Increasing access to and utilization of long-acting reversible contraceptives (LARC) can prevent unintended pregnancies and reduce unmet need for family planning in Ethiopia However, LARC uptake lags behind less effective contraceptive methods. This study aimed to analyze the geographical distribution and demographic factors associated to LARC uptake.

Antibiotic Self-Medication (ASM) is a major contributing factor to Antimicrobial Resistance (AMR) that can lead to both mortality and long-term hospitalizations. High provincial ASM proportions associated with mortality due to AMR have been observed in Thailand but there is a lack of studies on geographic factors contributing to ASM. The present study aimed to quantify the distribution of ASM in Thailand and its correlated factors. Socioeconomic and health services factors were included in the spatial analysis. Moran's I was performed to identify global autocorrelation with the significance level set at p=0.05 and spatial regression were applied to identify the factors associated with ASM, the proportion of which is predominant in the north-eastern, central and eastern regions with Phitsanulok Province reporting the highest proportion of Thailand's 77 provinces. Autocorrelation between Night-Time Light (NTL) and the proportion of ASM was observed to be statistically significant at p=0.030. The Spatial Lag Model (SLM) and the Spatial Error Model (SEM) were used with the latter providing both the lowest R2 and Akaike Information Criterion (AIC). It was demonstrated that the proportion of alcohol consumption significantly increased the proportion of ASM. The annual number of outpatient department visits and the average NTL decreased the proportion of ASM by 1.5% and 0.4%, respectively. Average monthly household expenditures also decreased the ASM proportion. Policies to control alcohol consumption while promoting healthcare visits are essential strategies to mitigate the burden of AMR in Thailand.

This study aimed to estimate a socio-spatial vulnerability index for type 2 diabetes mellitus (T2DM) at the municipal level in Mexico for 2020. It incorporated factors such as poverty, social backwardness, marginalization index, and human development index. This retrospective ecological study analyzed 317,011 incident cases of T2DM in 2020. Utilizing multi-criteria decision analysis, weighted values were assigned to each vulnerability criterion. A multiple linear regression model was developed, complemented by cluster and outlier analyses using Moran I's and the high-low clustering method. A clustered spatial autocorrelation of high values was found across 17.65% of Mexico, which was statistically significant (p < 0.001). Conversely, 37.78% of the territory showed a pattern of low values without significant evidence of groupings. The analysis revealed 117 nodes of very high vulnerability forming six focal areas, 172 nodes with high vulnerability across five areas, 168 nodes with medium vulnerability in two areas, 112 nodes with low vulnerability across 16 areas, and 152 nodes with very low vulnerability in 24 focal areas. This method proves to be robust and offers a technical-scientific basis for guiding T2DM prevention strategies and actions using a spatial/epidemiological approach. It is recommended that future strategies take into account factors such as poverty, social backwardness, marginalization index, and human development index to be effective.

This study aimed to analyze the epidemiological characteristics and spatiotemporal clustering of hepatitis A in mainland China from 2004 to 2019 and to evaluate the practical impact of integrating hepatitis A vaccines into the Expanded Program on Immunization (EPI). Spatial and temporal autocorrelation and spatiotemporal scanning statistics were used to perform spatial and temporal characterization to quantify the spatial similarity or degree of aggregation of geographic data, and Geographical and Temporal Weighted Regression (GTWR) models were used to reveal spatial and temporal heterogeneity in the relationships between variables to test for spatial and temporal outbreaks of disease and other factors, such as socio-economic factors. Spatially, the incidence rates exhibited a west-high and east-low spatial differentiation, with the High-High (HH) clusters predominantly located in the western regions, maintaining stability butgradually diminishing. Hepatitis A prevalence peaked during the initial study period (2004-2008) showing significant spatial clustering. However, since the inclusion of hepatitis A vaccine in the immunization program in 2008, the incidence rates of hepatitis A in mainland China significantly decreased demonstrating the positive impact of immunization strategies. In addition to the effects of vaccination, socio-economic factors such as education level, water resources and age groups showed significant associations with hepatitis A incidence rates. Increased vaccine coverage and improved social conditions are crucial for controlling hepatitis A in China.

Geographical accessibility is important for promoting health equity, and calculating it requires the locations of all existing healthcare facilities in a region. Authoritative location data collected by governments is accurate but mostly not publicly available, while point-of-interest (POI) data from online sources, such as Baidu Maps and AutoNavi Maps are easily accessible. However, the accuracy of the latter has not been thoroughly analyzed. Taking Baotou, a medium-sized city in China, as aneample, we assessed the suitability of using POI data for measuring geographic accessibility to healthcare facilities.We computedthe difference of geographic accessibility calculated based on POI data and that on authoritative data.Logistic regression and a multiple linear regression model was applied to identify factors related to the consistency between the two data sources. Compared to authoritative data, POI data exhibited discrepancies, with completeness of 54.9% and accuracy of 63.7%. Geographic accessibility calculated based on both data showed similar patterns, with good consistency for hospitals and in urban areas. However, large differences (>30 minutes) were shown in rural areas for primary healthcare facilities. The differences were small regarding to population- weighted average accessibility (with slight underestimation of 3.07 minutes) and population coverage across various levels of accessibility (with differences less than 1% of the population) for the entire area. In conclusion, POI data can be considered foruse in both urban areas and at the level of entire city; however, awareness should be raised in rural areas.

Objects for Targeted Surveillance (OTS) are infrastructure entities that can be considered as focal points and conduits for transmitting infectious animal diseases, necessitating ongoing epidemiological surveillance. These entities encompass slaughterhouses, meat processing plants, animal markets, burial sites, veterinary laboratories, etc. Currently, in Kazakhstan, a funded research project is underway to establish a Geographic Information System (GIS) database of OTSs and investigate their role in the emergence and dissemination of infectious livestock diseases. This initial investigation examined the correlation between brucellosis outbreaks in cattle and small ruminant farms in the southeastern region of Kazakhstan and the presence of OTSs categorized as "slaughterhouses," "cattle markets," and "meat processing plants. The study area (namely Qyzylorda, Turkestan, Zhambyl, Almaty, Zhetysu, Abay and East Kazakhstan oblasts), characterized by the highest livestock density in the country, covers 335 slaughterhouses (with varying levels of biosecurity), 45 livestock markets and 15 meat processing plants. Between 2020 and 2023, 338 cases of brucellosis were reported from livestock farms in this region. The findings of the regression model reveal a statistically significant (p<0.05) positive association between the incidence of brucellosis cases and the number of OTSs in the region. Conversely, meat processing plants and livestock markets did not exhibit a significant influence on the prevalence of brucellosis cases. These results corroborate the hypothesis of an elevated risk of brucellosis transmission in regions with slaughterhouses, likely attributable to increased animal movements within and across regions, interactions with vehicles and contact with slaughterhouse staff. These outcomes mark a pivotal advancement in the national agricultural development agenda. The research will be extended to encompass the entire country, compiling a comprehensive OTS database.

Access to healthcare is influenced by various socioeconomic factors such as income, population group, educational attainment and health insurance. This study used Geographically Weighted Regression (GWR) to investigate spatial variations in the association between socioeconomic factors and access to public healthcare facilities in the City of Tshwane, South Africa based on data from the Gauteng City-Region Observatory Quality of Life Survey (2020/2021). Socioeconomic predictors included population group, income, health insurance status and health satisfaction. The GWR model revealed that all socioeconomic factors combined explained the variation in access to healthcare facilities (R²=0.77). Deviance residuals, ranging from -2.67 to 1.83, demonstrated a good model fit, indicating the robustness of the GWR model in predicting access to healthcare facilities. Black African, low-income and uninsured populations had each a relatively strong association with access to healthcare facilities (R²=0.65). Additionally, spatial patterns revealed that socioeconomic relationships with access to health care facilities are not homogeneous, with significance of the relationships varying with space. This study highlights the need for a spatially nuanced approach to improving healthcare facilities access and emphasizes the need for targeted policy interventions that address local socio-environmental conditions.

Disease surveillance remains important for early detection of new COVID-19 variants. For this purpose, the World Health Organization (WHO) recommends integrating of COVID-19 surveillance with other respiratory diseases. This requires knowledge of areas with elevated risk, which in developing countries is lacking from the routine analyses. Focusing on Ghana, this study employed scan-statistic cluster analysis to uncover the spatial patterns of incidence and Case Fatality Rates (CFR) of COVID-19 based on reports covering the four pandemic waves in Ghana between 12 March 2020 and 28 February 2022. Applying flexible spatial scan statistic with restricted likelihood ratio, we examined the incidence and CFR clusters before and after adjustment for covariates. We used distance to the epicentre, proportion of the population aged ≥ 65, male proportion of the population and urban proportion of the population as the covariates. We identified 56 significant spatial clusters for incidence and 26 for CFR for all four waves of the pandemic. The Most Likely Clusters (MLCs) of incidence occurred in the districts in south-eastern Ghana, while the CFR ones occurred in districts in the central and the northeastern parts of the country. These districts could serve as sites for sentinel or genomic surveillance. Spatial relationships were identified between COVID-19 incidence covariates and the CFR. We observed closeness to the epicentre and high proportions of urban populations increased COVID-19 incidence, whiles high proportions of those aged ≥ 65 years increased the CFR. Accounting for the covariates resulted in changes in the distribution of the clusters. Both incidence and CFR due to COVID-19 were spatially clustered, and these clusters were affected by high proportions of the urban population, high proportions of the male population, high proportions of the population aged ≥ 65 years and closeness to the epicentre. Surveillance should target districts with elevated risk. Long-term control measures for COVID-19 and other contagious diseases should consider improving quality healthcare access and measures to reduce growth rates of urban populations.

This study used data from the second Nigeria Malaria Indicator Survey (NMIS) conducted in 2015 to investigate the spatial distribution of malaria prevalence in the country and identify its associated factors. Nigeria is divided into 36 states with 109 senatorial districts, most of which are affected by malaria, a major cause of morbidity and mortality in children under five years of age. We carried out an ecological study with analysis at the senatorial district level. A malaria prevalence map was produced combining geographic information systems data from the Nigeria Malaria Indicator Survey (NMIS) of 2015 with shape files from an open data-sharing platform. Spatial autoregressive models were fitted using a set of key covariates. Malaria prevalence in children under-five was highest in Kebbi South senatorial district (70.6%). It was found that poorest wealth index (β = 0.10 (95% CI: 0.01, 0.20), p = 0.04), mothers having only secondary level of education (β = 0.78 (95% CI: 0.05, 1.51), p = 0.04) and households without mosquito bed nets (β = 0.21 (95% CI: 0.02, 0.39), p = 0.03) were all significantly associated with higher malaria prevalence. Moran's I (54.81, p<0.001) showed spatial dependence of malaria prevalence across contiguous districts and spatial autoregressive modelling demonstrated significant spill-over effect of malaria prevalence. Maps produced in this study provide a useful graphical representation of the spatial distribution of malaria prevalence based on NMIS-2015 data. Clustering of malaria prevalence in certain areas further highlights the need for sustained malaria elimination interventions across affected regions in order to break the chain of transmission.

Chronic Obstructive Pulmonary Disease (COPD) is a significant global health issue, leading to high rates of sickness and death worldwide. In Thailand, there are over 3 million patients with the COPD, with more than a million patients admitted to hospitals due to symptoms of the disease. This study investigated factors influencing the incidence of acute exacerbations among COPD patients in Thailand, including the spatial autocorrelation between socioeconomic and environmental factors. We conducted a spatial analysis using Moran's I, Local Indicators Of Spatial Association (LISA), and spatial regression models, specifically the Spatial Lag Model (SLM) and the Spatial Error Model (SEM), to explore the relationships between the variables. The univariate Moran's I scatter plots showed a significant positive spatial autocorrelation of 0.606 in the incidence rate of COPD among individuals aged 15 years and older across all 77 provinces in Thailand. High-High (HH) clusters for the COPD were observed in the northern and southern regions, while Low-Low (LL) clusters were observed in the northern and north-eastern regions. Bivariate Moran's I indicated a spatial autocorrelation between various factors and acute exacerbation of COPD in Thailand. LISA analysis revealed 4 HH clusters and 5 LL clusters related to average income, 12 HH and 8 LL clusters in areas where many people smoke, 5 HH and 8 LL clusters in areas with industrial factory activities, 11 HH and 9 LL clusters associated with forested areas, and 6 LL clusters associated with the average rice field. Based on the Akaike information criterion (AIC). The SLM outperformed the SEM but only slightly so, with an AIC value of 1014.29 compared to 1019.56 and a Lagrange multiplier value of p<0.001. However, it did explain approximately 63.9% of the incidence of acute exacerbations of COPD, with a coefficient of determination (R² = 0.6394) along with a Rho (ρ) of 0.4164. The results revealed that several factors, including income, smoking, industrial surroundings, forested areas and rice fields are associated with increased levels of acute COPD exacerbations.

According to World Trade Organization (WTO) statistics, the incidence of seasonal influenza in China has been on the rise since 2018. The aim of this study was to identify and investigate the influence of factors related to the incidence of four common types of influenza viruses. Data of patients with common cold and associated virus infections are described, and a logistic regression model based on gender, age and season was established. The relationship between virus type and the above three factors was analyzed in depth and significant (p<0.05) associations noted. We noted a fluctuation trend, with the infection rate of influenza virus showing an upward trend from 2018 to 2019 and from 2021 to 2022 and a downward trend from 2019 to 2021. The total number of cases in adolescents aged 18-30 years was higher than that in the elderly. The impact of different types of influenza virus on the population ranked from large to small, with special roles played by Influenza B/Victoria, H3N2, Influenza A/H1N1 pdm and Influenza B/Yamagata.

Cartography, or geographical visualization of disease is an essential aspect of the field of GeoHealth, yet there is limited guidance on the visualization of spatiotemporal disease maps. In order to adequately contribute to understanding disease outbreaks, disease maps should be crafted carefully and according to relevant cartographic guidelines. This article aims to increase the understanding of space-time visualization techniques that are relevant to the field of GeoHealth, by providing a step-by-step framework for the creation of space-time disease visualizations. This study introduces a systematic approach to spatiotemporal disease mapping by integrating operations from the Generalized Space Time Cube (GSTC) Framework with established cartographic symbology guidelines. This resulted in an overview table that contains both the relevant GSTC operations and cartographic guidelines, as well as a step-by-step procedure that guides users through the process of creating informative spatiotemporal disease maps. The practical application of this step-by-step procedure is demonstrated with an example using Dutch COVID-19 data. By providing a clear, practical step by step procedure, this study enhances the capacity of public health professionals, policymakers, and researchers to monitor, understand, and respond to the spatial and temporal dynamics of diseases.

As extreme weather events increase in frequency and intensity, the health system faces significant challenges, not only from shifting patterns of climate-sensitive diseases but also from disruptions to healthcare infrastructure, supply chains and the physical systems essential for delivering care. This necessitates the strategic use of geospatial tools to guide the delivery of healthcare services and make evidence-informed priorities, especially in contexts with scarce human and financial resources. In this article, we highlight several published papers that have been used throughout the phases of the disaster management cycle in relation to health service delivery. We complement the findings from these publications with a rapid scoping review to present the body of knowledge for using spatial methods for health service delivery in the context of disasters. The main aim of this article is to demonstrate the benefits and discuss the challenges associated with the use of geospatial methods throughout the disaster management cycle. Our scoping review identified 48 articles employing geospatial techniques in the disaster management cycle. Most of them focused on geospatial tools employed for preparedness, anticipatory action and mitigation, particularly for targeted health service delivery. We note that while geospatial data analytics are effectively deployed throughout the different phases of disaster management, important challenges remain, such as ensuring timely availability of geospatial data during disasters, developing standardized and structured data formats, securing pre-disaster data for disaster preparedness, addressing gaps in health incidence data, reducing underreporting of cases and overcoming limitations in spatial and temporal coverage and granularity. Overall, existing and novel geospatial methods can bridge specific evidence gaps in all phases of the disaster management cycle. Improvement and 'operationalization' of these methods can provide opportunities for more evidence-informed decision making in responding to health crises during climate change.

In order to effectively cope with the situation caused by the COVID-19 pandemic, cases should be concentrated in designated medical institutions with full capability to deal with patients infected by this virus. We studied the location of such hospitals dividing the patients into two categories: ordinary and severe. Genetic algorithms were constructed to achieve a three-phase dynamic approach for the location of hospitals designated to receive and treat COVID-19 cases based on the goal of minimizing the cost of construction and operation isolation wards as well as the transportation costs involved. A dynamic location model was established with the decision variables of the corresponding 'chromosome' of the genetic algorithms designed so that this goal could be reached. In the static location model, 15 hospitals were required throughout the treatment cycle, whereas the dynamic location model found a requirement of only 11 hospitals. It further showed that hospital construction costs can be reduced by approximately 13.7% and operational costs by approximately 26.7%. A comparison of the genetic algorithm and the Gurobi optimizer gave the genetic algorithm several advantages, such as great convergence and high operational efficiency.

Stunting continues to be a significant health issue, particularly in developing nations, with Indonesia ranking third in prevalence in Southeast Asia. This research examined the risk of stunting and influencing factors in Indonesia by implementing various Bayesian spatial conditional autoregressive (CAR) models that include covariates. A total of 750 models were run, including five different Bayesian spatial CAR models (Besag-York-Mollie (BYM), CAR Leroux and three forms of localised CAR), with 30 covariate combinations and five different hyperprior combinations for each model. The Poisson distribution was employed to model the counts of stunting cases. After a comprehensive evaluation of all model selection criteria utilized, the Bayesian localised CAR model with three covariates were preferred, either allowing up to 2 clusters with a variance hyperprior of inverse-gamma (1, 0.1) or allowing 3 clusters with a variance hyperprior of inverse-gamma (1, 0.01). Poverty and recent low birth weight (LBW) births are significantly associated with an increased risk of stunting, whereas child diet diversity is inversely related to the risk of stunting. Model results indicated that Sulawesi Barat Province has the highest risk of stunting, with DKI Jakarta Province the lowest. These areas with high stunting require interventions to reduce poverty, LBW births and increase child diet diversity.

An expert panel discussion on achievements, current areas of rapid scientific progress, prospects, and critical gaps in geospatial health was organized as part of the 16thsymposium of the global network of public health and earth scientists dedicated to the development of geospatial health (GnosisGIS), held at the Faculty of Geo-Information Science and Earth Observation (ITC) of the University of Twente in The Netherlands in November 2023. The symposium consisted of a three-day scientific event that brought together an interdisciplinary group of researchers and health professionals from across the globe. The aim of the panel session was threefold: firstly, to reflect on the main achievements of the scientific discipline of geospatial health in the past decade; secondly, to identify key innovation areas where rapid scientific progress is currently made and thirdly, to identify critical gaps and associated research and education priorities to move the discipline forward. [...].

The purpose of this study was to determine whether there were any TB clusters in Aceh Province, Indonesia and their temporal distribution during the period of 2019-2021. A spatial geo-reference was conducted to 290 sub-districts coordinates by geocoding each sub-district's offices. By using SaTScan TM v9.4.4, a retrospective space-time scan statistics analysis based on population data and annual TB incidence was carried out. To determine the regions at high risk of TB, data from 1 January 2019 to 31 December 2021 were evaluated using the Poisson model. The likelihood ratio (LLR) value was utilized to locate the TB clusters based on a total of 999 permutations were performed. A Moran's I analysis (using GeoDa) was chosen for a study of both local and global spatial autocorrelation. The threshold for significance was fixed at 0.05. At the sub-district level, the spatial distribution of TB in Aceh Province from 2019-2021 showed 19 clusters (three most likely and 16 secondary ones), and there was a spatial autocorrelation of TB. The findings can be used to provide thorough knowledge on the spatial pattern of TB occurrence, which is important for designing effective TB interventions.

Cardiovascular Disease (CVD) is currently the major challenge to people's health and the world's top cause of death. In Tanzania, deaths due to CVD account for about 13% of the total deaths caused by the non-communicable diseases. This study examined the spatio-temporal clustering of CVDs from 2010 to 2019 in Tanzania for retrospective spatio-temporal analysis using the Bernoulli probability model on data sampled from four selected hospitals. Spatial scan statistics was performed to identify CVD clusters and the effect of covariates on the CVD incidences was examined using multiple logistic regression. It was found that there was a comparatively high risk of CVD during 2011-2015 followed by a decline during 2015-2019. The spatio-temporal analysis detected two high-risk disease clusters in the coastal and lake zones from 2012 to 2016 (p<0.001), with similar results produced by purely spatial analysis. The multiple logistic model showed that sex, age, blood pressure, body mass index (BMI), alcohol intake and smoking were significant predictors of CVD incidence.

Spatial cluster analyses of health events are useful for enabling targeted interventions. Spatial scan statistic is the stateof- the-art method for this kind of analysis and the Poisson Generalized Linear Model (GLM) approach to the spatial scan statistic can be used for count data for spatial cluster detection with covariate adjustment. However, its use for modelling is limited due to data over-dispersion. A Generalized Linear Mixed Model (GLMM) has recently been proposed for modelling this kind of over-dispersion by incorporating random effects to model area-specific intrinsic variation not explained by other covariates in the model. However, these random effects may exhibit a geographical correlation, which may lead to a potential spatial cluster being undetected. To handle the over-dispersion in the count data, this study aimed to evaluate the performance of a negative binomial- GLM in spatial scan statistic on real-world data of low birth weights in Khyber-Pakhtunkhwa Province, Pakistan, 2019. The results were compared with the Poisson-GLM and GLMM, showing that the negative binomial-GLM is an ideal choice for spatial scan statistic in the presence of over-dispersed data. With a covariate (maternal anaemia) adjustment, the negative binomial-GLMbased spatial scan statistic detected one significant cluster covering Dir lower district. Without the covariate adjustment, it detected two clusters, each covering one district. The district of Peshawar was seen as the most likely cluster and Battagram as the secondary cluster. However, none of the clusters were detected by GLMM spatial scan statistic, which might be due to the spatial correlation of the random effects in GLMM.

During the COVID-19 pandemic, a system was established in China that required testing of all residents for COVID-19. It consisted of sampling stations, laboratories capable of carrying out DNA investigations and vehicles carrying out immediate transfer of all samples from the former to the latter. Using Beilin District, Xi'an City, Shaanxi Province, China as example, we designed a genetic algorithm based on a two-stage location coverage model for the location of the sampling stations with regard to existing residencies as well as the transfer between the sampling stations and the laboratories. The aim was to estimate the minimum transportation costs between these units. In the first stage, the model considered demands for testing in residential areas, with the objective of minimizing the costs related to travel between residencies and sampling stations. In the second stage, this approach was extended to cover the location of the laboratories doing the DNAinvestigation, with the aim of minimizing the transportation costs between them and the sampling stations as well as the estimating the number of laboratories needed. Solutions were based on sampling stations and laboratories existing in 2022, with the results visualized by geographic information systems (GIS). The results show that the genetic algorithm designed in this paper had a better solution speed than the Gurobi algorithm. The convergence was better and the larger the network size, the more efficient the genetic algorithm solution time.

During the period 2013-2023, 917 cases of rabies among animals were registered in the Republic of Kazakhstan. Out of these, the number of cases in farm animals amounted to 515, in wild animals to 50 and in pets to 352. Data on rabies cases were obtained from the Committee for Veterinary Control and Supervision of Kazakhstan, as well as during expeditionary trips. This research was carried out to demonstrate the use of modern information and communication technologies, geospatial analysis technologies in particular, to identify and visualize spatio-temporal patterns of rabies emergence among different animal species in Kazakhstan. We also aimed to predict an expected number of cases next year based on time series analysis. Applying the 'space-time cube' technique to a time series representingcases from the three categories of animals at the district-level demonstrated a decreasing trend of incidence in most of the country over the study period. We estimated the expected number of rabies cases for 2024 using a random forest model based on the space-time cube in Arc-GIS. This type of model imposes only a few assumptions on the data and is useful when dealing with time series including complicated trends. The forecast showed that in most districts of Kazakhstan, a total of no more than one case of rabies should beexpected, with the exception of certain areas in the North and the East of the country, where the number of cases could reach three. The results of this research may be useful to the veterinary service in mapping the current epidemiological situation and in planning targeted vaccination campaigns among different categories of animals.