Tuberculosis (TB) is known as a disease that prone to spatial clustering. Recent development has seen a sharp rise in the number of epidemiologic studies employing Geographical Information System (GIS), particularly in identifying TB clusters and evidences of etiologic factors. The aim of this systematic review is to determine evidence of TB clustering, type of spatial analysis commonly used and the application of GIS in TB surveillance and control. A literature search of articles published in English language between 2000 and November 2015 was performed using MEDLINE and Science Direct using relevant search terms related to spatial analysis in studies of TB cluster. The search strategy was adapted and developed for each database using appropriate subject headings and keywords. The literature reviewed showed strong evidence of TB clustering occurred in high risk areas in both developed and developing countries. Spatial scan statistics were the most commonly used analysis and proved useful in TB surveillance through detection of outbreak, early warning and identifying area of increased TB transmission. Among others are targeted screening and assessment of TB program using GIS technology. However there were limitations on suitability of utilizing aggregated data such as national cencus that were pre-collected in explaining the present spatial distribution among population at risk. Spatial boundaries determined by zip code may be too large for metropolitan area or too small for country. Nevertheless, GIS is a powerful tool in aiding TB control and prevention in developing countries and should be used for real-time surveillance and decision making.
Tuberculosis cluster ; geographical information system ; spatial analysis

OBJECTIVETo review the transmission models of tuberculosis in heterogeneous population.

DATA SOURCESThe data used in this review were adopted mainly from the studies of models of tuberculosis reported from 1995 to 2006.

STUDY SELECTIONRelevant literature on transmission models of tuberculosis in heterogeneous populations are referenced.

RESULTSCasual/random factors and genetic factors are the main reasons for epidemics of tuberculosis in recent years. Mass public transport is playing the primary role in casually close contact which can facilitate the transmission of tuberculosis. Genetic susceptibility not only varies endemic prevalence levels, but also drastically alters the effects of treatment for tuberculosis patients. Detailed studies further exhibit that casual contact and genetic factor are responsible for over 30% - 40% of the total new cases in recent years. The prevalence of tuberculosis could double (from 33% to 60%) if a genetically susceptible phenotype is present in only 30% of the population. And some challenges have emerged along with these exciting results.

CONCLUSIONSCasual/random contact, public transport and genetic susceptibility are responsible for most new tuberculosis cases and a wide variation in endemic tuberculosis levels between regions. Hence, the transmission model of tuberculosis in a heterogeneous population can provide more clues to underlying mechanism of tuberculosis transmission than in a homogeneous population. However, many challenges remain for us in understanding transmission of disease.

Cluster Analysis ; Genetic Predisposition to Disease ; Humans ; Models, Theoretical ; Tuberculosis ; transmission

OBJECTIVETo study the tuberculosis clustering areas and the changing trend, from 2008 to 2010, so as to provider the reference for tuberculosis control.

METHODSGlobal spatial autocorrelation and SaTScan methods were used to detect and analyse the spatial clustering of total tuberculosis notification rate and the new smear-positive pulmonary tuberculosis notification rate, at the provincial level from 2008 to 2010.

RESULTSThe spatial clustering (SC) phenomenon was significant on total notification rate and new smear-positive pulmonary tuberculosis notification rate from 2008 to 2010 (P < 0.01). The coverages of clustering areas on total notification rate showed a reduction from 19 provinces to 14 provinces, distributed in the south, west and north-east areas of China. The coverages of clustering areas on new smear-positive pulmonary tuberculosis notification rate concentrated in 14 provinces which covered the south and north-east of China.

CONCLUSIONThe disease burden and the risk of transmission in the clustering areas of tuberculosis both located in the south and the north-east of China. The disease burden of tuberculosis was high in the west of China, but not the areas with high risk of transmission.

China ; epidemiology ; Cluster Analysis ; Humans ; Statistical Distributions ; Tuberculosis ; epidemiology

Objective: To understand the recent transmission of Mycobacterium tuberculosis (MTB), and to identify the influencing factors of recent transmission among pulmonary tuberculosis (TB) patients in Jing'an district, Shanghai. Methods: The genotypes and drug resistances of MTB isolated from TB patients registered in the TB designated hospitals in Jing'an district during 2010-2015 were analyzed through 12-loci Mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR)(QUB11b, QUB18, Mtub21, Miru26, QUB26, Mtub04, Miru31, Miru40, VNTR2372, VNTR3820, 3232, 4120), and tested for drug susceptibility as well. With the results of field epidemiological investigation, univariate and multivariate analyses were performed to analyze the distribution of the clusters and influencing factors on recent transmission. Results: This study enrolled 80 TB patients, 23 (28.75%) had a resistance to at least one anti-TB drug, and the prevalence of multidrug-resistant tuberculosis (MDR-TB) was 16.25%. A total of 65 genotypes were identified with 58 (72.50%, 58/80) being unique and 7 clusters with 2-10 isolated in each cluster. The proportion of clustering was 27.50% (22/80). Results from the multivariate analysis revealed that multidrug- resistance (OR=35.799, 95%CI: 4.239-302.346) and having comorbidity with TB (OR=7.695, 95%CI: 1.421-41.658) were independently associated with the clustering, which suggesting a recent transmission. The field investigation to the clustered cases proved that the patients in two clusters had epidemiological links, one was between family members, and the other contained 10 MDR-TB patients with 9 knowing each other which have a definite connection and 1 having the possible connection with them. Conclusion: Recent transmission of tuberculosis happened among TB patients in Jing'an district, with high risks among the MDR-TB patients.
China ; Cluster Analysis ; Genotype ; Humans ; Mycobacterium tuberculosis/isolation & purification* ; Tuberculosis ; Tuberculosis, Multidrug-Resistant/transmission* ; Tuberculosis, Pulmonary/transmission*

Objective: To analyze the spatial distribution and identify the high risk areas of pulmonary tuberculosis at the township level in Beijing during 2005-2015. Methods: Data on pulmonary tuberculosis cases was collected from the tuberculosis information management system. Global autocorrelation analysis, local indicators of spatial association and Kulldorff's Scan Statistics were applied to map the spatial distribution and detect the space-time clusters of the pulmonary tuberculosis cases during 2005-2015. Results: Spatial analysis on the incidence of pulmonary tuberculosis at the township level demonstrated that the spatial autocorrelation was positive during the study period. The values of Moran's I ranged from 0.224 3 to 0.291 8 with all the P values less than 0.05. Hotspots were primarily distributed in 8 towns/streets as follows: Junzhuang, Wangping, Yongding and Tanzhesi in Mentougou district, Yancun in Fangshan district, Wangzuo town in Fengtai district, Tianqiao street in Xicheng district and Tianzhu town in Shunyi district. Spatiotemporal clusters across the entire study period were identified by using Kulldorff's spatiotemporal scan statistic. The primary cluster was located in Chaoyang and Shunyi districts, including 17 towns/streets, as follows: Cuigezhuang, Maizidian, Dongfeng, Taiyanggong, Zuojiazhuang, Hepingjie, Xiaoguan, Xiangheyuan, Dongba, Jiangtai, Wangjing, Jinzhan, Jiuxianqiao, Laiguangying, Sunhe towns/streets in Chaoyang district, Houshayu and Tianzhu town in Shunyi district, during January to December 2005. Conclusion: Incidence rates of pulmonary tuberculosis displayed spatial and temporal clusterings at the township level in Beijing during 2005-2015, with high risk areas relatively concentrated in the central and southern parts of Beijing.
Beijing ; China ; Cluster Analysis ; Humans ; Incidence ; Spatial Analysis ; Spatio-Temporal Analysis ; Tuberculosis ; Tuberculosis, Pulmonary/ethnology*

OBJECTIVETo understand the spatial-temporal specific incidence of pulmonary tuberculosis (TB) in Gansu.

METHODSThe county-based incidence of pulmonary TB in Gansu from 2009 to 2013 was used to calculate Moran's I and G statistics, and analyze the spatial-temporal distribution of areas with different pulmonary TB incidences.

RESULTSThe spatial correlation in incidence of pulmonary TB was found in Gansu from 2009 to 2013 (P<0.001), and the hot spot areas were mainly in Hexi area, Linxia, part of Dingxi, while the cold spot areas were in Lanzhou, part of Dingxi, Tianshui, Pingliang and Qingyang. Spatial-temporal analysis showed that the clustering of high pulmonary TB incidence areas were most likely in the Hexi area during 2009-2010 (LLR=3,031.10, RR=2.27, P<0.001), and the clustering of low pulmonary TB incidence areas were most likely in Lanzhou during 2011-2013 (LLR=1,545.52, RR=0.37, P<0.001).

CONCLUSIONThe analysis on spatial and spatial-temporal specific incidences of pulmonary TB in Gansu from 2009 to 2013 indicated that Hexi area is the key area in pulmonary TB prevention and control in Gansu.

Biometry ; China ; epidemiology ; Cluster Analysis ; Humans ; Incidence ; Spatio-Temporal Analysis ; Tuberculosis, Pulmonary ; epidemiology

OBJECTIVETo evaluate four candidate variable number tandem repeat (VNTR) loci for genotyping Mycobacterium tuberculosis complex strains.

METHODSGenomic sequences for two M. tuberculosis strains (CCDC5079 and CCDC5180) were generated, and using published sequence data, four candidate VNTR loci were identified. The VNTRs were used to genotype 225 Chinese clinical M. tuberculosis complex strains. The discriminatory power of the VNTRs was evaluated using BioNumerics 5.0 software.

RESULTSThe Hunter-Gaston Index (HGI) for BJ1, BJ2, BJ3, and BJ4 loci was 0.634, 0.917, 0.697, and 0.910, respectively. Combining all four loci gave an HGI value of 0.995, thus confirming that the genotyping had good discriminatory power. The HGI values for BJ1, BJ2, BJ3, and BJ4, obtained from Beijing family strain genotyping, were 0.447, 0.878, 0.315, and 0.850, respectively. Combining all four loci produced an HGI value of 0.988 for genotyping the Beijing family strains. We observed unique patterns for M. bovis and M. africanum strains from the four loci.

CONCLUSIONWe have shown that the four VNTR loci can be successfully used for genotyping M. tuberculosis complex strains. Notably, these new loci may provide additional information about Chinese M. tuberculosis isolates than that currently afforded by established VNTR loci typing.

Cluster Analysis ; Genotyping Techniques ; Humans ; Minisatellite Repeats ; Mycobacterium bovis ; genetics ; Mycobacterium tuberculosis ; genetics

OBJECTIVETo investigate the spatial distribution characteristics of pulmonary tuberculosis in Zhejinag province in 2006 by G statistics, so as to find out the hotspot of occurrence of pulmonary tuberculosis in Zhejiang province and provide evidence for control and prevention on pulmonary tuberculosis.

METHODSThe data of pulmonary tuberculosis cases in 90 counties (districts) in Zhejiang province in 2006 were obtained from the tuberculosis surveillance system. Based on the same county field in digital maps and pulmonary tuberculosis incidence database, digital map of Zhejiang province was interrelated with the database of Zhejiang province pulmonary tuberculosis incidence to establish Zhejiang geographic information system database. General and local G statistics were developed to test for spatial heterogeneity by ArcGIS 9.2 software.

RESULTSA total of 43 467 cases of pulmonary tuberculosis were reported in Zhejiang province in 2006, and the reported incidence was 88.74/100,000.G statistics indicated that there were high occurrence of pulmonary tuberculosis (Getis-Ord Gi=0.0764, P<0.05). Local Getis-Ord Gi statistics analysis showed that there were statistically significant hotspots in Yuhuan county, Leqing city, Pingyang county, Dongtou county, Yongjia county, Ruian city and Ouhai, Longwan, Lucheng district in Wenzhou city, other areas were intergradational zone with Z(Gi) value fell in -1.96 to 1.96.

CONCLUSIONPulmonary tuberculosis in Zhejiang province present unrandomly distributed and geographically clustered.

China ; epidemiology ; Cluster Analysis ; Geographic Information Systems ; Geography ; Humans ; Incidence ; Models, Statistical ; Tuberculosis, Pulmonary ; epidemiology

Objective: To analyze the trend of epidemiological characteristics and spatiotemporal distribution of pulmonary tuberculosis (PTB) among smear-positive or other types of students in Guizhou Province from 2011 to 2020, and to provide a reference for improving prevention and control measures. Methods: Data were collected from the Chinese Information System's Notifiable Disease and Tuberculosis Management Information System for disease control and prevention, the Joinpoint 4.9.1.0 software was used to analyze the trend of registration rate; the ArcGIS 10.6 software was used to construct a ring map and to perform spatial autocorrelation analysis; the SaTScan 9.7 software was used for spatial-temporal scan statistics. Results: A total of 32 682 student PTB cases were reported in Guizhou Province from 2011 to 2020, including 5 949 (18.20%) smear-positive cases. Most cases occurred from high school students of 16 to 18 years old (43.99%, 14 376/32 682); the annual average registered rate was 36.22/100 000, the highest in 2018 (52.90/100 000), and the registration rate showed an increasing trend. Meanwhile, a similar trend of registration rate was observed among smear-positive or other types of students. The spatialtemporal heterogeneity was found that the "high-high" clustering patterns of smear-positive or other types were aggregated in Bijie City. Six spatialtemporal clusters with statistically significant (all P<0.001) were detected among smear-positive or other cases, respectively. Conclusions: Upward trend with spatial- temporal clusters of PTB cases reported in students from Guizhou Province from 2011 to 2020. Surveillance should be strengthened for high school students, and regular screening should be conducted in high-risk areas to control the source of infection and reduce the risk of transmission.
Humans ; Adolescent ; Tuberculosis, Pulmonary/epidemiology* ; Asian People ; Cluster Analysis ; Software ; Students

OBJECTIVEWe determined the genetic diversity of Mycobacterium tuberculosis (MTB) in a remote mountainous area of southwest China and evaluated the resolving ability of single nucleotide polymorphism (SNP) genotyping combined with variable number tandem repeat (VNTR) genotyping for Beijing family strains in association with drug resistance status.

METHODSThree hundred thirty-one MTB strains were isolated from patients living in mountainous regions of southwest China, and 8-loci SNP, VNTR-15 genotyping assays, and drug susceptibility testing of 9 drugs were performed.

RESULTSA total of 183 [55.29% (183/331)] strains were classified into the Beijing family. Of the 183 strains, 111 (60.66%) were defined as modern Beijing strains. The most predominant modern Beijing sub-lineage and ancient Beijing sub-lineage were Bmyc10 [39.34% (72/183)] and Bmyc25 [20.77% (38/183)], respectively. Of the isolates, 19.64% (65/331) were resistant to at least 1 of the 9 anti-TB drugs and 17 [4.98% (17/331)] MTB isolates were multi-drug resistant tuberculosis (MDR-TB). Two hundred sixty-one isolates showed a clustering rate of 14.18% (37/261) and a discriminatory index of 0.9990. The Beijing lineage exhibited a significantly higher prevalence of MDR-TB, as well as resistance to isoniazid (INH), rifampin (RIF), and para-aminosalicylic acid (PAS) when analyzed independently (P = 0.005, P = 0.017, P = 0.014, and P = 0.006 respectively). The Beijing lineage was not associated with genetic clustering or resistance to any drug. In addition, genetic clustering was not associated with drug resistance.

CONCLUSIONMTB strains demonstrate high genetic diversity in remote mountainous areas of southwest China. Beijing strains, especially modern Beijing strains, are predominant in remote mountainous area of China. The combination of 8-loci SNPs and VNTR-15 genotyping is a useful tool to study the molecular epidemiology of MTB strains in this area.

Antitubercular Agents ; pharmacology ; China ; epidemiology ; Cluster Analysis ; Drug Resistance, Bacterial ; Genotype ; Humans ; Mycobacterium tuberculosis ; drug effects ; genetics ; Phylogeny ; Polymorphism, Single Nucleotide ; Tuberculosis ; epidemiology ; microbiology