According to a survey on the coronavirus COVID-19 impact on rural China in late January 2020, national news and provincial TV channels were the major information sources of the epidemic for residents in the rural areas in the country. Around ********* of the respondents were first informed about the virus outbreak via online sources, such as social media and video streaming platforms.

Over the first eleven weeks of 2020, Chinese consumers spent a total of ***** minutes on average watching TV news, indicating nearly a double amount of time compared to that in the equivalent period of the previous year. To curb the spread of coronavirus COVID-19, the Chinese government imposed a lockdown in the epicenter Wuhan city from late January, 2020, and later expanded the measure across the country. Most of the Chinese people relied on national TV channels to receive the most updated epidemic information.

Abstract

JHU Coronavirus COVID-19 Global Cases, by country

Documentation

PHS is updating the Coronavirus Global Cases dataset weekly, Monday, Wednesday and Friday from Cloud Marketplace.

This data comes from the data repository for the 2019 Novel Coronavirus Visual Dashboard operated by the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE). This database was created in response to the Coronavirus public health emergency to track reported cases in real-time. The data include the location and number of confirmed COVID-19 cases, deaths, and recoveries for all affected countries, aggregated at the appropriate province or state. It was developed to enable researchers, public health authorities and the general public to track the outbreak as it unfolds. Additional information is available in the blog post.

Visual Dashboard (desktop): https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

Section 2

Included Data Sources are:

• World Health Organization (WHO): https://www.who.int/
• DXY.cn. Pneumonia. 2020. http://3g.dxy.cn/newh5/view/pneumonia.
• BNO News: https://bnonews.com/index.php/2020/02/the-latest-coronavirus-cases/
• National Health Commission of the People’s Republic of China (NHC): http://www.nhc.gov.cn/xcs/yqtb/list_gzbd.shtml
• China CDC (CCDC): http://weekly.chinacdc.cn/news/TrackingtheEpidemic.htm
• Hong Kong Department of Health: https://www.chp.gov.hk/en/features/102465.html
• Macau Government: https://www.ssm.gov.mo/portal/
• Taiwan CDC: https://sites.google.com/cdc.gov.tw/2019ncov/taiwan?authuser=0
• US CDC: https://www.cdc.gov/coronavirus/2019-ncov/index.html
• Government of Canada: https://www.canada.ca/en/public-health/services/diseases/coronavirus.html
• Australia Government Department of Health: https://www.health.gov.au/news/coronavirus-update-at-a-glance
• European Centre for Disease Prevention and Control (ECDC): https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases
• Ministry of Health Singapore (MOH): https://www.moh.gov.sg/covid-19
• Italy Ministry of Health: http://www.salute.gov.it/nuovocoronavirus
• 1Point3Arces: https://coronavirus.1point3acres.com/en
• WorldoMeters: https://www.worldometers.info/coronavirus/

%3C!-- --%3E

Section 3

**Terms of Use: **

This GitHub repo and its contents herein, including all data, mapping, and analysis, copyright 2020 Johns Hopkins University, all rights reserved, is provided to the public strictly for educational and academic research purposes. The Website relies upon publicly available data from multiple sources, that do not always agree. The Johns Hopkins University hereby disclaims any and all representations and warranties with respect to the Website, including accuracy, fitness for use, and merchantability. Reliance on the Website for medical guidance or use of the Website in commerce is strictly prohibited.

Section 4

**U.S. county-level characteristics relevant to COVID-19 **

Chin, Kahn, Krieger, Buckee, Balsari and Kiang (forthcoming) show that counties differ significantly in biological, demographic and socioeconomic factors that are associated with COVID-19 vulnerability. A range of publicly available county-specific data identifying these key factors, guided by international experiences and consideration of epidemiological parameters of importance, have been combined by the authors and are available for use:

https://github.com/mkiang/county_preparedness/

Key vaccine stocks like Moderna, Pfizer, and Novavax are rising as new coronavirus concerns emerge in China, highlighting a mixed day in the stock market with travel and tech sectors facing declines.

Context

Each workbook contains daily COVID-19 stats by each country affected. Additional sheets have also been added for more specific breakdown by different locations within Australia, Canada, China, and USA. Worked with BNO News to put this together. Additional credits include: Michael Van Poppel and Carlos Robles. Github updated every 24 hrs can be found here: https://github.com/jamesvalles/CORONAVIUS-COVID-19-DAILYSTATS

The study examines 30 consecutive episodes of Xinwen Lianbo (541 stories), 332 posts on “cctvxwlianbo” WeChat official account, 161 articles on the front page of People’s Daily newspaper in 30 days, as well as 1,015 hashtags and news articles on Sina Weibo’s two categories of ranking (top searched and real-time hot topic).

The study's methodology is related to these publications: An, Seon-Kyoung, and Karla K. Gower. “How Do the News Media Frame Crises? A Content Analysis of Crisis News Coverage.” Public Relations Review 35, no. 2 (2009): 107–12. https://doi.org/10.1016/j.pubrev.2009.01.010. Goodall, Catherine; Sabo, Jason; Cline, Rebecca & Egbert Nichole (2012), Threat, Efficacy, and Uncertainty in the First 5 Months of National Print and Electronic News Coverage of the H1N1 Virus, Journal of Health Communication, 17:3, 338-355, DOI: 10.1080/10810730.2011.626499

Context

Amidst the COVID-19 outbreak, the world is facing great crisis in every way. The value and things we built as a human race are going through tremendous challenges. It is a very small effort to bring curated data set on Novel Corona Virus to accelerate the forecasting and analytical experiments to cope up with this critical situation. It will help to visualize the country level out break and to keep track on regularly added new incidents.

COVID-19 Country Level Timeseries Dataset

This Dataset contains country wise public domain time series information on COVID-19 outbreak. The Data is sorted alphabetically on Country name and Date of Observation.

Column Descriptions

The data set contains the following columns:
ObservationDate: The date on which the incidents are observed country: Country of the Outbreak Confirmed: Number of confirmed cases till observation date Deaths: Number of death cases till observation date Recovered: Number of recovered cases till observation date New Confirmed: Number of new confirmed cases on observation date New Deaths: Number of New death cases on observation date New Recovered: Number of New recovered cases on observation date latitude: Latitude of the affected country longitude: Longitude of the affected country

Acknowledgements

This data set is a cleaner version of the https://www.kaggle.com/sudalairajkumar/novel-corona-virus-2019-dataset data set with added geo location information and regularly added incident counts. I would like to thank this great effort by SRK.

Original Data Source

Johns Hopkins University MoBS lab - https://www.mobs-lab.org/2019ncov.html World Health Organization (WHO): https://www.who.int/ DXY.cn. Pneumonia. 2020. http://3g.dxy.cn/newh5/view/pneumonia. BNO News: https://bnonews.com/index.php/2020/02/the-latest-coronavirus-cases/ National Health Commission of the People’s Republic of China (NHC): http://www.nhc.gov.cn/xcs/yqtb/list_gzbd.shtml China CDC (CCDC): http://weekly.chinacdc.cn/news/TrackingtheEpidemic.htm Hong Kong Department of Health: https://www.chp.gov.hk/en/features/102465.html Macau Government: https://www.ssm.gov.mo/portal/ Taiwan CDC: https://sites.google.com/cdc.gov.tw/2019ncov/taiwan?authuser=0 US CDC: https://www.cdc.gov/coronavirus/2019-ncov/index.html Government of Canada: https://www.canada.ca/en/public-health/services/diseases/coronavirus.html Australia Government Department of Health: https://www.health.gov.au/news/coronavirus-update-at-a-glance European Centre for Disease Prevention and Control (ECDC): https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases Ministry of Health Singapore (MOH): https://www.moh.gov.sg/covid-19 Italy Ministry of Health: http://www.salute.gov.it/nuovocoronavirus

This is the data for the 2019 Novel Coronavirus Visual Dashboard operated by the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE). Also, Supported by ESRI Living Atlas Team and the Johns Hopkins University Applied Physics Lab (JHU APL).Data SourcesWorld Health Organization (WHO): https://www.who.int/ DXY.cn. Pneumonia. 2020. http://3g.dxy.cn/newh5/view/pneumonia. BNO News: https://bnonews.com/index.php/2020/02/the-latest-coronavirus-cases/ National Health Commission of the People’s Republic of China (NHC): http://www.nhc.gov.cn/xcs/yqtb/list_gzbd.shtml China CDC (CCDC): http://weekly.chinacdc.cn/news/TrackingtheEpidemic.htm Hong Kong Department of Health: https://www.chp.gov.hk/en/features/102465.html Macau Government: https://www.ssm.gov.mo/portal/ Taiwan CDC: https://sites.google.com/cdc.gov.tw/2019ncov/taiwan?authuser=0 US CDC: https://www.cdc.gov/coronavirus/2019-ncov/index.html Government of Canada: https://www.canada.ca/en/public-health/services/diseases/coronavirus.html Australia Government Department of Health: https://www.health.gov.au/news/coronavirus-update-at-a-glance European Centre for Disease Prevention and Control (ECDC): https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-casesMinistry of Health Singapore (MOH): https://www.moh.gov.sg/covid-19Italy Ministry of Health: http://www.salute.gov.it/nuovocoronavirus

This is the replication data for paper "A Soft Power Challenge, or an Opportunity? A Big Data analysis on Chinese Soft Power During COVID-19 Pandemic." The files include 1) the metadata for the news texts we utilized for the analyses, 2) preprocessed tokens of the news texts, 3) the python code that we used to conduct the DMR analyses and keyword counting, and 4) the results tables.

This is the data for the 2019 Novel Coronavirus Visual Dashboard operated by the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE). Also, Supported by ESRI Living Atlas Team and the Johns Hopkins University Applied Physics Lab (JHU APL).Data SourcesWorld Health Organization (WHO): https://www.who.int/ DXY.cn. Pneumonia. 2020. http://3g.dxy.cn/newh5/view/pneumonia. BNO News: https://bnonews.com/index.php/2020/02/the-latest-coronavirus-cases/ National Health Commission of the People’s Republic of China (NHC): http://www.nhc.gov.cn/xcs/yqtb/list_gzbd.shtml China CDC (CCDC): http://weekly.chinacdc.cn/news/TrackingtheEpidemic.htm Hong Kong Department of Health: https://www.chp.gov.hk/en/features/102465.html Macau Government: https://www.ssm.gov.mo/portal/ Taiwan CDC: https://sites.google.com/cdc.gov.tw/2019ncov/taiwan?authuser=0 US CDC: https://www.cdc.gov/coronavirus/2019-ncov/index.html Government of Canada: https://www.canada.ca/en/public-health/services/diseases/coronavirus.html Australia Government Department of Health: https://www.health.gov.au/news/coronavirus-update-at-a-glance European Centre for Disease Prevention and Control (ECDC): https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-casesMinistry of Health Singapore (MOH): https://www.moh.gov.sg/covid-19Italy Ministry of Health: http://www.salute.gov.it/nuovocoronavirus

According to a survey on the coronavirus COVID-19 impact on rural China in late January 2020, around **** percent of the respondents living in small towns and villages were first informed about the outbreak between the *** and **** of January. The swift spread of the virus, first detected in early December 2019 in Wuhan, has caused a huge impact on the daily life of Chinese people.

Introduction

There are several works based on Natural Language Processing on newspaper reports. Mining opinions from headlines [ 1 ] using Standford NLP and SVM by Rameshbhaiet. Al.compared several algorithms on a small and large dataset. Rubinet. al., in their paper [ 2 ], created a mechanism to differentiate fake news from real ones by building a set of characteristics of news according to their types. The purpose was to contribute to the low resource data available for training machine learning algorithms. Doumitet. al.in [ 3 ] have implemented LDA, a topic modeling approach to study bias present in online news media.

However, there are not many NLP research invested in studying COVID-19. Most applications include classification of chest X-rays and CT-scans to detect presence of pneumonia in lungs [ 4 ], a consequence of the virus. Other research areas include studying the genome sequence of the virus[ 5 ][ 6 ][ 7 ] and replicating its structure to fight and find a vaccine. This research is crucial in battling the pandemic. The few NLP based research publications are sentiment classification of online tweets by Samuel et el [ 8 ] to understand fear persisting in people due to the virus. Similar work has been done using the LSTM network to classify sentiments from online discussion forums by Jelodaret. al.[ 9 ]. NKK dataset is the first study on a comparatively larger dataset of a newspaper report on COVID-19, which contributed to the virus’s awareness to the best of our knowledge.

2 Data-set Introduction

2.1 Data Collection

We accumulated 1000 online newspaper report from United States of America (USA) on COVID-19. The newspaper includes The Washington Post (USA) and StarTribune (USA). We have named it as “Covid-News-USA-NNK”. We also accumulated 50 online newspaper report from Bangladesh on the issue and named it “Covid-News-BD-NNK”. The newspaper includes The Daily Star (BD) and Prothom Alo (BD). All these newspapers are from the top provider and top read in the respective countries. The collection was done manually by 10 human data-collectors of age group 23- with university degrees. This approach was suitable compared to automation to ensure the news were highly relevant to the subject. The newspaper online sites had dynamic content with advertisements in no particular order. Therefore there were high chances of online scrappers to collect inaccurate news reports. One of the challenges while collecting the data is the requirement of subscription. Each newspaper required $1 per subscriptions. Some criteria in collecting the news reports provided as guideline to the human data-collectors were as follows:

The headline must have one or more words directly or indirectly related to COVID-19.

The content of each news must have 5 or more keywords directly or indirectly related to COVID-19.

The genre of the news can be anything as long as it is relevant to the topic. Political, social, economical genres are to be more prioritized.

Avoid taking duplicate reports.

Maintain a time frame for the above mentioned newspapers.

To collect these data we used a google form for USA and BD. We have two human editor to go through each entry to check any spam or troll entry.

2.2 Data Pre-processing and Statistics

Some pre-processing steps performed on the newspaper report dataset are as follows:

Remove hyperlinks.

Remove non-English alphanumeric characters.

Remove stop words.

Lemmatize text.

While more pre-processing could have been applied, we tried to keep the data as much unchanged as possible since changing sentence structures could result us in valuable information loss. While this was done with help of a script, we also assigned same human collectors to cross check for any presence of the above mentioned criteria.

The primary data statistics of the two dataset are shown in Table 1 and 2.

Table 1: Covid-News-USA-NNK data statistics

No of words per headline

7 to 20

No of words per body content

150 to 2100

Table 2: Covid-News-BD-NNK data statistics No of words per headline

10 to 20

No of words per body content

100 to 1500

2.3 Dataset Repository

We used GitHub as our primary data repository in account name NKK^1. Here, we created two repositories USA-NKK^2 and BD-NNK^3. The dataset is available in both CSV and JSON format. We are regularly updating the CSV files and regenerating JSON using a py script. We provided a python script file for essential operation. We welcome all outside collaboration to enrich the dataset.

3 Literature Review

Natural Language Processing (NLP) deals with text (also known as categorical) data in computer science, utilizing numerous diverse methods like one-hot encoding, word embedding, etc., that transform text to machine language, which can be fed to multiple machine learning and deep learning algorithms.

Some well-known applications of NLP includes fraud detection on online media sites[ 10 ], using authorship attribution in fallback authentication systems[ 11 ], intelligent conversational agents or chatbots[ 12 ] and machine translations used by Google Translate[ 13 ]. While these are all downstream tasks, several exciting developments have been made in the algorithm solely for Natural Language Processing tasks. The two most trending ones are BERT[ 14 ], which uses bidirectional encoder-decoder architecture to create the transformer model, that can do near-perfect classification tasks and next-word predictions for next generations, and GPT-3 models released by OpenAI[ 15 ] that can generate texts almost human-like. However, these are all pre-trained models since they carry huge computation cost. Information Extraction is a generalized concept of retrieving information from a dataset. Information extraction from an image could be retrieving vital feature spaces or targeted portions of an image; information extraction from speech could be retrieving information about names, places, etc[ 16 ]. Information extraction in texts could be identifying named entities and locations or essential data. Topic modeling is a sub-task of NLP and also a process of information extraction. It clusters words and phrases of the same context together into groups. Topic modeling is an unsupervised learning method that gives us a brief idea about a set of text. One commonly used topic modeling is Latent Dirichlet Allocation or LDA[17].

Keyword extraction is a process of information extraction and sub-task of NLP to extract essential words and phrases from a text. TextRank [ 18 ] is an efficient keyword extraction technique that uses graphs to calculate the weight of each word and pick the words with more weight to it.

Word clouds are a great visualization technique to understand the overall ’talk of the topic’. The clustered words give us a quick understanding of the content.

4 Our experiments and Result analysis

We used the wordcloud library^4 to create the word clouds. Figure 1 and 3 presents the word cloud of Covid-News-USA- NNK dataset by month from February to May. From the figures 1,2,3, we can point few information:

In February, both the news paper have talked about China and source of the outbreak.

StarTribune emphasized on Minnesota as the most concerned state. In April, it seemed to have been concerned more.

Both the newspaper talked about the virus impacting the economy, i.e, bank, elections, administrations, markets.

Washington Post discussed global issues more than StarTribune.

StarTribune in February mentioned the first precautionary measurement: wearing masks, and the uncontrollable spread of the virus throughout the nation.

While both the newspaper mentioned the outbreak in China in February, the weight of the spread in the United States are more highlighted through out March till May, displaying the critical impact caused by the virus.

We used a script to extract all numbers related to certain keywords like ’Deaths’, ’Infected’, ’Died’ , ’Infections’, ’Quarantined’, Lock-down’, ’Diagnosed’ etc from the news reports and created a number of cases for both the newspaper. Figure 4 shows the statistics of this series. From this extraction technique, we can observe that April was the peak month for the covid cases as it gradually rose from February. Both the newspaper clearly shows us that the rise in covid cases from February to March was slower than the rise from March to April. This is an important indicator of possible recklessness in preparations to battle the virus. However, the steep fall from April to May also shows the positive response against the attack. We used Vader Sentiment Analysis to extract sentiment of the headlines and the body. On average, the sentiments were from -0.5 to -0.9. Vader Sentiment scale ranges from -1(highly negative to 1(highly positive). There were some cases

where the sentiment scores of the headline and body contradicted each other,i.e., the sentiment of the headline was negative but the sentiment of the body was slightly positive. Overall, sentiment analysis can assist us sort the most concerning (most negative) news from the positive ones, from which we can learn more about the indicators related to COVID-19 and the serious impact caused by it. Moreover, sentiment analysis can also provide us information about how a state or country is reacting to the pandemic. We used PageRank algorithm to extract keywords from headlines as well as the body content. PageRank efficiently highlights important relevant keywords in the text. Some frequently occurring important keywords extracted from both the datasets are: ’China’, Government’, ’Masks’, ’Economy’, ’Crisis’, ’Theft’ , ’Stock market’ , ’Jobs’ , ’Election’, ’Missteps’, ’Health’, ’Response’. Keywords extraction acts as a filter allowing quick searches for indicators in case of locating situations of the economy,

Regression and moderation analyses on examining the effect of media attributes on the weight of severity frame.

Context

From World Health Organization - On 31 December 2019, WHO was alerted to several cases of pneumonia in Wuhan City, Hubei Province of China. The virus did not match any other known virus. This raised concern because when a virus is new, we do not know how it affects people.

So daily level information on the affected people can give some interesting insights when it is made available to the broader data science community.

Johns Hopkins University has made an excellent dashboard using the affected cases data. Data is extracted from the google sheets associated and made available here.

Edited

Now data is available as csv files in the Johns Hopkins Github repository. Please refer to the github repository for the Terms of Use details. Uploading it here for using it in Kaggle kernels and getting insights from the broader DS community.

Content 2019 Novel Coronavirus (2019-nCoV) is a virus (more specifically, a coronavirus) identified as the cause of an outbreak of respiratory illness first detected in Wuhan, China. Early on, many of the patients in the outbreak in Wuhan, China reportedly had some link to a large seafood and animal market, suggesting animal-to-person spread. However, a growing number of patients reportedly have not had exposure to animal markets, indicating person-to-person spread is occurring. At this time, it’s unclear how easily or sustainably this virus is spreading between people - CDC

This dataset has daily level information on the number of affected cases, deaths and recovery from 2019 novel coronavirus. Please note that this is a time series data and so the number of cases on any given day is the cumulative number.

The data is available from 22 Jan, 2020 to 30 Dec, 2020.

Sources

JHU confirmed covid datasets.

This project examines how China’s two major official media outlets—Xinhua News Agency and People’s Daily—cited scientific experts on their official social media accounts during the COVID-19 pandemic (2020–2022). The dataset documents which experts were cited, when they were cited, and what types of statements were quoted. It also includes thematic coding of the cited content to capture the communicative functions these expert quotations served.

IntroductionSocial media infuses modern relationships with vitality and brings a series of information dissemination with subjective consciousness. Studies have indicated that official Chinese media channels are transforming their communication style from didactic hard persuasion to softened emotional management in the digital era. However, previous studies have rarely provided valid empirical evidence for the communicational transformation. The study fills the gap by providing a longitudinal time-series analysis to reveal the pattern of communication of Chinese digital Chinese official media from 2019 to 2022.MethodThe study crawler collected 43,259 posts from the People’s Daily’s Weibo account from 2019 to 2021. The study analyzed the textual data with using trained artificial intelligence models.ResultsThis study explored the practices of the People’s Daily’s Weibo account from 2019 to 2021, COVID-19 is hardly normalized as it is still used as the justification for extraordinary measures in China. This study confirmed that People’s Daily’s Weibo account posts are undergoing softenization transformation, with the use of soft news, positive energy promotion, and the embedding of sentiment. Although the outburst of COVID-19 temporarily increased the media’s use of hard news, it only occur at the initial stage of the pandemic. Emotional posts occupy a nonnegligible amount of the People’s Daily Weibo content. However, the majority of posts are emotionally neutral and contribute to shaping the authoritative image of the party press.DiscussionOverall, the People’s Daily has softened their communication style on digital platforms and used emotional mobilization, distraction, and timely information provision to balance the political logic of building an authoritative media agency and the media logic of constructing audience relevance.

Crisis motivates people to track news closely, and this increased engagement can expose individuals to politically sensitive information unrelated to the initial crisis. We use the case of the COVID-19 outbreak in China to examine how crisis affects information seeking in countries that normally exert significant control over access to media. The crisis spurred censorship circumvention and access to international news and political content on websites blocked in China. Once individuals circumvented censorship, they not only received more information about the crisis itself but also accessed unrelated information that the regime has long censored. Using comparisons to democratic and other authoritarian countries also affected by early outbreaks, the findings suggest that people blocked from accessing information most of the time might disproportionately and collectively access that long-hidden information during a crisis. Evaluations resulting from this access, negative or positive for a government, might draw on both current events and censored history.

COVID-19 Trends MethodologyOur goal is to analyze and present daily updates in the form of recent trends within countries, states, or counties during the COVID-19 global pandemic. The data we are analyzing is taken directly from the Johns Hopkins University Coronavirus COVID-19 Global Cases Dashboard, though we expect to be one day behind the dashboard’s live feeds to allow for quality assurance of the data.Revisions added on 4/23/2020 are highlighted.Revisions added on 4/30/2020 are highlighted.Discussion of our assertion of an abundance of caution in assigning trends in rural counties added 5/7/2020. Correction on 6/1/2020Methodology update on 6/2/2020: This sets the length of the tail of new cases to 6 to a maximum of 14 days, rather than 21 days as determined by the last 1/3 of cases. This was done to align trends and criteria for them with U.S. CDC guidance. The impact is areas transition into Controlled trend sooner for not bearing the burden of new case 15-21 days earlier.Reasons for undertaking this work:The popular online maps and dashboards show counts of confirmed cases, deaths, and recoveries by country or administrative sub-region. Comparing the counts of one country to another can only provide a basis for comparison during the initial stages of the outbreak when counts were low and the number of local outbreaks in each country was low. By late March 2020, countries with small populations were being left out of the mainstream news because it was not easy to recognize they had high per capita rates of cases (Switzerland, Luxembourg, Iceland, etc.). Additionally, comparing countries that have had confirmed COVID-19 cases for high numbers of days to countries where the outbreak occurred recently is also a poor basis for comparison.The graphs of confirmed cases and daily increases in cases were fit into a standard size rectangle, though the Y-axis for one country had a maximum value of 50, and for another country 100,000, which potentially misled people interpreting the slope of the curve. Such misleading circumstances affected comparing large population countries to small population counties or countries with low numbers of cases to China which had a large count of cases in the early part of the outbreak. These challenges for interpreting and comparing these graphs represent work each reader must do based on their experience and ability. Thus, we felt it would be a service to attempt to automate the thought process experts would use when visually analyzing these graphs, particularly the most recent tail of the graph, and provide readers with an a resulting synthesis to characterize the state of the pandemic in that country, state, or county.The lack of reliable data for confirmed recoveries and therefore active cases. Merely subtracting deaths from total cases to arrive at this figure progressively loses accuracy after two weeks. The reason is 81% of cases recover after experiencing mild symptoms in 10 to 14 days. Severe cases are 14% and last 15-30 days (based on average days with symptoms of 11 when admitted to hospital plus 12 days median stay, and plus of one week to include a full range of severely affected people who recover). Critical cases are 5% and last 31-56 days. Sources:U.S. CDC. April 3, 2020 Interim Clinical Guidance for Management of Patients with Confirmed Coronavirus Disease (COVID-19). Accessed online. Initial older guidance was also obtained online. Additionally, many people who recover may not be tested, and many who are, may not be tracked due to privacy laws. Thus, the formula used to compute an estimate of active cases is: Active Cases = 100% of new cases in past 14 days + 19% from past 15-30 days + 5% from past 31-56 days - total deaths.We’ve never been inside a pandemic with the ability to learn of new cases as they are confirmed anywhere in the world. After reviewing epidemiological and pandemic scientific literature, three needs arose. We need to specify which portions of the pandemic lifecycle this map cover. The World Health Organization (WHO) specifies six phases. The source data for this map begins just after the beginning of Phase 5: human to human spread and encompasses Phase 6: pandemic phase. Phase six is only characterized in terms of pre- and post-peak. However, these two phases are after-the-fact analyses and cannot ascertained during the event. Instead, we describe (below) a series of five trends for Phase 6 of the COVID-19 pandemic.Choosing terms to describe the five trends was informed by the scientific literature, particularly the use of epidemic, which signifies uncontrolled spread. The five trends are: Emergent, Spreading, Epidemic, Controlled, and End Stage. Not every locale will experience all five, but all will experience at least three: emergent, controlled, and end stage.This layer presents the current trends for the COVID-19 pandemic by country (or appropriate level). There are five trends:Emergent: Early stages of outbreak. Spreading: Early stages and depending on an administrative area’s capacity, this may represent a manageable rate of spread. Epidemic: Uncontrolled spread. Controlled: Very low levels of new casesEnd Stage: No New cases These trends can be applied at several levels of administration: Local: Ex., City, District or County – a.k.a. Admin level 2State: Ex., State or Province – a.k.a. Admin level 1National: Country – a.k.a. Admin level 0Recommend that at least 100,000 persons be represented by a unit; granted this may not be possible, and then the case rate per 100,000 will become more important.Key Concepts and Basis for Methodology: 10 Total Cases minimum threshold: Empirically, there must be enough cases to constitute an outbreak. Ideally, this would be 5.0 per 100,000, but not every area has a population of 100,000 or more. Ten, or fewer, cases are also relatively less difficult to track and trace to sources. 21 Days of Cases minimum threshold: Empirically based on COVID-19 and would need to be adjusted for any other event. 21 days is also the minimum threshold for analyzing the “tail” of the new cases curve, providing seven cases as the basis for a likely trend (note that 21 days in the tail is preferred). This is the minimum needed to encompass the onset and duration of a normal case (5-7 days plus 10-14 days). Specifically, a median of 5.1 days incubation time, and 11.2 days for 97.5% of cases to incubate. This is also driven by pressure to understand trends and could easily be adjusted to 28 days. Source used as basis:Stephen A. Lauer, MS, PhD *; Kyra H. Grantz, BA *; Qifang Bi, MHS; Forrest K. Jones, MPH; Qulu Zheng, MHS; Hannah R. Meredith, PhD; Andrew S. Azman, PhD; Nicholas G. Reich, PhD; Justin Lessler, PhD. 2020. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Annals of Internal Medicine DOI: 10.7326/M20-0504.New Cases per Day (NCD) = Measures the daily spread of COVID-19. This is the basis for all rates. Back-casting revisions: In the Johns Hopkins’ data, the structure is to provide the cumulative number of cases per day, which presumes an ever-increasing sequence of numbers, e.g., 0,0,1,1,2,5,7,7,7, etc. However, revisions do occur and would look like, 0,0,1,1,2,5,7,7,6. To accommodate this, we revised the lists to eliminate decreases, which make this list look like, 0,0,1,1,2,5,6,6,6.Reporting Interval: In the early weeks, Johns Hopkins' data provided reporting every day regardless of change. In late April, this changed allowing for days to be skipped if no new data was available. The day was still included, but the value of total cases was set to Null. The processing therefore was updated to include tracking of the spacing between intervals with valid values.100 News Cases in a day as a spike threshold: Empirically, this is based on COVID-19’s rate of spread, or r0 of ~2.5, which indicates each case will infect between two and three other people. There is a point at which each administrative area’s capacity will not have the resources to trace and account for all contacts of each patient. Thus, this is an indicator of uncontrolled or epidemic trend. Spiking activity in combination with the rate of new cases is the basis for determining whether an area has a spreading or epidemic trend (see below). Source used as basis:World Health Organization (WHO). 16-24 Feb 2020. Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19). Obtained online.Mean of Recent Tail of NCD = Empirical, and a COVID-19-specific basis for establishing a recent trend. The recent mean of NCD is taken from the most recent fourteen days. A minimum of 21 days of cases is required for analysis but cannot be considered reliable. Thus, a preference of 42 days of cases ensures much higher reliability. This analysis is not explanatory and thus, merely represents a likely trend. The tail is analyzed for the following:Most recent 2 days: In terms of likelihood, this does not mean much, but can indicate a reason for hope and a basis to share positive change that is not yet a trend. There are two worthwhile indicators:Last 2 days count of new cases is less than any in either the past five or 14 days. Past 2 days has only one or fewer new cases – this is an extremely positive outcome if the rate of testing has continued at the same rate as the previous 5 days or 14 days. Most recent 5 days: In terms of likelihood, this is more meaningful, as it does represent at short-term trend. There are five worthwhile indicators:Past five days is greater than past 2 days and past 14 days indicates the potential of the past 2 days being an aberration. Past five days is greater than past 14 days and less than past 2 days indicates slight positive trend, but likely still within peak trend time frame.Past five days is less than the past 14 days. This means a downward trend. This would be an

COVID-19 Trends MethodologyOur goal is to analyze and present daily updates in the form of recent trends within countries, states, or counties during the COVID-19 global pandemic. The data we are analyzing is taken directly from the Johns Hopkins University Coronavirus COVID-19 Global Cases Dashboard, though we expect to be one day behind the dashboard’s live feeds to allow for quality assurance of the data.Reasons for undertaking this work:The popular online maps and dashboards show counts of confirmed cases, deaths, and recoveries by country or administrative sub-region. Comparing the counts of one country to another can only provide a basis for comparison during the initial stages of the outbreak when counts were low and the number of local outbreaks in each country was low. By late March 2020, countries with small populations were being left out of the mainstream news because it was not easy to recognize they had high per capita rates of cases (Switzerland, Luxembourg, Iceland, etc.). Additionally, comparing countries that have had confirmed COVID-19 cases for high numbers of days to countries where the outbreak occurred recently is also a poor basis for comparison.The graphs of confirmed cases and daily increases in cases were fit into a standard size rectangle, though the Y-axis for one country had a maximum value of 50, and for another country 100,000, which potentially misled people interpreting the slope of the curve. Such misleading circumstances affected comparing large population countries to small population counties or countries with low numbers of cases to China which had a large count of cases in the early part of the outbreak. These challenges for interpreting and comparing these graphs represent work each reader must do based on their experience and ability. Thus, we felt it would be a service to attempt to automate the thought process experts would use when visually analyzing these graphs, particularly the most recent tail of the graph, and provide readers with an a resulting synthesis to characterize the state of the pandemic in that country, state, or county.The lack of reliable data for confirmed recoveries and therefore active cases. Merely subtracting deaths from total cases to arrive at this figure progressively loses accuracy after two weeks. The reason is 81% of cases recover after experiencing mild symptoms in 10 to 14 days. Severe cases are 14% and last 15-30 days (based on average days with symptoms of 11 when admitted to hospital plus 12 days median stay, and plus of one week to include a full range of severely affected people who recover). Critical cases are 5% and last 31-56 days. Sources:U.S. CDC. April 3, 2020 Interim Clinical Guidance for Management of Patients with Confirmed Coronavirus Disease (COVID-19). Accessed online. Initial older guidance was also obtained online. Additionally, many people who recover may not be tested, and many who are, may not be tracked due to privacy laws. Thus, the formula used to compute an estimate of active cases is: Active Cases = 100% of new cases in past 14 days + 19% from past 15-30 days + 5% from past 31-56 days - total deaths.We’ve never been inside a pandemic with the ability to learn of new cases as they are confirmed anywhere in the world. After reviewing epidemiological and pandemic scientific literature, three needs arose. We need to specify which portions of the pandemic lifecycle this map cover. The World Health Organization (WHO) specifies six phases. The source data for this map begins just after the beginning of Phase 5: human to human spread and encompasses Phase 6: pandemic phase. Phase six is only characterized in terms of pre- and post-peak. However, these two phases are after-the-fact analyses and cannot ascertained during the event. Instead, we describe (below) a series of five trends for Phase 6 of the COVID-19 pandemic.Choosing terms to describe the five trends was informed by the scientific literature, particularly the use of epidemic, which signifies uncontrolled spread. The five trends are: Emergent, Spreading, Epidemic, Controlled, and End Stage. Not every locale will experience all five, but all will experience at least three: emergent, controlled, and end stage.This layer presents the current trends for the COVID-19 pandemic by country (or appropriate level). There are five trends:Emergent: Early stages of outbreak. Spreading: Early stages and depending on an administrative area’s capacity, this may represent a manageable rate of spread. Epidemic: Uncontrolled spread. Controlled: Very low levels of new casesEnd Stage: No New cases These trends can be applied at several levels of administration: Local: Ex., City, District or County – a.k.a. Admin level 2State: Ex., State or Province – a.k.a. Admin level 1National: Country – a.k.a. Admin level 0Recommend that at least 100,000 persons be represented by a unit; granted this may not be possible, and then the case rate per 100,000 will become more important.Key Concepts and Basis for Methodology: 10 Total Cases minimum threshold: Empirically, there must be enough cases to constitute an outbreak. Ideally, this would be 5.0 per 100,000, but not every area has a population of 100,000 or more. Ten, or fewer, cases are also relatively less difficult to track and trace to sources. 21 Days of Cases minimum threshold: Empirically based on COVID-19 and would need to be adjusted for any other event. 21 days is also the minimum threshold for analyzing the “tail” of the new cases curve, providing seven cases as the basis for a likely trend (note that 21 days in the tail is preferred). This is the minimum needed to encompass the onset and duration of a normal case (5-7 days plus 10-14 days). Specifically, a median of 5.1 days incubation time, and 11.2 days for 97.5% of cases to incubate. This is also driven by pressure to understand trends and could easily be adjusted to 28 days. Source used as basis:Stephen A. Lauer, MS, PhD *; Kyra H. Grantz, BA *; Qifang Bi, MHS; Forrest K. Jones, MPH; Qulu Zheng, MHS; Hannah R. Meredith, PhD; Andrew S. Azman, PhD; Nicholas G. Reich, PhD; Justin Lessler, PhD. 2020. The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Annals of Internal Medicine DOI: 10.7326/M20-0504.New Cases per Day (NCD) = Measures the daily spread of COVID-19. This is the basis for all rates. 100 News Cases in a day as a spike threshold: Empirically, this is based on COVID-19’s rate of spread, or r0 of ~2.5, which indicates each case will infect between two and three other people. There is a point at which each administrative area’s capacity will not have the resources to trace and account for all contacts of each patient. Thus, this is an indicator of uncontrolled or epidemic trend. Spiking activity in combination with the rate of new cases is the basis for determining whether an area has a spreading or epidemic trend (see below). Source used as basis:World Health Organization (WHO). 16-24 Feb 2020. Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19). Obtained online.Mean of Recent Tail of NCD = Empirical, and a COVID-19-specific basis for establishing a recent trend. The recent mean of NCD is taken from the most recent one third of case days. A minimum of 21 days of cases is required for analysis but cannot be considered reliable. Thus, a preference of 63 days of cases ensures much higher reliability. This analysis is not explanatory and thus, merely represents a likely trend. The tail is analyzed for the following:Most recent 2 days: In terms of likelihood, this does not mean much, but can indicate a reason for hope and a basis to share positive change that is not yet a trend. There are two worthwhile indicators:Last 2 days count of new cases is less than any in either the past five or 6-21 days. Past 2 days has only one or fewer new cases – this is an extremely positive outcome if the rate of testing has continued at the same rate as the previous 5 days or 6 to 21 days. Most recent 5 days: In terms of likelihood, this is more meaningful, as it does represent at short-term trend. There are five worthwhile indicators:Past five days is greater than past 2 days and past 6-21 days indicates the potential of the past 2 days being an aberration. Past five days is greater than past 6-21 days and less than past 2 days indicates slight positive trend, but likely still within peak trend timeframe.Past five days is less than the past 6-21 days. This means a downward trend. This would be an important trend for any administrative area in an epidemic trend that the rate of spread is slowing.If less than the past 2 days, but not the last 6-21 days, this is still positive, but is not indicating a passage out of the peak timeframe of the daily new cases curve.Past 5 days has only one or two new cases – this is an extremely positive outcome if the rate of testing has continued at the same rate as the previous 6 to 21 days. Most recent 6-21 days: Represents the full tail of the curve and provides context for the past 2- and 5-day trends.If this is greater than both the 2- and 5-day trends, then a short-term downward trend has begun. Mean of Recent Tail NCD in the context of the Mean of All NCD, and raw counts of cases:Mean of Recent NCD is less than 0.5 cases per 100,000 = high level of controlMean of Recent NCD is less than 1.0 and fewer than 30 cases indicate continued emergent trend.3. Mean of Recent NCD is less than 1.0 and greater than 30 cases indicate a change from emergent to spreading trend.Mean of All NCD less than 2.0 per 100,000, and areas that have been in epidemic trends have Mean of Recent NCD of less than 5.0 per 100,000 is a significant indicator of changing trends from epidemic to spreading, now going in the direction of controlled trend.Similarly, in the context of Mean of All NCD greater than 2.0