This is an archive of past discussions about COVID-19 pandemic. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Quadratic growth and comparison to previous research
A recent preprint paper models the infected and death statistics as highly regular quadratic growth. I added a paragraph about this and included the following sentence:
The regular quadratic growth is compared to previous research describing how Chinese organ donation statistics was falsified using a simple quadratic equation.
The sentence is not an accurate description of the paper. Reference 7 of the paper - Robertson, Hinde & Lavee 2019 - is only mentioned in passing in the Conclusions section of the paper; the abstract and main thrust of the paper are that the author argues that a quadratic model is epidemiologically justified, giving a theoretical justification of why it would make sense in the present context, and an empirical graphical argument that the model fits the data well, without any statistical analysis. He seems to refer to falsification, as per RHL2019's suggestion, as one possible explanation, but that his model provides a straightforward explanation.
The shifting of medical staff to handle the COVID-19 outbreak might right now be delaying organ transplants, so many of the prisoners may effectively be given a few extra weeks or a month or so longer to live before their organs are transplanted to transplantees. But getting evidence of this would require a whistleblower from medical staff normally responsible for removing the organs from the prisoners; the prisoners themselves and their families are presumably not informed in advance of their planned days of sacrifice, or of delays in the sacrifice dates. A whisteblower report published by a reliable source, e.g. sent to the NYT on their https :// www dot nytimes3xbfgragh dot onion/ service over Tor (anonymity network), and published by the NYT after their journalists judge the information credible, would most likely be notable enough for this article. Boud (talk) 00:45, 21 February 2020 (UTC)
Regarding the preprint itself: there is no explanation of which sources the Table 1 values come from and in what precise way they are extracted, except for a hint that they come from "daily news reports" and that WHO, Devex and Worldometer provide data. There's no explanation of the difference between lab-confirmed cases, clinically diagnosed cases, possible limits in lab-testing, and how he models all of these to interpret the data in an epidemiologically justified way. Also, the author fails to comment on the last 4-5 days of the cumulative count. Figure 2 shows that the e-folding time of the cumulative confirmed cases is huge - above the scale of the plot - during around 14-19 February; in other words, the cumulative number of confirmed cases is no longer following his quadratic fit since 4-5 days ago, so "does not show any sign of a decline" in the conclusions means that the author didn't notice what was in Figure 2 and what is obvious in File:NCoV20200209 Hubei China ROW cases.png. Boud (talk) 02:19, 21 February 2020 (UTC)
Unclear sentence
The grammar of the following sentence is so poor that I can not determine its meaning.
Please reinstate recovery column indicating maximum estimate of 12.5% mortality, minimum estimate 3%.
Please reinstate the recovery column. It is extremely useful for estimating upper and lower bounds for the death rate.
Recoveries outnumber deaths by 7 to 1, so the mortality is at most 12.5% using this morning's figures (Feb 19). (This assumes that dying on average takes just as long as recovering, which may not be the case. It is a maximal estimate because mostly only symptomatic cases are documented within mainland China).
Similarly the minimal death rate can be estimated from the non-Chinese cases: about 190 recoveries of which 6 have recovered, so a minimal mortality of 3%. (It is a minimal estimate because both symptomatic and asymptomatic cases are detected outside China via PCR analysis.)
In the coming weeks the minimal and maximal estimates will converge on the true value, in line with other Corona virus mortalities (SARS 2002 had 9% mortality, MERS 2012 had 30% mortality).
We have an extensive discussion about this with the graphs. Surveillance systems are not standardised worldwide, and comparing deaths to recoveries is near impossible at this stage. Hong Kong had many cases and only 2 recoveries from memory. That can't be compared in a graph to other countries - it is very misleading. If you follow my contribs you can see a fair bit of the argument --Almaty (talk) 15:09, 19 February 2020 (UTC)
It is inadvisable to compare the tiny sample of HongKong with anything, as you correctly say. But it is nonetheless highly informative for example that 8 out of 9 UK patients and 13 out of 16 German patients have been released after x number of days fully recovered, whereas the remainder are still in hospital after y days - that tells about the duration of recovery, which then allows us to calculate the mortality rate. There are so many useful things you can do with the data. We are all aware of the caveats and shortcomings. Please let us be the judge whether we can handle raw data, not you.86.161.82.254 (talk) 15:32, 19 February 2020 (UTC)
We have text about the CCDC paper describing the mortality rate. We don't need raw numbers if someone else, a reliable source, has done the work for us. Bondegezou (talk) 18:54, 19 February 2020 (UTC)
You are getting confucius. This thread is about the recovery column in the table, which was deleted this morning without much warning or support. (Perhaps someone felt the recoveries column made the Chinese government look bad, and should be suppressed? No, surely not.) 86.161.82.254 (talk) 19:15, 19 February 2020 (UTC)
I support reinstating the recoveries column for the reason stated above, namely estimation of bounds on the death rate. In addition, people who have been following the page may now wonder if these recoveries are now in question - as those previously "recovered" now appear to be "sick". If we are worried that readers will mistakenly compare between the percentages for China and other countries, we can add a disclaimer explaining why direct comparison of the data from different countries may not be strictly correct.77.125.11.104 (talk) 20:46, 19 February 2020 (UTC)
I also support reporting recoveries. Once the number of new recoveries and deaths starts to exceed the number of new cases being reported, that will indicate that the total number of infections is starting downwards. That will be a very good sign. EMS | Talk00:08, 20 February 2020 (UTC)
It is extremely misleading to do so, if this virus is anything like other coronaviruses the amount of infected people is 100 to 1000 time more then the confirmed case. There is no evidence that it's not the case here. There is up to 99.9% of the people who do not seek medical attention. By the look of things, this virus gonna stick around a decade or two. Iluvalar (talk) 21:17, 20 February 2020 (UTC)
A couple of days ago the media reported that the Egyptian "coronavirus" patient turned out to be negative after RNA/DNA testing. The diagnosis was in error. This would mean Egypt needs to be dropped from the league table and from the map. Can someone follow this up please? 86.161.83.159 (talk) 10:01, 20 February 2020 (UTC)
He is on his way to recovery and now tests negative for SARS-CoV-2. That means that his immune system got rid of the virus, or has almost got rid of it, to below detectable levels. That's not an error in diagnosis, and it's not a reason to drop the case from the tables. This table includes the Egyptian person as recovered. Boud (talk) 21:36, 20 February 2020 (UTC)
Thanks for the update. So we leave the table as is. (By the way, your source does agree with my source in saying "State TV earlier said a foreigner thought to be carrying the coronavirus had tested negative, raising questions about his initial infection." My hunch is thst someone in the lab got the RT-PCR wrong thus creating a false positive, and now they are pretending the patient's immune system has claared the alleged "virus" in just a couple of days and therefore the follow-up test is negative. This saves the reputation of the lab. I hope someone had the wit to store an aliquot of the original sample, for future investigation.) 86.161.54.108 (talk) 06:57, 21 February 2020 (UTC)
Mexico is clear
In Mexico, there are no longer suspected and confirmed cases of the Wuhan coronavirus. All were negative. Any questions? --МОДОКАУ06:38, 21 February 2020 (UTC)
9 day survival time for coronaviruses: literature review value
I had to restore this text in the prevention section on information from a review of medical research literature on the survival time of coronaviruses. Someone had left in a reference which has a brief quote by a scientist who says "a few hours" and removed the literature review estimate. The original reference is set earlier; for convenience, here it is: <ref name="GunterTPS_9days_persistence">{{cite journal | last1 = Kampf | first1 = Günter | last2 = Todt | first2 = Daniel |last3=Pfaender |first3=Stephanie |last4=Steinmann |first4=Eike | title = Persistence of coronaviruses on inanimate surfaces and its inactivation with biocidal agents | journal = [[Journal of Hospital Infection]] | volume = | issue = | pages = | publisher = Elsevier | date = 6 February 2020 | url = https://www.sciencedirect.com/science/article/pii/S0195670120300463/ | url-access = <!-- (subscription/registration/limited) default=free --> | issn = | doi = 10.1016/j.jhin.2020.01.022 | pmid = 32035997 | id = | biorxiv= | accessdate = 15 February 2020 | archiveurl= https://archive.ph/Ftr5w | archivedate= 15 February 2020 |url-status=live <!-- live|dead|unfit|usurped -->}}</ref>. [I said "review" here, because I'm not sure of the difference between a "review" and a "meta-review".] The source of this reference was from discussion on this page (search in the archives). Boud (talk) 18:53, 20 February 2020 (UTC)
It is unencyclopedic to provide medical advice that the virus only persists a few hours on surfaces, when reliable scientific and medical sources say it might persist over a week. The questionable low estimate could lead to someone assuming it is safe to move into a room or ride in a bus used by an infected person the very next day or 12 hours later without disinfecting it, resulting in new infections.. Edison (talk) 17:10, 21 February 2020 (UTC)
There is a huge need for more sources on "silent countries" and also more coverage on xenophobia
After epidemy outbreak in Iran, it is clear that the countries with less medical advanced medical equipment have the epidemy going on uncovered which is very disturbing. I believe the public needs to know more about it as it looks like we lost the window of opprtunity to curb the epidemy and it is going viral. I'm asking all the editors who know other languages to watch and follow reliable sources in their countires (Africa, Asia,Middle East) and share it here on the Talk page, so the other editors can double check their authencity.
Also, there is number of reports on rising sinophobia and general phobia of people who are afraid of those infected with coronovirus. I think it should be also covered in neutral tone. Here is the first alarming signal from Ukriane:
2020 coronavirus outbreak in Iran. There are refs for four COVID-19 deaths. WP:OR says that for the upper limit case-fatality ratio of about 2%, four deaths implies about 200 (give or take 100 or so) SARS-CoV-2 infections; a lower ratio (generally thought to be likely) implies a higher number of infections. So "750 casi sospetti" - 750 suspected cases - sounds quite realistic. This is a rather important quote from the NYT source: Already, cases of travelers from Iran testing positive for the virus have turned up in Canada and Lebanon.Boud (talk) 23:37, 21 February 2020 (UTC)
I agree that suspected cases might be worth tracking as well as people who have been quarantained or have outstanding test results. GuntramGraef (talk) 03:41, 22 February 2020 (UTC)
An ongoing epidemic of coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 started in December 2019. It was first identified in Wuhan, capital of Hubei province, China. It was originally a zoonosis, crossing over to humans from an animal, but subsequently started spreading between people.
Because when linked SARS-CoV-2, the link redirects to Severe acute respiratory syndrome coronavirus 2 and the link need to replace to long-form name instead and word "province" after Hubei needs to removed because it already known that Hubei is a province in China. 180.245.214.207 (talk) 00:00, 22 February 2020 (UTC)
The table says "Deaths on international conveyance: 2", but there is nothing else about it in the article or elsewhere. What's the scoop?24.108.56.72 (talk) 05:05, 20 February 2020 (UTC)
Needing expansion is one thing but one sentence sections ("Environmental impacts") or those that call into question if the article "reasonably covers the topic, and does not contain obvious omissions or inaccuracies", such as the incomplete "List of interprovincial medical aid teams that went into Hubei", is a concern per #2. The shut down casinos (Macau section) have been opened meaning the information is outdated. Otr500 (talk) 11:31, 21 February 2020 (UTC)
Thanks for the comments but there may be a contradiction. I also don't see the relevance that there are article issues and what seems to be a dismissal of concerns directly related to article assessment.
According to Wikipedia:Content assessment#Assessing articles (Who can assess articles?) "In general, anyone can add or change an article's rating." Some projects stipulate that "Any member is free to add or update the rating of an article", (below A-class that requires two editors) which might only involve joining the project, and some individual WikiProjects may also have more formal procedures for rating an article. Sometimes improvements can be achieved through the article talk page, without going to the projects or the more broad community.
My "goal" was not that my comments be summarily dismissed, seemingly that I have to go find a second editor that agrees, but to call attention to issues I observed and note them here in hopes someone will resolve them.
Of the five WikiProjects involved with this article all seems to follow the general assessment criteria. Sometimes an article can be promoted by mistake, by the addition of content that may degrade the article or addition of a maintenance tag, by a blanket promotion (one editor assessing several projects to match), or maybe by a bot.
There are citation needed tags on the article, that is a red flag to article assessment, so brought it up here on the talk page. It is as simple as that, and at a point a concerned editor (or maybe more than one) should probably take a look at it so article reassessment is not warranted. Thanks, Otr500 (talk) 09:27, 22 February 2020 (UTC)
Iranian Mayor
State TV in Iran initially reported that the mayor of Tehran’s district 13 tested positive for the virus. However, an official later denied the report. [3] Either way it’s a pretty big deal given the individual’s high profile. 63.231.141.132 (talk) 11:11, 22 February 2020 (UTC)
Please add a column for active cases (those who are not dead or recovered) to the charts. That gives a clearer picture of the current strain on the health systems and size of the known carrier populations. GuntramGraef (talk) 03:39, 22 February 2020 (UTC)
Why not configure indefinite pending changes protection in the article
Hello, I see in the article that this article is semi-protected so only registered users can edit. But for me semi-protected is too strong to prevent IPs and new users from editing the page. So, why not implementing indefinite pending changes to this article, so new users as well as IP users can "good faith" edit the article, but will subject to "autoconfirmed" review if the latest accepted revision was accepted by reviewer. For me, three times this article is semi-protected was fail to supress IP editors. — Preceding unsigned comment added by 110.137.187.159 (talk) 14:30, 22 February 2020 (UTC)
I don't really understand the state of policy for when pending changes may be used, but what I do get out of WP:SEMI and WP:PCPP is that PC is more appropriate for pages with a low editing rate as an alternative to semi-protection. As this is an ongoing current event, I think the pending changes policy says we should use semi-protection. 199.66.69.88 (talk) 21:31, 22 February 2020 (UTC)
New suggested content to the section: As of 18 February, most universities in South Korea had announced plans to postpone the start of the spring semester. This included 155 universities planning to delay the semester start by 2 weeks to 16 March, and 22 universities planning to delay the semester start by 1 week to 9 March. (Reference: University News Network (Published: February 8, 2020. Obtained: February 23, 2020): http://news.unn.net/news/articleView.html?idxno=226388)
New suggested content to the section: Following an outbreak in the administrative area of Daegu, the Daegu Office of Education decided to postpone the start of every school in the region by one week. (Money Today (Published: February 23, 2020. Obtained: February 23, 2020): https://news.mt.co.kr/mtview.php?no=2020022313237612860)
New suggested content to the section: Shinsegae, one of the major South Korean department stores, made the announcement on 23 February that one of its food sections in southern Seoul will close due to a coronavirus infected individual visiting the store's food section. (Reference: Yonhap News (Published: February 23, 2020. Obtained: February 23, 2020): https://en.yna.co.kr/view/AEN20200223000800320?section=coronavirus/news)
New suggested content to the section: Following the outbreak, Samsung decided to shut down the smartphone production facility in the city Gumi (Gyeongsangbuk-do), following an observed coronavirus case. Other employees who may have been infected by the individual are quarantined. (Reference: Yonhap News (Published: February 22, 2020. Obtained: February 23, 2020): https://en.yna.co.kr/view/AEN20200222002700320?section=coronavirus/news)
New suggested content to the section: On 22 February, the South Korean military manpower agency made an announcement that conscription from the Daegu, the administrative area of South Korea with the most observed cases of the coronavirus at the time, will temporarily be suspended after concerns of spreading coronasvirus within related government agencies. (Reference: Yonhap News (Published: February 21, 2020. Obtained: February 23, 2020): https://en.yna.co.kr/view/AEN20200221006900325?section=coronavirus/news)
I suggest adding the following image, as it relates the above statements to the presence of coronavirus in different administrative areas of South Korea: Thematic map of confirmed cases of COVID-19 in South Korea.Statistologist (talk) 07:45, 23 February 2020 (UTC)
In northern Italy in the last 48 hours there has been an explosion of contagions (73 with two deaths). The government has quarantined 11 municipalities in Lombardy and Veneto, closed the universities and suspended all sporting events in three northern regions. I would like to know if all this goes in the article and if yes where. thank you. Alex2006 (talk) 06:44, 23 February 2020 (UTC)
The Chinese phase seems to be over, sort of, and we are seeing a woldwide spread (probably under-estimated). Isn't time to ove to a more global perspective? — Preceding unsigned comment added by Frederic Y Bois (talk • contribs) 20:18, 23 February 2020 (UTC)
Given that there are c. 50 times more cases in China than in the rest of the world (population c. 5 times that of China), I don't think we can say the Chinese phase is over yet. If the Chinese reported counts can be trusted (and there are doubts), then the daily counts of newly reported cases in China are still c. 2.5 times those outside. Is there particular information you think is missing in the current presentation, that could be added? If today's news out of Italy (of an upsurge there) foreshadows similar activity elsewhere in the world, splitting this page somehow into national or regional pages might be useful, but I'm not in a rush to try that. - Inkwzitv (talk) 21:35, 23 February 2020 (UTC)
Date discrepancy regarding latest change in how Chinese count cases
Yeah, I would have to agree. I mean technically it is an outbreak, but more accurately it is an epidemic. I mean if you look at the list of epidemics on Wikipedia, COVID-19 is one of the strains causing ones. Dannelsluc (talk) 00:44, 23 February 2020 (UTC)
its numbers reflect an epidemic[4], it simply has a low case fatality rate (the original posting editor above mentioned the Ebola epidemic, for comparison the current Kivu Ebola epidemic/outbreak has a CSF of 67%)--Ozzie10aaaa (talk) 12:11, 23 February 2020 (UTC)
Should be declined, since this isn't a request for a specific change from "A" to "B", but it's always worth discussing the use of technical terminology. If there are specific concerns we can always include explanatory footnotes (see {{efn}} and {{notelist}} if unfamiliar). I see that the article as-is doesn't have any explanatory footnotes, and in something that bridges highly technical (virology and infectious diseases) and highly popular (public health, international relations, politics), it's usually not a bad idea to use explanatory footnotes to make the article accessible without breaking up its readability. 199.66.69.88 (talk) 18:57, 20 February 2020 (UTC)
The IP editor is incorrect. MEDRS approves of "guidelines or position statements from national or international expert bodies", which includes WHO and the CDC. Bondegezou (talk) 14:00, 21 February 2020 (UTC)
Those links are neither guidelines nor position statements. Guidelines means evidence-based medicine guidelines, that is, secondary or tertiary sources that provide recommendations citing sources. For instance, the Guidelines for drinking-water quality, https://www.who.int/water_sanitation_health/publications/2011/dwq_guidelines/en/. Position statements refers to a shorter form guideline on a very specific topic, such as a specific treatment or similar, that is also a form of evidence based medicine document. For instance, the Rabies vaccines position paper, https://www.who.int/immunization/policy/position_papers/rabies/en/. Public affairs documents, such as news releases and unsigned FAQs are neither guidelines nor position statements. Furthermore, the purposes for which MEDRS allows such documents is not for settling whether something is an outbreak or epidemic, but whether a treatment is efficacious and supported by the medical literature. It’s a bit like claiming that something written in the style of WebMD is a MEDRS-compliant source. 199.66.69.88 (talk) 14:41, 21 February 2020 (UTC)
Covid-19 is rather new, as you may have noticed. We're not going to get quite the same materials on it as we do with guidelines for drinking-water quality. But these WHO, CDC etc. documents are based on the best evidence available, and clearly satisfy MEDRS. They are not "public affairs documents": they are public health documents. Bondegezou (talk) 14:55, 21 February 2020 (UTC)
Did you even look at those links? Or have you ever read a medical research or policy document? These public-targeted information resources, like the pamphlets you find in a doctor’s office, and trying to shoehorn them into the same category as actual research and position papers is frankly offensive to everyone who actually works in healthcare. 199.66.69.88 (talk) 15:02, 21 February 2020 (UTC)
These resources are like the pamphlets you find in a doctor's office. But they are based on the best evidence available from reputable sources. And they are neither guidelines nor position statements. I am baffled by your quixotic objections [...] I could say the same. 199.66.69.88 (talk) 16:34, 21 February 2020 (UTC)
Officials from WHO and other medical services have used the term "epidemic" in relation to this, as have many other news sources. The WHO has only come out saying that it isn't a pandemic. Tsukide (talk) 07:23, 22 February 2020 (UTC)
I made a section on this talk page about this topic before where consensus was agreed against using simple English on this article.
Trying to read in simple English is cumbersome and doesn't flow easily. It conveys too little information across more words and is generally more vague. We are presuming that any reader of normal Wikipedia would have reasonable literacy skills and have come across healthcare information before, so there's no reason to remove "normal" words from the article. These complex words and phrases are similar to legal doublets in that they convey a complex meaning that would otherwise take many words to explain. I agree that there are words which can be too complex (PCR test was the example I used last time), but even these terms can be used if written within context.
I think that anyone who is unable to read the lead in simple English should be guided to the Simple English Wikipedia article since much of the information here is written in normal prose. The relevant articles on that encyclopedia need to be expanded by the users who constantly write simple English on this article. A URL could be placed at the top of the article to the Simple English Wikipedia article. Tsukide (talk) 07:17, 22 February 2020 (UTC)
Questions What your misunderstanding about this articl writting in Simple English. For me, written formal English, not colliqual can used to wrote this article. I have two questions for you. Are you doesn't fluent in English because i suspected that you only fluent in Japanese and Korean? Are you also editor in Simple English Wikipedia? If you are editor of that Wikipedia, i can understand. — Preceding unsigned comment added by 110.137.187.159 (talk) 07:34, 22 February 2020 (UTC)
I think that anyone who is unable to read the lead in simple English should be guided to the Simple English Wikipedia article This is more a general complaint that's applicable to any article where there's a Simple English Wikipedia version. I know the current "theme" kind of buries the interlanguage link a little more than perhaps it should be, but what I think Tsukide is calling for—basically a hatnote or box that essentially says "For a Simple English version of this article, click here"—is likely contrary to MOS:SELFREF, and in any case raises questions of, if we start putting interlanguage links for Simple English front and center, why not put others front and center?As Doc James has said, the language used in this article is not hopelessly technical. But even if it was, the solution would not be to link to the Simple English version, but to create an "Introduction to..." article as described in WP:MTAU. But, again, this article isn't hopelessly technical. 199.66.69.88 (talk) 18:16, 22 February 2020 (UTC)
The article is being restored to how it was. This is simple not correct "local transmissions have been reported throughout the world". Local transmission has been reported in a few locations.
Not sure why this was removed "Hand washing is recommended to prevent the disease." Fairly important.
Tsukide, why did you remove reference to hand washing with this edit? This is the primary line of defense against the disease. This is one of the most important pieces of information in the whole article! Bondegezou (talk) 22:09, 23 February 2020 (UTC)
i'm not sure it's particularly necessary, but when it states that researchers have published the genome i think it should link to this, which is the site containing said published genome
Missing info about spreading the virus. Does a person spread it when only breathing or talking?
The article currently only mentions that the virus can spread when a person sneezes or coughs, but doesn't say anything about whether or not it has been found whether simply breathing and talking can spread the new Corona virus. — Preceding unsigned comment added by 2A00:A040:186:F172:2DAE:239B:F556:5615 (talk) 03:33, 24 February 2020 (UTC)
I initially wrote that carriers can transmit the virus through respiratory droplets that people "sneeze, cough or exhale" based on a EU health advisory, combined with the information that asymptomatic carriers can also pass on the virus, but User:Doc James repeatedly simplifies the article and removes "exhale" from the text. Tsukide (talk) 06:24, 24 February 2020 (UTC)
Increased fatality rate
Pretty long article, but it is missing one of most important data, the fatality rate. And do you know why it is so sad, because it was in the main article, even in the 2nd paragraph say on Revision as of 07:01, 29 January 2020,
"The first confirmed death from the coronavirus infection occurred on January 9 and since then 132 deaths have been confirmed, with a case fatality rate of 2.2%."
And what is the current state of the art: 2466/78876, that is 3.12 percentage, so it is increasing, and don't come to me that it is original research, it is grade 2 maths (division of two integers). 87.97.82.120 (talk) 10:35, 23 February 2020 (UTC)
We should discuss the mortality rate in the lead and in the main body of the article. We used to say something in the lead. When we discuss the mortality rate, we should not be trying to calculate figures ourselves (because it isn't as easy as dividing one number by another), but we should be citing the journal articles and preprints that have come up with figures. Bondegezou (talk) 10:37, 23 February 2020 (UTC)
Complicated subject. Mortality rate varies by age, pre-existing conditions, quality of care, whether unapproved anti-virals are used, work while sick, etc... . Data from a large scale China study says mortality rate is 1% for under 50, 2% for 50-60 and 15% for over 80 people. Daniel.Cardenas (talk) 16:37, 23 February 2020 (UTC)
Not that complicated, see what is in for example at sars article: "outbreak of SARS in southern China caused an eventual 8,098 cases, resulting in 774 deaths reported in 17 countries (9.6% fatality rate)," you need to divide two numbers. You can further study this in different age groups/sex, but what matters is this fatality rate. 87.97.82.120 (talk) 18:43, 23 February 2020 (UTC)
SARS appears to have been completed eliminated. Case fatality rate: CFRs are most often used for diseases with discrete, limited time courses, such as outbreaks of acute infections. CFR can only be considered final when all the cases have been resolved (either died or recovered). We are nowhere near an elimination phase of SARS-CoV-2. Models have to be made to guess the most likely true numbers of cases (especially difficult in mainland China, for both political reasons and the fact that the Hubei health system is overwhelmed; and obviously unrealistic in Iran, where the naive (divide official values for the same date) case fatality rate started at 100% and has dropped to 20%, still about 10 times higher than everywhere else in the world. So far, case fatality rate does not give any formal definition of naive case fatality rate; I just made up this expression using common English. Even if epidemiologists/statisticians made a formal definition of the naive case fatality rate (handling incomplete, inhomogeneous data is always possible, but Bayesian hypotheses about the unknown probability distributions have to be made and mathematically this is not trivial), it would be more relevant to the sociology of how people react to incomplete information than to the outbreak itself. Convince some epidemiologists/statisticians to publish a formal paper on the naive case fatality rate, wait for it to be peer-reviewed, and then try again to see if you can convince Wikipedians that it's worth including. Boud (talk) 21:52, 23 February 2020 (UTC)
The "moving target" aspect of this is important. Plus the best the world (and Wikipedia) will ever be able to do is identify deaths/confirmed cases, but what about all the asymptomatic (or lightly symptomatic) cases that never get tested? (CDC Director Dr. Robert Redfield said 13 Feb that "What we don't know though is how much of the asymptomatic cases are driving transmission", so it looks like the fatality rate among all infected cases will be lower (maybe significantly) than that among confirmed cases.) Tho' I calculate the number myself in my head every couple days, unless we get it from WP:RS I wouldn't include it here. - Inkwzitv (talk) 22:15, 23 February 2020 (UTC)
The case fatality rate isn't a mortality rate, AFAIK there is no known mortality rate quoted cos it is so early. Any indication of the number of deaths compared to number of cases needs to come with that sort of disclaimer. The overall CFR changes and is a very crude underestimate of the mortality rate early in an epidemic but becomes more reliable as time progresses, dependingn on the spread. I made a graph on it under the RfC. --Almaty (talk) 08:44, 24 February 2020 (UTC)
The new chart is too misleading and could increase public panic
A combination chart of 2020 coronavirus cases, deaths and epidemic curve.
I am a creator of the chart. If "most reader" would interpret it as if the death toll is over 10,000, I will change it to left scale, but it will be unreadable. I will wait for more opinion here.―― Phoenix7777 (talk) 13:47, 22 February 2020 (UTC)
I think some tweaking of the graph should suffice to make clear which the death toll curve is. Perhaps labels could be superimposed over the plotting area and near the relevant lines, not just on the axes and reliant on colour matching? Bondegezou (talk) 15:37, 22 February 2020 (UTC)
I wondered about separating this into two graphs, but it's not clear what's the best way to do that is. You could have a cases graph, which still has 2 scales (for cumulative and daily) and a separate deaths graph, which solves Jabhilee's concern. Bondegezou (talk) 16:50, 22 February 2020 (UTC)
There is already a graph under Estimates that gives cumulative figure and death using a semi-log plot. The only thing different in the new chart is the number of new cases per day, which is also given in other plots. There might be too much overlap in the graphs already. Perhaps we should discuss how to merge graphs with the same data sets or split them with different data set so that there isn't too much duplication in the data presented. Hzh (talk) 17:24, 22 February 2020 (UTC)
Phoenix7777, I agree with the IP editor that the orange for "Cases (cumulative)" is too close visually to the red for "Deaths (cumulative)". I recommend that you change the orange to another contrasting color that cannot posdibly be confused with red by people with normal color perception. Otherwise, I find the graph useful. Cullen328Let's discuss it18:30, 22 February 2020 (UTC)
I just ran a quick check actually. The red ("Deaths (cumulative)"=#FF0000) and the orange ("Cases (cumulative)"=#ED7D31) have a contrast ratio of 1.443:1. The two shades of blue vertical bar do not contrast very well either. The darker shade ("Confirmed cases (daily)"=#5B9BD5) and the lighter shade ("Clinically diagnosed cases (daily)"=#BDD7EE) have a contrast ratio of 1.99:1.While MOS:COLOR is primarily concerned with text-to-background contrast, I believe the same principles for contrast ratio can be used as a rule of thumb. MOS:COLOR states that we should "Ensure the contrast of the text with its background reaches at least Web Content Accessibility Guidelines (WCAG) 2.0's AA level, and AAA level when feasible". For AA level, for text-on-background, 4.5:1 for smaller text, and 3:1 for larger text. For AAA level, again for text-on-background, 7:1 for smaller text, and 4.5:1 for larger text. Again, we're not talking about text, but I think we can use this as a rule of thumb or guidance.There's a convenient table of colors that can be used which are all AAA-compliant when used together that should provide sufficient variation to convey the same information. See Wikipedia:Manual of Style/Accessibility/Colors. 199.66.69.88 (talk) 20:19, 22 February 2020 (UTC)
I like the new graph. Many thanks for the effort. In answer to the concern above, I am not confused into thinking that there are 10,000 deaths - it is well publicised in the media that that the number of infected people is much greater than those who are overtly ill or dead, and therefore there is no possibility that I would think there ar 10,000 people dead. 86.161.82.81 (talk) 19:28, 22 February 2020 (UTC)
I agree that the use of the left and right-hand scales together can easily be confused. I think I was initially confused but solved the problem because I'm familiar with the general scales of the numbers. It's not reasonable to assume that the typical reader is already familiar with the scales of the numbers. The addition of the two values of the dotted curves definitely helps! I propose to also replace the legend Cases (cumulative) by Cases (cumulative; left axis). Since this is the only one of the four using the left-hand axis, the reader can guess that the other three use the right-hand axis. Boud (talk) 20:32, 22 February 2020 (UTC)
Two of the semi-log plots I added have been removed without adequate discussion by an unregistered user. I've added them back as different versions that contrast cumulative and daily cases and deaths by region. Semi-log plots allow visualisation of changes over orders of magnitude unlike linear plots. Straight lines on semi-log plots also are diagnostic of exponential growth or decline. I have added them back. When world cumulative totals differ meaningfully from cumulative totals in China I will add that data to the first graph. I would appreciate help with formatting as I am using the line break (br) to tidy presentation.Galerita (talk) 05:03, 23 February 2020 (UTC)
I find it hard to get my head around the new chart. It mixes daily and cumulative and different axes scales as well as line and bar charts and begs the question, "Where are the daily deaths?". Clinically diagnosed cases (daily) suddenly appear from behind the confirmed cases (daily), but then disappear. A lot of background knowledge is required to understand what the difference between the two is - it's due to changes in the case definition. And any linear plot becomes unreadable as numbers vary over orders of magnitude. Numbers before 24 January are impossible to estimate from the plot, but are important for understanding that this was the time of the fastest growth. Galerita (talk) 05:16, 23 February 2020 (UTC)
Okay have trimmed for now. No need to repeat the same thing in a slightly different format.
As expressed in detail when proposing other graphs, the current graph is not replicated in a reliable published source. Neither is this alternate. Both could increase panic to the general reader, IMO. I would prefer to simply have a standard epidemic curve for China, as that is in the most reliable sources, interpretable to the general reader and the epidemiologist. --Almaty (talk) 09:56, 23 February 2020 (UTC)
Thanks Almaty. The daily plots are epidemic curves, but on a log scale. For example cases in China are falling, especially outside Hubei. I've used a semi-log plot because the outbreak spans several orders of magnitude. On a linear plot the difference between 1 case and 10 cases is not discernible when the maximum is about 15,000 cases in one day. Similarly cases in China but outside Hubei have fallen to 11, at least officially, by the most recent count. If the plot were linear the continued decline would not be clear. The dot plots allow the comparison of several regions as well as a total. Chinese cases will soon fall below ROW cases so stacked bar plots would obscure this. Another issue is the outbreak is likely to span months. Standard epidemic curves are hard to read over long periods, unlike line plots. The semi-log plot avoids frightening readers because large increases in absolute terms appear smaller on a log scale. For example an increase from 10 to 20 will look the same as an increase from 100,000 to 200,000, as both are 100% increases. The cumulative cases and deaths will be reported in the press, but again the semi-log plot allows comparison of both cases and deaths from both China and the ROW, a key point of interest for the moment. I've chosen an approach that is as concise as possible. I'm a statistician and data presentation is an important aspect of my trade. I realise this is a non-standard presentation for epidemiologists, but I considered that. The main issue is that the semi-log plot is not a familiar device, although logarithms and semi-log plots would be part of the mathematics education of people who have completed high school. I've carefully labelled the y-axis and provided reference lines for the intermediate numbers, eg 10 and 100 are labelled and reference lines allow 20, 20, 40, 50, 60, 0, 80 and 90 to be read. I have received a lot of thanks from other readers saying the plots are the most useful on the outbreak they have seen. Galerita (talk) 03:21, 24 February 2020 (UTC)
A suggestion - perhaps it might be more easily readable using different colours for the plot? Just introduce one or two extra colours (rather than shades of grey or different geometric shapes) would make the plot and the crowded region easier to read. Hzh (talk) 11:33, 24 February 2020 (UTC)
A semi-log chart is a king of WP:OR. There is no semi-log epidemic curve in the world. You can see epidemic curve which WHO publishes in its daily situation report. Also an editor who says this chart is "the same thing in a slightly different format" seems to have no knowledge of epidemic curve.―― Phoenix7777 (talk) 11:49, 24 February 2020 (UTC)
No, a semi-log chart is not WP:OR, it's just a different way of representing data, commonly done when the range is too big [11], and I'm sure you can find many examples of semilog plots - [12],[13] (ebola cases in figure 1). You can also use various kinds of plot (e.g. log [a/1-a] vs time, which gives proportion of cases over time), but these different types of plots may be too complicated for the casual readers. Hzh (talk) 12:44, 24 February 2020 (UTC)
I said "There is no semi-log epidemic chart (curve) in the world". I know some cases charts (the range is very large) employ semi-log scale. A semi-log chart makes sharp changes of graph dull. So it is not recommended to use for a chart which requires a subtle change of a trend.―― Phoenix7777 (talk) 12:57, 24 February 2020 (UTC)
I assume it is done here because of the differences in order of magnitude between cases in China and the rest of the world as well as cases of deaths. I have no opinion on whether to use a semi-log plot or not, just stating that semi-log plots have been used before (see the examples I gave such as the ebola outbreak one), and normal for epidemic curves - [14]. Hzh (talk) 13:08, 24 February 2020 (UTC)
"Reporting in Barron's observed that the cumulative death statistics released by China "is described by a simple mathematical formula to a very high accuracy"[1] with much less variance than expected and quoted Melody Goodman, associate professor of biostatistics at New York University's School of Global Public Health as saying "As a statistician, it makes me question the data."[1]"
If the CDC was questioning the veracity of the numbers sure. But one biostats person at NYU. Meh.
Without a proper research paper (at least a preprint on ArXiv or BiorXiv), anyone reading the archived version of the source has to speculate quite a lot on what the quoted researchers are trying to say. My feeling is that there are more serious expressions of concern with the data, including from Chinese medical researchers/epidemiologists, to be found; my guess is that CDC, ECDC, WHO have some of these caveats (concerns/criticisms) listed. Boud (talk) 15:31, 24 February 2020 (UTC)
What is the estimated cost of diganostic tests ?
What is the capacity possible, the amount of tests done every day, or in another period of time?
For example, can they be cheap enough so that anoyone can do a test every day so that suspected cases and populations will be detected ealier ? — Preceding unsigned comment added by 2A00:A040:186:F172:2DAE:239B:F556:5615 (talk) 13:44, 23 February 2020 (UTC)
What is the estimated cost of diganostic tests ?, this is an interesting question you raise due to the fact that we(WikiProject Medicine) have been going thru a process of 'whether and how' to indicate medical pricing in our articles, so Im glad you brought this up--Ozzie10aaaa (talk) 00:44, 24 February 2020 (UTC)
Can someone help me remove Iraq from the table? The goverment has confirmed they have no case after the first suspected case was tested negative Vinhson27 (talk) 06:50, 24 February 2020 (UTC)
In the description we have "From 19 February 2020 onwards, only new lab-tested cases were counted towards the total (but clinically diagnosed cases counted earlier were not discarded)" but in these days color used is only orange which means "tested cases". It is inconsistent. It should be another color. Maybe mix of this and the "clinically diagnosed (C.D.) cases" color or another one. — Preceding unsigned comment added by 83.21.237.6 (talk) 04:47, 23 February 2020 (UTC)
Wow I see my suggestion was followed by a fix. :) I am happy to see I was not alone caring about consistency. New colors a little hm wilder, but maybe I will get used to them. Hm now probably this section should be closed. 83.21.237.6 (talk) 18:16, 24 February 2020 (UTC)
Go for it. I think this is a good idea because an international borders status map is very informative, gives a useful additional perspective, and provides us with a quick snapshot of the global containment situation. I recommend that you ask User:Pharexia to create this map since he has already done similar ones for this article. History DMZ (talk) 02:26, 25 February 2020 (UTC)
A preference for sources from the Asia-Pacific, scientific journals and WHO
Due to language barriers and geopolitics, health authorities and general media in the Asia-Pacific region tend to be more updated and thorough than media from the West. Generally, other than scientific sources such as The Lancet, media and health authorities in the East should be preferred to ones in the West (especially NHS, CDC...). Tsukide (talk) 07:55, 23 February 2020 (UTC)
This is such a dubious assertion. It really depends on what the information is, as far as general media goes, I would in no way prefer the ones in the Asia-Pacific region to those in the West, given that "general media in the Asia-Pacific region" would include those in China which tend to be highly deferential to the Chinese government, and often are no more than government mouthpiece. Hzh (talk) 16:06, 23 February 2020 (UTC)
Personally, I would never use the NHS as a source for health information. The US attracts a lot of talent from across the world so it's a good place to find research data, but the Asia-Pacific has some of the best healthcare systems in the world, far better than those in Europe for example. Tsukide (talk) 18:02, 23 February 2020 (UTC)
I have no idea why you are talking about NHS, if there is a specific issue with using an NHS source with regard to this case, then discuss that specific case. Trying to make a blanket comment about NHS, CDC, and other European health authorities is inadvisable. If you think China has a better health care system or medical research or health authority, then say so (but I doubt you can make the assertion stick), trying to use a blanket term like Asia-Pacific is extremely misleading (the difference between Japan and China is vast, even within China, there are huge differences between the top tier cities and other parts of the country). We could in fact argue that this outbreak has happened precisely because of the deficiency of Chinese health and political authorities. Hzh (talk) 20:49, 23 February 2020 (UTC)
@Tsukide: I agree with @Hzh:. I would exercise extreme caution and take any Mainland Chinese source with a heap of salt as they are heavily biased and downplay how serious the situation is. --Tenryuu (🐲 • 💬 • 🌟) 16:15, 23 February 2020 (UTC)
I agree with the others. This isn’t the dark ages. There’s no cause to discount the work of researchers or journalists in other advanced nations simply because the outbreak is farther away. Yes, if we were a newspaper and the subject required up-to-the-second tracking of developments I could see distance and language barrier being an obstacle. But Wikipedia is an encyclopedia. It should wait for new developments to become established and have enduring relevance in the overall story of a topic before including them.That said, I do think there’s cause to ensure our focus in the article reflects the reality that the overwhelming majority of cases and deaths are in East Asia, and particularly in Hubei—that’s a basic matter of understanding how WP:DUE works. But since we have a subarticle on the outbreak within particular geographic areas, we’re probably doing fine in this article. 199.66.69.88 (talk) 16:37, 23 February 2020 (UTC)
But I should also note, contrary to the suggestions above that we should actually discount regional sources because of known biases, I would advise caution with that as well. The rule for Wikipedia is verifiability, not truth. And while guidelines like WP:MEDRS provide guidance for determining whether something is verifiable for medical topics, that does not change the fact that we don’t usually second-guess sources on our own. Fortunately, there are plenty of sources to question PRC government-affiliated sources (and those linked to regimes friendly to the PRC regime), but this matter should be kept in mind when working on this article. We are not a public health information service, we are an encyclopedia. 199.66.69.88 (talk) 16:43, 23 February 2020 (UTC)
The Asia-Pacific covers about half of humanity (including the West: North America, Chile and Australia) with a vast variety of socio-political-economic situations and qualities of health agencies; I'm not convinced that any generalisation is useful here. It seems to me that we are generally favouring the sources that seem the most reliable, which in most cases are the local health authorities. As an example, Singapore detailed and untraced counts are almost all from the Singaporean Ministry of Health and are reported in great detail, including which case is linked to which other case. The fact that the untraced count is increasing very slowly and the overall total count seems to be slowing are consistent with the Singapore MOH being well in control of the situation, despite the number of untraceable cases being, in principle, a sign of loss of control. Iran is in the opposite situation right now, with the naive case-fatality rate dropping, very unrealistically, from 100% on day 1 = 19 Feb to 20% today = 23 Feb, just before and after the parliamentary election. Unless Iranians are much more easily killed by SARS-CoV-2 than non-Iranians (virus designed by the MEK?), the more standard rate of 2% would give the total count to about 400, so it's obvious that there should be about 400, give or take a few hundred, lab-confirmed cases. In this case non-Iranian-official sources could in principle be better, except that that would require Iranian hospital staff to be well-coordinated and to leak the information in a way that is reported by a Wikipedia reliable source. For the moment, we seem only to be using Iranian official sources, as reported either from Iran or via mainstream Western news media. Boud (talk) 21:37, 23 February 2020 (UTC)
Sure, but it's a guideline, not an absolute rule; the word "preferred" is itself non-absolute. There are some topics where the non-English-language sources are clearly more reliable and in-depth for a lot of encyclopedic type knowledge. Boud (talk) 14:10, 25 February 2020 (UTC)
Comment: An Asia-Pacific deputy health minister just appeared at a press conference yesterday coughing and sweating. Today he tested positive for SARS-CoV-2. His main public critic was requested to self-quarantine today after a temperature reading rather than participate in a closed parliamentary session with the Health Minister. He and two others with suspicious body temperature readings participated in the session anyway. (At least, that's what the sources say.) Generalisations about the Asia-Pacific are rather difficult. Boud (talk) 14:10, 25 February 2020 (UTC)
There is a clear consensus that SARS 2 is not yet the common name. There needs to be a clear shift in circumstances before this discussion can be reopened Benica11 (talk) 21:04, 25 February 2020 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Per WP:COMMONNAME I recommend a proposed change to SARS 2 coronavirus outbreak or something similar to make sure nobody gets confused. --burntbuttons (talk) 16:21, 19 February 2020
Neutral at this time, but I think we should look out for this possibly becoming a common term, but I can see arguments for this. Part of the issue is that the "official name" of the virus is indeed SARS-CoV 2, but WHO has (for reasons likely more political and diplomatic than scientific or medical) designated the disease caused by SARS-CoV 2 as COVID-19. The question is this: Is the outbreak of viral infections or of the disease? Considering we're talking about asymptomatic individuals at length—that is, people who do not have signs or symptoms of the disease—there's an argument to be made that we're talking about the spread of the virus rather than the disease. To an extent, COMMONNAME doesn't work so well in that situation, since "outbreak" becomes a qualifier rather than part of the name. But of course, the definition Wikipedia uses for outbreak is centered on the disease rather than the infectious agent. And so we wind up back here. I don't know. Another problem is that the popular press are using terms for the virus and disease interchangeably (I've seen news sources talking about a "virus" called "COVID-19", for example). So really getting at what the COMMONNAME is becomes very problematic—what do we do about misuses? 199.66.69.88 (talk) 22:50, 19 February 2020 (UTC)
Strongly Oppose The outbreak is of the disease not the virus! And the disease page is currently Coronavirus disease 2019. It is that title that this article's title needs to become consistent with, not the virus name. EMS | Talk00:03, 20 February 2020 (UTC)
Oppose The Daily Mail is unreliable and should not, ever, be used as a source, except in rare circumstances. please see WP:RSP. Do not trust whatever the Daily Mail says. Pancho507 (talk) 07:12, 20 February 2020 (UTC)
Oppose "SARS 2 outbreak" is not commonly used, while "coronavirus outbreak" is far more common and more recognisable [15], therefore it should not be changed. Changing it would also likely cause confusion with the SARS outbreak of 2003. Hzh (talk) 17:16, 20 February 2020 (UTC)
oppose we normally talk about an outbreak of a disease, such as flu or common cold. So the article should echo the name of the disease, not the virus--robertpedley (talk)09:34, 22 February 2020 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Talk:2020 coronavirus outbreak in Iran#WHO visit 25 Feb. If Iranian authorities follow WHO's advice, the lab-confirmed cases should quickly get up to 600-5000 or so in a few days as testing capacity and full reporting is improved, and to higher values for growth beyond the present naive-case-fatality-rate implied number of SARS-CoV-2 cases. Boud (talk) 23:33, 24 February 2020 (UTC)
This preprint is cited, "One study found that the usual incubation time was three days but may be as long as 24 days" while they refer to the range as 0.0 - 24.0 (what is that ".0" doing?) their sample of around 1000 cases comes from over 500 hospitals. I am reluctant to place much authority on the long tail here, because the potential for outliers due to poor methodology is huge, 522 institutions are likely to have widely varying isolation protocols, and indeed there is no discussion on how the incubation period is measured. All the best: RichFarmbrough (the apparently calm and reasonable) 15:09, 25 February 2020 (UTC).
Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. Rotideypoc41352 (talk · contribs) 05:27, 21 February 2020 (UTC)
.png)
Already done Benica11 (talk) 20:33, 23 February 2020 (UTC)
Not done: please provide reliable sources that support the change you want to be made.Benica11 (talk) 20:37, 23 February 2020 (UTC)
Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. Benica11 (talk) 00:45, 24 February 2020 (UTC)
Not done. Please follow the semi-protected edit request format of "Replace *X* with *Y*". RayDeeUx (talk) 02:40, 25 February 2020 (UTC)
