IEA International Epidemiological Association International Journal of Epidemiology, 2021, 390–390f doi: 10.1093/ije/dyab027 Data Resource Profile Data Resource Profile Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 Data Resource Profile: COVerAGE-DB: a global demographic database of COVID-19 cases and deaths Tim Riffe ,*,† Enrique Acosta†; and the COVerAGE-DB team‡ Laboratory of Population Health, Max Planck Institute for Demographic Research, Rostock, Germany *Corresponding author. Max Planck Institute for Demographic Research, Konrad-Zuse-Str. 1, 18057 Rostock, Germany. E-mail: riffe@demogr.mpg.de † Co-first authors ‡ The complete list of authors is given at the end of the article. Editorial decision 18 January 2021; Accepted 9 February 2021 Data resource basics contains daily counts of COVID-19 cases, deaths and tests performed, by age and sex, for 108 national and 371 sub- Information about pandemic dynamics is crucial to under- national populations around the world, depending on the stand the potential impacts on populations, design mitiga- tion strategies and evaluate the efficacy of their available data for each source. The date range available for implementation. Centralization, standardization and har- each country or subpopulation varies. In several country monization of data are critical to enable comparisons of series, the database includes the earliest confirmed cases in the demographic impact of COVID-19 which take into ac- January 2020. For most populations, the database includes count differences in the age and sex compositions of con- daily time series, beginning from an initial starting date firmed infections and deaths. The international data when the data were first released or collected by our team. landscape must keep pace with the global march of the Figure 1 displays a map of countries included in the data- pandemic, and researchers must work to triangulate the base, indicating at least one subnational population from available data to create comparable measures to monitor 13 countries. A detailed overview of data availability is and predict its demographic impacts. given in a searchable table: [https://bit.ly/3kVDrLD]. COVerAGE-DB aims to provide global coverage of key demographic aspects of the COVID-19 pandemic as it unfolds in an up-to-date, transparent and open-access for- Data collected mat. COVerAGE-DB offers data with standardized count Official counts of COVID-19 cases, deaths, and tests are measures by sex and harmonized age groups, which is a extracted from reports published by official governmental necessary but not sufficient condition to allow compari- institutions, such as health ministries and statistical offices. sons between populations at national and subnational Depending on the source, data are collected in a variety of scales. formats, including machine-readable files, pdf tables, html The database is currently under expansion through both tables, interactive dashboards, press releases, official the increase in coverage of national and subnational popu- announcements via Twitter, and in a few instances, from lations and the inclusion of more recent periods as the pan- digitized graphics. A full list of data sources is available in demic continues. At the time of writing, the database a dashboard view [https://bit.ly/2Qg1MxL]. C The Author(s) 2021. Published by Oxford University Press on behalf of the International Epidemiological Association. V 390 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. International Journal of Epidemiology, 2021, Vol. 50, No. 2 390a Generally, COVID-19 cases, deaths and tests in age Data resource use groups are reported as counts, but some sources report Since collection efforts began for COVerAGE-DB in late data in other metrics (fractions, percentages, ratios) or as March 2020, we are aware of 15 studies using the data, summary indicators such as case fatality ratios (CFRs) by many of which provide R code online and are fully repro- age. Reported age intervals vary by source, ranging from ducible. Broadly, these studies aim to measure the single ages to 30-year or greater age bands, and sometimes influence of demographic factors on mortality from reported age intervals change over time within sources. COVID-19,3,4 assess the pandemic impact on health and Usually data are reported as cross-sectional snapshots of mortality within5,6 and across populations,7–12 analyse Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 cumulative counts, but some sources give full time series of COVID-19 data availability and quality,13 propose meth- new cases or deaths, in which case we cumulate counts odological innovations that allow comparisons of CFRs14 over time. We also collect standard metadata on each of and the development of indirect methods to estimate infec- the sources to capture various characteristics of the col- tions in the population.15,16 The database is also used to lected data, such as the primary collection channels, defini- monitor COVID-19 impacts in particular age ranges. For tions used and notes on major disruptions or events. An instance, UNICEF has used the database for monitoring overview of key fields from these metadata is shared as a the burden of the pandemic on children around the spreadsheet [https://bit.ly/2FAmKFn]. world17 and the UN Department of Economic and Social Affairs has used it similarly to focus on older age groups.18 As an example of the analyses that COVerAGE-DB ena- Data production bles, Figure 2 displays changes in the relation between age- All source data are entered into standard spreadsheet tem- specific deaths and cases rates in Colombia, inspired by plates hosted in a central folder on Google Drive. Data en- Figure 1 of Dudel et al.14 We divide both cases and deaths try into the templates is either manual or automatic, in each age band by the respective population sizes. depending on the source. Diagonal lines indicate age-specific CFRs. The graph illus- R programs collect data from the source templates and trates a sharp increase in CFR over age for each sex, and compile the merged input database. The merged input file displays considerable sex differences. For instance, men is then subject to a series of automatic validity checks. aged 60–69 in Colombia have almost the same CFR (ap- Initial checks are carried out by the individual responsible proximately 12% risk of death after COVID-19 disease di- for data collection and entry, using an interactive applica- agnosis) as women aged 70–79. tion [https://mpidr.shinyapps.io/cleaning_tracker/]. Data We repeat this exercise to compare Colombia with are then harmonized to standard metrics (counts), meas- Mexico (see Figure 3), where standardizing by population ures (cases, deaths, tests) and age bands (5- and 10-year size is more justified. CFRs and death rates are much age intervals). Harmonization procedures include rescaling higher in Mexico than in Colombia in each age band— to ensure coherence between age distributions and reported around 2-fold—except for ages 80þ, which show a sub- total counts. Age group harmonization is done using the stantial reduction in the CFR difference, and much higher penalized composite link model for ungrouping1 which death rates for Colombia. was designed for splitting histograms of count data. This comparison between Colombia and Mexico allows Output data also include a file containing selected diagnos- us to illustrate several issues in data quality to be consid- tics of data quality, such as completeness of age reporting, ered when comparing COVID-19 outcomes between popu- for each source and date. lations in general. Besides the economic and sanitary The complete details on all steps of production are conditions that make Latin American countries more vul- available in the COVerAGE-DB Method Protocol, which nerable to the pandemic, the lack of unambiguous defini- is publicly available on the web.2 A table listing which tions of COVID-19 cases and deaths and the limited adjustments are applied to each population is available on testing capacity represent major challenges for data quality the project website [https://bit.ly/2E61BSV]. The merged assessment.19–21 We focus here on definitions and testing input database, the harmonized output and the data qual- strategies. ity files are uploaded daily as zipped csv files to an Open With respect to COVID-19 case and death definitions, Science Framework repository (OSF) [https://osf.io/mpwjq/]. criteria have varied since records started. At the time of A GitHub repository [https://bit.ly/2YbtPCJ], which is data retrieval, both countries use laboratory, clinical and linked to OSF, contains all R scripts used in the complete epidemiological criteria to confirm SARS-CoV-2 infec- production pipeline, including compilation, diagnostics and tions.22,23 However, the vast majority of COVID-19 cases harmonization. and deaths are confirmed with RT-PCR tests results in 390b International Journal of Epidemiology, 2021, Vol. 50, No. 2 Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 Figure 1 Availability of national and subnational information on COVID-19 cases, deaths, and tests in the countries included in the database as of 7 January, 2021. Figure 2 Relationship between deaths and cases per 100 000 population Figure 3 Relationship between deaths and cases per 100 000 population by age group and sex in Colombia, until 7 November 2020. Diagonal by age group in Mexico and Colombia, until 7 November 2020. lines indicate the case fatality ratio. Diagonal lines indicate the case fatality ratio. delays in diagnosis and death registration are retrospec- both populations (99.6% and 91.6% in Colombia and tively adjusted. Mexico, respectively).24,25 Regarding the definition of Differences in testing capacity and strategy between tests, whereas in Colombia it refers to laboratory samples countries are also key determinants for infection diagnosis. tested (4.5 M as of 7 November 2020), in Mexico it alludes Given both the magnitude of contagion and limited resour- to persons (2.3 M). Because individuals may be tested more ces in the region, Latin American countries have struggled than once, comparison between these two units is not to increase testing capacity proportionally to the spread of straightforward. Testing performance measures, such as the infection.28,29 Although with very limited capacity, the positive rates (e.g. 30% in Colombia and 45% in testing approach of Colombia has been to test as many sus- Mexico26), are essential for interpreting differences in cases pected cases as possible. In contrast, an important part of and deaths across populations, because they help to assess the test strategy in Mexico has focused on inferring the ex- the extent of infection under-reporting.27 However, differ- tent of contagion in the population by using nationally rep- ences in test definitions pose serious challenges for direct resentative samples (known as Centinela, which represent comparisons. Dates in both sources are comparable, corre- 36.5% of all confirmed infections at the date under obser- sponding to the occurrence of events. Since information vation), and it has gradually included a small proportion from both sources relies on individual-level databases, of suspected infections outside the Centinela system.23 International Journal of Epidemiology, 2021, Vol. 50, No. 2 390c On 7 November 2020, Colombia performed five times alongside the harmonized counts, as well as the code used to more tests per capita than Mexico. These differences in harmonize the different input measures, metrics and age testing regimes between both countries may account for a groups into comparable granular output metrics. All scripts substantial part of the CFR discrepancies observed in are written in the open-source R programming language.30 Figure 3. The data sources and limitations are documented for each The differences in definitions and testing strategies be- country in a standard metadata framework. tween populations highlight challenges in making compari- A limitation of the COVerAGE-DB is the heterogeneous sons and also the need to produce data with sufficient and difficult-to-evaluate quality of the underlying data. No Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 detail to adjust for biases. For this reason, alongside data single data source can currently claim accurate estimates of on cases, deaths and tests, COVerAGE-DB offers addi- COVID-19 incidence or fatalities. Age-specific case counts tional information on metadata and quality metrics that are highly dependent upon the testing capacity,31 testing are needed for a cautious interpretation of the data and strategy32 and differences in the definition of cases across their limitations. It is our view that researchers should tri- sources and over time. Recorded cases underestimate infec- angulate creatively from all available data rather than tions everywhere, with underestimation expected to vary by avoid difficult comparisons. age, given the relationship between age and case severity.33 The accuracy of diagnostic RT-PCR tests used to confirm infections is also known to vary.34 Furthermore, at any Strengths and weaknesses given date, cumulative counts are underestimated because of Since the beginning of the pandemic, it has been evident that the lag between infection and a positive test result.35 population characteristics are key to understanding the Death counts from COVID-19 are also likely underesti- prevalence, spread and fatality of COVID-19 across coun- mated for similar reasons and also due to various kinds of tries. However, data on cases, deaths and tests disaggregated delays in death registration. Media reports have circulated by age and sex are not easily comparable across countries, about intentional data manipulation in some of the official and sometimes not even accessible. The main strength of data covered in the database.36 Excess all-cause mortality COVerAGE-DB is to provide a centralized, open-access and has been observed across many regions.37–40 Although fully reproducible repository of age-and sex-specific case, some of these deaths likely are from postponing or forego- death and test counts from COVID-19, collected from offi- ing treatment from non-COVID-19-related causes, the cial sources and harmonized to standard output formats. magnitude of this excess is suggestive that numerous The data harmonization process is transparent, following a COVID-19-related deaths are classified under different strict protocol.2 The initial input data are provided causes. Populations also differ in whether deaths of Figure 4 View of the Open Science Framework (OSF) repository, File section [https://osf.io/mpwjq/files/]. To download data files, click on Data, and se- lect one of the files. 390d International Journal of Epidemiology, 2021, Vol. 50, No. 2 Table 1 The main data files, a description of their content, Data testing or excess mortality data,26 the COVID-19 dash- and their stable URLs board of Johns Hopkins,43 the World Population Prospects database44 and the Short Term Mortality Fluctuations data- Filename Description Stable URL base.40 Moreover, given that we have near-complete time se- 1. inputDB.zip Data in original metrics, [https://osf.io/ ries capturing the whole pandemic curve in some places, measures and age 9dsfk/] careful modelling of lag structures might allow some of these intervals data-driven biases to be estimated. 2. Output_5.zip Data with standardized [https://osf.io/ metrics and measures, 7tnfh/] Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 and harmonized age Data resource access groups in 5-year intervals Both merged input and harmonized output files can be 3. Output_10.zip Data with standardized [https://osf.io/ downloaded directly from the OSF site [https://osf.io/ metrics and measures, 43ucn/] mpwjq doi: 10.17605/OSF.IO/MPWJQ, which contains a and harmonized age folder called ‘Data’ with four files of primary data. groups in 10-year Figure 4 shows where to find the files in the OSF intervals 4. qualityMetrics.zip Selected data quality [https://osf.io/ repository. indicators by location qpfw5/] Each of the main data files has a stable link (see and date Table 1) which always points to the most recent version. Each file is a zipped csv file by the same name. For stable links to download particular versions, click on the version suspected COVID-19 cases are included in official statistics number in the Version column seen in Figure 4. Users can and in post-mortem practices when an infection is sus- note versions either by referring to timestamps provided in pected.41 Some populations only report deaths occurring in the headers of data files or by referring to OSF file version hospitals, neglecting a potentially sizeable proportion of numbers, which increment with each daily update. deaths occurring in institutional settings and at home.42 A data dictionary is given in both the OSF wiki [https:// Most populations currently report all deaths to confirmed osf.io/mpwjq/wiki/home/] and the Method Protocol.2 Files SARS-CoV-2 infections as COVID-19 deaths for this data- are shared in csv format to be as universally accessible as base, but the underlying cause of death eventually reported on the death certificate may differ in patients with severe comorbidities. To mitigate biases and misinterpretations Profile in a nutshell due to different practices and definitions, such information • COVerAGE-DB is an open-access database including is constantly updated and documented in the metadata of the database which are freely accessible to users. Further, a cumulative counts of confirmed COVID-19 cases, supplementary data quality metrics file contains a suite of deaths and tests by age and sex. Original data and data quality indicators that is easily merged with the main sources are provided alongside data and measures output data. Quality metrics include age-reporting com- in age-harmonized formats. • The database is in continuous development. It pleteness, some indicators on how aggressive age harmoni- zation is, and two positivity measures from Our World in includes data since January 2020, and as of 7 Data database on COVID-19 testing.26 January 2021, it includes 108 countries and 371 All of these issues compromise the comparability of the subnational areas. • The database also documents variations in definitions data contained within the COVerAGE-DB, both across pop- ulations at any given time and within populations over time. of all input data and indicators of reporting That is, the database enables direct calculation of age- completeness across sources and over time. • An international team, composed of more than 60 specific CFRs, but one must be careful when making com- parisons. Care must also be taken not to interpret calculated researchers, contributed to the collection of data and CFRs as infection fatality ratios, the latter of which include metadata in COVerAGE-DB from governmental both detected and undetected SARS-CoV-2 infections in the institutions, as well as to the design and denominator. Proper estimation of incidence and fatality, implementation of the data processing and validation and of total demographic impacts, will likely require triangu- pipeline. We encourage researchers interested in lating data across numerous sources as these become avail- supporting this project to send a message to the able. To this end, the COVerAGE-DB was designed to be email: [coverage-db@demogr.mpg.de]. easily merged with other databases such as the Our World In International Journal of Epidemiology, 2021, Vol. 50, No. 2 390e possible. A guide to getting started using the data in R is économique, France; 13University of Texas San Antonio, USA; 14 also provided [https://bit.ly/3g8nIVU], to merge Centre for Demographic Studies, Spain; 15Centre national de la recherche scientifique, France; 16Bursa Uludag University, Turkey; COVerAGE-DB with other databases, and tips for other 17 Federal University Oye Ekiti, Nigeria; 18University of the statistical packages may also be added. Users are encour- Witwatersrand, South Africa; 19Institut national d’études démo- aged to reach out for further information or advice on us- graphiques, France; 20SCOR Global Life, France; 21CIDEHUS, ing the database, or to express interest in the project at: Universidade de Évora, Portugal; 22University of Southern [coverage-db@demogr.mpg.de]. Denmark, Denmark; 23University of Liverpool, UK; 24Nasiriyah Technical Institute, Southern Technical University, Iraq; 25Tel Aviv University, Israel; 26Muséum national d’histoire naturelle, France; Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 Acknowledgements 27 Princeton University, USA; 28Universidade Federal de Minas We gratefully acknowledge the hard work of health ministries and Gerais, Brazil; 29Universidad de Buenos Aires/INDEC, Argentina statistical offices around the world in preparing and disseminating the data included in COVerAGE-DB. Conflict of interest Funding None declared. AvR, JS, and MRN received funding from European Research Council Starting Grant #716323. EA received funding from Social References Sciences and Humanities Research Council (Canada) - Postdoctoral 1. Rizzi S, Gampe J, Eilers PHC. Efficient estimation of smooth dis- grant #756-2019-0768. STL received funding from European tributions from coarsely grouped data. Am J Epidemiol 2015; Research Council - Starting Grant #864616. FU received funding 182:138–47. from the Eunice Kennedy Shriver National Institute of Child Health 2. Riffe T, Rizzi S, Dudel C et al. Method Protocol for the & Human Development of the National Institutes of Health COVerAGE-DB. Report No. 1. 2020. https://osf.io/jcnw3/ (28 #P2CHD047879. RS received funding from Fonds de recherche du January 2021, date last accessed). Québec – Société et culture #2019-B2Z-257115. CL received fund- 3. Medford A, Trias-Llimós S. Population age structure only par- ing from European Research Council - Starting Grant #834103. tially explains the large number of COVID-19 deaths at the old- est ages. Demogr Res2020;43:533–44. 4. Arpino B, Bordone V, Pasqualini M. No clear association Complete list of authors and affiliations emerges between intergenerational relationships and COVID-19 Tim Riffe,1 Enrique Acosta,1 José Manuel Aburto,2 Diego Alburez- fatality rates from macro-level analyses. Proc Natl Acad Sci U S Gutierrez,1 Anna Altová,3 Ainhoa Alustiza,1 Ugofilippo Basellini,1 A 2020;117:19116–21. Simona Bignami,4 Didier Breton,5 Eungang Choi,6 Jorge 5. Uchikoshi F. COVerAGE-JP: COVID-19 deaths by age and sex Cimentada,1 Gonzalo De Armas,7 Emanuele Del Fava,1 Alicia in Japan. SocArXiv, 19 August 2020. doi:10.31235/osf.io/cpqrt. Delgado,8 Viorela Diaconu,1 Jessica Donzowa,1 Christian Dudel,1 Preprint: not peer reviewed. Antonia Fröhlich,1 Alain Gagnon,4 Mariana Garcia-Crisóstomo,9 6. Verdery AM, Smith-Greenaway E, Margolis R, Daw J. Tracking Victor M. Garcia-Guerrero,9 Armando González-Dı́az,9 Irwin the reach of COVID-19 kin loss with a bereavement multiplier Hecker,5 Dagnon Eric Koba,4 Marina Kolobova,1 Mine Kühn,1 applied to the United States. Proc Natl Acad Sci U S A 2020;117: Mélanie Lépori,5 Chia Liu,10 Andrea Lozer,1 M adalina Manea,11 17695–701. 12 13 Lilian Marey, Muntasir Masum, Ryohei Mogi,14 Céline 7. Hulı́ková Tesárková K. Demographic aspects of the COVID-19 Monicolle, Saskia Morwinsky, Ronald Musizvingoza,16 Mikko 15 1 pandemic in Italy, Spain, Germany, and South Korea. Geografie Myrskylä,1 Marı́lia R. Nepomuceno,1 Michelle Nickel,1 Natalie 2020;125:139–70. Nitsche,1 Anna Oksuzyan,1 Samuel Oladele,17 Emmanuel 8. Kashnitsky I, Aburto JM. COVID-19 in unequally ageing Olamijuwon,18 Oluwafunke Omodara,17 Soumaila Ouedraogo,19 European regions. World Dev 2020;136:105170. Mariana Paredes,7 Marius D. Pascariu,20 Manuel Piriz,7 Raquel 9. Sasson I. Age and COVID-19 mortality: A comparison of Pollero,7 Larbi Qanni,1 Federico Rehermann,7 Filipe Ribeiro,21 Gompertz doubling time across countries and causes of death. Silvia Rizzi,22 Francisco Rowe,23 Adil R. Sarhan,24 Isaac Sasson,25 Demogr Res 2021; 44:379–96. Erez Shomron,25 Jiaxin Shi,1 Rafael Silva-Ramirez,4 Cosmo 10. Gonzalez-Garcia N, Miranda-Lora AL, Mendez-Galvan J et al. Strozza,22 Catalina Torres,26 Sergi Trias-Llimos,14 Fumiya International heterogeneity in coronavirus disease 2019 pediatric Uchikoshi,27 Alyson van Raalte,1 Paola Vazquez-Castillo,9 Estev~ ao A. Vilela,28 Muhammad Ali Waqar,1 Iván Williams,29 Virginia mortality rates. Boletı́n Médico del Hospital Infantil de México Zarulli.22 2021; 78:24–28 Affiliations:1Max Planck Institute for Demographic Research, 11. Pifarré I Arolas H, Acosta E, Casasnovas GL et al. Years of life Germany; 2University of Oxford, UK; 3Faculty of Science, Charles lost to COVID-19 in 81 countries. Sc Rep 2021; 11. doi: University, Czechia; 4Université de Montréal, Canada; 5Université 10.1038/s41598-021-83040-3 de Strasbourg, France; 6Ohio State University, USA; 7Universidad de 12. Kippen R. Australian age-sex-specific COVID-19 mortality in in- la República, Uruguay; 8Pontificia Universidad Católica del ternational comparative perspective, to June 2020. Aust Popul Ecuador, Ecuador; 9El Colegio de México, México; 10St Andrews Stud 2020;4:33–38. University, UK; 11Research Institute for the Quality of Life, 13. Lloyd-Sherlock P, Sempe L, McKee M, Guntupalli A. Problems of Romania; 12École nationale de la statistique et de l’administration data availability and quality for COVID-19 and older people in 390f International Journal of Epidemiology, 2021, Vol. 50, No. 2 low- and middle-income countries. Gerontologist 2020; 61: 26. Hasell J, Mathieu E, Beltekian D et al. A cross-country database 141–144 of COVID-19 testing. Sci Data 2020;7:345. 14. Dudel C, Riffe T, Acosta E, van RA, Strozza C, Myrskylä M. 27. Kigozi SP, Kigozi RN, Sserwanga A et al. Malaria burden Monitoring trends and differences in COVID-19 case-fatality through routine reporting: relationship between incidence and rates using decomposition methods: Contributions of age struc- test positivity rates. Am J Trop Med Hygiene 2019;101:137–47. ture and age-specific fatality. PloS One 2020;15:e0238904. 28. Burki T. COVID-19 in Latin America. Lancet Infect Dis 2020; 15. Bohk-Ewald C, Dudel C, Myrskylä M. A demographic scaling 20:547–48. model for estimating the total number of COVID-19 infections. 29. Benı́tez MA, Velasco C, Sequeira AR, Henrı́quez J, Menezes FM, Int J Epidemiol 2021;49:1963–71. Paolucci F. Responses to COVID-19 in five Latin American Downloaded from https://academic.oup.com/ije/article/50/2/390/6277233 by Universitat Autònoma de Barcelona user on 18 May 2021 16. Louca S. COVID-19 prevalence in 161 countries and over time. countries. Health Policy Technol 2020;9:525–59. medRxiv, 2 December 2020. doi:10.1101/2020.12.01.2024 30. R Core Team. R: A Language and Environment for Statistical 1539. Preprint: not peer reviewed. Computing. R Foundation for Statistical Computing. 2020. 17. UNICEF. Averting a Lost COVID Generation. A Six-Point Plan http://www.R-project.org (28 January 2021, date last accessed). to Respond, Recover and Reimagine a Post-Pandemic World for 31. Cohen J, Kupferschmidt K. Countries test tactics in ‘war’ against Every Child. 2020. https://www.unicef.org/reports/averting- COVID-19. Science 2020;367:1287–88. lost-generation-covid19-world-childrens-day-2020-brief (28 32. Bi Q, Wu Y, Mei S et al. Epidemiology and transmission of January 2021, date last accessed). COVID-19 in 391 cases and 1286 of their close contacts in 18. United Nations Department of Economic and Social Affairs, Shenzhen, China: a retrospective cohort study. Lancet Infect Dis Population Division. World Population Ageing 2020 Highlights. 2020;20:911–19. Report No.: ST/ESA/SER.A/451. 2020. https://www.un.org/de 33. Verity R, Okell LC, Dorigatti I et al. Estimates of the severity of velopment/desa/pd/sites/www.un.org.development.desa.pd/files/ coronavirus disease 2019: a model-based analysis. Lancet Infect undesa_pd-2020_world_population_ageing_highlights.pdf (28 Dis 2020;20:669–77. January, 2021, date last accessed) 34. Tang Y-W, Schmitz JE, Persing DH, Stratton CW. Laboratory 19. França EB, Ishitani LH, Teixeira RA et al. Óbitos por COVID- diagnosis of COVID-19: current issues and challenges. J Clin 19 no Brasil: quantos e quais estamos identificando? [Deaths due Microbiol 2020;58:e00512–20. to COVID-19 in Brazil: how many are there and which are being 35. Backer JA, Klinkenberg D, Wallinga J. Incubation period of identified?] Rev Bras Epidemiol 2020;23. doi: 2019 novel coronavirus (2019-nCoV) infections among travel- 10.1590/1980-549720200053. lers from Wuhan, China. Euro Surveill 2020;25:2000062. 20. Rao C. Bulletin of the World Health Organization. Medical 36. Leon DA, Shkolnikov VM, Smeeth L, Magnus P, Pechholdová Certification of Cause of Death for COVID-19. 2020. http://www. M, Jarvis CI. COVID-19: a need for real-time monitoring of who.int/bulletin/volumes/98/5/20-257600/en/ (10 November weekly excess deaths. Lancet 2020;395:e81. 2020, date last accessed). 37. Wu J, McCann A, Katz J, Peltier E. New York Times, 21 April 21. Pe~na R. Y al Tercer Dı́a, Resucitó [And on the third day, he resus- 2020. The Pandemic’s Hidden Toll: Half a Million Deaths. citated]. 2020 https://www.etcetera.com.mx/opinion/y-al-tercer- https://www.nytimes.com/interactive/2020/04/21/world/corona dia-resucito/ (10 November 2020, date last accessed). virus-missing-deaths.html (28 January 2021, date last accessed). 22. Instituto Nacional de Salud. Orientaciones para la Vigilancia en 38. The Economist, 15 July 2020. Tracking COVID-19 Excess Salud Pública de la Covid19 [Guidelines for Public Health Deaths Across Countries. https://www.economist.com/graphic- Surveillance of Covid19]. 2020. http://www.ins.gov.co/Noticias/ detail/2020/07/15/tracking-covid-19-excess-deaths-across-coun Coronavirus/Estrategia%20VSP%20COVID-19%2023072020. tries (28 January 2021, date last accessed). pdf (10 November 2020, date last accessed). 39. Mølbak K, Mazick A. European monitoring of excess mortality 23. Secretarı́a de Salud de México. Dirección General de for public health action (EuroMOMO)Kåre Mølbak. Eur J Public Epidemiologı́a. Lineamiento Estandarizado para la Vigilancia Health 2013. doi: http://doi.org/10.1093/eurpub/ckt126.1133. Epidemiológica y por Laboratorio de la Enfermedad 40. The Human Mortality Database, University of California, Respiratoria Viral [Standardized Guidelines for Epidemiological Berkeley, Max Planck Institute for Demographic Research. and Laboratory Surveillance of Viral Respiratory Disease]. Short-term Mortality Fluctuations (STMF) Data Series. 2020. 2020. https://coronavirus.gob.mx/wp-content/uploads/2020/09/ http://www.mortality.org/ (28 January 2021, date last accessed). Lineamiento_VE_y_Lab_Enf_Viral_Ago-2020.pdf (10 41. Institut national d’études démographiques. Demographics of November 2020, date last accessed). COVID-19 Deaths. 2020. https://dc-covid.site.ined.fr/en/ (17 24. Ministerio de Salud y Protección Social. Casos positivos de August 2020, date last accessed). COVID-19 en Colombia [Positive cases of COVID-19 in 42. Meslé F, Pison G; The Conversation. Comment la France Colombia]. 2020. https://www.datos.gov.co/Salud-y-Protecci-n- compte-t-elle ses morts? [How does France count its dead?]. Social/Casos-positivos-de-COVID-19-en-Colombia/gt2j-8ykr/data 2020. http://theconversation.com/comment-la-france-compte-t- (10 November 2020, date last accessed). elle-ses-morts-135586 (17 August 2020, date last accessed). 25. Secretarı́a de Salud. Base de Datos de COVID-19 - Datos 43. Dong E, Du H, Gardner L. An interactive web-based dashboard to Abiertos Dirección General de Epidemiologı́a [COVID-19 track COVID-19 in real time. Lancet Infect Dis 2020;20:533–34. Database - Open Data General Directorate of Epidemiology]. 44. Population Division, 2019, United Nations. World Population 2020. http://www.gob.mx/salud/documentos/datos-abiertos- Prospects. https://population.un.org/wpp/ (17 August 2020, date 152127 (10 November 2020, date last accessed). last accessed).