Selection of the Maximum Spatial Cluster Size of the Spatial Scan Statistic by Using the Maximum Clustering Set-Proportion Statistic

Spatial distribution and associated factors of cesarean section in Ethiopia using mini EDHS 2019 data: a community based cross-sectional study

Maternal health is a major public health tricky globally. Cesarean section delivery reduces morbidity and mortality when certain complications occur throughout pregnancy and labor. Cesarean section subjected to the availability and use of essential obstetric services in regional factors in Ethiopia. There was a scarcity of studies that assess the spatial distribution and associated factors of cesarean section. Consequently, this study aimed to assess the spatial variation of cesarean section and associated factors using mini EDHS 2019 national representative data. A community based cross-sectional study was conducted in Ethiopia from March to June 2019. A two-stage stratified sampling design was used to select participants. A Global Moran's I and Getis-Ord Gi* statistic hotspot analysis was used to assess the spatial distribution. Kuldorff's SaTScan was employed to determine the purely statistically significant spatial clusters. A multilevel binary logistic regression model fitted to identify factors. A total of 5753 mothers were included. More than one-fourth of mothers delivered through cesarean section at private health institutions and 54.74% were not educated. The proportion of cesarean section clustered geographically in Ethiopia and hotspot areas were observed in Addis Ababa, Oromia, Tigray, Derie Dewa, Amhara, and SNNR regions. Mothers' age (AOR = 1.07, 95% CI 1.02-1.12), mother's had secondary education (AOR = 2.113, 95% CI 1.414, 3.157), mother's higher education (2.646, 95% CI 1.724, 4.063), Muslim religion followers (AOR = 0.632, 95% CI 0.469, 0.852), poorer (AOR = 1.719, 95% CI 1.057, 2.795), middle wealth index (AOR = 1.769, 95% CI 1.073, 2.918), richer (AOR = 2.041, 95% CI 1.246, 3.344), richest (AOR = 3.510, 95% CI 2.197, 5.607), parity (AOR = 0.825, 95% CI 0.739, 0.921), and multiple pregnancies (AOR = 4.032, 95% CI 2.418, 6.723) were significant factors. Therefore, geographically targeted interventions are essential to reduce maternal and infant mortality with WHO recommendations for those Muslim, poorest and not educated mothers.

Assessing the short-term effects of PM2.5 and O3 on cardiovascular mortality using high-resolution exposure: a time-stratified case cross-over study in Southwestern China

Air pollution is a major risk factor of cardiovascular disease (CVD). To date, limited studies have estimated the effects of ambient air pollution on CVD mortality using high-resolution exposure assessment, which might fail to capture the spatial variation in exposure and introduce bias in results. Besides, the three-year action plan (TYAP, 2018-2020) was released; thus, the constitution and health effect of air pollutants may have changed. In this study, we estimated the short-term effect exposed to particulate matters with parameter less than 2.5 µm (PM2.5) and ozone (O3) with 0.05° × 0.05° resolution on CVD mortality and measured the influence of TYAP in the associations. We used random forest models with spatial weight matrices to attain high-resolution pollutant concentrations and conditional Poisson regression to assess the relationship between air pollution and cardiovascular mortality. With an increase of 10 µg/m3 in PM2.5 and O3 during 2018-2021 in the Sichuan Basin (SCB), CVD mortality increased 1.0134 (95% CI 1.0102, 1.0166) and 1.0083 (95% CI 1.0060, 1.0107), respectively, using high-resolution air pollutant concentration, comparing to 1.0070 (95% CI 1.0052, 1.0087) and 1.0057 (95% CI 1.0037, 1.0078) using data from air quality monitoring stations (AQMs). After TYAP, the relative risk (RR) due to PM2.5 rose up to 1.0149 (95% CI 1.0054, 1.0243), and the RR due to O3 rose up to 1.0089 (95% CI 1.0030, 1.0148) in Sichuan Province. We found significantly positive association of cardiovascular mortality and air pollution in Sichuan Province. And using high-resolution exposure would be more accurate to estimate the effect of air pollution on CVD. After TYAP, the cardiovascular mortality risk estimation due to PM2.5 decreased in elderly in SCB, and the risk due to O3 increased in Sichuan Province.

Detecting the spatial clustering of exposure-response relationships with estimation error: a novel spatial scan statistic

Detecting the spatial clustering of the exposure-response relationship (ERR) between environmental risk factors and health-related outcomes plays important roles in disease control and prevention, such as identifying highly sensitive regions, exploring the causes of heterogeneous ERRs, and designing region-specific health intervention measures. However, few studies have focused on this issue. A possible reason is that the commonly used cluster-detecting tool, spatial scan statistics, cannot be used for multivariate spatial datasets with estimation error, such as the ERR, which is often defined by a vector with its covariance estimated by a regression model. Such spatial datasets have been produced in abundance in the last decade, which suggests the importance of developing a novel cluster-detecting tool applicable for multivariate datasets with estimation error. In this work, by extending the classic scan statistic, we developed a novel spatial scan statistic called the estimation-error-based scan statistic (EESS), which is applicable for both univariate and multivariate datasets with estimation error. Then, a two-stage analytic process was proposed to detect the spatial clustering of ERRs in practical studies. A published motivating example and a simulation study were used to validate the performance of EESS. The results show that the clusters detected by EESS can efficiently reflect the clustering heterogeneity and yield more accurate ERR estimates by adjusting for such heterogeneity.

Optimizing the maximum reported cluster size for the multinomial-based spatial scan statistic

BACKGROUND

Correctly identifying spatial disease cluster is a fundamental concern in public health and epidemiology. The spatial scan statistic is widely used for detecting spatial disease clusters in spatial epidemiology and disease surveillance. Many studies default to a maximum reported cluster size (MRCS) set at 50% of the total population when searching for spatial clusters. However, this default setting can sometimes report clusters larger than true clusters, which include less relevant regions. For the Poisson, Bernoulli, ordinal, normal, and exponential models, a Gini coefficient has been developed to optimize the MRCS. Yet, no measure is available for the multinomial model.

RESULTS

We propose two versions of a spatial cluster information criterion (SCIC) for selecting the optimal MRCS value for the multinomial-based spatial scan statistic. Our simulation study suggests that SCIC improves the accuracy of reporting true clusters. Analysis of the Korea Community Health Survey (KCHS) data further demonstrates that our method identifies more meaningful small clusters compared to the default setting.

CONCLUSIONS

Our method focuses on improving the performance of the spatial scan statistic by optimizing the MRCS value when using the multinomial model. In public health and disease surveillance, the proposed method can be used to provide more accurate and meaningful spatial cluster detection for multinomial data, such as disease subtypes.

Spatial analysis of Varroa destructor and the relationship with surrounding landscape types in Southern Ontario

Elevated colony losses have continued to be an issue for Canadian beekeepers for more than a decade. Numerous studies have identified unmanaged Apis mellifera colony infestation by the Varroa destructor mite as a main cause of the problem. V. destructor spread externally of the hive through a phoretic stage in their life cycle. Consequently, their movement outside the hive is influenced by honey bee flight behaviours, which can range to multiple kilometers from the originating hive in any direction. V. destructor are therefore of regional concern as neighboring colonies and yards share nearby forage which can serve as fomites. Additionally, mites can be transmitted through bee behaviours such as robbing and drifting, thus impacting surrounding colonies. Understanding the distribution of mites across a population is key for surveillance and equitable allocation of resources. Spatial patterns of V. destructor infestations in Southern Ontario, Canada, were investigated using a combination of cluster analysis, scan statistics, and geostatistical modelling, using 5 years of provincial apiary inspection data, from 2015 to 2019. A collection of disease clusters of V. destructor infestations was identified and found to be stable over multiple years with several other individual clusters occurring sporadically throughout Southern Ontario during the same study period. Universal kriging was applied to the V. destructor data in combination with regional colony density, and land use data as covariates, producing an isopleth map of the prevalence risk for V. destructor infestation. No substantial link between V. destructor infestation and environmental factors was found. This study highlights the need for more data and investigation to determine the cause of the identified clusters and areas of elevated risk. These results are hypothesis-generating but simultaneously provide information for government agencies, industry organizations, and beekeepers into the spatial distribution of V. destructor at a macro scale.

Spatiotemporal trends and socioecological factors associated with Lyme disease in eastern Ontario, Canada from 2010-2017

Currently, there is limited knowledge about socioeconomic, neighbourhood, and local ecological factors that contribute to the growing Lyme disease incidence in the province of Ontario, Canada. In this study, we sought to identify these factors that play an important role at the local scale, where people are encountering ticks in their communities. We used reported human Lyme disease case data and tick surveillance data submitted by the public from 2010–2017 to analyze trends in tick exposure, spatiotemporal clusters of infection using the spatial scan statistic and Local Moran’s I statistic, and socioecological risk factors for Lyme disease using a multivariable negative binomial regression model. Data were analyzed at the smallest geographic unit, consisting of 400–700 individuals, for which census data are disseminated in Canada. We found significant heterogeneity in tick exposure patterns based on location of residence, with 65.2% of Lyme disease patients from the city of Ottawa reporting tick exposures outside their health unit of residence, compared to 86.1%—98.1% of patients from other, largely rural, health units, reporting peri-domestic exposures. We detected eight spatiotemporal clusters of human Lyme disease incidence in eastern Ontario, overlapping with three clusters of Borrelia burgdorferi-infected ticks. When adjusting for population counts, Lyme disease case counts increased with larger numbers of Borrelia burgdorferi-infected ticks submitted by the public, higher proportion of treed landcover, lower neighbourhood walkability due to fewer intersections, dwellings, and points of interest, as well as with regions of higher residential instability and lower ethnic concentration (Relative Risk [RR] = 1.25, 1.02, 0.67–0.04, 1.34, and 0.57, respectively, p < .0001). Our study shows that there are regional differences in tick exposure patterns in eastern Ontario and that multiple socioecological factors contribute to Lyme disease risk in this region.

Reclaiming independence in spatial-clustering datasets: A series of data-driven spatial weights matrices

Most spatial models include a spatial weights matrix (W) derived from the first law of geography to adjust the spatial dependence to fulfill the independence assumption. In various fields such as epidemiological and environmental studies, the spatial dependence often shows clustering (or geographic discontinuity) due to natural or social factors. In such cases, adjustment using the first‐law‐of‐geography‐based W might be inappropriate and leads to inaccuracy estimations and loss of statistical power. In this work, we propose a series of data‐driven Ws (DDWs) built following the spatial pattern identified by the scan statistic, which can be easily carried out using existing tools such as SaTScan software. The DDWs take both the clustering (or discontinuous) and the intuitive first‐law‐of‐geographic‐based spatial dependence into consideration. Aiming at two common purposes in epidemiology studies (ie, estimating the effect value of explanatory variable X and estimating the risk of each spatial unit in disease mapping), the common spatial autoregressive models and the Leroux‐prior‐based conditional autoregressive (CAR) models were selected to evaluate performance of DDWs, respectively. Both simulation and case studies show that our DDWs achieve considerably better performance than the classic W in datasets with clustering (or discontinuous) spatial dependence. Furthermore, the latest published density‐based spatial clustering models, aiming at dealing with such clustering (or discontinuity) spatial dependence in disease mapping, were also compared as references. The DDWs, incorporated into the CAR models, still show considerable advantage, especially in the datasets for common diseases.

Detection of temporal, spatial and spatiotemporal clustering of malaria incidence in northwest Ethiopia, 2012–2020

Malaria is one of Ethiopia's most targeted communicable diseases for elimination. Malaria transmission varies significantly across space and time; and Ethiopia had space–time disparity in its transmission intensities. Considering heterogeneity and transmission intensity at the district level could play a crucial role in malaria prevention and elimination. This study aimed to explore temporal, spatial, and spatiotemporal clusters of malaria incidence in northwest Ethiopia. The analysis is based on monthly malaria surveillance data of districts and collected from the Amhara public health institute. The Kulldorff's retrospective space–time scan statistics using a discrete Poisson model were used to detect temporal, spatial, and space–time clusters of malaria incidence with and without adjusting the altitude + LLIN arm. Monthly malaria incidence had seasonal variations, and higher seasonal indices occurred in October and November. The temporal cluster occurred in the higher transmission season between September and December annually. The higher malaria incidence risk occurred between July 2012 and December 2013 (LLR = 414,013.41, RR = 2.54, P < 0.05). The purely spatial clustering result revealed that the most likely cluster occurred in the north and northwest parts of the region while secondary clusters varied in years. The space–time clusters were detected with and without considering altitude + LLIN arm. The most likely space–time cluster was concentrated in northwestern and western parts of the region with a high-risk period between July 2012 and December 2013 (LLR = 880,088.3, RR = 5.5, P < 0.001). We found eight significant space–time clusters using the altitude + LLIN arm. The most likely space–time cluster occurred in the western and northwestern parts of the region in July 2012–December 2013 (LLR = 886,097.7, RR = 5.55, P < 0.05). However, secondary clusters were located in eastern, northwestern, western parts of regions, which had different cases and relative risks in each cluster. Malaria transmission had temporal, spatial, and space–time variation in the region at the district level. Hence, considering these variations and factors contributing to malaria stratification would play an indispensable role in preventing and controlling practices that ultimately leads to malaria eliminations.

Spatiotemporal cluster analysis of COVID-19 and its relationship with environmental factors at the city level in mainland China

This study sought to identify the spatial, temporal, and spatiotemporal clusters of COVID-19 cases in 366 cities in mainland China with the highest risks and to explore the possible influencing factors of imported risks and environmental factors on the spatiotemporal aggregation, which would be useful to the design and implementation of critical preventative measures. The retrospective analysis of temporal, spatial, and spatiotemporal clustering of COVID-19 during the period (January 15 to February 25, 2020) was based on Kulldorff's time-space scanning statistics using the discrete Poisson probability model, and then the logistic regression model was used to evaluate the impact of imported risk and environmental factors on spatiotemporal aggregation. We found that the spatial distribution of COVID-19 cases was nonrandom; the Moran's I value ranged from 0.017 to 0.453 (P < 0.001). One most likely cluster and three secondary likely clusters were discovered in spatial cluster analysis. The period from February 2 to February 9, 2020, was identified as the most likely cluster in the temporal cluster analysis. One most likely cluster and seven secondary likely clusters were discovered in spatiotemporal cluster analysis. Imported risk, humidity, and inhalable particulate matter PM_2.5 had a significant impact on temporal and spatial accumulation, and temperature and PM₁₀ had a low correlation with the spatiotemporal aggregation of COVID-19. The information is useful for health departments to develop a better prevention strategy and potentially increase the effectiveness of public health interventions.

Descriptive Spatial Analysis of Human-Elephant Conflict (HEC) Distribution and Mapping HEC Hotspots in Keonjhar Forest Division, India

Escalation of human-elephant conflict (HEC) in India threatens its Asian elephant (Elephas maximus) population and victimizes local communities. India supports 60% of the total Asian elephant population in the world. Understanding HEC spatial patterns will ensure targeted mitigation efforts and efficient resource allocation to high-risk regions. This study deals with the spatial aspects of HEC in Keonjhar forest division, where 345 people were killed and 5,145 hectares of croplands were destroyed by elephant attacks during 2001–2018. We classified the data into three temporal phases (HEC1: 2001–2006, HEC2: 2007–2012, and HEC3: 2013–2018), in order to (1) derive spatial patterns of HEC; (2) identify the hotspots of HEC and its different types along with the number of people living in the high-risk zones; and (3) assess the temporal change in the spatial risk of HEC. Significantly dense clusters of HEC were identified in Keonjhar and Ghatgaon forest ranges throughout the 18 years, whereas Champua forest range became a prominent hotspot since HEC2. The number of people under HEC risk escalated from 14,724 during HEC1 and 34,288 in HEC2, to 65,444 people during HEC3. Crop damage was the most frequent form of HEC in the study area followed by house damage and loss of human lives. Risk mapping of HEC types and high priority regions that are vulnerable to HEC, provides a contextual background for researchers, policy makers and managers.

Spatial pattern of perinatal mortality and its determinants in Ethiopia: Data from Ethiopian Demographic and Health Survey 2016

BACKGROUND

The perinatal mortality rate in Ethiopia is among the highest in Sub Saharan Africa. The aim of this study was to identify the spatial patterns and determinants of perinatal mortality in the country using a national representative 2016 Ethiopia Demographic and Health Survey (EDHS) data.

METHODS

The analysis was completed utilizing data from 2016 Ethiopian Demographic and Health Survey. This data captured the information of 5 years preceding the survey period. A total of 7230 women who at delivered at seven or more months gestational age nested within 622 enumeration areas (EAs) were used. Statistical analysis was performed by using STATA version 14.1, by considering the hierarchical nature of the data. Multilevel logistic regression models were fitted to identify community and individual-level factors associated with perinatal mortality. ArcGIS version 10.1 was used for spatial analysis. Moran's, I statistics fitted to identify global autocorrelation and local autocorrelation was identified using SatSCan version 9.6.

RESULTS

The spatial distribution of perinatal mortality in Ethiopia revealed a clustering pattern. The global Moran's I value was 0.047 with p-value <0.001. Perinatal mortality was positively associated with the maternal age, being from rural residence, history of terminating a pregnancy, and place of delivery, while negatively associated with partners' educational level, higher wealth index, longer birth interval, female being head of household and the number of antenatal care (ANC) follow up.

CONCLUSIONS

In Ethiopia, the perinatal mortality is high and had spatial variations across the country. Strengthening partner's education, family planning for longer birth interval, ANC, and delivery services are essential to reduce perinatal mortality and achieve sustainable development goals in Ethiopia. Disparities in perinatal mortality rates should be addressed alongside efforts to address inequities in maternal and neonatal healthcare services all over the country.

Spatial inequality, characteristics of internal migration, and pulmonary tuberculosis in China, 2011-2017: a spatial analysis

Background

Human migration facilitate the spread of tuberculosis (TB). Migrants face an increased risk of TB infection. In this study, we aim to explore the spatial inequity of sputum smear-positive pulmonary TB (SS + PTB) in China; and the spatial heterogeneity between SS + PTB and internal migration.

Methods

Notified SS + PTB cases in 31 provinces in mainland China were obtained from the national web-based PTB surveillance system database. Internal migrant data were extracted from the report on China’s migrant population development. Spatial autocorrelations were explored using the global Moran’s statistic and local indicators of spatial association. The spatial variation in temporal trends was performed using Kulldorff’s scan statistic. Fixed effect and spatial autoregressive models were used to explore the spatial inequity between SS + PTB and internal migration.

Results

A total of 2 380 233 SS + PTB cases were reported in China between 2011 and 2017, of which, 1 716 382 (72.11%) were male and 663 851 (27.89%) were female. Over 70% of internal migrants were from rural households and had lower income and less education. The spatial variation in temporal trend results showed that there was an 9.9% average annual decrease in the notification rate of SS + PTB from 2011 to 2017; and spatial clustering of SS + PTB cases was mainly located in western and southern China. The spatial autocorrelation results revealed spatial clustering of internal migration each year (2011–2017), and the clusters were stable within most provinces. Internal emigration, urban-to-rural migration and GDP per capita were significantly associated with SS + PTB, further, internal emigration could explain more variation in SS + PTB in the eastern region in mainland. However, internal immigration and rural-to-urban migration were not significantly associated with SS + PTB across China.

Conclusions

Our study found the spatial inequity between SS + PTB and internal migration. Internal emigration, urban-to-rural migration and GDP per capita were statistically associated with SS + PTB; the negative association was identified between internal emigration, urban-to-rural migration and SS + PTB. Further, we found those migrants with lower income and less education, and most of them were from rural households. These findings can help stakeholders to implement effective PTB control strategies for areas at high risk of PTB and those with high rates of internal migration.

Epidemiological and aetiological characteristics of hand, foot, and mouth disease in Sichuan Province, China, 2011-2017

Hand, foot, and mouth disease (HFMD) remains a threat to the Asia-Pacific region. The epidemiological characteristics and pathogen spectrum of HFMD vary with space and time. These variations are crucial for HFMD interventions but poorly understood in Sichuan Province, China, particularly after the introduction of the EV-A71 vaccine. Using descriptive methods, regression analyses, spatial autocorrelation analysis, and space-time scan statistics, we analysed the epidemiological and aetiological characteristics of HFMD surveillance data in Sichuan Province between 2011 and 2017 to identify spatio-temporal variations. The dominant serotypes of HFMD have changed from enterovirus 71 and coxsackievirus A16 to other enteroviruses since 2013. The seasonal pattern of HFMD showed two peaks generally occurring from April to July and November to December; however, the seasonal pattern varied by prefecture and enterovirus serotype. From 2011 to 2017, spatio-temporal clusters were increasingly concentrated in Chengdu, with several small clusters in northeast Sichuan. The clusters observed in southern Sichuan from 2011 to 2015 disappeared in 2016–2017. These findings highlight the importance of pathogen surveillance and vaccination strategies for HFMD interventions; future prevention and control of HFMD should focus on Chengdu and its vicinity.

Using the maximum clustering heterogeneous set-proportion to select the maximum window size for the spatial scan statistic

The spatial scan statistic has been widely used to detect spatial clusters that are of common interest in many health-related problems. However, in most situations, different scan parameters, especially the maximum window size (MWS), result in obtaining different detected clusters. Although performance measures can select an optimal scan parameter, most of them depend on historical prior or true cluster information, which is usually unavailable in practical datasets. Currently, the Gini coefficient and the maximum clustering set-proportion statistic (MCS-P) are used to select appropriate parameters without any prior information. However, the Gini coefficient may be unstable and select inappropriate parameters, especially in complex practical datasets, while the MCS-P may have unsatisfactory performance in spatial datasets with heterogeneous clusters. Based on the MCS-P, we proposed a new indicator, the maximum clustering heterogeneous set-proportion (MCHS-P). A simulation study of selecting the optimal MWS confirmed that in spatial datasets with heterogeneous clusters, the MWSs selected using the MCHS-P have much better performance than those selected using the MCS-P; moreover, higher heterogeneity led to a larger advantage of the MCHS-P, with up to 538% and 69.5% improvement in the Youden's index and misclassification in specific scenarios, respectively. Meanwhile, the MCHS-P maintains similar performance to that of the MCS-P in spatial datasets with homogeneous clusters. Furthermore, the MCHS-P has significant improvements over the Gini coefficient and the default 50% MWS, especially in datasets with clusters that are not far from each other. Two practical studies showed similar results to those obtained in the simulation study. In the case where there is no prior information about the true clusters or the heterogeneity between the clusters, the MCHS-P is recommended to select the MWS in order to accurately identify spatial clusters.

Spatial and temporal clustering analysis of tuberculosis in the mainland of China at the prefecture level, 2005-2015

Background

Tuberculosis (TB) is still one of the most serious infectious diseases in the mainland of China. So it was urgent for the formulation of more effective measures to prevent and control it.

Methods

The data of reported TB cases in 340 prefectures from the mainland of China were extracted from the China Information System for Disease Control and Prevention (CISDCP) during January 2005 to December 2015. The Kulldorff’s retrospective space-time scan statistics was used to identify the temporal, spatial and spatio-temporal clusters of reported TB in the mainland of China by using the discrete Poisson probability model. Spatio-temporal clusters of sputum smear-positive (SS+) reported TB and sputum smear-negative (SS-) reported TB were also detected at the prefecture level.

Results

A total of 10 200 528 reported TB cases were collected from 2005 to 2015 in 340 prefectures, including 5 283 983 SS- TB cases and 4 631 734 SS + TB cases with specific sputum smear results, 284 811 cases without sputum smear test. Significantly TB clustering patterns in spatial, temporal and spatio-temporal were observed in this research. Results of the Kulldorff’s scan found twelve significant space-time clusters of reported TB. The most likely spatio-temporal cluster (RR = 3.27, P <  0.001) was mainly located in Xinjiang Uygur Autonomous Region of western China, covering five prefectures and clustering in the time frame from September 2012 to November 2015. The spatio-temporal clustering results of SS+ TB and SS- TB also showed the most likely clusters distributed in the western China. However, the clustering time of SS+ TB was concentrated before 2010 while SS- TB was mainly concentrated after 2010.

Conclusions

This study identified the time and region of TB, SS+ TB and SS- TB clustered easily in 340 prefectures in the mainland of China, which is helpful in prioritizing resource assignment in high-risk periods and high-risk areas, and to formulate powerful strategy to prevention and control TB.

Electronic supplementary material

The online version of this article (10.1186/s40249-018-0490-8) contains supplementary material, which is available to authorized users.

Spatiotemporal Distribution and Geospatial Diffusion Patterns of 2013 Dengue Outbreak in Jhapa District, Nepal

This study describes spatiotemporal distribution and geospatial diffusion patterns of dengue outbreak of 2013 in Jhapa district, Nepal. Laboratory-confirmed dengue cases were collected from the District Public Health Office, Government of Nepal. Choropleth mapping technique, Global Moran's Index, SaTScan, and standard deviational ellipse were used to map and quantify the outbreak dynamics. The results revealed heterogeneous distribution and globally autocorrelated patterns. Local clusters were observed in 3 major urban centers. The standard deviational ellipse demonstrated the outbreak occurred from the east and diffused to the west along the east-west highway in different weeks. The results of this study could be useful to public health authorities to plan and execute dengue control strategies.

Using the Kulldorff's scan statistical analysis to detect spatio-temporal clusters of tuberculosis in Qinghai Province, China, 2009-2016

BACKGROUND

Although the incidence of tuberculosis (TB) in most parts of China are well under control now, in less developed areas such as Qinghai, TB still remains a major public health problem. This study aims to reveal the spatio-temporal patterns of TB in the Qinghai province, which could be helpful in the planning and implementing key preventative measures.

METHODS

We extracted data of reported TB cases in the Qinghai province from the China Information System for Disease Control and Prevention (CISDCP) during January 2009 to December 2016. The Kulldorff's retrospective space-time scan statistics, calculated by using the discrete Poisson probability model, was used to identify the temporal, spatial, and spatio-temporal clusters of TB at the county level in Qinghai.

RESULTS

A total of 48,274 TB cases were reported from 2009 to 2016 in Qinghai. Results of the Kulldorff's scan revealed that the TB cases in Qinghai were significantly clustered in spatial, temporal, and spatio-temporal distribution. The most likely spatio-temporal cluster (LLR = 2547.64, RR = 4.21, P < 0.001) was mainly concentrated in the southwest of Qinghai, covering seven counties and clustered in the time frame from September 2014 to December 2016.

CONCLUSION

This study identified eight significant space-time clusters of TB in Qinghai from 2009 to 2016, which could be helpful in prioritizing resource assignment in high-risk areas for TB control and elimination in the future.