Analyze and plot regional EU mortality data



Build Status

Coverage Status


This package can be used to analyze cause-specific mortality in NUTS regions in the European Union, and, more specifically, to explore the relative importance of different causes of death. It uses the table with crude age-specific death rates in NUTS 2 regions averaged over 3 years, hlth_cd_ycdr2, which is available via the Eurostat database. The table, converted to CSV, is included in the packages at data/hlth_cd_ycdr2.csv. For data about the different NUTS regions, the file data/NUTS_AT_2013.csv (included in the shapefile archive for the default plots, as described below) is used.

To define an array of the NUTS 2 regions in the Nordic countries with data in hlth_cd_ycdr2 (Denmark, Finland, Norway and Sweden):

using EUregmort
nordnuts2 = vcat(map((x)->nuts2ids(x), ["DK"; "FI"; "NO"; "SE"])...)

Use this array to plot the correlation between female and male proportion of mortality from circulatory causes in the Nordic countries (see the data/CL_ICD10_20170129_155451.csv, based on Eurostat metadata, for information about causes of death):

caprop_regsexplot(nordnuts2, "TOTAL", "I")

Using cartopy, it is also possible to plot maps showing regions with a lower or higher proportion of deaths from a given cause. To do this, it is necessary to have shapefiles with the different regions. You can download files with 1:3 million scale as a ZIP archive, which can be unzipped in the data directory. However, the shapefiles from Eurostat use the EPSG:4258 projection, which is not suited for use with Cartopy. To convert them to EPSG:3034 projection (default settings for the package), use e.g. GDAL, and run, in the data/NUTS_2013_03M_SH/data directory:

ogr2ogr -f "ESRI Shapefile" -t_srs EPSG:3034 -s_srs EPSG:4258 NUTS_RG_03M_2013_3034.shp

To plot a map of female proportion of deaths due to circulatory causes in the Nordic countries:

caprop_mapplot(nordnuts2, "F", "TOTAL", "I")

By default, death rates from all causes are used as denominator. However, you can give another cause as the fourth argument. For example, to plot a map of male deaths due to circulatory causes relative to neoplasms in the Nordic countries:

caprop_mapplot(nordnuts2, "M", "TOTAL", "I", ca2 = "C00-D48")

You can also plot the death rates themselves by specifying "pop" as denominator:

caprop_mapplot(nordnuts2, "M", "TOTAL", "I", ca2 = "pop")

When plotting death rates, it is interesting to compare average rates over age groups. Use the function meanrate to define an alternative dataframe. Age groups with average rates are prefixed YM. To plot a map of regions with higher average female mortality from neoplasms over the ages from 45 to 64.

caprop_mapplot(nordnuts2, "F", "YM45-64", "C00-D48", ca2 = "pop",
inframe = meanrate("Y45-49", "Y60-64"))

First Commit


Last Touched

over 1 year ago


14 commits

Used By: