Documentation

Health Security Net is one of a suite of free resources from the Georgetown University Center for Global Health Science and Security that makes data and information about pandemic planning and oversight centralized and publicly accessible. Built on an integrated data architecture to support cross-platform analysis, this global health security library is a searchable and filterable database designed to enable ready access to warnings, evaluations, oversight efforts, strategies, and other documents concerning pandemic risk and related issues, including documents from governmental, international, and non-governmental organizations. The library is a work in progress and continues to be updated as additional resources are identified. Please contact us with any questions or additions at healthsecuritynet@georgetown.edu.

The library includes documents from global, regional, and national-level sources dated from 1972-present. National sources are mostly drawn from the United States, with additional sources reflecting national pandemic influenza planning from other countries. Academic journal literature as it relates to coronaviruses as a global infectious disease threat is also represented. This is an ongoing research project and the library may be updated to further reflect pre-1995 sources, additional countries, and other elements.

The site includes:

1. A searchable, filterable database of all documents in the dataset. The complete dataset can be downloaded in an Excel file format directly from the site.
2. A written analysis from the Georgetown University Center for Global Health Science and Security reflecting key observations about the dataset.

This work and underlying dataset is available for use under the Creative Commons Attribution 4.0 International Public License (https://creativecommons.org/licenses/by/4.0/), with appropriate reference and acknowledgement of the original research teams, as listed on this site.

This documentation includes:

1. The Library and the methodology used to populate it
2. The data coding process
3. The glossary of terms as applied to the Library

Library

Data collection

The modern biothreat can be viewed as a tripartite construct that consists of intentional acts, emerging infectious diseases (EID) from nature, and accidents. Preparedness for that threat may be specific to one of those elements, to individual pathogens of concern, or may be approached generally as “all-hazards.” This library is designed to primarily capture documents relating to federal/national and global officials’ awareness of the EID or naturally-occurring pandemic threat and the risk from that threat. The researchers did not specifically seek information related to intentional acts and biological accidents, but did not exclude such documents where they were relevant to global health security and pandemic preparedness. All-hazards items captured by the search were included only if they also directly addressed the nexus between all-hazards preparedness and infectious disease. The library includes items published from 1972 – present.

Global

The research team collected documents online from multiple global sources and excluded those meeting any of the following criteria: 

• Items published prior to January 1, 1995, except where expert review judged the item foundational to the development and understanding of global health security efforts.
• Items solely focused on endemic diseases or those typically addressed by the global health (as opposed to global health security) community, e.g., HIV, TB, malaria, and noncommunicable diseases.
• Fact sheets on diseases.
• Items strictly about bioterrorism and/or biosafety/biological accidents, except where noted in this Documentation.

Global sources were reviewed for relevant documents. The search strategy varied among them, for instance depending on the sophistication of their websites’ search functionality, or whether the site provided its own relevant terms for grouping. Global sources currently included in Health Security Net include:

1. Food and Agriculture Organization of the United Nations (FAO): FAO holds biennial conferences to discuss and review programs and provide guidance for Member States on health-related activities. Resolutions and decisions from each conference are published in the conference report at http://www.fao.org/unfao/govbodies/gsbhome/conference/conference-reports/en/. Because resolutions and decisions are not published separately from the full conference report and the search function is not sufficiently advanced to search only for terms within the resolutions and decisions, the research team was unable to employ the comprehensive list of search terms used for other entities (e.g., see United States hearing section). The research team therefore reviewed the table of contents of each report from 1995-2019 for the titles of the resolutions and decisions to determine their relevance to pandemics. If the title included the search terms, it was captured; if it contained terms that may otherwise be relevant to infectious disease in the judgement of the researcher, the contents were reviewed to determine inclusion or exclusion. 
2. United Nations General Assembly (UNGA): The United Nations Digital Library (https://digitallibrary.un.org/?ln=en&as=1) was searched in the primary search field for “pandemic” using All of the words and any field. Full text search was by default not toggled on, resulting in 142 hits for UNGA; despite this, the search appeared to capture documents with the search term in the body of the document, not just the title. This search methodology as applied to the Digital Library was not perfectly sensitive; in some cases, upon reading the captured resolutions, the researchers found references to additional resolutions about pandemics, and these were searched for directly and added to the database if relevant. (In some cases, the captured record did not have the word “pandemic” at all.) Once the dataset was ready, exclusion criteria were applied.
3. United Nations Security Council (UNSC): Based on the experience and judgement of Center faculty, the following UNSC resolutions were deemed relevant to global health security: 620, 1308, 1540, 1673, 1810, 1977, 2176, 2177, and 2439. Each of these was logged into the database. The text of each was also reviewed for mentions of other resolutions, exclusion criteria were applied, and any that remained were captured into the database. (All such entries met exclusion criteria and thus are not represented in the database.)
4. World Organisation for Animal Health (OIE): OIE holds annual conferences to discuss and review programs and provide guidance for Member States on health-related activities. At the time of research, OIE published a dedicated website with resolutions and decisions from each conference, which the research team used to access these records. This site is no longer available, but the documents can still be accessed at https://www.oie.int/en/who-we-are/structure/framework/#ui-id-4. Because resolutions and decisions are not published separately from the full conference report and the search function is not sufficiently advanced to search only for terms within the resolutions and decisions, the research team was unable to employ the comprehensive list of search terms used for other entities (e.g., see United States hearing section). The research team therefore reviewed the table of contents of each report from 1995-2019 for the titles of the resolutions and decisions to determine their relevance to pandemics. If the title included the search terms, it was captured; if it contained terms that may otherwise be relevant to infectious disease in the judgement of the researcher, the contents were reviewed to determine inclusion or exclusion. 
5. World Health Assembly (WHA): WHA holds annual conferences to discuss and review programs and provide guidance for Member States on health-related activities. Resolutions and decisions from each conference are published at https://apps.who.int/gb/index.html. Because resolutions and decisions are not published separately from the full conference report and the search function is not sufficiently advanced to search only for terms within the resolutions and decisions, the research team was unable to employ the comprehensive list of search terms used for other entities (e.g., see United States hearing section). The research team therefore reviewed the table of contents of each report from 1995-2019 for the titles of the resolutions and decisions to determine their relevance to pandemics. If the title explicitly included the search terms, it was captured; if it contained terms that may otherwise be relevant to infectious disease in the judgement of the researcher, it was reviewed to determine inclusion or exclusion.

6. World Health Organization (WHO): WHO documents located in the Institutional Repository for Information Sharing (IRIS) database were searched (https://apps.who.int/iris/). Because many of our desired search terms (see United States—Hearings section below) returned thousands of results, and IRIS can only export 500 at a time, the research team instead used the “MeSH” subject categories that IRIS uses to organize its topics. In the IRIS database, we manually searched the “MeSH subjects” category and chose relevant subjects to review.

   “Weekly Update” reports and country-specific items were excluded from the captured sources; regionally-oriented WHO reports were excluded in the first round of data collection. Others that in the researchers’ expert judgement were irrelevant to health security were also excluded. During our data collection starting February 2021, we gathered WHO regional organization documents using the same methodology. We used all of the same MeSH terms and filters.

National (United States)

The research team compiled United States documentation in the form of congressional hearings, government reports and other documents, and third-party reports and other documents. In most cases, a list of 25 search terms were deployed to capture a broad array of hits relevant to health emergency preparedness (see details below). The team excluded documents meeting any of the following criteria:

• Items prior to January 1, 1995, except where expert review judged the item foundational to the development and understanding of global health security efforts.
• Items solely focused on endemic diseases or those typically addressed by the global health (as opposed to global health security) community, e.g., HIV, TB, malaria.
• Fact sheets on diseases.
• Items strictly about bioterrorism and/or biosafety/biological accidents were excluded except where noted in this Documentation.

The research team sought information from the following sources:

1. Hearings: The website www.congress.gov was used as a primary source of information on congressional activity by way of the Congressional Record (CR). The CR captures committee activity in the form of committee hearings, briefings, and business meetings. Records were sought using the following keyword searches:

   biodefense; biological threat; biopreparedness; biosurveillance; biothreat; CBRN; chemical, biological, radiological, and nuclear; coronavirus; Ebola; emerging infectious disease; global infectious disease; health security; infectious disease epidemic; influenza; MCM; medical countermeasure; medical preparedness; medical readiness; medical supply chain security; MERS; middle east respiratory syndrome; pandemic; public health response; SARS; severe acute respiratory syndrome; zika

   Hits were reviewed for the two types of records that contain information on committee activity, the Daily Digest and Senate Committee Meetings. Each record was reviewed and the activity it represented categorized as a hearing, briefing, or business meeting. Hearings were the primary unit of interest, as they represent one of the most public forms of congressional oversight and information-gathering. Briefings, of which only a small minority are noticed in the CR, and business meetings, which almost always represent markups of legislation, were excluded. Exclusion criteria were applied after all hearings consistent with the search terms were identified. Some hearings used the topic of bioterrorism to discuss broader preparedness efforts also relevant to pandemics, and these were included. Some hearings identified from early in the timeframe parameters related to Department of Defense (DoD) budgets use terms like “health security” in a way different from the contemporary usage, or cover the topic of “health readiness” more broadly than this library is meant to capture; these were excluded. When the search occasionally resulted in an announcement for a hearing but no actual notice confirming that the hearing took place, additional research through www.congress.gov, www.gpo.gov, or committee websites was undertaken to find a notice confirming that the hearing was held. Rarely no such notice can be found, and these hearings were excluded on the presumption that they were cancelled. On occasion when the research team was aware of relevant hearings not captured by any of the keywords, these were included for comprehensiveness. All hearing titles were copied from the relevant hearing report.

   In addition to their inclusion in the library, the list of included hearings has also been made available as a supplemental file containing additional metadata. In this document, witness names and affiliations for each hearing were noted from the CR and other sources, including official committee reports available at www.govinfo.gov and committee websites. Witness names were manually standardized to resolve spelling or other discrepancies. Each hearing was tagged as having occurred during either Democrat or Republican control of a given chamber.¹² Some committee names changed across the data capture period; these were grouped with their current names to permit proper committee-level analysis of activity.³ Two hearings were held jointly by multiple committees; for analysis purposes, these were treated as though they were held by separate, unique committees.

2. Government reports. The research team sought reports from sources to which decision-makers in the Executive and/or Legislative branches could or should reliably be exposed. Three primary groups were considered: independent advisory bodies such as federal advisory committees; departments and agencies with significant operational jurisdiction over and spending on EID-relevant activity;⁴⁵ and government bodies whose specific role is to advise decision-makers.

   Rosters of federal advisory committees for the Department of Health and Human Services (HHS) (specifically for the Centers for Disease Control and Prevention, National Institutes of Health, and Office of the Assistant Secretary for Preparedness and Response), the Department of Homeland Security (DHS), and DoD were reviewed; those with some responsibility for biodefense were further researched for all published reports, and those consistent with the same parameters used for hearing adjudication were retained. Reports issued by the President’s Council of Advisors on Science and Technology were included in the same manner. Because the body of work issued by each of these entities is relatively small and not contained in searchable databases, we did not apply the hearing search terms but rather examined the available report titles, and content where necessary, to judge whether reports were relevant.

   All Worldwide Threat Assessments issued by the Director of National Intelligence (and their precursors by a different name) were identified; those that reported an infectious disease threat or risk were included. Reports from Inspectors General (IG) of DHS, DoD, HHS, U.S. Agency for International Development (USAID), and Department of Veterans Affairs (VA) were sought on their respective websites; the search terms (same as those used for hearings) were input into the search bar for DHS, DoD, HHS, and VA. The HHS IG additionally provides a pre-populated category of “emerging infectious disease preparedness and response” reports and these reports were also captured.⁶ USAID presents its reports by category; from the “Global Health” page we manually reviewed all entries to extract relevant reports. The HHS Public Health Emergency Medical Countermeasures Enterprise (PHEMCE) collates and makes a variety of reports available on its website, and from these we pulled the organization’s multi-year budgets. For all of these searches, the standing exclusion criteria were applied; those deemed topically irrelevant or focused on areas like expenditure audits were also omitted.

   While these and other departments and agencies have issued many other EID-relevant reports throughout the timeframe of interest, these reports are often one-offs and are not available in any systematized fashion to the authors’ knowledge. The research team captured some of these by the methods outlined in United States Step 5.

   Two legislative branch agencies that provide analyses to Congress were also included: the Congressional Research Service (CRS) and the Government Accountability Office (GAO). The same search terms and date parameters used for hearings were applied. For CRS, the primary search bar in the public database (crsreports.congress.gov) was employed, and for GAO, the primary search bar on its website (www.gao.gov) was used. CRS reports are sometimes updated and re-released at a later date; in these cases, we documented the most recent date and uploaded the most recent report.

   We also sought to identify federal strategies and implementation plans relevant to possible biothreats. We used the following sources for this information: the HHS Public Health Emergency website,⁷ which compiles several relevant strategies and plans, and two published collations of key strategies, executive orders, and related documents.⁸⁹

   In a small number of cases, the researchers became aware of reports that were not captured by the search strategy, for instance, when a captured GAO testimony is based on a series of prior GAO reports, some of which may not themselves have been captured; these were then included for comprehensiveness.

3. Third-party reports. This category includes non-governmental organization, academic, and private sector reports. Researchers consulted the University of Pennsylvania TTSCP Global Think Tank Index, published annually, to compile a list of all institutions that have appeared among the top ten in the “Top Think Tanks in the United States” category in any year from 2008 through 2019.¹⁰ From this list, we excluded the Pew Research Center, the National Bureau of Economic Research, and the Peterson Institute for International Economics due to their specialized research focuses on other topics. The final list consisted of the following organizations: the American Enterprise Institute, the Atlantic Council, the Brookings Institution, the Carnegie Endowment for International Peace, the Cato Institute, the Center for American Progress, the Council on Foreign Relations, the Center for Strategic and International Studies (CSIS), the Heritage Foundation, the Hoover Institution, the Hudson Institute, Human Rights Watch, RAND Corporation, the Urban Institute, and the Wilson Center. The team consulted the website of each think tank to view its published reports. In addition to the standing exclusion criteria, we excluded commentaries, blog posts, and other informal products (as opposed to formal reports). For each site, we searched for “pandemic” and manually reviewed titles of resulting reports, as well as texts when needed, to determine relevancy.
   
   The RAND Corporation and Council on Foreign Relations sites categorize reports by topic and have particularly relevant categories for global health security (e.g., “Public Health Threats and Pandemics”). We manually reviewed all reports in these relevant categories in lieu of using a site search. We also manually reviewed all reports in relevant categories from the CSIS and Hudson Institute sites, but due to concerns about the comprehensiveness of these sites’ categories, the search term “pandemic” was also applied to these two sites.
   
   Conducting these searches and manual reviews and applying exclusion criteria led to document inclusion from the following institutions: the American Enterprise Institute, Brookings Institution, the Center for American Progress, the Council on Foreign Relations, CSIS, the Heritage Foundation, the Hoover Institution, the Hudson Institute, and RAND Corporation.
   
   For the National Academies of Sciences, Engineering, and Medicine (NASEM), the National Academies Press reports (www.nap.edu) were searched within the topics “Biology and Life Sciences” and “Heath and Medicine,” dated from 1995-2019; the same search terms and date parameters used for hearings were applied. The researchers were aware from experience about the existence of one relevant NASEM report not captured by any of the keywords and included it for comprehensiveness. All reports from NASEM (including Institute of Medicine or other such sub-NASEM publishers) were titled as “National Academies of Sciences, Engineering, and Medicine” as the publishing organization.
   
   To capture additional influential reports from non-governmental organizations, researchers compiled a list of non-governmental organizations who had published at least one document already included in the database, but whose sites had not yet been fully searched. (These organizations’ documents had been included through the methods described in 5: Other methods and 2: Simulations and exercises below.). The resulting list consisted of the following organizations: the American Hospital Association, the Association of Public Health Laboratories, the Association of State and Territorial Health Officials, the Bill and Melinda Gates Foundation, the Center for Transatlantic Relations at Johns Hopkins University, the Council of State and Territorial Epidemiologists, the Institute for Defense Analyses, the Johns Hopkins Center for Civilian Biodefense Studies, the Johns Hopkins University Center for Health Security, the Memorial Institute for the Prevention of Terrorism, the National Association of County and City Health Officials, the National Governors Association, The Lewin Group, the Transatlantic Biosecurity Network, Trust for America's Health, and the World Economic Forum. The team consulted the website of each listed organization to view its published reports, if any were publicly available.
   
   For the majority of sites that featured publicly available reports, we manually reviewed all reports in relevant topic categories provided by the website. For the Johns Hopkins University Center for Health Security, we manually reviewed all reports. For the Institute for Defense Analyses, we searched “pandemic” and “infectious disease” and reviewed the search results. For the Association of State and Territorial Health Officials, we reviewed all reports in the category of “infectious disease,” and also searched for “pandemic” and reviewed the search results. For the World Economic Forum, we reviewed all reports in the category of “Shaping the Future of Health and Healthcare,” and also searched for “pandemic” and “infectious disease” and reviewed the search results. In all cases when reviewing reports, we first reviewed report titles, and then full texts where needed to determine relevancy. In addition to the standing exclusion criteria, we excluded commentaries, blog posts, and other informal products (as opposed to formal reports). We also excluded links to works published elsewhere (for example, links to academic journal articles), and also excluded congressional testimonies, in order to avoid duplication of existing information on congressional hearings (see 1: Hearings above).
   
   Conducting these searches and manual reviews and applying exclusion criteria led to document inclusion from the following institutions: the Association of Public Health Laboratories, the Association of State and Territorial Health Officials, the Institute for Defense Analyses, the Johns Hopkins University Center for Health Security, the National Governors Association, Trust for America’s Health, and the World Economic Forum.
   
   In some cases, the publishers of third-party reports do not provide permission for posting the report on other websites; in these cases, information about the report was still provided for comprehensiveness, along with a hyperlink for accessing it elsewhere.
4. Executive Orders and Presidential Directives. This category includes presidential review directives, presidential decision directives, and executive orders issued by the president of the United States. For this category of documents alone, we expanded our inclusion criteria to also include documents that (1) only directly discussed biological events of intentional and/or accidental origin, but (2) nevertheless provided insight to the priorities and frameworks guiding American health security policy at the time.
   
   To gather executive orders, we consulted the Federal Register, which provides an online archive of all executive orders issued since 1994.¹¹ We examined the titles and, where necessary, the full texts of all executive orders issued from 1995-2021 to determine relevancy.
   
   To gather presidential directives, we consulted the online collection of unclassified and declassified presidential directives maintained by the Federation of American Scientists (FAS).¹² For each review directive and decision directive issued from 1995-2021 for which the FAS had obtained a declassified or unclassified copy, we reviewed the title and, where necessary, the full text to determine relevancy.
5. Other methods. To capture additional relevant documents that were not already captured by other methods, documents referenced by GAO reports already included in the Library (see U.S. Step 3) were reviewed. We used document titles and, if needed, document contents to assess relevancy. In addition to the standing exclusion criteria, articles in academic journals, situation reports, laws (including treaties), entries in the Federal Register, and items that could not be located online using an internet search were excluded. In some cases, GAO reports included in the Library by the methodology specified in U.S. Step 3 referenced another GAO report that we deemed relevant; in these cases, we included the referenced GAO report, but did not in turn consider the second report’s references for inclusion.

After all documents were captured, duplicates and those that based on the researchers’ expert judgement were irrelevant to health security and/or pandemics were excluded.

Other

1. Academic journal literature. Journal literature was searched using Ovid Medline (https://ovidsp.ovid.com/). The following Boolean search was performed using the Advanced Search tool: (exp Coronavirus Infections/ or exp Coronavirus) AND (exp Risk Assessment) as limited to articles in English published prior to January 1, 1995. The research team reviewed each article to exclude any that met the following exclusion criteria (in addition to the standing exclusion criteria) regarding article focus: clinical management; non-human species; bioterrorism; or biowarfare. Articles behind paywalls were not posted; open access articles were posted if the research team received permission to do so. When the paper was unavailable or behind a paywall, the research team reviewed the piece and summarized the item to reflect the comprehensive search results. On occasion, journal articles were also captured through other means—specifically, as a result of the search of the WHO IRIS database (see Global Step 6), CDC Morbidity and Mortality Weekly Report, or when a journal record was an editorial that flagged one or more relevant papers on coronavirus within that issue.
2. Simulations and exercises. Databases and other sources that might have records of simulations or exercises were initially identified by a Google search of the keywords “infectious,” “disease,” “tabletop,” “exercise,” “report,” “readout,” and “simulation” in various combinations. The research team looked for records that described a national or international simulation or exercise that was conducted, and ideally provided a read-out. Exercises were dated according to the date the exercise itself was conducted, rather than according to the date the read-out was released. The team additionally collected infectious disease tabletop exercise toolkits or templates. These toolkits were coded as “Report.”
   
   Through this method, key databases were identified, such as the Homeland Security Digital Library (https://www.hsdl.org/c/), the National Institute of Health Emergency and Disaster Response Training Exercises (https://dr2.nlm.nih.gov/training-exercises), the Health and Human Services TRACIE website (https://asprtracie.hhs.gov/technical-resources/7/exercise-program/1#after-action-reports) and the United Nations Office for Disarmament Affairs’ list of current activities involving the Implementation Support Unit (https://www.un.org/disarmament/biological-weapons/implementation-support-unit/relevant-activities-overseen-by-the-isu). The same keyword searches were performed in these databases and the hits reviewed for relevance. To review for any potentially missed records, the Department of Homeland Security website (https://www.dhs.gov/) was examined via a parallel search in its primary search bar. The exercise summaries for those exercises mentioned in Maxmen and Tollefson 2020¹³ were also sought.
   
   Subject matter experts within the Center for Global Health Science and Security provided suggestions for additional websites and exercises to review. This included the UN Geneva website (https://www.ungeneva.org/en), within which the team performed a search for any mention of “simulation” or “tabletop exercise” and identified relevant exercises that conformed with Health Security Net’s criteria. The research team additionally performed explicit searches to find specific exercises it had learned about from the experts or through online reading.
3. National pandemic influenza plans. National and regional pandemic plans were sought using the following sources and methods:
   • Plans were identified in the research paper “Inclusion of Veterinary Services in national emergency management plans,"¹⁴ which contained a compiled list of all national emergency plans from OIE member countries.
   • An archived WHO influenza plan repository¹⁵ was used to search for additional plans and countries that may not have been included in that resource; previously unidentified plans were then manually searched for via Google. The public availability of these plans was determined by searching “Country X pandemic influenza plan” into the Google search bar to see if it yielded an appropriate link within the first three search result pages.
   • WHO regional offices (Africa, Americas, South-East Asia, Europe, Eastern Mediterranean, and Western Pacific) were each reviewed for a list or repository of publicly available pandemic influenza plans of their Member Countries. For offices that listed plans but did not have proper links to them, the records were sought online via the same methodology described above.
   • Additional plans were gathered from pandemic and avian influenza plan repositories provided by the European Centre for Disease Prevention and Control¹⁶ and the Asia-Pacific Alliance for the Control of Influenza.¹⁷
   • Academic literature analyzing national influenza plans^181920212223 was reviewed to capture the titles of additional plans, which were sought online via the previously described methodology.

   In cases when a country produced an updated plan some time after an original plan had been published, the most recent plan was used to create the record; these cases are noted in the record’s description. If a plan addressed influenza plus other infectious diseases, it was included. Only documents written in English were selected, but additional reports in other languages may be added at a later date. If the plan stated that it was developed in direct response to 2005 H5N1 or 2009 H1N1, it was coded as being tied to these outbreak events. Pandemic planning documents for the United States were captured via search strategies delineated in the “Domestic” section of this methodology.

4. Non-WHO Regional Organizations. This category includes records from regional organizations that collaborate through multilateral action. The organizations searched encompassed Council of Ministers of Health of Central America and the Dominican Republic (COMISCA), the Caribbean Public Health Agency (CARPHA), Southern Common Market (MERCOSUR), Central American Integration System (SICA), Caribbean Community (CARICOM), African Union, European Union, Council of Europe, Association of Southeast Asian Nations (ASEAN), Community of Latina American and Caribbean States (CELAC), Pacific Alliance, Arab League, Pacific Islanders Forum (PIF), Organization of American States (OAS), and Economic Community of West African States (ECOWAS). Documents matching search criteria were only found on COMISCA’s and ASEAN’s websites. Each organization’s website was manually searched using the same search terms described above to identify documents, each of which was manually scanned to ensure that the terms were not only brief mentions but reflected records relevant to the library; those that met exclusion criteria were omitted from further analysis.
5. Emergency declarations for COVID-19. National emergency declarations from the United States and member states of the European Union and state-level emergency declarations from the United States with regard to the COVID-19 pandemic were captured using the following sources and methods:
   • Initial state and federal declarations were identified via COVID-AMP, using the policy category filter with "Emergency declarations" selected. Extensions of the original policies, amendments, and termination documents were also collected where applicable.
   • For states and countries whose emergency declarations were not accessible on COVID-AMP, the relevant governmnet's official website and online legal archives were reviewed for the document.
   • Where possible, both English and official language copies were captured. If an English version was not available, the researcher attempted to find an English-language press release announcing the emergency declaration.
6. Risk and policy communication for COVID-19 response. Several response initiatives specific to COVID-19 in the United States were captured in an effort to document what information was made public and which actions were taken, in the context of what scientific evidence was available at the time. The categories of documents for inclusion were brainstormed by the research team. Items were captured via manual review of U.S. governmental and third-party websites, including statements from public officials in cities with initial outbreaks; evolving guidance from the CDC about disease transmission and protective measures for the public; travel restrictions and repatriation efforts; Strategic National Stockpile mobilization; Centers for Medicare and Medicaid Services Emergency Rules and policies related to coverage of costs of testing and medical care; press releases and guidance from the American Hospital Association and American Academy of Pediatrics; industry guidance from the International Air Transport Association, International Civil Aviation Organization, and American Society of Heating, Refrigerating, and Air-Condition Engineers; handbooks from the Department of Education on school re-openings; National Governor’s Association public health guidance; CDC Advisory Committee on Immunization Practices guidance and FDA Emergency Use Authorizations; U.S. state vaccination incentive programs; and employer rules and regulations related to masking, testing, and vaccination.
7. Investigative journalism on COVID-19. High-impact investigative journalism on key topics throughout the pandemic were captured to fill gaps in how the outbreak unfolded. Investigative journalism was defined by the research team as pieces that primarily reported on information that was not easily accessible or publicly available, or that attempted to expose flaws in preparedness or response to the pandemic based on what was previously available.
   
   Investigative journalism pieces were identified by manually reviewing websites with thorough coverage of the COVID-19 pandemic dating back to January 2020, including the New York Times, the Atlantic, STAT News, the New Yorker, and the Washington Post. Articles from AP, Politico, and other sites were included as well if they were the first to release an exclusive investigation. While capturing investigative reports was highly subjective and there are thousands of impactful pieces available, key topics that were searched for included supply chain shortages, healthcare system preparedness, health disparities and disproportionate impacts on vulnerable groups, masking and testing guidance, and intelligence failures. The pieces selected may critique government figures, but on the basis of their response to the pandemic or actions that impacted its trajectory, rather than partisan objections.

Data coding

The team developed a custom data taxonomy and data dictionary to define key metadata and organize the dataset. The data were populated into Airtable and transferred via API into a database on Amazon Web Services. The complete dataset including definitions and a glossary of terms can be downloaded here. The Glossary below provides the guidance that researchers used to code the data.

Glossary