key: cord-0864258-kn2pyoew authors: Tragaki, Alexandra; Richard, Jean‐Luc title: First wave of SARS‐COV2 in Europe: Study and typology of the 15 worst affected European countries date: 2021-10-26 journal: Popul Space Place DOI: 10.1002/psp.2534 sha: d42e300fc99b0f87f502a7a5e88f371a3f48d35d doc_id: 864258 cord_uid: kn2pyoew Since 11 March 2020 when officially declared a global pandemic, Covid‐19 (or SARS‐COV2) has turned out to be a multifaceted disease differently affecting countries and individuals. What makes certain countries more vulnerable than others has attracted the interest of scientists from various disciplines. This paper intends to compare the impact of demographic parameters, population health conditions and policy actions on prevalence and fatality levels of Covid‐19 during the first 3 months since its declaration of global pandemic. A country‐level exploratory analysis has been conducted in order to assess how demography, national health conditions and measures taken interact and condition the disease outcomes. Analysis relies on publicly available data on Covid‐19 reported cases, deaths and number of persons tested. Those data are combined with demographic parameters (sex ratio, mean age, population density and life expectancy), health data (cardiovascular death rate, diabetes prevalence, share of smokers among males and females and number of hospital beds) and information about relative national policies aiming the management of the pandemic (lockdown timing and duration). Our analysis confirms the diversity of factors and the complexity of their interaction in explaining the propagation and fatality of the disease across Europe. Our findings question some well‐established attitudes concerning the role of demographic variables and public health conditions in the spread of the disease. health to demography-is anything but surprising. It displays the necessity of interdisciplinary approaches should the nature and the mechanisms of this virus are to be fully understood and its severe social and economic implications are to be efficiently addressed. Uncertainties remain around the epidemiological and clinical characteristics as well as about the determinant factors behind critical cases. However, one persistent pattern has emerged. Data from different countries suggest that, although the probability to be infected does not vary with age or sex, fatality rates are significantly higher among men and persons above 65 years of age (Sobotka et al., 2020) . While the gender imbalance remains largely unexplained, the vulnerability of elder is somehow related to chronic health conditions, another aggravating factor at individual level (Romero Starke et al., 2020) . Cardiovascular diseases, asthma and diabetes have been put forward to defend the 'underlying diseases' argument. Moreover, personal lifestyle choices, mainly obesity and smoking, are also associated with the disease outcome. Different studies point obesity as a culprit in Covid-19 deaths, mainly among males below 60 years of age (Dietz & Santos-Burgoa, 2020; Tartof et al., 2020) . On the other hand, research about smoking as an independent risk factor remains inconclusive. Some first findings report a significant inverse relationship between current smoking and Covid-19 mortality rates whereas others suggest a non-significant positive association (WHO, 2020) . The reasons why some patients sail through the disease whereas for others implications are overwhelming and lead to death are still to be identified. However determinant those parameters may be at individual level, they do not necessarily suffice to justify significant differences across countries. The aim of this work is to explore plausible explanatory factors behind different pandemic outcomes across European countries. This study intends to provide a complementary analysis by analysing the role of demographic parameters, national health conditions and policy actions on risk for death due to the novel coronavirus. The rest of this paper is structured as follows: Section 2 describes the data and methods applied in our analysis, Section 3 presents and discusses the results and Section 4 summarises the most important findings. Analysis relies on country-specific epidemiological results referring to the first 100 days of the pandemic. At a first level, analysis covers 41 European countries with more than 150,000 inhabitants whereas at a second stage, the discussion is narrowed down to the 15 worst affected countries. This work combines data from different databases. Official country-level Covid-19 data (namely, confirmed cases, reported deaths and number of persons tested) are retrieved from national government reports and databases; their values refer to 20 June 2020. This date, 100 days away from the pandemic declaration, has an additional significance as the end of the so-called first wave of the pandemic in Europe. Thereafter, but only for a couple of months, most European countries managed to retain new virus-related death rates at low levels. We opt to focus on the first wave, a period when very little was known about the pandemic and how to deal with it, so as to better identify the role of the demographic parameters, social conditions and the readiness of policy on the spread of Covid-19. This choice is further justified by the fact that the countries were unprepared and variants of the Covid-19 virus had not emerged by the time. Demographic data comprise population size and density, sex ratio, median age and share of above 70 years of age in total population, as provided by the Eurostat (2020) . Health statistics including cardiovascular death rate and diabetes prevalence refer to the year 2017 and are supplied by the World Bank. Data about policy actions, such as date of strict lockdown and its duration (in days), have been compiled by official governmental reports. Official data about regional and national mortality trends during the 22 first weeks of 2020 have also been used to produce the map (Illustration 1). The later data refer to (Dowd et al., 2020) . At the beginning of the pandemic, it has been repeatedly mentioned that older persons and males are disproportionally affected by the new virus (Dudley & Lee, 2020; Sasson, 2020; Verity et al., 2020) . The age structure argument has been put forward to explain the huge outbreak in Italy, the first European country to be harshly hit by the pandemic (Omori et al., 2020) . As shown on Figure 2 , the share of above 70 years of age in a population does not explain the distinction between more and less harshly hit European countries. Among the countries the most affected, the share of elder in the total population ranges from as low as 8% (in N. Macedonia and Ireland) to above 17% (in Italy). Things get slightly more straightforward when it comes to the gender composition. The countries more affected by the pandemic have a sex ratio close or above to 100, indicating a relatively higher share of males. France is one of the two countries with a high level of Covid-19 mortality in spite of medium sex ratio. The high level of mortality in nursing houses where 73% of residents are women explains this specific position (Tragaki & Richard, 2020) . Against common wisdom, high living standards do not provide a national shield against Covid-19. Figure (Nishiga et al., 2020; Ren et al., 2020; Sabatino et al., 2020) . However, this positive association valid at individual level is not confirmed at an aggregate level. Our data suggest a strong statistically significant inverse relationship between Covid-19 and national CVD death rates. More precisely, for an additional 1000 CVD deaths, the expected decrease in the number of Covid-19 per million inhabitants ranges from 44 to 141 (95% confidence level). An inverse, though not statistically significant, correlation (b = À0.293, n = 41, t STAT = À1.987 < 2.0227) between Covid-19 death rates and diabetes prevalence across European countries has been observed ( Figure 4 ). Those really interesting findings are counter-intuitional as longevity is correlated to increased cardiovascular disease rates and CVD deaths are the first cause of death among elder persons (Kollia et al., 2018; . Such findings are presented for the first time in F I G U R E 2 Correlation between demographic parameters (sex ratio and share of above 70 years of age) and Covid-19 mortality rates (per 1M of population). Note: Vertical axis refers to the number of men for every 100 women; horizontal axis refers to the share of above 70 in total population. Each dot refers to a country; the colour of each dot illustrates how high the number of Covid-19 deaths/million persons is. In bold the 15 hardest hit European countries (as mentioned in our analysis) HAC is a useful tool to identify differences and similarities across countries in respect to a number of parameters that can be linked with the dynamic of the pandemic. The first HAC we run takes into Obviously, this divide cannot justify the pandemic dynamics shown on Figure 6 . Demographic variables though important are not decisive in conditioning the diffusion and severity of the disease. In fact, density often appears as more pertinent if densities are examined at regional rather than national level. When HAC is applied on the 15 hardest hit European countries taking into consideration the health conditions, grouping is different to the demographic one but still far from sufficiently explaining the pandemic dynamics. Countries varied substantially in terms of how their healthcare system provide life-saving services: Several countries were less able to rapidly enhance capacity, partly related to uneven health and social care spending, responded less effectively to increased healthcare needs. These countries are characterised by low per capita spending in ICU beds (Kontis et al., 2020) . In several countries, IC capacity in beds delayed admission of patients with COVID-19 or even led to patients' triage. 1 As early evidence suggested that the main way the new virus spreads is either by respiratory droplets among people who are in close contact with each other or aerosol transmission that can occur in specific settings, restrictions were applied in most areas of everyday life. Sooner or later, all European countries implemented social F I G U R E 5 Dendrogram of a hierarchical ascendant classification of 15 hardest hit European countries in respect to their demographic variables. Note: Countries have been classified in respect to the following variables: population density, sex ratio, share of above the age of 70 and life expectancy. Clustering method here used was the nearest neighbour combined with the squared Euclidean distance distancing measures, strongly encouraged populations to follow the 'stay at home' calls and ultimately most of them enforced a total lockdown. Differences concerning the timing and duration of lockdown across countries depend on societal norms regarding accepted levels of risk (Glynn, 2020) . Despite its very serious side effects (mostly economic and psychological), the effect of lockdown in containing the spread of the pandemic has been tested and found positive (Atalan, 2020; Kennelly et al., 2020) . Clustering countries in respect to the date, precocity and duration of lockdown suggests different groups of countries, if compared with demographic and national health variables. It seems however that this grouping better explains the ranking of countries in Figure 1 . The last dendrogram (Figure 7) The nature of policy measures imposed to contain the pandemic has also been examined, this time across the 15 worst affected countries. It seems that the timing the restrictions are introduced as well as the length of period measures remain effective may provide an additional explanatory factor. Though the rule is not general, the fewer the cases and deaths at the time of measures, the higher their efficiency, even among the 15 hardest hit countries. There are some limitations in this study, as it is the case in every work on an unfolding issue. There are discrepancies in the way outcomes are measured across countries and conflicting statements about their measures, the duration and peoples' compliance to them. Moreover, confounding factors, others to those here studied, may turn out to be crucial in the study of the pandemic. International comparisons are useful to make out decisive factors that may affect the outcome. Although nothing guarantees that the second wave experience will be the same, and similar measures will suffice, there are definitely things to be learnt from those countries that did not manage to efficiently address the disease. Data used for figures are available upon request. ORCID Alexandra Tragaki https://orcid.org/0000-0003-0768-3345 Jean-Luc Richard https://orcid.org/0000-0002-6723-4958 ENDNOTES 1 Countries with low level of ICU beds have not necessarily been the most affected by the pandemic. It needs, however, to be reminded that here we focus on the 15 countries most hit by the pandemics. 2 Sweden and Switzerland have been excluded from this clustering for they never proceeded with a total lockdown. 3 The pandemic in North Macedonia followed a different timeline. Three months after the reference date (20 June), the number of deaths was three times higher (cumulated total of 233 deaths on 20 June and 721 deaths on 21 September). Is the lockdown important to prevent the COVID-9 pandemic? Effects on psychology, environment and economy-perspective Obesity and its implications for COVID-19 mortality Demographic science aids in understanding the spread and fatality rates of COVID-19 Disparities in age-specific morbidity and mortality from SARS-CoV-2 in China and the Republic of Korea Protecting workers aged 60-69 years from COVID-19. The Lancet Infectious Diseases The COVID-19 pandemic in Ireland: An overview of the health service and economic policy response Trends of cardiovascular disease mortality in relation to population aging Magnitude, demographics and dynamics of the effect of the first wave of the COVID-19 pandemic on all-cause mortality in 21 industrialized countries COVID-19 and cardiovascular disease: From basic mechanisms to clinical perspectives SARS-CoV-2 elimination, not mitigation, creates best outcomes for health, the economy, and civil liberties The age distribution of mortality from novel coronavirus disease (COVID-19) suggests no large difference of susceptibility by age Association of the insulin resistance marker TyG index with the severity and mortality of COVID-19 The age-related risk of severe outcomes due to COVID-19 infection: A rapid review, metaanalysis, and meta-regression Impact of cardiovascular risk profile on COVID-19 outcome. A meta-analysis Aging and COVID-19 mortality: A demographic perspective Age, gender and COVID-19 infections Obesity and mortality among patients diagnosed with Covid-19: Results from an integrated health care organization Population ageing and cardiovascular health: The case of Greece Premiers mois de l'épidémie de coronavirus COVID19 dans deux pays aux trajectoires différentes, la Grèce et la France, Documents de travail de l'Observatoire démographique de la Méditerranée Estimates of the severity of coronavirus disease 2019: A model-based analysis. The Lancet Infectious Diseases Smoking and Covid-19 Monitoring transmissibility and mortality of COVID-19 in Europe First wave of SARS-COV2 in Europe: Study and typology of the 15 worst affected European countries. Population, Space and Place, e2534