COVID-19: Spatial Resolution of Excess Mortality in Germany and Italy

Stang et al.1.Stang A. Standl F. Kowall B. B. Brune, J. Böttcher, M. Brinkmann et al., Excess mortality due to COVID-19 in Germany.J Infect. 2020 Sep 19; Abstract Full Text Full Text PDF PubMed Scopus (89) explored age-specific numbers of weekly deaths in Germany from 2016 to June 2020. We wish to complement their results and conclusion that excess mortality existed for two months with analyses of higher spatial resolution. Importantly, analysing weekly deaths can offer virus test-independent information on mortality effects during the SARS-CoV-2/COVID-19 pandemic. Thus, the authors contribute to a fuller epidemiological picture. However, looking at countries alone can miss relevant information on the pandemic's course and toll in smaller spatial units. To exemplify, let us look at effects of SARS-CoV-2/COVID-19 on mortality in Germany and Italy2.Morfeld P. Erren T.C. Deaths in nine regions of Italy in February/March 2020: "mortality excess loupe" for SARS-CoV-2/COVID-19-epidemiology in Germany.Gesundheitswesen. 2020; 82: 400-406Crossref PubMed Scopus (10) Google Scholar in the first six months of 2020. Assuming otherwise constant determinants of death, we chose Standardized Mortality Ratio (SMR) methodology3.Morfeld P. Erren T.C. Mortality and attributable fraction in COVID-19 analysis: avoiding research waste and negligence.Am J Public Health. 2020; 110: 1644-1645Crossref PubMed Scopus (8) Google Scholar to analyse excess mortality by state in Germany and by region in Italy. Monthly and weekly all-cause mortality data from January 2016 to June 2020, published by the Federal Statistical Office in Germany and the National Institute of Statistics in Italy, respectively, for all age groups, <65 years and ≥ 65 years and individual states or regions, were used for our explorations. SMRs were evaluated by comparing the index year 2020 with our reference years 2016–2019 with a focus on trend detection.4.Breslow N.E. Day N.E. Statistical methods in cancer research. Volume II – the design and analysis of cohort studies. International Agency for Research on Cancer, Lyon1987Google Scholar, 5.Rothman K.J. Greenland S. Lash T.L. Modern epidemiology.3rd ed. Lippincott Williams & Wilkins, Philadelphia2008Google Scholar, 6.Sutton A.J. Abrams K.R. Jones D.R. Sheldon T.A. Song F. Methods for meta‐analysis in medical research. Wiley, Chichester, U.K2000Google Scholar In analyses for Germany, higher mortality in April was followed by a decline: For January-June, we calculated the SMR as 1.00 (95% CI: 0.97–1.04), including an SMR of 1.10 in April (95% CI: 1.06–1.13). As one noticeable – as yet unappreciated – result in individual states, in Mecklenburg-Western Pomerania, reduced SMRs in January turned monotonically into significantly increased SMRs until June with an SMR of 1.07 (95% CI: 0.99–1.16). Poisson trend models estimated an SMR increase of 2.5% per month, observed in both age groups (always p<0.05). SMRs were most increased in April in Hamburg (1.24; 95%CI: 1.12–1.37), Bavaria (1.21; 95%CI: 1.17–1.26) and Bremen (1.20; 95%CI: 1.06–1.37). In Italy, SMRs of 1.43 (95%CI: 1.39–1.47) were observed in March and 1.31 (95%CI: 1.3–1.33) in April with highest SMRs in Lombardy in March (2.89; 95%CI: 2.79–3.00) and April (2.12; 95%CI: 2.07–2.17), Valle d'Aoste (April: 1.71; 95%CI: 1.47–1.99) and Emilia-Romagna (March: 1.69; 95%CI 1.61–1.77). Importantly, in the same time windows with significant excess mortality there were also federal states and regions with no elevated SMRs: In Germany, in April for instance in Schleswig-Holstein (0.96; 95%CI: 0.91–1.02) or Saxony-Anhalt (0.98; 95%CI: 0.93–1.04). In Italy, for instance in Basilicata (March: 0.83; 95%CI: 0.72–0.95), Friuli Venezia (March: 0.86; 95%CI: 0.78–0.94; April: 0.88; 95%CI: 0.83–0.94) or Latium (March: 0.97; 95%CI: 0.93–1.0; April: 0.91; 95%CI: 0.88–0.94). Clearly, the variation of SMRs may hold important clues regarding effects of the pandemic, and of counter-measures, which can vary over space and time. This heterogeneity should be explored rather than remain masked by exclusively looking at the countries as a whole. Overall, when Stang et al.1.Stang A. Standl F. Kowall B. B. Brune, J. Böttcher, M. Brinkmann et al., Excess mortality due to COVID-19 in Germany.J Infect. 2020 Sep 19; Abstract Full Text Full Text PDF PubMed Scopus (89) point out that the course of the pandemic across Europe is different and that pooling of mortality data7.Vestergaard L.S. Nielsen J. Richter L. Schmid D. Bustos N. Braeye T. et al.Excess all-cause mortality during the COVID-19 pandemic in Europe - preliminary pooled estimates from the EuroMOMO network, March to April 2020.Euro Surveill. 2020; 25 (2001214)https://doi.org/10.2807/1560-7917.ES.2020.25.26.2001214Crossref Scopus (167) Google Scholar may mask relevant differences at national levels we do agree. But exclusively focusing on Germany, or Italy, as a whole can also mask relevant effects in individual states or regions. In this vein, extending the authors' aim to provide estimates of excess mortality during the "first wave" of the pandemic would offer added value: SMR analyses with appropriate spatial resolution are needed to directly compare the burden of disease not only between but also within countries. They should be used to assess desired, and undesired, effects of measures against SARS-CoV-2/COVID-19. In conclusion, to "fly with high visibility"8.Erren T.C. Morfeld P. COVID-19-Mortalität: Mit viel Sicht fliegen.Dtsch Ärztebl. 2020; 117 (A-1010 / B-850, https://www.aerzteblatt.de/archiv/213856/COVID-19-Mortalitaet-Mit-viel-Sicht-fliegen)Google Scholar when working to cope with the pandemic, SMR analyses – with appropriate spatial resolution – are needed in Germany, and in other countries, on a regular basis. For independent analyses, national authorities should expedite the publication of raw data on mortality and populations – expanded by detailed information on age, sex and causes of death.9.Leon D.A. Shkolnikov V.M. Smeeth L. Magnus P. Pechholdova M. Jarvis C.I. COVID-19: a need for real-time monitoring of weekly excess deaths.Lancet. 2020; 395: e81Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. S0163-4453(20)30596-X. doi:10.1016/j.jinf.2020.09.012. Online ahead of print.