A Bayesian spatial ecological analysis of social conditions, infection burden and COVID-19 mortality across mainland Scottish councils 2020-2021
Geospatial Health · 10.06.2026 · 21 (1)
DOI: 10.4081/gh.2026.1477 ↗Kurz gefasst
Fachartikel in der Zeitschrift Geospatial Health, online erschienen am 10.06.2026. Verfasst von Zhiyuan Luo und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Sociology and Anthropology, Sun Yat-sen University, Guangzhou. Hauptthemen laut MEDLINE: COVID-19, Spatial Analysis, Social Conditions. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The COVID-19 pandemic has generated substantial spatial and social inequalities in mortality, yet within-country council-level variations remain incompletely understood. Scotland offers a useful case, with a severe epidemic but rich administrative data and marked socio-economic gradients. We examined how infection burden, hospital-related indicators and selected area-level social indicators were associated with COVID-19 mortality across mainland Scotland 2020-2021. We assembled a two-year panel for 29 mainland councils, using deaths involving COVID-19, population denominators, COVID-19 testing and hospital activity, council-level indicators of marriage, unemployment, smoking cessation and migration. A Bayesian Poisson log-linear model with conditional autore- gressive random effects was used to estimate area-level relative risks and covariate associations. Given the short panel, the analysis is interpreted primarily as a spatial ecological analysis conducted over two successive pandemic years. Crude mortality showed consistently higher COVID-19 mortality in the central belt than in many northern and rural councils. After adjustment, spatial differences remained but were modest: no council in 2021 had a posterior probability above 0.10 of exceeding the national mean mortality risk. Positive test burden was positively associated with mortality and is interpreted primarily as a proximal epidemiological indicator of infection burden. Social determinants were also important. Higher unemployment was associated with increased risk, whereas higher marriage counts were linked to lower mortality. Smoking quit rates showed a positive association with deaths, likely reflecting residual confounding by underlying deprivation and historical smoking prevalence. Hospital utilisation and migration indicators showed weaker and more uncertain effects. Together, these findings indicate that crude spatial disparities in COVID-19 mortality across mainland Scottish councils became more moderate after adjustment. Given the ecological design and the two-year panel, the findings should be interpreted as area-level associations rather than as evidence of specific local social mechanisms.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Sociology and Anthropology, Sun Yat-sen University, Guangzhou.
Zitieren
APA
Luo, Z., Liu, Y., & Xu, W. (2026). A Bayesian spatial ecological analysis of social conditions, infection burden and COVID-19 mortality across mainland Scottish councils 2020-2021. Geospatial Health, 21(1). https://doi.org/10.4081/gh.2026.1477
Vancouver
Luo Z, Liu Y, Xu W. A Bayesian spatial ecological analysis of social conditions, infection burden and COVID-19 mortality across mainland Scottish councils 2020-2021. Geospatial Health. 2026;21(1). doi:10.4081/gh.2026.1477
BibTeX
@article{Luo2026,
author = {Luo, Zhiyuan and Liu, Yancheng and Xu, Wenqi},
title = {A Bayesian spatial ecological analysis of social conditions, infection burden and COVID-19 mortality across mainland Scottish councils 2020-2021},
journal = {Geospatial Health},
year = {2026},
volume = {21},
number = {1},
doi = {10.4081/gh.2026.1477}
}
EUVIMED