Geospatial Health最新文献

This article examines three spatiotemporal methods used for analyzing of infectious diseases, with a focus on COVID-19 in the United States. The methods considered include inverse distance weighting (IDW) interpolation, retrospective spatiotemporal scan statistics and Bayesian spatiotemporal models. The study covers a 12-month period from May 2020 to April 2021, including monthly data from 49 states or regions in the United States. The results show that the spread of COVID-19 pandemic increased rapidly to a high value in winter of 2020, followed by a brief decline that later reverted into another increase. Spatially, the COVID-19 epidemic in the United States exhibited a multi-centre, rapid spread character, with clustering areas represented by states such as New York, North Dakota, Texas and California. By demonstrating the applicability and limitations of different analytical tools in investigating the spatiotemporal dynamics of disease outbreaks, this study contributes to the broader field of epidemiology and helps improve strategies for responding to future major public health events.

This research aims to uncover how the association between social determinants of health and COVID-19 cases and fatality rate have changed across time and space. To begin to understand these associations and show the benefits of analysing temporal and spatial variations in COVID-19, we utilized Geographically Weighted Regression (GWR). The results emphasize the advantages for using GWR in data with a spatial component, while showing the changing spatiotemporal magnitude of association between a given social determinant and cases or fatalities. While previous research has demonstrated the merits of GWR for spatial epidemiology, our study fills a gap in the literature, by examining a suite of variables across time to reveal how the pandemic unfolded across the US at a county-level spatial scale. The results speak to the importance of understanding the local effects that a social determinant may have on populations at the county level. From a public health perspective, these results can be used for an understanding of the disproportionate disease burden felt by different populations, while upholding and building upon trends observed in epidemiological literature.

This research proposes a 'temporal attention' addition for long-short term memory (LSTM) models for dengue prediction. The number of monthly dengue cases was collected for each of five Malaysian states i.e. Selangor, Kelantan, Johor, Pulau Pinang, and Melaka from 2011 to 2016. Climatic, demographic, geographic and temporal attributes were used as covariates. The proposed LSTM models with temporal attention was compared with several benchmark models including a linear support vector machine (LSVM), a radial basis function support vector machine (RBFSVM), a decision tree (DT), a shallow neural network (SANN) and a deep neural network (D-ANN). In addition, experiments were conducted to analyze the impact of look-back settings on each model performance. The results showed that the attention LSTM (A-LSTM) model performed best, with the stacked, attention LSTM (SA-LSTM) one in second place. The LSTM and stacked LSTM (S-LSTM) models performed almost identically but with the accuracy improved by the attention mechanism was added. Indeed, they were both found to be superior to the benchmark models mentioned above. The best results were obtained when all attributes were included in the model. The four models (LSTM, S-LSTM, A-LSTM and SA-LSTM) were able to accurately predict dengue presence 1-6 months ahead. Our findings provide a more accurate dengue prediction model than previously used, with the prospect of also applying this approach in other geographic areas.

Coronary heart disease is a non-communicable disease whose treatment is closely related to infrastructure, such as diagnostic imaging equipment visualizing arteries and chambers of the heart (cath lab) and infrastructure that supports access to healthcare. This research is intended as a preliminary geospatial study to carry out initial measurements of health facility coverage at the regional level, survey available supporting data and provide input on problems in future research. Data on cath lab presence was gathered through direct survey, while population data was taken from an open-source geospatial system. The cath lab service coverage was obtained by analysis based on a Geographical Information System (GIS) specific tool to evaluate travel time from the sub-district centre to the nearest cath lab facility. The number of cath labs in East Java has increased from 16 to 33 in the last six years and the 1-hour access time increased from 24.2% to 53.8%. However, accessibility remains a problem as16.5% of the total population of East Java cannot access a cath lab even within 2 hours. Thus, additional cath lab facilities are required to provide ideal healthcare coverage. Geospatial analysis is the tool to determine the optimal cath lab distribution.

A study of 2,569,617 Thailand citizens diagnosed with COVID-19 from January 2020 to March 2022 was conducted with the aim of identifying the spatial distribution pattern of incidence rate of COVID-19 during its five main waves in all 77 provinces of the country. Wave 4 had the highest incidence rate (9,007 cases per 100,000) followed by the Wave 5, with 8,460 cases per 100,000. We also determined the spatial autocorrelation between a set of five demographic and health care factors and the spread of the infection within the provinces using Local Indicators of Spatial Association (LISA) and univariate and bivariate analysis with Moran's I. The spatial autocorrelation between the variables examined and the incidence rates was particularly strong during the waves 3-5. All findings confirmed the existence of spatial autocorrelation and heterogenicity of COVID-19 with the distribution of cases with respect to one or several of the five factors examined. The study identified significant spatial autocorrelation with regard to the COVID-19 incidence rate with these variables in all five waves. Depending on which province that was investigated, strong spatial autocorrelation of the High-High pattern was observed in 3 to 9 clusters and of the Low-Low pattern in 4 to 17 clusters, whereas negative spatial autocorrelation was observed in 1 to 9 clusters of the High-Low pattern and in 1 to 6 clusters of Low-High pattern. These spatial data should support stakeholders and policymakers in their efforts to prevent, control, monitor and evaluate the multidimensional determinants of the COVID-19 pandemic.

Positive and negative economic growth is closely related to the suicide rate. To answer the question whether economic development has a dynamic impact on this rate, we used a panel smooth transition autoregressive model to evaluate the threshold effect of economic growth rate on the persistence of suicide. The research period was from 1994 to 2020, and the results show that the suicide rate had a persistent effect, which varied over time depending on the transition variable within different threshold intervals. However, the persistent effect was manifested in different degrees with the change in the economic growth rate and as the lag period of the suicide rate increased, the effect of the influence gradually decreased. We investigated different lag periods and noted that the impact on the suicide rate was the strongest in the first year after an economic change and then reduced to be only marginal after three years. This means that the growth momentum of the suicide rate within the first two years after a change in the economic growth rate, should be included in policy considerations of how to prevent suicides.

Pulmonary tuberculosis (PTB) remains a serious public health problem, especially in areas of developing countries. This study aimed to explore the spatial-temporal clusters and associated risk factors of PTB in south-western China. Space-time scan statistics were used to explore the spatial and temporal distribution characteristics of PTB. We collected data on PTB, population, geographic information and possible influencing factors (average temperature, average rainfall, average altitude, planting area of crops and population density) from 11 towns in Mengzi, a prefecture-level city in China, between 1 January 2015 and 31 December 2019. A total of 901 reported PTB cases were collected in the study area and a spatial lag model was conducted to analyse the association between these variables and the PTB incidence. Kulldorff's scan results identified two significant space-time clusters, with the most likely cluster (RR = 2.24, p < 0.001) mainly located in northeastern Mengzi involving five towns in the time frame June 2017 - November 2019. A secondary cluster (RR = 2.09, p < 0.05) was located in southern Mengzi, covering two towns and persisting from July 2017 to December 2019. The results of the spatial lag model showed that average rainfall was associated with PTB incidence. Precautions and protective measures should be strengthened in high-risk areas to avoid spread of the disease.

The rapid increase in out-of-pocket expenditures regressively raises the issue of equity in medical access opportunities according to income class and negatively affects public health. Factors related to out-of-pocket expenses have been analyzed in previous studies using an ordinary regression model (Ordinary Least Squares [OLS]). However, as OLS assumes equal error variance, it does not consider spatial variation due to spatial heterogeneity and dependence. Accordingly, this study presents a spatial analysis of outpatient out-of-pocket expenses from 2015 to 2020, targeting 237 local governments nationwide, excluding islands and island regions. R (version 4.1.1) was used for statistical analysis, and QGIS (version 3.10.9), GWR4 (version 4.0.9), and Geoda (version 1.20.0.10) were used for the spatial analysis. As a result, in OLS, it was found that the aging rate and number of general hospitals, clinics, public health centers, and beds had a positive (+) significant effect on outpatient out-of-pocket expenses. The Geographically Weighted Regression (GWR) suggests regional differences exist concerning out-of-pocket payments. As a result of comparing the OLS and GWR models through the Adj. R² and Akaike's Information Criterion indices, the GWR model showed a higher fit. This study provides public health professionals and policymakers with insights that could inform effective regional strategies for appropriate out-of-pocket cost management.

Antimicrobial resistance (AMR) is a global major health concern. Spatial analysis is considered an invaluable method in health studies. Therefore, we explored the usage of spatial analysis in Geographic Information Systems (GIS) in studies on AMR in the environment. This systematic review is based on database searches, a content analysis, ranking of the included studies according to the preference ranking organization method for enrichment evaluations (PROMETHEE) and estimation of data points per km2. Initial database searches resulted in 524 records after removal of duplicates. After the last stage of full text screening, 13 greatly heterogeneous articles with diverse study origins, methods and design remained. In the majority of studies, the data density was considerably less than one sampling site per km2 but exceeded 1,000 sites per km2 in one study. The results of the content analysis and ranking showed a variation between studies that primarily used spatial analysis and those that used spatial analysis as a sec ondary method. We identified two distinct groups of GIS methods. The first was focused on sample collection and laboratory testing, with GIS as supporting method. The second group used overlay analysis as the primary method to combine datasets in a map. In one case, both methods were combined. The low number of articles that met our inclusion criteria highlights a research gap. Based on the findings of this study we encourage application of GIS to its full potential in studies of AMR in the environment.

COVID-19 is the most severe health crisis of the 21st century. COVID-19 presents a threat to almost all countries worldwide. The restriction of human mobility is one of the strategies used to control the transmission of COVID-19. However, it has yet to be determined how effective this restriction is in controlling the rise in COVID-19 cases, particularly in small areas. Using Facebook's mobility data, our study explores the impact of restricting human mobility on COVID-19 cases in several small districts in Jakarta, Indonesia. Our main contribution is showing how the restriction of human mobility data can give important information about how COVID-19 spreads in different small areas. We proposed modifying a global regression model into a local regression model by accounting for the spatial and temporal interdependence of COVID-19 transmission across space and time. We applied Bayesian hierarchical Poisson spatiotemporal models with spatially varying regression coefficients to account for non-stationarity in human mobility. We estimated the regression parameters using an Integrated Nested Laplace Approximation. We found that the local regression model with spatially varying regression coefficients outperforms the global regression model based on DIC, WAIC, MPL, and R2 criteria for model selection. In Jakarta's 44 districts, the impact of human mobility varies significantly. The impacts of human mobility on the log relative risk of COVID-19 range from -4.445 to 2.353. The prevention strategy involving the restriction of human mobility may be beneficial in some districts but ineffective in others. Therefore, a cost-effective strategy had to be adopted.