I was surprised too, and for the other countries in the sample the result is the way you would expect. (But it seems that the Netherlands is an outlier exactly

I was surprised too, and for the other countries in the sample the result is the way you would expect. (But it seems that the Netherlands is an outlier exactly because the cost of end of life care is very expensive compared to "obesity care". But it would probably be much less surprising if we would see the data on graphs, as health problems - so associated costs too - tend to spike waaay up above 80, so in this model it seems that ~5 year difference makes up for a lot of money...)