RESUMO
En la actualidad, los análisis de distribución espacial mediante el uso de técnicas de clusters para enfermedades crónicas como el cáncer de mama, son relevantes para la identificación de patrones espaciales de la mortalidad por cáncer según áreas geográficas. Identificar clústeres espaciales de la mortalidad por cáncer de mama en mujeres a nivel de las provincias del Ecuador, entre 2004 al 2018. Estudio observacional, de tipo descriptivo, ecológico multigrupal que compara a nivel espacio temporal las tasas de mortalidad por cáncer de mama en mujeres según las provincias del Ecuador, utilizando el índice de Móran para el análisis de autocorrelación y el algoritmo de k-medias para el análisis de agrupamiento en períodos quinquenales mediante el programa informático ArcGIS versión 10.5. Resultados. En el Ecuador, el 86,5% de las muertes por cáncer de mama en mujeres se registraron en el área urbana, dichas muertes tienen un patrón no aleatorio según el índice de Morán, distinto al área rural que tiene un patrón aleatorio; se identificó diferencia en el agrupamiento de la mortalidad por cáncer de mama en las provincias urbanas y rurales, donde se obtuvo para el área urbana, clústeres con altas, media-altas, media-baja y bajas tasas de mortalidad, mientras que en lo rural se obtuvieron solo clústeres con altas, medias y bajas tasas de mortalidad. La distribución espacial y el análisis de agrupamiento identificó clústeres de la mortalidad por cáncer de mama en el Ecuador, evidenciando entre lo urbano y rural diferencias en los clústeres obtenidos, siendo esta información de utilidad para la implementación de estrategias de control del cáncer en el país.
Currently spatial distribution analyzes through the use of cluster techniques for chronic diseases such as breast cancer are revealing for the identification of spatial patterns of cancer mortality according to geographic areas. Objective. Identify spatial clusters of breast cancer mortality in women at the level of the provinces of Ecuador, between 2004 to 2018. We used an observational, descriptive, ecological multigroup study that compares at a Spatio-temporal level the rates of breast cancer mortality in women according to the provinces of Ecuador, using the Moran index for the autocorrelation analysis and the k-, means algorithm for cluster analysis in five-year periods using the ArcGIS version 10.5 software. Results. In Ecuador, 86.5% of breast cancer deaths in women were recorded in the urban area, these deaths have a non-random pattern according to the Morán Index different from the rural area that has a random pattern; difference was identified in the grouping of breast cancer mortality in urban and rural provinces, where it was obtained for urban areas, clusters with high, medium. high, medium-low and low mortality rates. While in rural areas only clusters with high, medium and low mortality rates were obtained. Conclusions. The spatial distribution and cluster analysis identified clusters of breast cancer mortality in Ecuador; evidencing between urban and rural differences in the clusters obtained, this information is useful for the development of cancer control strategies in the country.
Assuntos
Humanos , Feminino , Neoplasias da Mama/mortalidade , Análise por Conglomerados , Zona Rural , Demografia , Área Urbana , Equador/epidemiologia , Análise EspacialRESUMO
Resumen Los sentidos humanos tienen una importante capacidad para detectar patrones espaciales, pero sus limitaciones son enormes, como lo han demostrado la psicología cognitiva y la Gestalt desde hace mucho tiempo. Por tanto, se requieren instrumentos más precisos y confiables para identificar patrones y actuar en consecuencia. El análisis espacial ofrece diversas alternativas para identificar patrones territoriales estimando su significancia estadística, con lo que se minimiza la posibilidad de percibir patrones ilusorios. Este trabajo utiliza desarrollos de punta en materia conceptual, metodológica y de tecnología para: a) Identificar con estadística espacial los clusters de inmuebles dañados por el sismo del 19S-2017 en la CDMX. La estrategia se basa en una secuencia de zooms a diversas escalas geográficas: desde la escala global para toda la CDMX, pasando por las escalas de delegación, colonia y manzanas, hasta llegar a la escala mínima de inmueble; b) localizar Unidades Móviles de Emergencia mediante modelos de localización-asignación, y c) comparar los patrones espaciales de inmuebles colapsados y dañados por los grandes sismos de 1985 y 2017. Los resultados de este trabajo podrán orientar los esfuerzos de reconstrucción, atención e investigación hacia zonas prioritarias espacial y estadísticamente significativas.
Abstract Our senses have an important capacity to detect spatial patterns, but their limitations are enormous, as Cognitive Psychology and Gestalt have shown for a long time. Therefore, more accurate and reliable instruments are required than our pure senses to identify patterns and act accordingly. Spatial analysis offers several alternatives to identify territorial patterns, estimating their statistical significance, minimizing the possibility of perceiving illusory patterns. This work uses cutting edge developments in conceptual, methodological and technology matters to: a) identify with spatial statistics the clusters of damaged buildings by the earthquake of 19S-2017 in Mexico City (CDMX). The strategy is based on a sequence of zooms at various geographic scales: from the global scale for the entire México City (CDMX), through delegation, neighborhood and block scales, until reaching the minimum scale: buildings; b) locate Emergency Mobile Units using location-allocation models; and c) compare the spatial patterns of collapsed and damaged buildings by the great earthquakes of 1985 and 2017. The results of this work may guide reconstruction, policy actions and research efforts towards spatially and statistically significant priority areas.
RESUMO
Objective To analyze the spatial and temporal distribution of smear positive pulmonary tuberculosis (PTB) in Liangshan Yi autonomous prefecture in Sichuan province from 2011 to 2016. Methods The registration data of PTB in 618 townships of Liangshan from 2011 to 2016 were collected from"Tuberculosis Management Information System of National Disease Prevention and Control Information System". Software ArcGIS 10.2 was used to establish the geographic information database and realize the visualization of the analysis results. Software OpenGeoda 1.2.0 was used to conduct the analyses on global indication of spatial autocorrelation (GISA) and local indication of spatial autocorrelation (LISA). Software SaTScan 9.4.1 was used for spatio-temporal scanning analysis. Results From 2011 to 2016, the registration rate of smear positive PTB in Liangshan declined from 56.97/100000 (2666 cases) to 21.11/100000 (1038 cases). The global spatial autocorrelation coefficient Moran's I ranged from 0.25 to 0.45 and the difference was significant (all P=0.000). Local autocorrelation analysis showed that"high-high"area covered 43, 34, 37, 34, 42 and 61 townships from 2011 to 2016, respectively, mainly in Leibo county. Spatial temporal clustering analysis found one class Ⅰ clustering in the area around Bagu township of Meigu county and two class Ⅱ clustering in the areas around Liumin and Hekou township of Huili county, respectively (all P=0.000). Conclusion Obvious spatial temporal clustering of smear positive PTB distribution was found in Liangshan from 2011-2016. Hot spot areas with serious smear positive PTB epidemic and high spread risk were mainly found in northeastern Liangshan, including townships in Leibo and Meigu counties. Targeted TB prevention and control should be conducted in these areas.
RESUMO
Objective To analyze the epidemiological characteristics of temporal-spatial distribution on varicella in Guangxi Zhuang Autonomous Region (Guangxi) during 2014 to 2016.Methods Incidence data on varicella was collected from the National Notifiable Infectious Disease Reporting Information System (NNIDRIS) of the Center for Disease Control and Prevention (CDC)while geographic information data was from the national CDC.ArcGIS 10.2 software was used to analyze global and local spatial auto correlation on spatial clusters.SaTScan v9.1.1 was used to conduct temporal-spatial scan for exploring the areas of temporal-spatial clusters.Results The overall incidence rates of varicella during 2014 to 2016 were 32.48/100 000,43.56/100 000 and 61.56/100 000 respectively.Incidence of varicella showed a positive spatial auto correlation at the county level (the value of Moran's I was between 0.24 to 0.35,P＜0.01),with consistent high morbidity.High-high cluster areas were seen and mainly concentrated in the north-western areas of Guangxi.Result from the temporal-spatial scan showed that temporal cluster of varicella occurred mainly between October and next January while the type I cluster area was mainly distributed in all of the counties in Hechi city and most counties of Baise city,with most counties being covered in the north-western areas of Guangxi,during 2014-2016.When comparing to data from the last two years,two type Ⅱ cluster areas with larger scales were formed in the north-eastern area of Guanyang county and Haicheng county of southem area in Guangxi,in 2016.Conclusions Incidence on Varicella seemed on the rise,and the distribution of cases showed clustered features,both on time and space.Strategies regarding control and prevention on Varicella should focus on high-high clustered areas,namely north-western areas of the province,including surrounding areas during the high onset season.
RESUMO
Objective To analyze the spatial and temporal distribution of smear positive pulmonary tuberculosis (PTB) in Liangshan Yi autonomous prefecture in Sichuan province from 2011 to 2016. Methods The registration data of PTB in 618 townships of Liangshan from 2011 to 2016 were collected from"Tuberculosis Management Information System of National Disease Prevention and Control Information System". Software ArcGIS 10.2 was used to establish the geographic information database and realize the visualization of the analysis results. Software OpenGeoda 1.2.0 was used to conduct the analyses on global indication of spatial autocorrelation (GISA) and local indication of spatial autocorrelation (LISA). Software SaTScan 9.4.1 was used for spatio-temporal scanning analysis. Results From 2011 to 2016, the registration rate of smear positive PTB in Liangshan declined from 56.97/100000 (2666 cases) to 21.11/100000 (1038 cases). The global spatial autocorrelation coefficient Moran's I ranged from 0.25 to 0.45 and the difference was significant (all P=0.000). Local autocorrelation analysis showed that"high-high"area covered 43, 34, 37, 34, 42 and 61 townships from 2011 to 2016, respectively, mainly in Leibo county. Spatial temporal clustering analysis found one class Ⅰ clustering in the area around Bagu township of Meigu county and two class Ⅱ clustering in the areas around Liumin and Hekou township of Huili county, respectively (all P=0.000). Conclusion Obvious spatial temporal clustering of smear positive PTB distribution was found in Liangshan from 2011-2016. Hot spot areas with serious smear positive PTB epidemic and high spread risk were mainly found in northeastern Liangshan, including townships in Leibo and Meigu counties. Targeted TB prevention and control should be conducted in these areas.
RESUMO
Objective To analyze the epidemiological characteristics of temporal-spatial distribution on varicella in Guangxi Zhuang Autonomous Region (Guangxi) during 2014 to 2016.Methods Incidence data on varicella was collected from the National Notifiable Infectious Disease Reporting Information System (NNIDRIS) of the Center for Disease Control and Prevention (CDC)while geographic information data was from the national CDC.ArcGIS 10.2 software was used to analyze global and local spatial auto correlation on spatial clusters.SaTScan v9.1.1 was used to conduct temporal-spatial scan for exploring the areas of temporal-spatial clusters.Results The overall incidence rates of varicella during 2014 to 2016 were 32.48/100 000,43.56/100 000 and 61.56/100 000 respectively.Incidence of varicella showed a positive spatial auto correlation at the county level (the value of Moran's I was between 0.24 to 0.35,P＜0.01),with consistent high morbidity.High-high cluster areas were seen and mainly concentrated in the north-western areas of Guangxi.Result from the temporal-spatial scan showed that temporal cluster of varicella occurred mainly between October and next January while the type I cluster area was mainly distributed in all of the counties in Hechi city and most counties of Baise city,with most counties being covered in the north-western areas of Guangxi,during 2014-2016.When comparing to data from the last two years,two type Ⅱ cluster areas with larger scales were formed in the north-eastern area of Guanyang county and Haicheng county of southem area in Guangxi,in 2016.Conclusions Incidence on Varicella seemed on the rise,and the distribution of cases showed clustered features,both on time and space.Strategies regarding control and prevention on Varicella should focus on high-high clustered areas,namely north-western areas of the province,including surrounding areas during the high onset season.