The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.


Petter Pilesjö



Analysis of spatial co-occurrence between cancer and cardiovascular disease mortality and its spatial variation among the Swedish elderly (2010–2015)


  • Augustus Aturinde
  • Ali Mansourian
  • Mahdi Farnaghi
  • Petter Pilesjö
  • Kristina Sundquist
  • Gilbert Maiga

Summary, in English

CVD and cancer are the two leading causes of death worldwide. Improvement in cancer early detection and treatment has resulted in an increased number of cancer survivors. However, many of the survivors tend to develop CVD often leading to their demise. Conversely, people with pre-existing CVD conditions, especially the elderly, have increased chances of developing cancer and dying from the same. The World Health Organization, consequently, recommends joint management of both diseases. However, in Sweden, as with many other countries, few studies have explored the nature of the associations between the two disease mortalities and their spatial variation at a population level. This study uses correlation, global Moran's index and global bivariate Moran's index to investigate national trends of cancer and CVD crude mortality rates in the Swedish elderly. Spatial scan statistics, spatial overlay and local entropy maps were used to analyse for spatial co-occurrence, local joint spatial clustering and associations in the 2010–2015 cancer and CVD crude mortality rates for the Swedish elderly (65+ years). Mortality data were obtained from the Swedish Healthcare Registry. Our results showed that throughout the years of study, the correlation between cancer and CVD crude mortality rates was averagely positive. Spatial correlation analysis (univariate and bivariate) showed that the contribution of the neighbourhood mortality rates to the observed mortality rates was weak, though significant. From cluster analysis, the cancer and CVD crude mortality rates showed differences in clustering spatial scales with CVD clustering at a smaller scale. Finally, local entropy maps showed that cancer and CVD crude mortality rates were not always related across Sweden, but whenever they were, the relationship was mainly positive and linear. This study contributes to cancer and CVD public health efforts in Sweden by identifying areas where the two causes of death spatially co-occur, and where the two exhibit no spatial overlap. This provides a valuable starting ground for more focused studies to identify local drivers and/or informs coordinated targeted intervention in both causes of death.


  • Dept of Physical Geography and Ecosystem Science
  • Centre for Geographical Information Systems (GIS Centre)
  • MECW: The Middle East in the Contemporary World
  • Centre for Advanced Middle Eastern Studies
  • Family Medicine and Clinical Epidemiology
  • EpiHealth: Epidemiology for Health

Publishing year





Applied Geography



Document type

Journal article




  • Public Health, Global Health, Social Medicine and Epidemiology
  • Physical Geography


  • Cancer
  • CVD
  • Local entropy map (LEM)
  • Spatial scan statistics
  • Spatial variation
  • Swedish elderly



Research group

  • Family Medicine and Clinical Epidemiology


  • ISSN: 0143-6228