Spatiotemporal analysis of the relationship between socioeconomic factors and stroke in the Portuguese mainland population under 65 years old

  • André Oliveira | miguel.oliveira@ihmt.unl.pt Department of International Public Health and Biostatistics, Tropical Hygiene and Medicine Institute, Universidade Nova de Lisboa, Lisbon, Portugal.
  • António J.R. Cabral Department of International Public Health and Biostatistics, Tropical Hygiene and Medicine Institute, Universidade Nova de Lisboa, Lisbon, Portugal.
  • Jorge M. Mendes Nova Information Management School, Universidade Nova de Lisboa, Lisbon, Portugal.
  • Maria R.O. Martins Department of International Public Health and Biostatistics, Tropical Hygiene and Medicine Institute, Universidade Nova de Lisboa, Lisbon, Portugal.
  • Pedro Cabral Nova Information Management School, Universidade Nova de Lisboa, Lisbon, Portugal.

Abstract

Stroke risk has been shown to display varying patterns of geographic distribution amongst countries but also between regions of the same country. Traditionally a disease of older persons, a global 25% increase in incidence instead was noticed between 1990 and 2010 in persons aged 20-≤64 years, particularly in low- and medium-income countries. Understanding spatial disparities in the association between socioeconomic factors and stroke is critical to target public health initiatives aiming to mitigate or prevent this disease, including in younger persons. We aimed to identify socioeconomic determinants of geographic disparities of stroke risk in people <65 years old, in municipalities of mainland Portugal, and the spatiotemporal variation of the association between these determinants and stroke risk during two study periods (1992-1996 and 2002-2006). Poisson and negative binomial global regression models were used to explore determinants of disease risk. Geographically weighted regression (GWR) represents a distinctive approach, allowing estimation of local regression coefficients. Models for both study periods were identified. Significant variables included education attainment, work hours per week and unemployment. Local Poisson GWR models achieved the best fit and evidenced spatially varying regression coefficients. Spatiotemporal inequalities were observed in significant variables, with dissimilarities between men and women. This study contributes to a better understanding of the relationship between stroke and socioeconomic factors in the population <65 years of age, one age group seldom analysed separately. It can thus help to improve the targeting of public health initiatives, even more in a context of economic crisis.

Dimensions

Altmetric

PlumX Metrics

Downloads

Download data is not yet available.
Published
2015-11-04
Info
Issue
Section
Original Articles
Keywords:
Stroke, Working-age population, Geographically weighted regression, Multi-temporal analysis, Portugal
Statistics
  • Abstract views: 2148

  • PDF: 882
  • HTML: 1195
How to Cite
Oliveira, A., Cabral, A. J., Mendes, J. M., Martins, M. R., & Cabral, P. (2015). Spatiotemporal analysis of the relationship between socioeconomic factors and stroke in the Portuguese mainland population under 65 years old. Geospatial Health, 10(2). https://doi.org/10.4081/gh.2015.365