Figure 2 - uploaded by André Redert
Content may be subject to copyright.
Weekly mortality in The Netherlands and its city Rotterdam, by observed , estimated baseline and indicator . Note the different time scales due to data availability.
Source publication
This report investigates the natural variability of mortality, which determines the thin line between excesses and normal variation within expectations. I propose a model for mortality’s weekly variability based on a Poisson model, driven by potential non-stationary influences that act nation-wide and fast, on time scales of a week.
Results reveal...
Context in source publication
Context 1
... Netherlands is a country with a population of 17M people, and Rotterdam is one of its cities with 0.6M. Figure 2 shows mortality from 2010-2022 for The Netherlands, and 2019-2022 for Rotterdam due to data availability, excluding the first few weeks of 2019 due to the processing tail of filter in (4) and (9). Estimated baselines neatly follow the overall shape of mortality . ...
Citations
... My study on the natural variability of mortality [Red4] found is well-modeled by: ...
... The first term in (10) represents wellknown Poisson noise, the 2nd represents an omnipresent but sofar not-understood component with 3% for weekly mortality [Red4]. Possibly, the proposed temperature model explains part of mortality fluctuations, leading to lower . ...
... Possibly, the proposed temperature model explains part of mortality fluctuations, leading to lower . Higher may also be encountered here, as [Red4]'s method to determine automatically compensated for any kind of abnormal mortality lasting a month or longer. In this report, anything-not-predictableby-temperature-during-normal-years will add to , such as influenza-winter-variability. ...
A new mortality model is presented for more objective quantification and localization in time of excess mortality.
The model is just one equation, M = W*atT, with source data of mortality and temperature. Parameters represent trends over decades, seasonal/weekly variability, delay between cause and death, frailty regulating delay between excess and deficit, lifetime lost/saved (LTL/LTS), and natural mortality variability. Pandemic excesses are examined by integrating determinants of positive covid tests and vaccinations. Measured parameters are covid Case Fatality Rate (CFR), Vaccine-dose Fatality Rate (VFR), Vaccine Effectivity (VE) against covid mortality, and LTL/LTS by covid and vaccination.
Experiments involve 10 EU countries (344M people) and age-stratified datasets in The Netherlands over years 2000-2023. Predictions and excess measurements by the proposed model are fivefold more accurate than those of the national baseline. Covid is found to target much frailer people than natural causes do, and vaccine fatal events target younger and healthier people. Measured lifetime lost to vaccinations is of the same order as lifetime saved: Germany has LTL 450-4600ky (kiloyear), Netherlands 50-500ky.
The model is very different from the state of the art and many options for further research are outlined. A disturbing relation is found between vaccinations and the persisting excess mortality among the young and healthy population.
Keywords: Mortality, temperature, frailty, pandemic, excess, deficit, covid, vaccine, QALY, lifetime
