International Journal of Health Geographics最新文献

Life space-the geographical extent of people's movement within their communities-is closely linked to the physical and cognitive functions of older adults. It serves as a potential marker for mobility and health among aging populations, corresponding to the ability to age in place. Central to maintaining and expanding older adults' life spaces are their out-of-home travel and activities. In a study with 1118 community-dwelling older adults in Singapore, participants' travel patterns were tracked over a period of two weeks to identify how and when they travel, where they travel to, and why they travel (or what they do at out-of-home locations). Data was collected via a mobile application with precise location tracking and an accompanying log for participant input. This paper tests the feasibility and comprehensiveness of a list of activities that older adults engage in when they visit places outside their homes, using it to characterize travel motivations amongst older persons. Furthermore, it identifies the specific activities that are associated with larger life spaces, while controlling for sociodemographic differences in a hierarchical linear regression model. The findings suggest that the proposed list adequately represents older adults' motivations for travel and activities within the community, making it suitable for applications in future research and analyses. Crucially, the results indicate that although trips to meet day-to-day needs were performed more frequently, trips for employment and social activities were the key drivers of larger life spaces or greater extents of travel.

Background: Equitable access to general practitioner services remains a persistent challenge for health systems and is critical for reducing health inequalities, particularly between urban and rural regions. Understanding the spatial and temporal dynamics of primary care provision is vital for informed healthcare planning and policy.

Methods: Spatial and temporal disparities in the supply of general practitioners across Lower Saxony, Germany, were assessed over a 21-year period (2000-2021). Data from the Association of Statutory Health Insurance Physicians and municipal population statistics were used to develop the General Practitioner Regional Index - a composite measure incorporating both the availability and accessibility of general practitioners. Non-parametric statistical tests were applied to identify significant trends at the municipal level.

Results: Pronounced geographic inequities in general practitioner supply were identified. More than 40% of the population resides in areas with declining supply, while only 5% experience improvements. Urban centres and their peripheries consistently exhibited higher and mostly stable levels of general practitioner supply. In contrast, rural small towns and villages demonstrated both lower baseline accessibility and more frequent negative trends. The primary driver of supply losses in affected areas was physician retirement, while relocation played a secondary role and typically occurred within similar regional types, thereby limiting redistribution effects.

Conclusions: The results underscore the persistence of urban-rural disparities in general practitioner availability and highlight physician retirements as the principal factor behind declining supply, with limited offsetting effects from physician migration. The findings indicate a need for spatially sensitive, succession-focused workforce strategies and innovative primary care models to mitigate rural undersupply and promote health equity.

Trial registration: Not applicable. This study does not report the results of a health care intervention on human participants.

Background: This study is the first in Taiwan to apply the enhanced two-step floating catchment area (E2SFCA) method to evaluate the spatial accessibility of primary care. Traditional physician-to-population ratios by administrative region overlook cross-boundary healthcare-seeking and travel distance barriers. This study accounts for these limitations and further examines the impact of accessibility on healthcare utilization and outcomes.

Methods: We used national health insurance claims, physician registry data, and GIS-based road networks to measure accessibility with the E2SFCA method, defining it as the number of primary care physicians per 10,000 residents within a 30-minute travel time. A retrospective cohort of 2 million adults was analyzed. Generalized estimating equations with appropriate regression models assessed associations between accessibility and healthcare utilization, expenditures, avoidable emergency department (ED), and avoidable hospitalizations.

Results: Spatial analysis identified 15 townships (114,915 residents, 0.49%) with no primary care physicians and another 15 townships (114,430 residents, 0.49%) with low accessibility. These underserved areas were concentrated in central and eastern Taiwan, whereas metropolitan regions had sufficient resources. Higher accessibility was significantly associated with fewer ED visits (ratio = 0.994; 95% CI: 0.990-0.997, P< 0.001), ED expenditures (ratio = 0.993; 95% CI: 0.989-0.997, P< 0.001), the odds of avoidable ED visits (odds ratio = 0.993; 95% CI: 0.988-0.998, P = 0.005), and the number of avoidable ED visits (ratio = 0.993; 95% CI: 0.988-0.998, P = 0.004). Accessibility also reduced the odds of avoidable hospitalization (odds ratio = 0.995; 95% CI: 0.990-0.999, P = 0.017).

Conclusion: Greater spatial accessibility to primary care was linked to reductions in ED visits, ED costs, avoidable ED use, and avoidable hospitalization. The E2SFCA method provides a more accurate tool for identifying underserved regions and can inform equitable allocation of healthcare resources. Telemedicine and mobile services should be expanded to address shortages in remote areas.

Many studies linking spatial environmental exposures to health outcomes rely on small statistical geography units, such as Lower-layer Super Output Areas (LSOAs), to estimate exposure. However, these units commonly vary in size, particularly between urban and rural areas, leading to potential exposure misclassification. This study proposes a new method for better capturing environmental exposure at the small statistical geography unit level. Using the Living England Habitat Map as an example, we combined LSOA and postcode-level data to account for varying area sizes and mitigate edge effects. We compared our method with the typical approach, which calculates an average at the small geography unit level. Overall, our proposed method resulted in lower exposure to non-built-up areas compared to averaging across entire LSOAs, whereas exposure to built-up areas was higher by 8-10%. However, these patterns varied based on region, urban/rural classification, land cover type, and LSOA size class. We suggest that this proposed method offers a more consistent approach to estimating neighbourhood exposure to nature.

Background: To date, few HIV-related population-based data are available for refugee populations. Household surveys typically require reliable population counts and well-defined geographic areas, which are often not available for refugee settlements. We describe the gridded population sampling approach as an option for conducting such a survey in Uganda and describe its application for a household survey in Uganda and assess its utility among refugee populations.

Methods: The Uganda Refugee Population-based HIV Impact Assessment (RUPHIA) 2021 was a cross-sectional, population-based HIV survey among refugee households in Ugandan settlements, excluding Kampala. We collected shapefiles and population counts for the refugee settlements. These shapefiles from the various geographic areas of interest represented the aggregated refugee settlement zones (including all settlements with available zone shapefiles) and served as the base for creating the sample frame. The sample frame was constructed by disaggregating United Nations High Commission for Refugees population counts from large refugee settlement zones into 100 × 100 m grid cells using WorldPop's peanutButter-Disaggregate app that uses building footprint information to distribute the population into the grid cells. We then utilized a gridded population sampling approach which redistributed the population into manageable-sized areas of contiguous grid cells based on their estimated population size, forming enumeration area-like sampling units using the publicly available GridEZ algorithm.

Results: The resulting gridded population dataset had 43,193 100 m x 100 m cells with an estimated mean of 31 people per cell and a range from 2 to 1028. The final gridded population sample frame had 2636 GridEZ units with an average population of 500 ranging from 178 to 1531. The sample frame performed well for survey activities, with few issues encountered in the field, although the size measures for number of households had some inaccuracies, due to issues such as compounds having multiple structures.

Conclusions: Gridded population sampling was successfully utilized for this refugee study, saving time and money that would have been needed if enumeration of all the refugee settlements had been required. Gridded population sampling is a useful tool when census data are outdated or unavailable or when the population is dynamic, such as with refugees or other mobile or at-risk populations for surveillance or as part of a humanitarian response.

Background: Mental-health geomatics require reliable ways to convert high-frequency GPS trajectories into meaningful place types that support indicators such as homestay, location entropy, and spatial extent of daily activities. Raw coordinates are typically noisy and carry little semantic information. We introduce H3-MOSAIC(H3-based Multimodal OSM-and-Satellite AI for Classification), a multimodal generative framework that fuses OpenStreetMap (OSM) building text and satellite imagery on H3 grids to infer place semantics from high-frequency GPS.

Methods: Raw GPS was smoothed by minute-level speed filtering, then assigned to Level 10 H3 hexagons. Cells were retained if the mean speed was ≤ 1.2 m/s and the cumulative duration was ≥ 15 min, contiguous cells were merged, and home was defined as the cell with the longest dwell from 23:45 to 06:00. We compared text-only OSM classification with image-based and fused approaches across open-source models (DeepSeek, CLIP, LLaVA, Qwen-VL) and proprietary models (GPT-4o-mini, Gemini-2.5-flash-lite). Performance was assessed by accuracy, Cohen's kappa, precision, recall, F-measure, and confusion matrices. Day level associations between H3 semantic exposures and stress were examined by a random forest model and explainable methods.

Results: Multimodal methods outperformed single-modality baselines. In the 11-class task, accuracies were: CLIP 0.179, LLaVA 0.269, Qwen-VL 0.565, GPT-4o-mini 0.779, and Gemini-2.5-flash-lite 0.790. In the 5-class consolidation, accuracies rose to 0.702 (Qwen-VL), 0.849 (GPT-4o-mini), and 0.858 (Gemini-2.5-flash-lite). Text-only OSM baselines were lower (≈ 0.60-0.68). Across 3,845 hexagons with OSM text, closed-source models agreed on 79% of labels; disagreements concentrated in mixed-use, office, and green classes. Error modes reflected area-dominant versus keyword-triggered reasoning, hybrid-parcel ambiguity, tag sparsity, and symbolic artifacts. Stabilized semantics support more robust computation of homestay, entropy, and activity space and are suitable for privacy-aware, cross-city reuse. In a day-level case study, minutes at Home related to lower stress; Green showed a U-shaped pattern.

Conclusions: H3-MOSAIC provides a scalable, auditable pipeline for semantic place detection from high-frequency GPS. Multimodal fusion markedly improves accuracy and consistency. Proprietary models are most robust on hard classes and open-source models are practical for coarse taxonomies. H3 day level exposures show stress patterns consistent with established mental health pathways, supporting face validity. The framework enables downstream exposure analyses with reduced misclassification and improved interpretability.

Background: Geospatial studies that consider the relationships between the built environment and health typically rely on researcher-led 'objective' measurement of geospatial attributes of the built environment. Some studies can fail to find expected associations between environments and health outcomes where the geospatial measures do not reflect the experiences or perceptions of people themselves. We took a participatory approach to work with older adults with a concern for falling to assess the built environment in order that we could understand how their assessments relate to researcher assessments. We also wanted to assess whether specific demographic characteristics explained differences in assessments of the built environment between participants. Age-friendly environments can contribute to healthy active ageing. Falling and a fear of falling can lead to restricted outdoor activity. Therefore, understanding how the built environment contributes to fear of falling is important for age-friendly environments.

Methods: The study is a cross-sectional retrospective observational study of the built environment. We worked with older adults in workshop settings to undertake community audits of the built environment in Google Street View. They assessed locations where a fall had occurred. Researchers separately audited the same locations. We used descriptive statistics and ordinal regression cumulative link mixed models to estimate the odds that community members would rank a location one level higher than the researchers.

Results: There are significant differences in researcher and community auditor assessments of locations of attractiveness. Site related and individual attributes explain variation in how difficult locations were rated for walking, and for concern about falling. Only individual attributes explained variation in site attractiveness. Locations with more trip hazards and steeper slopes were rated as being more difficult to walk and were associated with greater concern for falling.

Conclusions: Attributes of the built environment influence perceptions of difficulty walking and concern or falling at specific locations. Furthermore, there are some differences in how researchers and community auditors assess the same locations, meaning that geospatial studies which rely only on researcher assessments may be prone to bias. Involving older people in geospatial studies that measure age-friendly environments can make measurement more reflective of their experiences.

Background: Malaria, whose parasites are transmitted by Anopheles mosquitoes, remains a major public health burden in Madagascar despite the control measures led by the National Malaria Control Program. Understanding the population dynamics of Anopheles mosquitoes is therefore essential to optimize malaria surveillance and control. This study aimed to develop a model incorporating environmental, climatic and agricultural determinants of Anopheles abundance to predict their spatiotemporal distribution.

Methods: We developed a model of spatiotemporal dynamics for four Anopheles species, vectors of malaria parasite in Madagascar: Anopheles arabiensis, Anopheles coustani, Anopheles funestus and Anopheles gambiae. This model was based on the life cycle of Anopheles and accounted for both the aquatic and aerial phases of their development. It used a system of differential equations to estimate the number of Anopheles mosquitoes at each stage of development. The Ocelet language, dedicated to the modeling of spatial dynamics, was used to produce simulations based on climate and environmental data. The model explicitly integrates the agricultural calendar to adjust the environmental carrying capacity of larval habitats. Model outputs were validated with entomological data collected in Vohimasy (Farafangana districts, 2014-2017).

Results: 24 simulation outputs, from three Anopheles species and eight sites, were obtained and the validation revealed a significant correlation between field observations and model predictions: the correlation coefficients obtained ranged from 0.70 to 0.76. The predicted abundance of host-seeking Anopheles varied seasonally influenced by precipitation, temperature and environmental carrying capacity. The model exhibited robustness across sites with diverse climates and accurately reproduced interannual dynamics. The integration of the agricultural calendar significantly reduced the overestimation of the density of host-seeking adult females.

Conclusion: The developed Anopheles dynamics model provides a valuable tool for predicting mosquito abundance and distribution over time and space. It correctly predicted the abundance at villages with contrasting climates and reproduced interannual dynamics well. A distinctive aspect of this work lies in the explicit integration of seasonal agricultural practices into the estimation of larval habitat availability. This allows for a more accurate and transferable modeling of Anopheles population dynamics.