in medical 17. Malaysia’s health ministry calls for ban on Eid
geography: A review. Soc Sci Med. 1986 Jan travel amid surge in COVID-19 cases. Reuters.
1;23(10):963–73. 2021;1–9.
15. Torok M. Epidemic Curves Ahead. 2021. 18. Gile M. 850 COVID-19 Cases in Malaysia Linked
16. Sarawak Government. Sarawak Population. to Eid Festival. 2021;1–17.
Sarawak Government [Internet]. 2014;7–10. 19. Bernama. “Please don’t come back for Gawai.”
Available from: http://www.sarawak.gov.my/web/ 2021;2021–4.
home/article_view/240/175/

Mal J Med Health Sci 19(SUPP10): 85-91, Sept 2023 91