This analysis quantifies the potential impact of different pandemic responses on COVID-19 mortality in six Northwestern European countries during the first pandemic wave in February through June 2020. It highlights that in the rapidly growing first COVID-19 wave\u2014infection rates initially doubled every 2\u20133\u00a0days\u2014small differences in initial epidemiological situations between countries, together with small disparities in the timing and effectiveness of adopting COVID-19 response from neighboring countries, result in large variations in mortality rates. A mere 3-day delay in the response was estimated to result in approximately doubling mortality during a single wave.<\/p>\n
For any of the six countries, mortality would have differed substantially, had the response as implemented in another country been adopted. The order of the resulting cumulative COVID-19 deaths per million, from lowest to the highest, was found for the responses of the Netherlands, Belgium, Denmark, the UK, Germany, and Sweden. This order differs from the observed rates of confirmed COVID-19 deaths, where Denmark and Germany reported the fewest deaths per million before the Netherlands, Sweden, Belgium, and the UK. Actual observed per-capita death rates are determined not only by the response but also by underreporting and the epidemiological situation in the early phase of the pandemic wave: for example, the incidence of infection on 13 March 2020 and the reproduction number before this time varied. In early March 2020, COVID-19 mortality trajectories in the Netherlands, UK, and Belgium slightly outpaced those in Germany, Denmark, and Sweden. This implies, for instance, that a marginally lower incidence of infection in Denmark compared to Netherlands allowed for a slower decline in Rt, while still resulting in a lower observed mortality rate. Furthermore, large-scale implementation of response measures may lead to the fastest reduction of the reproduction number Rt in countries with the highest viral transmission, as in countries with low transmission, the virus may emerge in settings where transmission is harder to control.<\/p>\n
Another aspect is that the reproduction number Rt without control measures differed between countries, being estimated higher in the Netherlands, Belgium, and the UK compared to Germany, Denmark, and Sweden. A higher reproduction number Rt without control measures requires a more effective response to bring the reproduction number Rt below 1. This explains also the substantial differences between approaches with transferring the relative reduction in Rt without control measures (reduction in transmission intensity) compared to transferring the absolute Rt (transmission intensity). Countries with a lower Rt without control measures tended to perform relatively better when using the transposition of absolute Rt. However, we considered the relative reduction in Rt to be the preferred approach, as it approximates the reduction in contact rates due to NPIs and provides a straightforward mechanistic interpretation. Besides, this approach is regarded as one of the recommended practices for measuring the effectiveness of non-pharmaceutical response measures [6<\/a>]. Nonetheless, estimation of Rt without control measures could also be influenced by seeding of infections, a factor that diminished after control measures discouraged or even banned international travel.<\/p>\n Our finding that small fluctuations in the reproduction number during a fast-growing epidemic can significantly impact mortality rates aligns with previous studies. For example, a UK study estimated a potential 73% decrease in COVID-19 deaths if lockdown measures were implemented 1\u00a0week earlier in spring 2020 [15<\/a>], while similar analyses for Sweden suggested a 34\u201340% reduction in deaths by May 2020 with a lockdown similar to Denmark or Norway [16<\/a>, 17<\/a>]. Our analysis builds upon the previous study of Mishra et al. [5<\/a>], incorporating three additional countries: the Netherlands, Belgium, and Germany. This expansion extends the intercountry comparisons from 6 to 30, facilitating the comparison of neighboring countries with similar initial trajectory of COVID-19 mortality in the initial wave (Germany similar to Denmark and Sweden, and Netherlands similar to Belgium and the UK). Our finding that the mortality does not only depend on the response of a donor country but also on the characteristics and epidemiological situation of the recipient country was only possible through this larger-scale intercountry comparison.<\/p>\n Our analysis comes with several limitations. Firstly, we applied the same delay between infection and death across countries. While the median delay time from symptom onset to death in the Netherlands was previously estimated at 11\u00a0days [18<\/a>], consistent with the data from England used in this study, this assumption may not hold across all studied countries. Secondly, we relied on exchanging reproduction numbers derived from time series of confirmed COVID-19 deaths, which can be influenced by reporting quality and case definitions. However, alternative national vital statistics data usually report deaths on a weekly basis, while daily data is needed to accommodate the small differences in timing of interventions between the selected countries. Moreover, the reproduction number is a relative measure, meaning that comparisons between mortality rates remain consistent if the level of underreporting is stable over time. The Netherlands had about 35% underreporting of COVID-19 deaths compared to excess deaths in the first wave [18<\/a>], while Belgium had no underreporting [19<\/a>]; nonetheless, the Netherlands also peaked earlier than Belgium in hospitalization rates [1<\/a>], indicating consistent findings across different outcomes.<\/p>\n Careful distinction between the counterfactual assessments and the actual implementation of a different response in another country is essential. The counterfactual assessments reflect the response as measured by the reduction in the reproduction number Rt, which is broader than a policy response, as it also encompasses adherence to voluntary recommendations and associated rules. Moreover, implementing a policy response in another country requires the local ability to facilitate measures for discretionary rules and enforce mandatory rules, similar to how it is done in the original country. Therefore, quantifying the implementation of an actual policy response from another country should ideally account for a range of factors, such as variations in healthcare systems, legal systems, culture, public trust in governmental institutions, socioeconomic status, and the nature of the workforce (e.g., possibilities for remote working), which is rather complex. Additionally, time-varying aspects, such as responses in neighboring countries or the potential risk of exceeding healthcare capacities, play a role in public support for control measures.<\/p>\n Our analysis does not account for the active steering of control measures based on the epidemiological situation in each country. Continuous epidemic monitoring will lead to intensified control measures when the current set proves insufficient to curb rising mortality, and to relaxation of measures when mortality is low or control appears overly stringent. However, inherent to delays in monitoring the impact of these measures, from implementation to decreasing rates of hospitalization or mortality due to COVID-19, we believe that adaptive control likely will not alter the presented outcomes.<\/p>\n These findings should also be interpreted considering the limited duration of the study period and available knowledge at the time. For instance, during the first COVID-19 wave, it was unknown that an effective vaccine would become available within a year, that individuals with mild infections could suffer from post-COVID conditions, and that several new, more transmissible variants would emerge within 2\u00a0years, each with different illness severity. Moreover, different responses affect the speed of (herd) immunity buildup, potentially leading to varied outcomes when evaluating the same strategies over a longer period.<\/p>\n Our study contributes to discussions about the merits of the different approaches taken in European countries. They demonstrate that the outcome of response is determined not only by the response itself but also to a large extent by small differences in the initial epidemiological situation in each country. Some countries had relatively low mortality rates for any of the six responses evaluated here, and these countries could afford a response that was less stringent; other countries faced relatively high mortality rates for any of the six response evaluated here, and these countries could ill afford a less stringent response. This underscores that a proper response has to be carefully tailored to the epidemiological situation in each country.<\/p>\n","protected":false},"excerpt":{"rendered":"This analysis quantifies the potential impact of different pandemic responses on COVID-19 mortality in six Northwestern European countries…\n","protected":false},"author":2,"featured_media":56071,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5310],"tags":[3967,29580,1117,29578,2000,299,3968,1824,20181,29581,2985,29579,21466],"class_list":{"0":"post-56070","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-germany","8":"tag-biomedicine","9":"tag-counterfactual","10":"tag-covid-19","11":"tag-effectiveness","12":"tag-eu","13":"tag-europe","14":"tag-general","15":"tag-germany","16":"tag-medicine-public-health","17":"tag-modeling","18":"tag-mortality","19":"tag-response-strategies","20":"tag-sars-cov-2"},"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@uk\/114412714973754433","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts\/56070","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/comments?post=56070"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts\/56070\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/media\/56071"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/media?parent=56070"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/categories?post=56070"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/tags?post=56070"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}