The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy, Hotspot and Risk Factors using Geographically Weighted Regression in 2020 - 2022

Tris Eryando; Fatma Lestari; Tiopan Sipahutar; Hendy Risdianto; Juliana Sutanto

doi:10.12688/f1000research.149041.1

Home Browse The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy,...

ALL Metrics

-

Views

-

Downloads

Get PDF

Get XML

Export

▬

✚

Research Article

The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy, Hotspot and Risk Factors using Geographically Weighted Regression in 2020 - 2022

[version 1; peer review: 1 approved with reservations]

Tris Eryando ¹, Fatma Lestari², Tiopan Sipahutar³, Hendy Risdianto⁴, Juliana Sutanto⁵

Tris Eryando ¹, Fatma Lestari², [...] Tiopan Sipahutar³, Hendy Risdianto⁴, Juliana Sutanto⁵

PUBLISHED 26 Mar 2026

Author details Author details

¹ Department of Biostatistics and Population, Universitas Indonesia, Depok, West Java, 16424, Indonesia
² Department of Occupational Health & Safety, Universitas Indonesia, Depok, West Java, 16424, Indonesia
³ Department of Biostatistics and Population, Universitas Indonesia, Depok, West Java, 16424, Indonesia
⁴ Health Informatic Research Centre, Faculty of Public Health, Universitas Indonesia, Depok, West Java, 16424, Indonesia
⁵ Department of Human Centred Computing, Monash University, Melbourne, Victoria, Australia

Tris Eryando
Roles: Conceptualization, Formal Analysis, Methodology, Resources, Supervision, Writing – Review & Editing

Fatma Lestari
Roles: Supervision

Tiopan Sipahutar
Roles: Conceptualization, Data Curation, Formal Analysis, Methodology, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Hendy Risdianto
Roles: Supervision

Juliana Sutanto
Roles: Supervision

OPEN PEER REVIEW

REVIEWER STATUS

This article is included in the Global Public Health gateway.

Abstract

Background

COVID-19 reported has declined in almost all countries, yet it remains critical to extract lessons from the pandemic experience. This study aimed to examine the spatial distribution of COVID-19 in Indonesia, identify hotspot areas, and model the key risk factors influencing transmission across provinces.

Method

An ecological study design was applied using aggregate data from 33 provinces in Indonesia between 2020 and 2022. COVID-19 case (dependent variable) data were sourced from the Indonesia COVID-19 Task Force, while potential explanatory variables (independent variable)—such as second-dose vaccination coverage, internet usage rates, the number of agglomeration areas, and population density—were retrieved from the Central Bureau of Statistics (BPS). Spatial analyses included global spatial autocorrelation (Moran’s Index), hotspot detection using Local Indicators of Spatial Association (LISA), and geographically weighted regression (GWR) to explore spatially varying relationships. All analyses were conducted using R i386 software (version 3.6.1).

Result

Findings revealed that COVID-19 cases were heavily concentrated in Java Island, particularly in DKI Jakarta, West Java, Central Java, Banten, and East Java Province. The spatial pattern indicated a non-random distribution, with significant clustering of high case counts in neighbouring provinces—defining Java as a national hotspot. Among the examined risk factors, the proportion of internet users consistently showed a statistically significant association with COVID-19 incidence across all provinces.

Conclusion

In conclusion, Java Island emerged as the core hotspot for COVID-19 in Indonesia, likely due to its high population density and economic centrality. Policy recommendations include prioritizing Java Island in future pandemic preparedness and improving digital infrastructure to support adaptive public health responses.

Keywords

COVID-19; health policy; Indonesia; GWR analysis; hotspot; COVID-19 risk factors

Corresponding author: Tris Eryando

Competing interests: No competing interests were disclosed.

Grant information: The author(s) declared that no grants were involved in supporting this work.

Copyright: © 2026 Eryando T et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Eryando T, Lestari F, Sipahutar T et al. The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy, Hotspot and Risk Factors using Geographically Weighted Regression in 2020 - 2022 [version 1; peer review: 1 approved with reservations]. F1000Research 2026, 15:440 (https://doi.org/10.12688/f1000research.149041.1) First published: 26 Mar 2026, 15:440 (https://doi.org/10.12688/f1000research.149041.1) Latest published: 26 Mar 2026, 15:440 (https://doi.org/10.12688/f1000research.149041.1)

Introduction

The global COVID-19 pandemic has shown a downward trend based on the 136th issue of the COVID-19 Weekly Epidemiological Update in March 2023 yet the spread of COVID-19 still exists in several countries and there are even some countries that have reported a significant increase in cases.¹ However, from the end of 2023 until today, we cannot see the number of cases worldwide representation due to the reduction in testing and reporting globally.²

Since December 2022, the Indonesian government has announced the end of the policy of enforcing restrictions on community activities (PPKM). As a result, the community is no longer required to wear masks both indoors and outdoors. However, hand washing is still recommended as part of common behaviour for clean and healthy living.³

The COVID-19 pandemic has provided lessons learned for all countries to improve the health sector and even involve many sectors. It is highly expected that the world can be better prepared to prevent or manage with the possibility of the next pandemic threat that can disrupt global health security. Various studies have been conducted by many countries to find out about the COVID-19 virus in terms of medicine and also public health in order to strengthen the global health security.

This valuable lesson is related to a country's governance, communication and financial management systems. The success that is generally seen is how the policy product carried out by a country is directly related to the pandemic product, they are mortality, morbidity and mortality rates. The variations of these products occur not only between countries, but also between regions within a country; such as in Indonesia which an archipelago with 33 provinces and 514 districts/cities that have flexibility in practicing their countermeasures.

This study is expected to provide important information about COVID-19 trends, the virus distribution pattern, and specific risk factors according to the location and geographical conditions of each area. This information can be a valuable reference for the preparedness management of a diseases with pandemic potential. The objectives of this study are:

1. To describe the trend distribution of COVID-19 cases in 33 provinces in Indonesia in 2020-2022
2. To identify the COVID-19 hotspot provinces and its trend in 2020-2022
3. To model COVID-19 risk factors in 33 province using geographically weighted regression (GWR)
4. To identify the main risk factors of COVID-19 in each province as the basic information of public health policy for infectious diseases.

Methods

Study design dan data source

This study employs an ecological study design using aggregate data representing 33 provinces in Indonesia from 2020 to 2022. This study aims to analyse spatial trends and risk factor of COVID-19 across different regions in Indonesia. This study was using secondary data which taken from various sources, thus there is no informed consent needed for this study.

Data source

COVID-19 case reports obtained from Indonesia COVID-19 Task Force (https://covid19.go.id/) and socioeconomic and demographic data generated from the Central Bureau of Statistics of Indonesia (BPS) (https://www.bps.go.id/id).

Study variables

The dependent variable is the aggregate COVID-19 cases (2020-2022). Independent variables include 2nd vaccination coverage (Prop_vacc2), proportion of internet users (Prop_internet), number of agglomeration areas in the province (Aglo), population density at the province level (density).^4–16 Agglomeration area defined as a spatial concentration of economic activity in urban settings which influenced by economies of proximity. The area includes the urban centre and surrounding buffer districts/area.¹⁷

Statistical analysis

This study follows a multi-step spatial analysis approach to determine the spatial distribution, hotspots, and risk factors for COVID-19. First step of the analysis is to determine the neighbour definition of each area. In this study, we used k-nearest neighbour with k=2.^18,19 Second step is to assess spatial autocorrelation among areas based on COVID-19 cases using Moran Index test to identify the autocorrelation. The null hypothesis in this analysis is that there is no spatial autocorrelation between regions (I=0). The third step is to determine the COVID-19 hotspot area using Moran's Scatter Plot.¹⁸ The hotspot areas definition in this study are all areas located in high-high (HH) and high low (HL) quadrant in the Moran’s scatter plot. HH describes an area with high cases surrounded by areas with high cases, while HL describes an area with high cases surrounded by areas with low cases.²⁰

In terms of finding the risk factors in each of the 33 provinces, researchers used geographically weighted regression (GWR) analysis.²¹ The steps of GWR analysis are (1) test of residual normality assumption using Anderson-Darling test (H0=the residual distribution is under the normal curve distribution), (2) residual independency assumption using Run-test (H0= the residuals are independent each other), (3) homogeneity assumption using Breusch-Pagan (H0= the residuals are homogeneous),²² (4) multicollinearity test using VIF value where VIF < 10 is acceptable, (5) determine the bandwidth value using cross validation method (used to identify the optimum bandwidth value), (6) GWR analysis using Kernel Gaussian and Kernel Bi-Square weight method.²¹

Software ad tools

All statistical analysis were conducted using R i386 (version 3.6.1) (https://www.r-project.org/).

Context of study area

During the COVID-19 pandemic in Indonesia, agglomeration areas were restricted from population movement because they were considered more vulnerable due population density and business activities carried out in the area. The data were taken from Central Bureau of Statistics of Indonesia report.

Indonesia consists of 34 provinces spread into seven major island groups: Sumatra, Jawa, Kalimantan, Sulawesi, Bali, Nusa Tenggara Timur (NTT) and Nusa Tenggara Barat (NTB), Maluku Islands, and Papua. Details of the provinces on each island are shown in Table 1.

Table 1. List of Indonesia’s provinces.

No	Island	Province
1.	Sumatera	Aceh
		Sumatera Utara
		Sumatera Barat
		Riau
		Kepulauan Riau (Riau Island)
		Bengkulu
		Jambi
		Sumatera Selatan
		Lampung
		Kepulauan Bangka Belitung (Bangka Belitung Island)
2.	Jawa	Banten
		DKI Jakarta
		Jawa Barat
		Jawa Tengah
		Jawa Timur
		DI Yogyakarta
3.	Kalimantan	Kalimantan Barat
		Kalimantan Timur
		Kalimantan Tengah
		Kalimantan Utara
		Kalimantan Selatan
4.	Sulawesi	Sulawesi Selatan
		Sulawesi Utara
		Gorontalo
		Sulawesi Barat
		Sulawesi Tenggara
		Sulawesi Tengah
5.	Bali NTT NTB	Bali
		NTT
		NTB
6.	Maluku Island	Maluku Utara
6.	Maluku Island	Maluku
7.	Papua	Papua
7.	Papua	Papua Barat

Results

Figure 1 describe the distribution of COVID-19 cases in each province in three years which most occurred in Jawa Island in particular in DKI Jakarta, Jawa Barat, Jawa Tengah, Banten, and Jawa Timur.

Figure 1. Trend of COVID-19 cases 2020-2022.

DKI Jakarta, Jawa Barat, Jawa Tengah, Jawa Timur, DI Yogyakarta, and Banten Provinces which are located in Jawa Island are province with highest density of population in Indonesia. In term of vaccination coverage, Jawa Barat, Jawa Timur, Jawa Tengah, DKI Jakarta, and Sumatera Utara are the highest. The highest percentage of internet users are in DKI Jakarta and Riau Island Province, while the lowest are in Papua and Nusa Tenggara Timur ( Figure 2).

Figure 2. Descriptive statistic of COVID-19 risks factors.

Spatial autocorrelation

Moran test found that in 2020 to 2022, there was autocorrelation among areas based on COVID-19 cases. It means that COVID-19 does not occur randomly but interrelated between one area to its neighbors ( Table 2). Since there is autocorrelation, the hotspot of COVID-19 can be determined further using Moran’s scatter plot ( Figure 3).

Figure 3. The COVID-19 hotspot areas in Moran’s Scatter Plot 2020 – 2022.

The hotspot areas are located in HH and HL quadrant.

Table 2. Index Moran value 2020 – 2022.

Year	Moran’s index	p-value	Decision
2020	0.3687	0.0047	There is spatial autocorrelation
2021	0.5743	4.159e-05	There is spatial autocorrelation
2022	0.7674	1.039e-07	There is spatial autocorrelation

The hotspot areas of COVID-19 in Indonesia

Over the pandemic, certain provinces consistently emerged as COVID-19 hotspots. Notably, DKI Jakarta, West Java, Central Java, and East Java were identified as hotspots across all three years of observation ( Table 3). The hotspot provinces in 2022 are Banten, DKI Jakarta, Jawa Barat, Jawa Tengah, Jawa Timur, and DI Yogyakarta ( Figure 4).

Table 3. The COVID-19 hotspot provinces in Indonesia 2020-2022.

Year	Total number of hotspot area (province)	Hotspot area (province)
2020	4	DKI Jakarta
		Jawa Barat
		Jawa Tengah
		Jawa Timur
2021	6	Banten
		DKI Jakarta
		Jawa Barat
		Jawa Tengah
		Jawa Timur
		DI Yogyakarta
2022	6	Banten
		DKI Jakarta
		Jawa Barat
		Jawa Tengah
		Jawa Timur
		DI Yogyakarta

Figure 4. The map of COVID-19 hotspot provinces in Indonesia 2022.

COVID-19 risk factors in 33 provinces in Indonesia

Before conducting the GWR test, all the assumption tests should be fulfilled using classic linear regression. The first assumption test found that the residuals did not distribute under the normal distribution curve ( Table 4). Hence, the researcher did data transformation using natural logarithm transformation method. Afterwards, we did the second assumption test and found that all assumptions have fulfilled ( Table 5).

Table 4. First assumption test result.

Assumption test	Value	p-value	Decision
Normality (Anderson-Darling)	0,91	0,02	Residual distribution was not under the normal curve distribution
Independency (Run-Test)	18	1	Residual independent
Homogeneity (Breusch-Pagan)	4,47	0,34	Residual homogeneity
Multicollinearity (VIF value) to all variable
Internet users	2,93		No multicollinearity among independent variables
Agglomeration area	1,11
Population density	1,58
Vaccine coverage	3,49

Table 5. Assumption test result with transformed data.

Test	Value	p-value	Decision
Normality (Anderson-Darling)	0,56	0,14	Residual distribution was under the normal curve
Independency (Run-Test)	18	1	Residual independent
Homogeneity (Breusch-Pagan)	2,87	0,58	Residual homogeneity
Multicollinearity (VIF value) to all variable
Internet users	2,93		No multicollinearity among independent variables
Agglomeration area	1,11
Population density	1,58
Vaccine coverage	3,49

Assumption test result toward transformed data showed that the residual of COVID-19 cases already distributed under the normal distribution curve, independent, homogeneous, and there was no multicollinearity among all variables ( Table 5). Thus, further analysis could be run to identify risk factors in each area.

COVID-19 risk factors in each province

The dependent variable in this study is the total number of cases from 2020 to 2022. GWR analysis was done to identify the risk factors. It began with determining bandwidth value by using Kernel Gaussian and Kernal Bi-Square method. These two kinds of bandwidth will be compared in terms of their contribution to the model performance created. The model performance will be seen by comparing the AIC and R² and Adj-R² values ( Table 6).

Table 6. The comparison of bandwidth value using Kernell Gaussian dan Bi-square.

Component	Regression model	Kernel Gaussian	Kernel Bi-square
AIC	47.3533	38.8130	35.9460
R²	0.6463	0.6680	0.7172
Adj - R²	0.5975	0.5922	0.5982

Table 6 clearly shows that GWR model with Kernel Bi-square had a better model performance due to the smaller AIC value and the higher R² and Adj-R² value. Thus, risk factors modeling with GWR analysis employed Kernel Bi-square bandwidth. The COVID-19 risk factors model of each province is presented in Table 7.

Table 7. COVID-19 risk factors model of all provinces in Indonesia 2022.

Province	Intercept		Aglo		Prop_internet		Prop_vacc2		Density
Province	Coeff.	p-value	Coeff.	p-value	Coeff.	p-value	Coeff.	p-value	Coeff.	p-value
Aceh	4,7139	<0.001	-0,0161	0,632	0,0520	0,008	0,6997	0,662	<0.001	0,715
Bali	5,0940	<0.001	0,0017	0,96	0,0394	0,03	1,0732	0,479	<0.001	0,637
Bangka Belitung Island	4,6905	<0.001	-0,0146	0,654	0,0480	0,008	1,0634	0,48	<0.001	0,795
Banten	4,6290	<0.001	-0,0185	0,574	0,0449	0,011	1,3984	0,365	<0.001	0,851
Bengkulu	4,6322	<0.001	-0,0181	0,586	0,0480	0,009	1,1397	0,461	<0.001	0,809
Gorontalo	5,8064	<0.001	0,0535	0,251	0,0456	0,009	-0,6195	0,584	<0.001	0,265
DKI Jakarta	4,6423	<0.001	-0,0177	0,588	0,0447	0,012	1,3937	0,366	<0.001	0,849
Jambi	4,6628	<0.001	-0,0166	0,615	0,0490	0,008	1,0196	0,507	<0.001	0,786
Jawa Barat	4,6463	<0.001	-0,0179	0,585	0,0437	0,013	1,4659	0,345	<0.001	0,858
Jawa Tengah	4,7084	<0.001	-0,0152	0,642	0,0418	0,019	1,5190	0,336	<0.001	0,851
Jawa Timur	4,9000	<0.001	-0,0088	0,79	0,0407	0,022	1,3011	0,396	<0.001	0,744
Kalimantan Barat	4,8871	<0.001	-0,0009	0,977	0,0483	0,009	0,7530	0,61	<0.001	0,745
Kalimantan Selatan	5,1778	<0.001	0,0142	0,681	0,0439	0,015	0,6166	0,679	<0.001	0,624
Kalimantan Tengah	5,0233	<0.001	0,0067	0,839	0,0462	0,011	0,6956	0,644	<0.001	0,702
Kalimantan Timur	5,3541	<0.001	0,0328	0,399	0,0460	0,013	0,1563	0,918	<0.001	0,564
Kalimantan Utara	5,3905	<0.001	0,0376	0,356	0,0479	0,013	-0,0467	0,976	<0.001	0,562
Riau Island	4,8886	<0.001	-0,0045	0,888	0,0522	0,006	0,4273	0,772	<0.001	0,666
Lampung	4,6362	<0.001	-0,0178	0,589	0,0464	0,009	1,2641	0,41	<0.001	0,831
Maluku	6,0401	<0.001	0,0568	0,228	0,0444	0,012	-0,9590	0,386	<0.001	0,138
Maluku Utara	6,0113	<0.001	0,0595	0,221	0,0456	0,011	-1,0031	0,368	<0.001	0,16
Nusa Tenggara Barat	5,3552	<0.001	0,0127	0,721	0,0404	0,020	0,5713	0,667	< 0.0001	0,468
Nusa Tenggara Timur	5,7056	<0.001	0,0356	0,386	0,0407	0,019	-0,0762	0,948	< 0.0001	0,270
Papua	6,1180	<0.001	0,0596	0,231	0,0442	0,016	-1,0960	0,341	0.0001	0,128
Papua Barat	6,0811	<0.001	0,0594	0,225	0,0447	0,014	-1,0616	0,349	0.0001	0,135
Riau	4,6901	<0.001	-0,0157	0,635	0,0503	0,007	0,8767	0,570	< 0.0001	0,756
Sulawesi Barat	5,5474	<0.001	0,0364	0,369	0,0436	0,011	-0,0034	0,998	< 0.0001	0,406
Sulawesi Selatan	5,6257	<0.001	0,0410	0,332	0,0427	0,013	-0,0811	0,948	< 0.0001	0,360
Sulawesi Tengah	5,7113	<0.001	0,0483	0,28	0,0445	0,010	-0,3659	0,757	< 0.0001	0,320
Sulawesi Tenggara	5,7639	<0.001	0,0453	0,293	0,0434	0,012	-0,3763	0,741	< 0.0001	0,259
Sulawesi Utara	5,9043	<0.001	0,0564	0,233	0,0458	0,009	-0,8134	0,460	0.0001	0,205
Sumatera Barat	4,6623	<0.001	-0,0170	0,608	0,0498	0,008	0,9566	0,539	< 0.0001	0,772
Sumatera Selatan	4,6523	<0.001	-0,0169	0,608	0,0478	0,008	1,1315	0,459	< 0.0001	0,808
Sumatera Utara	4,6990	<0.001	-0,0161	0,63	0,0512	0,008	0,7850	0,618	< 0.0001	0,734
DI Yogyakarta	4,7123	<0.001	-0,0156	0,635	0,0411	0,022	1,5640	0,325	< 0.0001	0,850

The modeling results show that only variable of proportion of internet users consistently has a positive and statistically significant relationship with COVID-19 cases in all provinces, while other variables do not have a significant relationship with total cases ( Table 7). It is known from the GWR model that areas with a high proportion of internet users are an indication that these regions have high cumulative cases of COVID-19.

Discussion

The statistically significant Moran index value proves that there is autocorrelation between provinces based on COVID-19 cases. This illustrates that COVID-19 cases in a province does not occur randomly, but have a relationship with COVID-19 cases in neighboring areas.²³ This is in line with the theory states that attribute values in an area will tend to be the same as areas that are closer than those farther away which is accordance with the basic concept of geography (Tobler’s Law 1) which states everything is related to everything else, but near things are more related than distant thing.^18,24,25

The number of COVID-19 hotspot provinces in Indonesia in 2020 was originally only four⁴ on Jawa, but in 2021 along with the increasing of COVID-19 cases in the community, the hotspot area became six⁶ provinces up to 2022. Based on the map in Figure 4, the hotspot area is clustered on Jawa, which means that Jawa Island is a COVID-19 hotspot in Indonesia. This finding is consistent with previous research in Indonesia.²³ Jawa Island has a higher population density than other regions in Indonesia ( Figure 2) and also has many agglomeration areas that are centers of economic activity. As such, population mobility in the region will tend to be higher than in other provinces across Indonesia.^23,26 The more densely populated and the more mobile the population in an area, the higher the risk of COVID-19 transmission as seen and proved in this study. It applies not only to the transmission of COVID-19, but also to other infectious diseases with a similar transmission mode.²⁷

The only risk factor that consistently had a statistically significant association with COVID-19 across all provinces in Indonesia was the number of internet users ( Table 7). The enactment of mobility restriction policies in all regions particularly in agglomeration areas (which almost all urban areas) has led most citizens to do work and school activities from home and some also use the internet to spend time at home. This finding is consistent with several studies conducted in other countries.^28–31 In Indonesia, the distribution of internet is not equal across Indonesia, there is still a gap between rural and urban area, between provinces and even within a province. However, generally, internet is more accessible in urban/city compared to rural areas. Unfortunately, the data cannot represent rural and urban since it was a cumulative number of provinces in which there are urban and rural areas within. Basically, internet can be a strong means for the community to get the correct information but instead it could be a risk to the public health due to misinformation.³²

Limitation

The data can no represent urban and rural area in Indonesia. For information, there are still gaps between areas in Indonesia regarding social characteristic, internet facility, health facility including COVID-19 test equipment which resulted bias of cases due to unreported and undetected cases. The level of province may not represent all conditions in one province due to the variation of areas within a province. This study also did not measure the level of compliance of the population with the movement restriction policy, mask wearing, and hand washing due to data limitations. We assume that in agglomeration area, the population have higher mobilization level.

Conclusion and recommendation

COVID-19 distribution was most concentrated in Pulau Jawa (Jawa Island) particularly in DKI Jakarta, Jawa Barat, Jawa Tengah, Banten, and Jawa Timur. Geographic weighted regression analysis showed that the distribution of COVID-19 in a province does not occur randomly but has a relationship with COVID-19 cases in the neighboring areas. The hotspot area forms a dense cluster on Jawa Island, meaning that the Jawa Island is a COVID-19 hotspot in Indonesia. Jawa has the highest population density in Indonesia, as well as an agglomeration of economic activities center. The risk factor that is statistically significant consistent with COVID-19 across all provinces in Indonesia is the proportion of internet users.

The study results are very useful to serve as lessons learned in responding to events similar to the COVID-10 pandemic. Indonesia, a vast archipelago, is a unique region that also requires unique or region-specific actions. With the knowledge that the hotspots of the COVID-19 pandemic in Indonesia are areas located on the Jawa Island, if there is an infectious disease outbreak or on a large scale, such as pandemic, then the Jawa Island must be a priority area that must first be addressed and protected. This may be related to its areas which are mostly with higher population density and are the centre of the economy (agglomeration) activities which has an impact on the high level of community activity and mobility. The restricted movement among the community inevitably makes people adapt, especially in terms of technology utilization. The community can quickly shift to be more technology-friendly although this has not occurred evenly throughout Indonesia due to socio-economic disparities in many parts of Indonesia. However, this shifting can provide an opportunity for the government to utilise technology for interventions. To that, the government should improve the infrastructure especially related to the internet provision and other technologies or tools without any doubt since the community in fact are able to adapt.

Ethical considerations

During the COVID-19 pandemic, the Government of Indonesia developed a publicly accessible web-based dashboard to monitor confirmed cases at both the provincial and district levels. This platform, which provides real-time data on case numbers disaggregated by province, is freely available to the general public (https://covid19.go.id/ which later integrated into https://infeksiemerging.kemkes.go.id/dashboard/covid-19). Leveraging the availability of these spatially referenced data, the research team initiated a study proposal employing spatial analysis techniques to explore the geographical distribution and determinants of COVID-19. The proposal was submitted on June 29, 2022, to the Research Ethics Committee of the Faculty of Public Health, Universitas Indonesia, through an online submission system (https://kajietik.fkm.ui.ac.id). The committee, chaired by Prof. Dr. Ratna Djuwita, MPH, with Prof. Dr. L. Meily Kurniawidjaja, M. Sc, Sp.OK serving as secretary, granted ethical clearance for the study on August 25, 2022. The approval remains valid until August 25, 2023 with letter number is No. Ket-532/UN2.F10.D11/PPM.00.02/2022.

This study was using secondary data which taken from various sources, thus there is no informed consent needed for this study.

Data availability

The dataset employed in this study is available in Figshare under a Creative Commons 4.0 (CC BY 4.0 Public Domain Dedication) license (http://creativecommons.org/publicdomain/zero/1.0/)

Data set is available at DOI: https://doi.org/10.6084/m9.figshare.25346761.v1³³

Figures are available DOI: https://doi.org/10.6084/m9.figshare.25347328.v1³⁴

Acknowledgements

The researcher appreciated for the hard work of the government of Indonesia to handle the pandemic and Pandemic Researcher Team of Universitas Indonesia (University of Indonesia), including provided the data publicly. The contribution of the team mentioned above have been acknowledged with their consent in this research article.

References

1. WHO et al.: COVID-19 Weekly Epidemiological Update. World Heal Organ; 2023; vol. 58. : 1–23. Reference Source
2. World Health Organization: COVID-19 Epidemiological Update.2024.
3. Menteri dalam Negeri Republik Indonesia: Instruksi Menteri dalam Negeri Nomor 51 Tahun 2022 tentang Pemberlakuan Pembatasan Kegiatan Masyarakat pada Kondisi Corona Virus Disease 2019 di Wilayah Sumatera, Nusa Tenggara, Kalimantan, Sulawesi, Maluku, dan Papua.2022. Reference Source
4. Ehlert A: The socio-economic determinants of COVID-19: A spatial analysis of German county level data. Socio Econ. Plan. Sci. 2021; 78(July 2020): 101083. PubMed Abstract | Publisher Full Text | Free Full Text
5. Kwok CYT, Wong MS, Chan KL, et al.: Spatial analysis of the impact of urban geometry and socio-demographic characteristics on COVID-19, a study in Hong Kong. Sci. Total Environ. 2021; 764: 144455. PubMed Abstract | Publisher Full Text | Free Full Text
6. Sun F, Matthews SA, Yang TC, et al.: A spatial analysis of the COVID-19 period prevalence in U.S. counties through June 28, 2020: where geography matters?. Ann. Epidemiol. 2020; 52: 54–59.e1. PubMed Abstract | Publisher Full Text | Free Full Text
7. Wang Q, Dong W, Yang K, et al.: Temporal and spatial analysis of COVID-19 transmission in China and its influencing factors. Int. J. Infect. Dis. 2021; 105: 675–685. PubMed Abstract | Publisher Full Text | Free Full Text
8. Dutta I, Basu T, Das A: Spatial analysis of COVID-19 incidence and its determinants using spatial modeling: A study on India. Environ Challenges. 2021; 4(November 2020): 100096. PubMed Abstract | Publisher Full Text | Free Full Text
9. Ramírez IJ, Lee J: COVID-19 emergence and social and health determinants in Colorado: A rapid spatial analysis. Int. J. Environ. Res. Public Health. 2020; 17(11): 1–15.
10. Ramírez-Aldana R, Gomez-Verjan JC, Bello-Chavolla OY, et al.: Spatial analysis of COVID-19 spread in Iran: Insights into geographical and structural transmission determinants at a province level. PLoS Negl. Trop. Dis. 2020; 14(11): 1–15.
11. Xie Z, Qin Y, Li Y, et al.: Spatial and temporal differentiation of COVID-19 epidemic spread in mainland China and its influencing factors. Sci. Total Environ. 2020; 744: 140929. PubMed Abstract | Publisher Full Text | Free Full Text
12. DiMaggio C, Klein M, Berry C, et al.: Black/African American Communities are at highest risk of COVID-19: spatial modeling of New York City ZIP Code–level testing results. Ann. Epidemiol. 2020; 51: 7–13. PubMed Abstract | Publisher Full Text | Free Full Text
13. Rahman MH, Zafri NM, Ashik FR, et al.: Identification of risk factors contributing to COVID-19 incidence rates in Bangladesh: A GIS-based spatial modeling approach. Heliyon. 2021; 7(2): e06260. PubMed Abstract | Publisher Full Text | Free Full Text
14. Sannigrahi S, Pilla F, Basu B, et al.: Examining the association between socio-demographic composition and COVID-19 fatalities in the European region using spatial regression approach. Sustain. Cities Soc. 2020; 62(July): 102418. PubMed Abstract | Publisher Full Text | Free Full Text
15. Bayode T, Popoola A, Akogun O, et al.: Spatial variability of COVID-19 and its risk factors in Nigeria: A spatial regression method. Appl. Geogr. 2022; 138(July 2020): 102621. PubMed Abstract | Publisher Full Text | Free Full Text
16. Sung B: A spatial analysis of the effect of neighborhood contexts on cumulative number of confirmed cases of COVID-19 in U.S. Counties through October 20 2020. Prev Med (Baltim). 2021; 147(November 2020): 106457. PubMed Abstract | Publisher Full Text | Free Full Text
17. Mauleny AT: Aglomerasi, Perubahan Sosial Ekonomi, Dan Kebijakan Pembangunan Jakarta. J Ekon Kebijak Publik. 2015; 6(2): 147–162.
18. Grekousis G: Spatial analysis methods and practice: describe, explore, explain through GIS.2020; 400.
19. Tiopan S, Tris E: Does stunting randomly occurred in Indonesia: Spatial analysis of Indonesia’s basic health survey 2018. Indian J Public Heal Res Dev. 2019; 10(10): 764–768.
20. Eryando T, Sipahutar T, Meiwita PB, et al.: Spatial analysis of stunting determinants in 514 Indonesian districts/cities: Implications for intervention and setting of priority. Geospat. Health. 2022; 17(1). Publisher Full Text
21. Fotheringham S, Brunsdon C, Charlton M: Geographically Wighted Regression. England: 2002.
22. Kleinbaum DG, Kupper LL, Muller KE, et al.: Applied Regression Analysis and Other Multivariable Methods. Third Edit. Forest Lodge Road Pacific Grove, CA: Duxbury Press; 1998.
23. Eryando T, Sipahutar T, Rahardiantoro S: The Risk Distribution of COVID-19 in Indonesia: A Spatial Analysis. Asia-Pacific J Public Heal. 2020; 32(8): 450–452.
24. Souris M: Epidemiology and Geography. Great Britain and United States: John Wiley & Sons, Ltd; 2019.
25. Bhunia GS, Shit PK: Geospatial Analysis of Public Health. Switzerland: Springer Nature Switzerland; 2019.
26. Handayani W, Insani TD, Fisher M, et al.: Effects of COVID-19 restriction measures in Indonesia: A comparative spatial and policy analysis of selected urban agglomerations. Int J Disaster Risk Reduct. 2022; 76(April): 103015. PubMed Abstract | Publisher Full Text | Free Full Text
27. Dehghani MH, Karri RR, Roy S: Environmental and Health Management of Novel Coronavirus Disease (COVID-19).Dehghani MH, Kariri RR, Roy S, editors. Environmental and Health Management of Novel Coronavirus Disease (COVID-19). London: 2021; 1–488 p.
28. Zeng B, Rivadeneira NA, Wen A, et al.: The Impact of the COVID-19 Pandemic on Internet Use and the Use of Digital Health Tools: Secondary Analysis of the 2020 Health Information National Trends Survey. J. Med. Internet Res. 2022; 24(9): 1–13.
29. van Deursen AJAM : Digital inequality during a pandemic: Quantitative study of differences in COVID-19-related internet uses and outcomes among the general population. J. Med. Internet Res. 2020; 22(8): 1–14. Publisher Full Text
30. Huang Q, Chen X, Huang S, et al.: Substance and Internet use during the COVID-19 pandemic in China. Transl. Psychiatry. 2021; 11(1): 1–9. PubMed Abstract | Publisher Full Text | Free Full Text
31. Gjoneska B, Potenza MN, Jones J, et al.: Problematic use of the internet during the COVID-19 pandemic: Good practices and mental health recommendations. Compr. Psychiatry. 2022; 112: 152279. Publisher Full Text
32. Cuan-Baltazar JY, Muñoz-Perez MJ, Robledo-Vega C, et al.: Misinformation of COVID-19 on the internet: Infodemiology study. JMIR Public Heal Surveill. 2020; 6(2): 1–9.
33. Sipahutar T: DATA COVID-19_indonesia.xlsx. Dataset. figshare. 2024. Publisher Full Text
34. Sipahutar T: LIST OF FIGURES.pdf. figshare. Figure. 2024. Publisher Full Text

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 26 Mar 2026

Author details Author details

¹ Department of Biostatistics and Population, Universitas Indonesia, Depok, West Java, 16424, Indonesia
² Department of Occupational Health & Safety, Universitas Indonesia, Depok, West Java, 16424, Indonesia
³ Department of Biostatistics and Population, Universitas Indonesia, Depok, West Java, 16424, Indonesia
⁴ Health Informatic Research Centre, Faculty of Public Health, Universitas Indonesia, Depok, West Java, 16424, Indonesia
⁵ Department of Human Centred Computing, Monash University, Melbourne, Victoria, Australia

Tris Eryando
Roles: Conceptualization, Formal Analysis, Methodology, Resources, Supervision, Writing – Review & Editing

Fatma Lestari
Roles: Supervision

Tiopan Sipahutar
Roles: Conceptualization, Data Curation, Formal Analysis, Methodology, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Hendy Risdianto
Roles: Supervision

Juliana Sutanto
Roles: Supervision

Competing interests

No competing interests were disclosed.

Grant information

The author(s) declared that no grants were involved in supporting this work.

Article Versions (1)

version 1

Published: 26 Mar 2026, 15:440

https://doi.org/10.12688/f1000research.149041.1

Copyright

© 2026 Eryando T et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

	Views	Downloads
F1000Research	-	-
PubMed Central Data from PMC are received and updated monthly.	-	-

Citations

0

SEE MORE DETAILS

CITE

how to cite this article

Eryando T, Lestari F, Sipahutar T et al. The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy, Hotspot and Risk Factors using Geographically Weighted Regression in 2020 - 2022 [version 1; peer review: 1 approved with reservations]. F1000Research 2026, 15:440 (https://doi.org/10.12688/f1000research.149041.1)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Version 1

VERSION 1

PUBLISHED 26 Mar 2026

Views

1

Reviewer Report 09 Jun 2026

Onsel Gurel Bayrali, Binghamton University, New York, New York, USA

Approved with Reservations

https://doi.org/10.5256/f1000research.163444.r485526

This manuscript examines the spatial distribution of COVID-19 cases in Indonesia between 2020 and 2022 using Moran’s I, hotspot analysis, and geographically weighted regression (GWR). The topic is relevant and potentially valuable for understanding spatial variation in pandemic outcomes and ... Continue reading

This manuscript examines the spatial distribution of COVID-19 cases in Indonesia between 2020 and 2022 using Moran’s I, hotspot analysis, and geographically weighted regression (GWR). The topic is relevant and potentially valuable for understanding spatial variation in pandemic outcomes and informing future public health preparedness. The manuscript also benefits from the use of publicly available data and the application of spatial analytical techniques. However, several conceptual, methodological, and interpretive issues limit confidence in the findings. I therefore recommend major revision before the manuscript can be considered for indexing.
Major Comments

Outcome Variable Specification

The most important concern relates to the dependent variable. The analysis models the cumulative number of COVID-19 cases between 2020 and 2022 at the provincial level. Using cumulative case counts rather than incidence rates creates a fundamental comparability problem because provinces with larger populations will naturally report more cases even if infection risk is similar or lower. Population density does not adequately address this issue because density and population size capture different dimensions of population structure.
The authors should justify why cumulative case counts were selected instead of cumulative incidence per 100,000 population, annual incidence rates, log incidence rates, or count models with population offsets. Without population adjustment, it is difficult to determine whether the observed spatial patterns reflect infection risk or simply population concentration.

Temporal Ordering and Potential Reverse Causality

The manuscript uses cumulative COVID-19 cases from 2020–2022 while including second-dose vaccination coverage as an explanatory variable. This creates potential temporal ordering problems because vaccination campaigns occurred after substantial portions of COVID-19 transmission had already occurred. Areas experiencing high case burdens may have received greater vaccination efforts, creating reverse causality.
The authors should clarify the year from which vaccination data were obtained, whether vaccination coverage corresponds to 2021 or 2022, and why it is theoretically appropriate to use vaccination coverage as a predictor of cumulative cases over the entire pandemic period.

Suitability of GWR for a Small Sample

The study estimates geographically weighted regression using only 33 provincial units. GWR is generally most informative when a relatively large number of spatial units is available. With only 33 observations, local parameter estimates may be unstable and highly sensitive to bandwidth selection.
The manuscript should provide a stronger justification for using GWR in this setting, discuss potential overfitting concerns, and consider whether alternative spatial modeling approaches such as spatial lag or spatial error models would be more appropriate.

Definition of Hotspots

The manuscript defines hotspot areas as provinces located in both the High-High (HH) and High-Low (HL) quadrants of Moran’s scatter plot. This definition is unusual. In the spatial analysis literature, HH observations are typically interpreted as hotspots, LL observations as cold spots, while HL and LH observations are generally considered spatial outliers.
The authors should provide a stronger methodological justification for treating HL observations as hotspot areas.

Interpretation of Internet Use as a Risk Factor

The central substantive finding is that internet use is the only variable that remains statistically significant across all provinces. However, the interpretation presented in the discussion appears stronger than warranted.
Internet access likely proxies for urbanization, economic development, testing capacity, surveillance quality, healthcare access, and reporting completeness. Therefore, the observed association does not necessarily imply that internet use increases COVID-19 transmission risk. The discussion should be revised to emphasize that internet use may represent broader socioeconomic and reporting characteristics rather than a direct causal mechanism.

Residual Spatial Dependence

The manuscript demonstrates significant spatial autocorrelation in COVID-19 cases using Moran’s I. However, it is unclear whether residual spatial autocorrelation remains after estimation of the GWR model.
The authors should report Moran’s I for model residuals and provide additional diagnostics demonstrating that the GWR model adequately accounts for spatial dependence.

Policy Conclusions

The conclusion recommends prioritizing Java Island in future pandemics because it emerged as the principal hotspot during COVID-19. While the descriptive evidence clearly shows higher case concentrations in Java, the analysis remains ecological and descriptive. The evidence supports identifying spatial concentration, but it does not fully support broader policy conclusions regarding future resource allocation strategies. The policy implications should therefore be presented more cautiously.
Reproducibility and Transparency
The authors should be commended for making their dataset and figures publicly available. However, full reproducibility would be improved by providing:

R scripts used for analysis;
Details on spatial weights construction;
Bandwidth values selected during GWR estimation;
Province coordinate definitions and spatial objects;
Exact transformations applied to variables.

At present, the study is only partially reproducible.
Minor Comments

The manuscript repeatedly refers to 33 provinces. Given changes in Indonesia’s administrative structure over time, the authors should clarify why 33 provinces were included and whether any provinces were excluded.
Several figures, particularly the Moran scatter plots, would benefit from improved readability and clearer province labels.
The manuscript would benefit from substantial English-language editing. Numerous grammatical and stylistic issues reduce readability throughout the paper.
The authors should explicitly state which variables were transformed using the natural logarithm, provide the transformation formula, and clarify how transformed coefficients should be interpreted.
The manuscript discusses bandwidth selection but does not report the selected bandwidth values. These values should be reported to improve transparency and reproducibility.

Overall Recommendation
This manuscript addresses an important topic and contains potentially useful descriptive spatial evidence regarding the geographic distribution of COVID-19 in Indonesia. However, concerns regarding outcome specification, temporal ordering, the suitability of GWR for a small sample, hotspot classification, interpretation of internet use, and model diagnostics need to be addressed before the findings can be considered robust. I therefore recommend Major Revision.

Is the work clearly and accurately presented and does it cite the current literature?

Partly
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: public health, computational methods, research design

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Respond or Comment

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 26 Mar 2026

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1
Version 1 26 Mar 26	read

Onsel Gurel Bayrali, Binghamton University, New York, USA

Comments on this article

All Comments(0)

Add a comment

Sign up for content alerts

Browse by related subjects

Back to all reports

Reviewer Report

1 Views

09 Jun 2026 | for Version 1

Onsel Gurel Bayrali, Binghamton University, New York, New York, USA

1 Views Cite this report Responses(0)

Approved With Reservations

This manuscript examines the spatial distribution of COVID-19 cases in Indonesia between 2020 and 2022 using Moran’s I, hotspot analysis, and geographically weighted regression (GWR). The topic is relevant and potentially valuable for understanding spatial variation in pandemic outcomes and informing future public health preparedness. The manuscript also benefits from the use of publicly available data and the application of spatial analytical techniques. However, several conceptual, methodological, and interpretive issues limit confidence in the findings. I therefore recommend major revision before the manuscript can be considered for indexing.
Major Comments

Outcome Variable Specification

The most important concern relates to the dependent variable. The analysis models the cumulative number of COVID-19 cases between 2020 and 2022 at the provincial level. Using cumulative case counts rather than incidence rates creates a fundamental comparability problem because provinces with larger populations will naturally report more cases even if infection risk is similar or lower. Population density does not adequately address this issue because density and population size capture different dimensions of population structure.
The authors should justify why cumulative case counts were selected instead of cumulative incidence per 100,000 population, annual incidence rates, log incidence rates, or count models with population offsets. Without population adjustment, it is difficult to determine whether the observed spatial patterns reflect infection risk or simply population concentration.

Temporal Ordering and Potential Reverse Causality

The manuscript uses cumulative COVID-19 cases from 2020–2022 while including second-dose vaccination coverage as an explanatory variable. This creates potential temporal ordering problems because vaccination campaigns occurred after substantial portions of COVID-19 transmission had already occurred. Areas experiencing high case burdens may have received greater vaccination efforts, creating reverse causality.
The authors should clarify the year from which vaccination data were obtained, whether vaccination coverage corresponds to 2021 or 2022, and why it is theoretically appropriate to use vaccination coverage as a predictor of cumulative cases over the entire pandemic period.

Suitability of GWR for a Small Sample

The study estimates geographically weighted regression using only 33 provincial units. GWR is generally most informative when a relatively large number of spatial units is available. With only 33 observations, local parameter estimates may be unstable and highly sensitive to bandwidth selection.
The manuscript should provide a stronger justification for using GWR in this setting, discuss potential overfitting concerns, and consider whether alternative spatial modeling approaches such as spatial lag or spatial error models would be more appropriate.

Definition of Hotspots

The manuscript defines hotspot areas as provinces located in both the High-High (HH) and High-Low (HL) quadrants of Moran’s scatter plot. This definition is unusual. In the spatial analysis literature, HH observations are typically interpreted as hotspots, LL observations as cold spots, while HL and LH observations are generally considered spatial outliers.
The authors should provide a stronger methodological justification for treating HL observations as hotspot areas.

Interpretation of Internet Use as a Risk Factor

The central substantive finding is that internet use is the only variable that remains statistically significant across all provinces. However, the interpretation presented in the discussion appears stronger than warranted.
Internet access likely proxies for urbanization, economic development, testing capacity, surveillance quality, healthcare access, and reporting completeness. Therefore, the observed association does not necessarily imply that internet use increases COVID-19 transmission risk. The discussion should be revised to emphasize that internet use may represent broader socioeconomic and reporting characteristics rather than a direct causal mechanism.

Residual Spatial Dependence

The manuscript demonstrates significant spatial autocorrelation in COVID-19 cases using Moran’s I. However, it is unclear whether residual spatial autocorrelation remains after estimation of the GWR model.
The authors should report Moran’s I for model residuals and provide additional diagnostics demonstrating that the GWR model adequately accounts for spatial dependence.

Policy Conclusions

The conclusion recommends prioritizing Java Island in future pandemics because it emerged as the principal hotspot during COVID-19. While the descriptive evidence clearly shows higher case concentrations in Java, the analysis remains ecological and descriptive. The evidence supports identifying spatial concentration, but it does not fully support broader policy conclusions regarding future resource allocation strategies. The policy implications should therefore be presented more cautiously.
Reproducibility and Transparency
The authors should be commended for making their dataset and figures publicly available. However, full reproducibility would be improved by providing:

R scripts used for analysis;
Details on spatial weights construction;
Bandwidth values selected during GWR estimation;
Province coordinate definitions and spatial objects;
Exact transformations applied to variables.

At present, the study is only partially reproducible.
Minor Comments

The manuscript repeatedly refers to 33 provinces. Given changes in Indonesia’s administrative structure over time, the authors should clarify why 33 provinces were included and whether any provinces were excluded.
Several figures, particularly the Moran scatter plots, would benefit from improved readability and clearer province labels.
The manuscript would benefit from substantial English-language editing. Numerous grammatical and stylistic issues reduce readability throughout the paper.
The authors should explicitly state which variables were transformed using the natural logarithm, provide the transformation formula, and clarify how transformed coefficients should be interpreted.
The manuscript discusses bandwidth selection but does not report the selected bandwidth values. These values should be reported to improve transparency and reproducibility.

Overall Recommendation
This manuscript addresses an important topic and contains potentially useful descriptive spatial evidence regarding the geographic distribution of COVID-19 in Indonesia. However, concerns regarding outcome specification, temporal ordering, the suitability of GWR for a small sample, hotspot classification, interpretation of internet use, and model diagnostics need to be addressed before the findings can be considered robust. I therefore recommend Major Revision.

Is the work clearly and accurately presented and does it cite the current literature?

Partly
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

public health, computational methods, research design

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Respond to this report

Responses (0)

[1] 1. WHO et al.: COVID-19 Weekly Epidemiological Update. World Heal Organ; 2023; vol. 58. : 1–23. Reference Source

[2] 2. World Health Organization: COVID-19 Epidemiological Update.2024.

[3] 3. Menteri dalam Negeri Republik Indonesia: Instruksi Menteri dalam Negeri Nomor 51 Tahun 2022 tentang Pemberlakuan Pembatasan Kegiatan Masyarakat pada Kondisi Corona Virus Disease 2019 di Wilayah Sumatera, Nusa Tenggara, Kalimantan, Sulawesi, Maluku, dan Papua.2022. Reference Source

[4] 4. Ehlert A: The socio-economic determinants of COVID-19: A spatial analysis of German county level data. Socio Econ. Plan. Sci. 2021; 78(July 2020): 101083. PubMed Abstract | Publisher Full Text | Free Full Text

[5] 5. Kwok CYT, Wong MS, Chan KL, et al.: Spatial analysis of the impact of urban geometry and socio-demographic characteristics on COVID-19, a study in Hong Kong. Sci. Total Environ. 2021; 764: 144455. PubMed Abstract | Publisher Full Text | Free Full Text

[6] 6. Sun F, Matthews SA, Yang TC, et al.: A spatial analysis of the COVID-19 period prevalence in U.S. counties through June 28, 2020: where geography matters?. Ann. Epidemiol. 2020; 52: 54–59.e1. PubMed Abstract | Publisher Full Text | Free Full Text

[7] 7. Wang Q, Dong W, Yang K, et al.: Temporal and spatial analysis of COVID-19 transmission in China and its influencing factors. Int. J. Infect. Dis. 2021; 105: 675–685. PubMed Abstract | Publisher Full Text | Free Full Text

[8] 8. Dutta I, Basu T, Das A: Spatial analysis of COVID-19 incidence and its determinants using spatial modeling: A study on India. Environ Challenges. 2021; 4(November 2020): 100096. PubMed Abstract | Publisher Full Text | Free Full Text

[9] 9. Ramírez IJ, Lee J: COVID-19 emergence and social and health determinants in Colorado: A rapid spatial analysis. Int. J. Environ. Res. Public Health. 2020; 17(11): 1–15.

[10] 10. Ramírez-Aldana R, Gomez-Verjan JC, Bello-Chavolla OY, et al.: Spatial analysis of COVID-19 spread in Iran: Insights into geographical and structural transmission determinants at a province level. PLoS Negl. Trop. Dis. 2020; 14(11): 1–15.

[11] 11. Xie Z, Qin Y, Li Y, et al.: Spatial and temporal differentiation of COVID-19 epidemic spread in mainland China and its influencing factors. Sci. Total Environ. 2020; 744: 140929. PubMed Abstract | Publisher Full Text | Free Full Text

[12] 12. DiMaggio C, Klein M, Berry C, et al.: Black/African American Communities are at highest risk of COVID-19: spatial modeling of New York City ZIP Code–level testing results. Ann. Epidemiol. 2020; 51: 7–13. PubMed Abstract | Publisher Full Text | Free Full Text

[13] 13. Rahman MH, Zafri NM, Ashik FR, et al.: Identification of risk factors contributing to COVID-19 incidence rates in Bangladesh: A GIS-based spatial modeling approach. Heliyon. 2021; 7(2): e06260. PubMed Abstract | Publisher Full Text | Free Full Text

[14] 14. Sannigrahi S, Pilla F, Basu B, et al.: Examining the association between socio-demographic composition and COVID-19 fatalities in the European region using spatial regression approach. Sustain. Cities Soc. 2020; 62(July): 102418. PubMed Abstract | Publisher Full Text | Free Full Text

[15] 15. Bayode T, Popoola A, Akogun O, et al.: Spatial variability of COVID-19 and its risk factors in Nigeria: A spatial regression method. Appl. Geogr. 2022; 138(July 2020): 102621. PubMed Abstract | Publisher Full Text | Free Full Text

[16] 16. Sung B: A spatial analysis of the effect of neighborhood contexts on cumulative number of confirmed cases of COVID-19 in U.S. Counties through October 20 2020. Prev Med (Baltim). 2021; 147(November 2020): 106457. PubMed Abstract | Publisher Full Text | Free Full Text

[17] 17. Mauleny AT: Aglomerasi, Perubahan Sosial Ekonomi, Dan Kebijakan Pembangunan Jakarta. J Ekon Kebijak Publik. 2015; 6(2): 147–162.

[18] 18. Grekousis G: Spatial analysis methods and practice: describe, explore, explain through GIS.2020; 400.

[19] 19. Tiopan S, Tris E: Does stunting randomly occurred in Indonesia: Spatial analysis of Indonesia’s basic health survey 2018. Indian J Public Heal Res Dev. 2019; 10(10): 764–768.

[20] 20. Eryando T, Sipahutar T, Meiwita PB, et al.: Spatial analysis of stunting determinants in 514 Indonesian districts/cities: Implications for intervention and setting of priority. Geospat. Health. 2022; 17(1). Publisher Full Text

[21] 21. Fotheringham S, Brunsdon C, Charlton M: Geographically Wighted Regression. England: 2002.

[22] 22. Kleinbaum DG, Kupper LL, Muller KE, et al.: Applied Regression Analysis and Other Multivariable Methods. Third Edit. Forest Lodge Road Pacific Grove, CA: Duxbury Press; 1998.

[23] 23. Eryando T, Sipahutar T, Rahardiantoro S: The Risk Distribution of COVID-19 in Indonesia: A Spatial Analysis. Asia-Pacific J Public Heal. 2020; 32(8): 450–452.

[24] 24. Souris M: Epidemiology and Geography. Great Britain and United States: John Wiley & Sons, Ltd; 2019.

[25] 25. Bhunia GS, Shit PK: Geospatial Analysis of Public Health. Switzerland: Springer Nature Switzerland; 2019.

[26] 26. Handayani W, Insani TD, Fisher M, et al.: Effects of COVID-19 restriction measures in Indonesia: A comparative spatial and policy analysis of selected urban agglomerations. Int J Disaster Risk Reduct. 2022; 76(April): 103015. PubMed Abstract | Publisher Full Text | Free Full Text

[27] 27. Dehghani MH, Karri RR, Roy S: Environmental and Health Management of Novel Coronavirus Disease (COVID-19).Dehghani MH, Kariri RR, Roy S, editors. Environmental and Health Management of Novel Coronavirus Disease (COVID-19). London: 2021; 1–488 p.

[28] 28. Zeng B, Rivadeneira NA, Wen A, et al.: The Impact of the COVID-19 Pandemic on Internet Use and the Use of Digital Health Tools: Secondary Analysis of the 2020 Health Information National Trends Survey. J. Med. Internet Res. 2022; 24(9): 1–13.

[29] 29. van Deursen AJAM : Digital inequality during a pandemic: Quantitative study of differences in COVID-19-related internet uses and outcomes among the general population. J. Med. Internet Res. 2020; 22(8): 1–14. Publisher Full Text

[30] 30. Huang Q, Chen X, Huang S, et al.: Substance and Internet use during the COVID-19 pandemic in China. Transl. Psychiatry. 2021; 11(1): 1–9. PubMed Abstract | Publisher Full Text | Free Full Text

[31] 31. Gjoneska B, Potenza MN, Jones J, et al.: Problematic use of the internet during the COVID-19 pandemic: Good practices and mental health recommendations. Compr. Psychiatry. 2022; 112: 152279. Publisher Full Text

[32] 32. Cuan-Baltazar JY, Muñoz-Perez MJ, Robledo-Vega C, et al.: Misinformation of COVID-19 on the internet: Infodemiology study. JMIR Public Heal Surveill. 2020; 6(2): 1–9.

[33] 33. Sipahutar T: DATA COVID-19_indonesia.xlsx. Dataset. figshare. 2024. Publisher Full Text

[34] 34. Sipahutar T: LIST OF FIGURES.pdf. figshare. Figure. 2024. Publisher Full Text

The Phenomenon of the COVID-19 Pandemic for Indonesian Health Policy, Hotspot and Risk Factors using Geographically Weighted Regression in 2020 - 2022

Abstract

Background

Method

Result

Conclusion

Keywords

Introduction

Methods

Study design dan data source

Data source

Study variables

Statistical analysis

Context of study area

Table 1. List of Indonesia’s provinces.

Results

Figure 1. Trend of COVID-19 cases 2020-2022.

Figure 2. Descriptive statistic of COVID-19 risks factors.

Spatial autocorrelation

Figure 3. The COVID-19 hotspot areas in Moran’s Scatter Plot 2020 – 2022.

Table 2. Index Moran value 2020 – 2022.

The hotspot areas of COVID-19 in Indonesia

Table 3. The COVID-19 hotspot provinces in Indonesia 2020-2022.

Figure 4. The map of COVID-19 hotspot provinces in Indonesia 2022.

COVID-19 risk factors in 33 provinces in Indonesia

Table 4. First assumption test result.

Table 5. Assumption test result with transformed data.

COVID-19 risk factors in each province

Table 6. The comparison of bandwidth value using Kernell Gaussian dan Bi-square.

Table 7. COVID-19 risk factors model of all provinces in Indonesia 2022.

Discussion

Limitation

Conclusion and recommendation

Ethical considerations

Data availability

Acknowledgements

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Browse by related subjects

Competing Interests Policy

Stay Updated