|Ahead of print
|Spatiotemporal analysis, hotspot mapping, and clustering of confirmed cases of COVID-19 in the initial phase of the pandemic in Qom province, Iran
Abolfazl Mohammadbeigi1, Shahram Arsang-Jang2, Ehsan Sharifipour3, Alireza Koohpaei4, Mostafa Vahedian5, Narges Mohammadsalehi1, Masoud Jafaresmaeili6, Moharam Karami6, Siamak Mohebi7
1 Department of Epidemiology and Biostatistics, Research Center for Environmental Sciences, Faculty of Health, Qom University of Medical Sciences, Qom, Iran
2 Department of Biostatistics and Epidemiology, Faculty of Medicine, Zanjan University of Medical Sciences, Zanjan, Iran
3 Neuroscience Research Center, Department of Neurology, Faculty of Medicine, Qom University of Medical Sciences, Qom, Iran
4 Department of Occupational Health, Faculty of Health, Qom University of Medical Sciences, Qom, Iran
5 Department of Social Medicine, Faculty of Medicine, Qom University of Medical Sciences, Qom, Iran
6 Health Vic Chancellor, Qom University of Medical Sciences, Qom, Iran
7 Department of Health Education, Faculty of Health, Qom University of Medical Sciences, Qom, Iran
Click here for correspondence address and email
|Date of Submission||07-Mar-2021|
|Date of Decision||14-Jun-2021|
|Date of Acceptance||30-Jun-2021|
|Date of Web Publication||15-Jul-2021|
Objective: To identify the incidence rate, relative risk, hotspot regions and incidence trend of COVID-19 in Qom province, northwest part of Iran in the first stage of the pandemic.
Methods: The study included 1 125 officially reported PCR-confirmed cases of COVID-19 from 20 February 2020 to 20 April 2020 in 90 regions in Qom city, Iran. The Bayesian hierarchical spatial model was used to model the relative risk of COVID-19 in Qom city, and the segmented regression model was used to estimate the trend of COVID-19 incidence rate. The Poisson distribution was applied for the observed number of COVID-19, and independent Gamma prior was used for inference on log-relative risk parameters of the model.
Results: The total incidence rate of COVID-19 was estimated at 89.5 per 100 000 persons in Qom city (95% CI: 84.3, 95.1). According to the results of the Bayesian hierarchical spatial model and posterior probabilities, 43.33% of the regions in Qom city have relative risk greater than 1; however, only 11.11% of them were significantly greater than 1. Based on Geographic Information Systems (GIS) spatial analysis, 10 spatial clusters were detected as active and emerging hotspot areas in the south and central parts of the city. The downward trend was estimated 10 days after the reporting of the first case (February 7, 2020); however, the incidence rate was decreased by an average of 4.24% per day (95%CI:-10.7, -3.5).
Conclusions: Spatial clusters with high incidence rates of COVID-19 in Qom city were in the south and central regions due to the high population density. The GIS could depict the spatial hotspot clusters of COVID-19 for timely surveillance and decision-making as a way to contain the disease.
Keywords: 2019 coronavirus disease; Geographic information science; Incidence rates; Spatial cluster; Spatial hotspot; Mapping
|How to cite this URL:|
Mohammadbeigi A, Arsang-Jang S, Sharifipour E, Koohpaei A, Vahedian M, Mohammadsalehi N, Jafaresmaeili M, Karami M, Mohebi S. Spatiotemporal analysis, hotspot mapping, and clustering of confirmed cases of COVID-19 in the initial phase of the pandemic in Qom province, Iran. J Acute Dis [Epub ahead of print] [cited 2022 Jan 19]. Available from: http://www.jadweb.org/preprintarticle.asp?id=320963
| 1. Introduction|| |
The Islamic Republic of Iran reported its first 2019 novel coronavirus (COVID-19) cases on 19th February 2020 in Qom, northwest of the country. Until to 4 June 2021, COVID-19 is spreaded in 220 countries and regions worldwide, and the number of confirmed cases and deaths is 179974 279 and 3 719 003 worldwide as 2 954 309 and 80 813 in Iran, respectively. As it poses a huge threat to the human community,, the COVID-19 outbreak was declared a pandemic by World Health Organization on March 11th, 2020.
Spatial modeling is an effective approach to understand the structural and socio-demographic factors that affect COVID-19 spread in different regions. Up to today, rare studies have been designed to evaluate the spatial spread of the COVID-19 in Qom province. However, understanding the spatial spread of the COVID-19 outbreak is critical to comprehend the outbreak patterns and accordingly develop prevention policies during the early stages of the pandemic and for outbreaks down the line.
In this context, spatial analysis is necessary to depict the current situation of the outbreak and to prevent the spread of the disease,. Understanding the spatiotemporal patterns of the COVID-19 epidemic is critical in effectively preventing and controlling the pandemic,. Moreover, assessment of the spatial spread of the COVID-19 outbreak is also crucial to identify the dynamic of further transmission,. To this end, we used data of COVID-19 confirmed cases based on PCR results to determine the incidence rate, relative risk of the incidence, hotspot regions, and the trend of incidence in the initial phase of the pandemic in Qom city. The results of the current study may provide valuable information about the incidence of COVID-19 and the transmission of the disease, for prevention at both the individual and organization levels.
| 2. Materials and methods|| |
2.1. Ethical consideration
This study was approved by the Ethical Committee of Qom University of Medical Sciences with approved No IR.MUQ. REC.1398.154 at 2020.03.10.
2.2. Study setting and subjects
Qom city located in the northerwest part of Iran, which has 90 subordinate regions. In this study, we used data of confirmed cases in the initial phase of the pandemic (from 20 February 2020 to 20 April 2020) in this city. Moreover, data of population census of the city and each region were obtained from the most recent data available. All data of confirmed cases were collected from the official database of the health Vic chancellor and its reports. Consequently, data of 1 125 cases were collected and entered into the further analysis.
2.3. Statistical analysis
Microsoft Excel 2017 (Microsoft Corporation, Redmond, WA, USA) was used to analyze data in this study. The geographical data including X and Y were received based on the residency place of cases and imported into Geographic Information System (GIS). The shapefiles of Qom city were used for ArcGIS (Environmental Systems Research Institute, Redlands, WA, USA) analysis. The segmented regression model was used to estimate the trend of the COVID-19 incidence rate. The Bayesian hierarchical spatial model implementation of the Besag, York and Mollié (BYM) model was used to model the relative risk (RR) of COVID-19 in Qom city. The Poisson distribution was used for the observed number of COVID-19, and independent Gamma prior for inference on log-relative risk parameters of the BYM model. The province-specific effects are assumed to follow conditional autoregressive normal prior distributions. The Bayesian posterior probability was used to test the difference of area-specific RR from 1. The Gelman-Rubin, trace, autocorrelation, and posterior-density plots were used to check the convergence of fitted models. The Gibbs samplers for generating samples from the posterior distribution were implemented in OpenBUGS.
| 3. Results|| |
A total of 1 125 confirmed cases were included in the current study from 20 February 2020 to 20 April 2020 in 90 regions in Qom city, Iran. The mean age of the cases was (57.76±18.42) years, ranging from 1 to 98 years. The sex distribution showed that 58% (653 cases) were male and 42% (472 cases) were female. From all confirmed cases 35.8% (403 cases) died, and 64.2% (722 cases) recovered.
[Figure 1] shows the estimated incidence rates for different regions in Qom city. The total incidence rate of COVID-19 was estimated at 89.6 per 100 000 persons in Qom city (95% CI: 84.3, 95.1). [Figure 2] shows the results for RR of COVID-19 incidence using the BYM model. According to posterior probabilities, 43.33% (39/90) of the regions in Qom city have RR greater than 1, however, 11.11% (10/90) of them were significantly greater than 1. In addition, 15.55% of the regions have RR significantly lower than 1 [Figure 2].
|Figure 1: The estimated incidence rates of different regions in Qom city.|
Click here to view
|Figure 2: Bayesian hierarchical spatial model showing the relative risk of COVID-19 incidence.|
Click here to view
Based on the results of GIS spatial analysis 10 spatial clusters was determined as active and emerging hotspot areas in the south and central of the city. The regions including Neshat (P13, RR=2.56), Hafte-Tir (P63, RR=1.977), Ammar-Yaser (P38, RR=1.704), Zanbilabad (P36, RR=1.682), Zaviyeh (P12, RR=1.682), Bajak-2 (P76, RR=1.67), Eram (P16, RR=1.575), Keyvanfar (P11, RR=1.567), 22 Bahman (P74, RR=1.462), and Shah-Ahmad Ghasem (P79, RR=1.444). All of them showed RR significantly greater than 1 [Figure 3]. Appendix 1 [Additional file 1] shows the region’s number (Pi).
|Figure 3: The hotspot regions for COVID-19 in Qom city based on the relative risk.|
Click here to view
The segmented regression model estimated one change point for the incidence rate trend of COVID-19 in Qom. The downward trend was estimated after February 7th, 2020 (10 d after the first case confirmation with COVID-19), however, the incidence rate was decreased by an average of 4.24 % per day (95% CI: -10.7, -3.5) [Figure 4].
|Figure 4: The incidence rate of COVID-19 trend in Qom from 20 February 2020 to 20 April 2020. APC: Annual precent change. : The annual precent change is significantly different from zero at the alpha=0.05 level.|
Click here to view
| 4. Discussion|| |
As a result, COVID-19 is a major threat to human health, production, life, social functioning, and international relations. The total incidence rate of COVID-19 in Qom city was 89.5 per 100 000 persons and this incidence varied among different regions from 13.9 to 274.2 per 100 000 persons. Our results showed that the RR of COVID-19 morbidity was greater than 1 in more than 43% of regions in Qom city, while, only in 15.5% of regions, the RR of morbidity was lower than 1, significantly. Moreover, in 11.11% of all regions of Qom city, the risk of morbidity (RR) was significantly higher than 1. These results showed that the risk of COVID-19 in Qom city was high and the spread of disease in the city was expanding.
Faced with the COVID-19 pandemic, GIS could play an important role in prediction, prevention, and health policy-making. Its functions include but not limited to rapid visualization of epidemic information, spatial tracking of confirmed cases, prediction of regional transmission, spatial segmentation of the epidemic risk and prevention level, balancing and management of the supply and demand of material resources, and social-emotional guidance and panic elimination,.
This hotspot mapping in our study showed that the spread of COVID-19 within Qom city was spatially correlated. According the map, a cluster of regions showed significantly higher rates of COVID-19 incidence in the center of the city, which can be explained by the fact that these regions have a higher population density. More crowded and near contact to the first reffering hospital center (Kamkar Hospital) of COVID-19 in Qom beside the older population structures were the most probable factors of higher incidence of COVID-19 in this area.
According to the segmented regression model, we found that the increasing trend of COVID-19 in Qom reversed to a downward trend 10 d after the reporting of the first case, and the incidence rate was decreased approximately 4.24% per day. This is due to the slowdown dynamic of the disease and the decrease of the basic reproduction number (R0) due to effective interventions. A study by Fang et al. showed an upward trend of R0 in the beginning and then followed by a downward trend, a temporary rebound, and another continuous decline for the COVID-19 incidence. According to another study, rigorous government control policies were associated with a slower increase in the infected population.
Though Qom is the first city to report the COVID-19 case in Iran, some religious factors hindered the promotion and implementation of the prevention measures. Nevertheless, besides treatment, isolation and protective procedures such as social distancing and domestic quarantine were effective measures for containing the disease and decreasing R0 to lower 1. The R0 values of COVID-19 are very different among studies due to the dynamic of disease and time of the investigation. However, a review study estimated that the overall R0 estimate was 3.38±1.40, with a range of 1.90 to 6.49, higher than the data released by the World Health Organization. Nevertheless, it was estimated from 2.24 to 3.58 in one study and largely followed the exponential growth, while it estimated 3.2 to 3.9 in another study in China.
Regarding the higher risk of morbidity in the central area of Qom city, we suggest that policies calling on social distancing, protecting older adults and other vulnerable populations, as well as promoting health literacy, should be developed and promoted to constrain the spread of COVID-19. Besides, Our approach could be applied to model COVID-19 outbreaks in other cities and provinces of Iran. By doing so, early detection, isolation, and treatment for suspected cases can be efficiently done as a way to control COVID-19.
This study was one of the first spatiotemporal analysis of COVID-19 in Iran. Therefore, it is exposed to some limitations. First the diagnosis based on PCR was limited because limitation in diagnosios kit. Second, the high incidence of diseases forced us to maping the confirmed cases to detect the hotspot areas. Due to these limitations, it is suggested for future studies one years after initial of COVID-19 pandemic.
| 5. Conclusions|| |
Spatial clusters with high incidence rates of COVID-19 in Qom city were in the south and central regions due to the high population density. The GIS could depict the spatial hotspot clusters of COVID-19 for timely surveillance and decision-making as it seeks to control the disease. Therefore, increasing the early diagnosis and treatment of patients besides controlling the population migration are effective approaches to prevent and control the regional outbreak of the epidemic.
Conflict of interest statement
The authors report no conflict of interest.
Study concept and design: A.M., and S.A.; Analysis and interpretation of data: M.V., and E.S, and A.K.; Drafting of the manuscript: S.M.; Critical revision of the manuscript for important intellectual content: M.K., M.E and S,M., and E.Sh; Statistical analysis: A.M and M.V.
| References|| |
Ramírez-Aldana R, Gomez-Verjan JC, Bello-Chavolla OY. Spatial analysis of COVID-19 spread in Iran: Insights into geographical and structural transmission determinants at a province level. PLoS Negl Trop Dis
Kang D, Choi H, Kim JH, Choi J. Spatial epidemic dynamics of the COVID-19 outbreak in China. Int J Infect Dis
Andrade LA, Gomes DS, Góes MAO, Souza MSF, Teixeira DCP, Ribeiro CJN, et al. Surveillance of the first cases of COVID-19 in Sergipe using a prospective spatiotemporal analysis: the spatial dispersion and its public health implications. Rev Soc Bras Med Trop
Zhou CH, Su FZ, Pei T, Zhang A, Du YY, Luo B, et al. COVID-19: Challenges to GIS with big data. Geogr Sustain
Kang DY, Choi H, Kim JH, Choi J. Spatial epidemic dynamics of the COVID-19 outbreak in China. Int J Infect Dis
Yang WT, Deng M, Li CK, Huang JC. Spatio-temporal patterns of the 2019-nCoV epidemic at the county level in Hubei province, China. Int J Environ Res Public Health
Adegboye OA, Adekunle AI, Gayawan E. Early transmission dynamics of novel coronavirus (COVID-19) in Nigeria. Int J Environ Res Public Health
Sarwar S, Waheed R, Sarwar S, Khan A. COVID-19 challenges to Pakistan: Is GIS analysis useful to draw solutions? Sci Total Environ
Adekunle IA, Onanuga AT, Akinola OO, Ogunbanjo OW. Modelling spatial variations of coronavirus disease (COVID-19) in Africa. Sci Total Environ
Fang YQ, Nie YT, Penny M. Transmission dynamics of the COVID-19 outbreak and effectiveness of government interventions: A data-driven analysis. J Med Virol
Zhao S, Lin QY, Ran JJ, Musa SS, Yang GP, Wang WM, et al. Preliminary estimation of the basic reproduction number of novel coronavirus (2019-nCoV) in China, from 2019 to 2020: A data-driven analysis in the early phase of the outbreak. Int J Infect Dis
Alimohamadi Y, Taghdir M, Sepandi M. Estimate of the basic reproduction number for covid-19: A systematic review and meta-analysis. J Prev Med Public Health
Zhou T, Liu QH, Yang ZM, Liao JY, Yang KX, Bai W, et al. Preliminary prediction of the basic reproduction number of the Wuhan novel coronavirus 2019-nCoV. J Evid Based M
Lipsitch M, Cohen T, Cooper B, Robins JM, Ma S, James L, et al. Transmission dynamics and control of severe acute respiratory syndrome. Science
Department of Biostatistics and Epidemiology, Faculty of Medicine, Zanjan University of Medical Sciences, Zanjan
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
| Article Access Statistics|
| Viewed||329 |
| PDF Downloaded||10 |