Overestimated Numbers

Rommel et al. report that in 2020, persons who died of COVID-19 in Germany lost 9.6 years of life on average (1). This number is an overestimate:

For each deceased person, the authors equate the years of life lost with the residual life expectancy that average persons of the respective age have according to the period life tables. This approach ignores the fact that persons who died of COVID-19 are in a significantly worse state of health than average persons of the same age. Assuming that 28.7% of the deceased are nursing home residents (2), and the residual life expectancy after admission to a nursing home is a maximum of 1 year, i.e. 0.5 years for an average home resident (3), this results in an estimated loss of life years of only 7.0 years (0.287 × 0.5 + 0.713 × 9.6). This estimate is still too high for three reasons: not all those who died of/with COVID-19 were reported to the Robert Koch Institute (RKI) by the nursing homes. Clients of outpatient care services are not taken into account. Even deceased persons without outpatient/inpatient care may often have had pre-existing conditions and thus a reduced life expectancy.

Taking into account the rapid aging of the society, there was a reduced mortality in Germany during the first wave of SARS-CoV-2, as compared to the corresponding period of 2016–2019; our own (still unpublished) analyses also show this for the entire year 2020. This indicates that the number of years of life lost may be lower than the estimated number given above.

The authors clearly state that pre-existing conditions were not taken into account. However, the media reports do not mention this limitation at all (Süddeutsche Zeitung) or mention it only marginally (Welt, Spiegel), thus failing to convey the limitations of this approach to the reader.