Effect of census-based correction of population figures on mortality rates in Germany.

Background: The population figures in Germany are obtained by updating the results of the latest census with information from the statistics on birth, deaths and migration statistics. The Census 2011 in Germany corrected population figures, which have only been updated over a long period of time. The aim of this work is to show the effect of the census-based correction of the population figures on the magnitude of mortality rates in Germany 2011-2013.

Methods: We compared mortality rates (total, cancer, and cardiovascular disease) for the period 2011-2013 based on the uncorrected and Census 2011 corrected population figures. We also compared the effect of the choice of different standard populations in the age standardization of rates on the difference in uncorrected and corrected mortality rates.

Results: There is a clear decline in age-specific cancer mortality among men aged 90 and over when using the uncorrected population figures, which is reversed as soon as the corrected population figures are used. Among women, there is hardly any difference between the uncorrected and corrected mortality rates. The correction of the population figures does not lead to a qualitatively different pattern in the mortality rates for cardiovascular diseases and myocardial infarction, but it increases the magnitude of the rates, particularly for elderly men. Standard populations with higher weights at older ages produced larger corrections in mortality rates.

Conclusions: Even though the Census 2011 corrected nationwide mortality rates without age stratification differed only slightly from the uncorrected rates, there were noticeable increases in mortality, particularly in the city states of Hamburg and Berlin and in old age. Due to the particularly large error in the population figures in the older age range, an age standard that assigns lower weights at older ages should be used for age standardization of rates wherever possible.