Talk:COVID-19/Archive 8

Archive 5

Archive 6

Archive 7

Archive 8

Archive 9

Archive 10

→

Archive 15

Original research

This section appears to be trying to calculate the IFR themselves. Yah it is one way to look at the numbers but I am not sure we should be doing this ourselves. Expecially the table. Doc James (talk · contribs · email) 22:13, 17 April 2020 (UTC)

We are basing the 66% figure on 60 people who were tested in the population. How do we know this is a representative sample User:Jmv2009? These numbers are all very fussy. How many people from that community remain in ICU or are critically ill?

Yes I get it. We all really want to know the risk of dying if you become infected. But we do not clearly know the answer. Yes a lot of low quality sources are making guesses but I am not sure we should include all these. Doc James (talk · contribs · email) 22:28, 17 April 2020 (UTC)

@Doc James: From the first source: "I donatori di Castiglione d'Adda non sono stati scelti a caso (The donors of Castiglione d'Adda were not chosen at random)". You might as well stop there because any attempts to use statistical methods to extrapolate from the data into the background population are doomed by the lack of a random sample. All you can say is "of the 60 people chosen by Avis to be screened, 40 had antibodies." And nothing else. --RexxS (talk) 22:47, 17 April 2020 (UTC)

Adjusted to reflect that. Doc James (talk · contribs · email) 23:04, 17 April 2020 (UTC)

It's not just us. I've found a similar problem with the figures from Robbio, where 2,000 of the 5,900 inhabitants were screened and 14% of those 2,000 tested positive for antibodies. A source in English ('National Interest' blog) made the mistake of extrapolating that 14% of the group who were screened to "14% of residents", which you simply can't do. The best you can say from the data is that somewhere between 5% and 70% of residents are positive for antibodies – assuming the tests were accurate, of course. We really have got to stop using blogs and similar poor quality sources for our health information. --RexxS (talk) 10:36, 18 April 2020 (UTC)

Antibody tests

	Antibodies	Deaths	Ratio deaths/antibodies	Remarks
Castiglione d'Adda	(40 of 60) 67%	1.7%	>1.7%, 2.7%	All deaths
Gangelt	14%	0.06%	0.4%	False positives issue
Netherlands	3%	0.018%x2	1.2%	Incl excess deaths. Plausible spatial distribution
Santa Clara	2.8%	0.004%	0.14%	False positives statistics issue
NYC	19.9%	0.25%	1.26%	Incl excess deaths
NY	12.3%	0.083%	0.67%	Incl excess deaths. Plausible spatial distribution
Geneva	5.5%	0.036%	0.65%
Italy	10%	0.067%	0.67%
Los Angeles	4.1%	0.022%	0.56%	False positives statistics issue. Latest death toll
Belgium	4.3%	0.066%	1.5%
Nembro	?	1.1%	>1.1%	Incl excess deaths
Miami-Dade	6%	0.013%	0.22%	False positives statics issue. Latest death count

While not all infected people develop antibodies, the presence of antibodies may provide information about how many people have been infected.

In the epicentre of the outbreak in Italy, Castiglione d'Adda, a small village of 4500, 80 (1.8%) are already dead. Most people in the village appear to have developed antibodies and possible immunity, most did so without being diagnosed, and many did not have symptoms.

In the German region of Gangelt, where 0.06% of the population has died, 14% have antibodies (15% have been infected and 2% were currently infectious). In Gangelt, the disease was spread by Carnival festivals, and spread to younger people, causing a relatively lower mortality, and not all COVID-19 deaths may have been formally classified as such. Furthermore, the German health system has not been overwhelmed.

In the Netherlands, about 3% may have antibodies, as assessed from blood donors. There, the confirmed deaths from the disease is 0.018% of the population, however the excess deaths with respect to normal circumstances is about twice as high as not all COVID-19 deaths are recorded as such.

In Santa Clara county, 2.8% appear to have developed antibodies. 69 (0.004% of the population) have died from COVID-19.

References

Routine calculation

[1] Please review, as it is somewhat non-trivial. Jmv2009 (talk) 11:05, 19 April 2020 (UTC)

Chance of picking <3 out of 401 when actual likelyhood is 0.0179 is 0.025. Solve[CDF[BinomialDistribution[401, x], 2] == 0.025] [Mathematica/Wolfram Alpha] 1 - InverseBetaRegularized[0.025, 399, 3] Both give 0.0179 [2] For Stanford Santa Clara study.

Now there is a confirmation of this: Joerg Stoye in the comments of the "peer review". Jmv2009 (talk) 14:56, 19 April 2020 (UTC)

It really helps when you actually give links to what you want to discuss. Assuming that "Stanford Santa Clara study" means https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1 you need to be aware of the problems of using a single study to extrapolate to the population in general. The statistical methods for estimating a population value from a measurement of the corresponding sample value rely completely on having a random sample. I would worry that using Facebook ads as a route for collecting a sample may well introduce more than a little unwanted bias to the sample, and if I were refereeing, I'd be somewhat unhappy with the "population-weighted prevalence" being almost twice the raw value. But that's just me. The other question is naturally whether the "lateral flow immunoassay" is actually specific enough for the results to be meaningful, and I think the jury's still out on that one. As usual, I'd say wait for more studies similar to this, to see if the "50-85-fold more than the number of confirmed cases" result is reproducible; and then wait a little longer for some secondary sources to analyse those studies and give us some usable data. --RexxS (talk) 18:45, 19 April 2020 (UTC)

Yes, this study was widely anticipated, and is getting widespread attention losing the statistical disclaimers and other issues in the process, unfortunately. (Apparently even going on FOX "news" television). One can ignore it, or be honest about it. By the way, In Chelsea they appear to have similar results now, although with x10 stronger signals, but unfortunately even less systematically sound. Jmv2009 (talk) 19:10, 19 April 2020 (UTC)

What amaze me the most is that people are surprised about it ? 100x was the rough number back in 2009 when I contributed on H1N1 and it was the same for coronaviruses. I guess we went out of our way this year to test more people so 50-85x sounds about the most predictable outcome. Everyone is so conditioned by media to expect the worst case scenario that the most rational surprise people. Iluvalar (talk) 20:27, 19 April 2020 (UTC)

Don't know what you are saying exactly. I think 1.2% IFR is still reasonable, based on Castiglione d'Adda. They are probably almost done. ~7% deduction of 1.7% is background death rate, 20% deduction of 1.7% is old population, 30% deduction is due to overwhelmed health care. Or something like that. Add 20% for not yet infected. Jmv2009 (talk) 20:43, 19 April 2020 (UTC)

The CFR in california is 3.4%. If there is 50x more infected people the IFR would be 0.068%. Which is coherent with coronaviruses. Iluvalar (talk) 21:03, 19 April 2020 (UTC)

We need to use proper sources. We have a lowest possible IFR in Lombardy of 0.12% (12,213 dead/10,078,012 people) and NYC 0.16% (13,240 dead / 8,398,748 people). These numbers require every single person in these regions to be infected already and no more to die. Nearly 1000 in that region in Italy are still in ICU and lots are still in ICU in NYC.

We have people who are doing small studies and than making extrapolations that are very highly improbable. This is why we require high quality secondary sources for medical claims. To hit an IFR of 0.12% in NCY reincarnation would need to be discovered and more than 3,000 dead people brought back to life. Doc James (talk · contribs · email) 21:38, 19 April 2020 (UTC)

I'll take the blame here. But the population you took is NYC not New-York state. New York have twice as much vulnerable people then average in america. The lack of bed and the panic may explain a slight increase in IFR and with 20% of the population who got the virus in the last months, we'd be facing a large amount of comorbidity. If we are at the peak of the outbreak right now, would you agree that we risk to face twice as much death ? Which would bring your finale estimate to 0.32% ? Or maybe we go 0.16% IFR * 70%/20% = 0.56% IFR ? Don't worry, I know better then going in the page and making homemade OR. But I think this new study, if confirmed, would shrink the worst case scenario by an order of magnitude isn't it ? Iluvalar (talk) 01:42, 20 April 2020 (UTC)

EDIT: To be clear, I DO expect 40% infected before the end, which is much higher then usual, so I'm not just wearing pink glasses here. Iluvalar (talk) 01:56, 20 April 2020 (UTC)

Yes the deaths and population is for NYC. Was not aware NYC was so different than the rest of the USA with respect to risks. What I do know is this is sufficient bad to take out a healthcare system. We are already low on blood were I am. That means if you have a postpartum bleed or MVC things could be tight. If some dies from a lack of blood it is not a COVID death though caused by COVID. How this will all shake out in the end I do not know. And the best avaliable sources say they do not know. Doc James (talk · contribs · email) 06:03, 20 April 2020 (UTC)

I'm sorry but I have to ask to be sure. Does Covid patients require a lot of blood transfusion I'm not aware of ? Or is it more the lack of donor and distribution problem in cause ? I guess you're pointing to another issue: There is IFR and IFR. I think we are not interested by the same. Iluvalar (talk) 19:04, 20 April 2020 (UTC)

Issues

"In Wuhan, a city of 11.08 million, where 3869 (0.03%) have died,"China increases death toll in outbreak city by 50%". BBC News. 2020-04-17. Retrieved 2020-04-19. 2.0 to 3.0% of hospital employees and patients appear to have developed antibodies."Wuhan anitbody tests show herd immunity is a long way off". HotAir. Retrieved 2020-04-19."

The first source does not mention antibodies and the second source is really really poor. Doc James (talk · contribs · email) 21:22, 19 April 2020 (UTC)

This is also very poorly sourced "In Denmark blood donors, out of 1487, 22 (1.5%) were positive, suggesting an infection mortality of 0.16% in a rapidly evolving situation, as the number of acknowledged deaths has since almost doubled." Doc James (talk · contribs · email) 21:30, 19 April 2020 (UTC)

References

Coronavirus_disease_2019#Antibody_tests

This section is very much undue weight. It is based on poor sources and trying to lead our readers to make conclusions that the sources are not strong enough to make. Doc James (talk · contribs · email) 21:29, 19 April 2020 (UTC)

Here is an overview by the NYTs on April 19th 2020.[3] Doc James (talk · contribs · email) 22:28, 20 April 2020 (UTC)

Speaking solely from knowledge of Wikipedia process and not any medical knowledge of this disease (which I have only from reading published papers plus Wikipedia and news), my instinct is WP:NOTNEWS and we should wait until the picture on the antibody tests is clearer, or just flatly state "it is unclear", with source. Reading the current Antibody section, that is not the sense I get, and perhaps it should be. --Calthinus (talk) 01:20, 21 April 2020 (UTC)

Published estimates on asymptomatic case rate

Thoughts about any of these being included:

• Mizumoto et al at Stanford [4] -- currently with 101 citations per Google -- estimates 17.9% of cases on the Diamond Princess were asymptomatic.
• According to classified Chinese data that the South China Morning Post apparently got access to, 1/3 of cases were considered asymptomatic. [5]
• Imperial College COVID-19 Response Team, experts advising the British government [6] : Analyses of data from China as well as data from those returning on repatriation flights suggest that 40-50% of infections were not identified as cases12. This may include asymptomatic infections, mild disease and a level of under-ascertainment. -- so not just asymptomatic, but also those mild enough to fall under the radar/ get mistaken for a different disease, it seems. This one is the one I am least inclined to add (Mizumoto I am most inclined.).

Also, possibly for the pandemic page, (recent) historical estimates on the role of undocumented (not necessarily asymptomatic) cases in transmission:

• A Stanford study on antibody prevalence on Santa Clara County, California [7] -- not yet peer reviewed afaik -- covered here in the Guardian [8] -- n=3330, argues the overall prevalence is much higher than previoulsy thought, affecting 2.5-4.16% of the pop'l, 50 to 85 times more than official figures.
• 266 citations for this paper in Science, estimate on the spread of coronavirus via undocumented cases in its early phases in China [9]: We estimate 86% of all infections were undocumented (95% CI: [82%–90%]) prior to 23 January 2020 travel restrictions. Per person, the transmission rate of undocumented infections was 55% of documented infections ([46%–62%]), yet, due to their greater numbers, undocumented infections were the infection source for 79% of documented cases.

Thoughts? --Calthinus (talk) 01:06, 21 April 2020 (UTC)

Pinging some active users who seem knowledgeable of the topic @Doc James, Gtoffoletto, and Cinadon36:. --Calthinus (talk) 01:10, 21 April 2020 (UTC)

WHO on April 2nd said "There are few reports of laboratory-confirmed cases who are truly asymptomatic, and to date, there has been no documented asymptomatic transmission. This does not exclude the possibility that it may occur. Asymptomatic cases have been reported as part of contact tracing efforts in some countries."[10]

Of course presymptomatic transmission has been documented.

The Santa Clara study is very tentative and we mention it already. Doc James (talk · contribs · email) 01:16, 21 April 2020 (UTC)

We have to be careful. Here is a paper for example were they do not understand the difference between presymptomatic and asymptomatic.[11] Even the academic press gets these confused. Doc James (talk · contribs · email) 01:22, 21 April 2020 (UTC)

This review states "Based on the current data, we do not know whether these patients are only asymptomatic initially after contracting the disease or if they are asymptomatic throughout the course of the disease."[12] but is a little old now. Doc James (talk · contribs · email) 01:26, 21 April 2020 (UTC)

Perhaps the Lai et al study above is itself worth adding? It would seem to be illuminating to the lay man about "what the nature of our uncertainty is". What is your analysis specifically of Mizumoto et al's Diamond Princess analysis? --Calthinus (talk) 01:28, 21 April 2020 (UTC)

We have this from the CDC[13] "We now know from recent studies that a significant portion of individuals with coronavirus lack symptoms (“asymptomatic”) and that even those who eventually develop symptoms (“pre-symptomatic”) can transmit the virus to others before showing symptoms." I think simple saying "A portion of people with coronavirus do not develop symptoms". I am not sure we need an exact number as it is currently unknown.

With respect to the Diamond Princess it is not really a representative sample. I am not entirely against including it just should not get to much weight. Doc James (talk · contribs · email) 01:31, 21 April 2020 (UTC)

But maybe we should include. Here is another CDC page "There have been multiple reports to date of children with asymptomatic SARS-CoV-2 infection.3,6,14,15 In one study, up to 13% of pediatric cases with SARS-CoV-2 infection were asymptomatic.16 The prevalence of asymptomatic SARS-CoV-2 infection and duration of pre-symptomatic infection in children are not well understood, as asymptomatic individuals are not routinely tested."[14] Doc James (talk · contribs · email) 01:34, 21 April 2020 (UTC)

We already have the sentence " The proportion of infected people who do not display symptoms is currently unknown and being studied, with the Korea Centers for Disease Control and Prevention (KCDC) reporting that 20% of all confirmed cases remained asymptomatic during their hospital stay.[50][51]" User:Calthinus is the proposal to change this and how? Doc James (talk · contribs · email) 01:37, 21 April 2020 (UTC)

Doc James In light of all we have discussed... Various studies (cite Miyamoto et al, and Bendavid et al on S Clara -- while removing its existing use) and authorities (cite CDC above, KCDC, and Imperial College study) argue a significant minority of cases may be asymptomatic. Asymptomatic individuals tend to not be tested, and currently our understanding of the roles asymptomatic and presymptomatic cases are not known(cite CDC above, plus Lai et al). As of April 2, the WHO held that there was no documented cases of asymptomatic transmission. -- would this be a good way to summarize the current situation while giving those readers who want to read more the sources to do so? I'm not sure either the Diamond Princess, who-gets-hospitalized-in-South-Korea, and tested cases in China are representative, so significant minority as accurately describing the estimates of 18%, 20%, and 33% would seem to suffice? --Calthinus (talk) 01:46, 21 April 2020 (UTC)

Would condense it further to A minority of cases do not develop symptoms at any point in time. These cases tend not to be tested, and currently our understanding of the roles asymptomatic cases are not known(cite CDC above, plus Lai et al). As of April 2, the WHO found no evidence of spread from these cases but that may just reflect the lack of study. Doc James (talk · contribs · email) 01:55, 21 April 2020 (UTC)

Sounds good. Will do.--Calthinus (talk) 05:23, 21 April 2020 (UTC)

Working it into the section as it stands at present is more challenging. I am just adding refs at present. At present it would seem that based on the above, I am concerned that the estimate that in China four fifths (!!!) of infections were asymptomatic is UNDUE. The source is in BMJ, but at present it has only 5 citations. @Doc James: Should I remove it? --Calthinus (talk) 17:38, 21 April 2020 (UTC)

Yah 4 5th is undue IMO. Have adjusted a bit. Doc James (talk · contribs · email) 22:07, 21 April 2020 (UTC)

IFR

So the whole point of antibody testing is to get at the IFR. I have replaced our discussion based on the popular press, pre prints and our own analysis by secondary sources from Our World in Data, the World Health Organization, and CEBM. Doc James (talk · contribs · email) 02:25, 23 April 2020 (UTC)

It's an improvement for sure. Good job. Is the source from the WHO dating from February is still relevant, knowing that they had about 2 thousand cases at that moment where now we have 2 million ? This being said, I still wait to see the data from those antibody tests, we can wait a bit more I guess. They inform much more than just the IFR. Iluvalar (talk) 05:37, 24 April 2020 (UTC)

All estimates and assessments are preliminary. We are providing them to guide decision making in the absence of better information, and they will continue to be revised or be superceded by the work of others as evidence warrants. [15]. The bold is part of the paper. Iluvalar (talk) 16:11, 24 April 2020 (UTC)

Also from same study : "we estimate that only 2.9 (1.3, 8.0) percent of infections had been reported as confirmed cases through January 25." Iluvalar (talk) 16:17, 24 April 2020 (UTC)

This review is pretty great:

[16]However, the LA confidence range is wrong. [17]

They appear to have used the same test as Stanford for the now notorious Santa Clara study.

As discussed, the results are slightly different from the leaked version[18] but the confidence range is still screwed up: No longer skew to low number of cases. Symmetric error bars. The size and non-skewing of the error bars apparently only account for the shot noise of the positives of the current population, not of the false positives on pre-covid blood. Expect an update on the error bars. Effectively, there are no valid scientific papers yet. Jmv2009 (talk) 18:32, 25 April 2020 (UTC)

The Miami-Dade survey was not included yet, and also seems off (low IFR). They actually don't calculate IRF. I guess the spring break folks are even younger than the Carnival folks in Gangelt. (Hopefully joking, as they were expected to have flown out again) But they may undercount deaths significantly. They may have done the math on specificity wrong again. Jmv2009 (talk) 18:49, 25 April 2020 (UTC)

Miami-Dade : 12 our of 128 false positives. So that study is all screwed up as well, whatever they did to correct for this. Confidence ranges will include IRF of infinity [19] Jmv2009 (talk) 04:41, 27 April 2020 (UTC)

Overprecision

We should not be stating precision in figures that do not justify them. In Coronavirus disease 2019 #Infectious fatality rate, a NYT source says "More than 21 percent of around 1,300 people in New York City who were tested for coronavirus antibodies this week were found to have them, Gov. Andrew M. Cuomo said on Thursday."

First of all that's third-hand commentary (NYT quoting Cuomo, who is quoting an unnamed study), a really poor source.

Secondly, the 1,300 people were grocery shoppers, so not representative of any population other than grocery shoppers. Please remember that the denominator of the IFR includes those who died, and none of those were out grocery shopping.

Finally, even if we consider the sample of 1,300 to represent a random sample of grocery shoppers, a calculated fraction of about 20% has a 95% confidence interval of ±2%, so writing "about 21.2%", as our article did, is nonsensically over-precise.

Can we please stop using third-hand newspaper sources, and then extrapolating from them, and just wait for good quality reliable sources who understand statistics to give us usable figures? --RexxS (talk) 21:07, 23 April 2020 (UTC)

Not really. See Talk:2019–20_coronavirus_pandemic#IFR. Those anti-body tests (as flawed as they can be) are giving similar results in many countries and states. Nature and Reuters... Not just the NYT. As such, the evidence start to stack pretty well on the 20% antibody test worldwide. And I don't know why it surprise people. Iluvalar (talk) 21:34, 23 April 2020 (UTC)

Yes really. When we have a WP:MEDRS-compliant source telling us what the IFR is, then we'll have something to write about. Until then, make sure you confine your OR and speculation to the talk page. --RexxS (talk) 21:52, 23 April 2020 (UTC)

Ok wait sorry, I don't mind the rounding. And now that I looked at where and how the source was used, I'm very confused. I better, leave someone else reply to you if needed. Iluvalar (talk) 23:19, 23 April 2020 (UTC)

Proposal: break out article

I propose that we break out COVID-19 antibody testing. All the issues addressed above can then be dealt with on their own talk page. Between the numbers that we do have, the issues with the tests thus developed, and the various proposals to use antibody testing as a "passport" to allow people to return to work, I think there is now enough to support a separate article. BD2412 T 20:06, 30 April 2020 (UTC)

Symptoms in the infobox

I propose we keep it to the three most prominent symptoms (fever, cough, and shortness of breath). Sure there are a bunch more and we have them in the picture at the top. Sure onset of symptoms can vary a bit and we discuss this in the body of the text. Doc James (talk · contribs · email) 22:06, 20 April 2020 (UTC)

Have added "others" to clarify that these do not represent an exhaustive list. Doc James (talk · contribs · email) 01:27, 21 April 2020 (UTC)

But we’re currently not going with the three most prominent ones.

"Interim Clinical Guidance for Management of Patients with Confirmed Coronavirus Disease (COVID-19)". Centers for Disease Control and Prevention. 2020-04-06. Archived from the original on 2 March 2020. Retrieved 2020-04-19.

Symptom [CDC Interim Guidance]	Range
Fever (sometimes comes later)	83–99%
Cough	59–82%
Loss of Appetite	40–84%
Fatigue	44–70%
Shortness of breath	31–40%
Coughing up sputum	28–33%
Loss of smell	15 [other source] to 30% [two connected sources]
Muscle aches and pains	11–35%

Yes, I’ve seen CDC and/or WHO prominently list “shortness of breath” (don’t a ton of different respiratory conditions cause this? Of course). But all the same, I think CDC’s Interim Clinical Guidance is the most specific recent source we have. FriendlyRiverOtter (talk) 17:39, 21 April 2020 (UTC)

@Doc James: as a medical professional, you know that the onset of symptoms can vary, but the intelligent lay person may not. And their reaction might be, why didn’t you just tell me? So, with fever, we should say “sometimes comes later” or equivalent, at least in the listing of symptoms. And if we ever get a diagram with references (hopefully!), I’d really prefer both. FriendlyRiverOtter (talk) 21:19, 21 April 2020 (UTC)

I have moved it to the body of that section. We do not provide that level of detail in that table for any of the others. Plus it makes the table wider. Doc James (talk · contribs · email) 22:09, 21 April 2020 (UTC)

If we take away the qualifier on fever, in light of:

Interim Clinical Guidance, U.S. CDC, April 6, 2020.

” . . and older adults and persons with medical comorbidities may have delayed presentation of fever and respiratory symptoms.[10],[11] In one study of 1,099 hospitalized patients, fever was present in only 44% at hospital admission but later developed in 89% during hospitalization.[1] . . ”

I mean, if they’re talking about only 44% of hospitalized patients having fever when first admitted, and then we’re giving a figure most of which is in the 90s — going all the way up to 99% which, of course, will really jump out at people! — yep, I’d say we are indeed failing to accurately communicate.

[1] Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ, He JX, et al. (February 2020). "Clinical Characteristics of Coronavirus Disease 2019 in China". The New England Journal of Medicine. Massachusetts Medical Society. doi:10.1056/nejmoa2002032. PMC 7092819. PMID 32109013.

Conclusions [in Abstract]

” . . Patients often presented without fever, . . ”

Results [in body]

” . . Fever was present in 43.8% of the patients on admission but developed in 88.7% during hospitalization. . ”

Discussion [in body]

” . . The absence of fever in Covid-19 is more frequent than in SARS-CoV (1%) and MERS-CoV infection (2%), so afebrile patients may be missed if the surveillance case definition focuses on fever detection. . ”

And yes, it really impresses me that The New England Journal of Medicine felt strongly enough to mention this in the abstract.

Alright, let me suggest an experiment. Let’s just get rid of the table and go with text, and try to be shorter and snappier. I’ve thought about bracketing the section beginning and ending with a mention of asymptomatic. But people generally know there can be asymptomatc patients. Let’s try just including it at the end. FriendlyRiverOtter (talk) 20:47, 22 April 2020 (UTC)

I found the table useful. The table is just a list of symptoms not the timing of the onset of each. That timing can go in the body including stuff like how lost of smell often presents early. Doc James (talk · contribs · email) 01:43, 23 April 2020 (UTC)

I really thought that for professional publications, any photo, graph, table, etc, has to be able to stand entirely on its own, as does the text. Is this not the case? FriendlyRiverOtter (talk) 16:00, 23 April 2020 (UTC)

And I’m thinking we actually might be better off with the old version of the table:

In cases with symptoms
Symptoms	%
Fever [sometimes comes later]	88
Dry cough	68
Fatigue	38
Sputum production	33
Loss of smell	15 to 30
Shortness of breath	19
Muscle or joint pain	15
Sore throat	14
Headache	14
Chills	11
Nausea or vomiting	5
Nasal congestion	5
Diarrhoea	4 to 31
Haemoptysis	0.9
Pink eyes	0.8

This is based on:

Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19) (PDF) (Report). World Health Organization (WHO). 16–24 February 2020. Archived (PDF) from the original on 29 February 2020. Retrieved 21 March 2020.

And I’d still want to include a qualifier on fever, basically because it has two sources above saying it needs a qualifier. FriendlyRiverOtter (talk) 20:09, 23 April 2020 (UTC)

————

Yang, Xiaobo; Yu, Yuan; Xu, Jiqian; et al. (24 February 2020). "Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study". Lancet. doi:10.1016/S2213-2600(20)30079-5. PMC 7102538.

Methods

“ . . 52 critically ill adult patients with SARS-CoV-2 pneumonia who were admitted to the intensive care unit (ICU) of Wuhan Jin Yin-tan hospital (Wuhan, China) . . “

Results

”The most common symptoms were fever (98%), cough (77%), and dyspnoea [difficult breathing] (63·5%; . . ) . “

Discussion

“We also found that fever was not detected at the onset of illness in six (11·5%), and that it was in fact detected 2–8 days later. The delay of fever manifestation hinders early identification of patients infected with SARS-CoV-2—if patients are asymptomatic identification of suspected cases is more difficult.”

So, one more source saying that fever is not always at the beginning. This is footnote [10] from CDC’s “Interim Clinical Guidance.” FriendlyRiverOtter (talk) 23:28, 23 April 2020 (UTC)

The source says "The signs and symptoms of COVID-19 present at illness onset vary, but over the course of the disease, most persons with COVID-19 will experience the following1,4-9: Fever (83–99%) Cough (59–82%) Fatigue (44–70%) Anorexia (40–84%) Shortness of breath (31–40%) Sputum production (28–33%) Myalgias (11–35%)"

So the source supports the table. Doc James (talk · contribs · email) 01:52, 24 April 2020 (UTC)

Again, as a physician, you’re very used to symptoms varying at onset. Someone else reading this may or may not know this. There’s no particular reason they would assume this, unless we go ahead and tell them. Plus, I have a way to make the table skinnier, basically using < br > , with no spaces of course, for a page break. FriendlyRiverOtter (talk) 02:09, 24 April 2020 (UTC)

All the qualifiers belong in the body of the text. We do not describe when all the other symptoms start. That loss of smell is often early. We already say in the lead that some people initially have no symptoms. And we describe in the first sentence of that section that some people develop fever later. Doc James (talk · contribs · email) 05:19, 24 April 2020 (UTC)

@Doc James: I can only say that fever needs a qualifier in a way that fatigue, for example, does not. And don’t most professional publications have a standard that graphics must be self-contained and stand or fall on their own? Because many readers skim. I think it’s a heck of a good idea, even if we don’t have a quote-unquote rule to that effect here at Wiki.

If we can do it with less cost. My idea is to go with the old table, which has the advantage of having the same numbers as the body image, with references. And I have an idea for a qualifier with less of a width issue.

I think you’ll be okay with the result, maybe even like it. If not, I’m very open to improvements.

I mean, New Engl J Med’s Clinical Characteristics (Feb. 2020) found that only 44% of Coronavirus patients had fever at hospital admission. Stunning. Phenomenal. No way we can ignore that. FriendlyRiverOtter (talk) 15:04, 24 April 2020 (UTC)

"Stunning. Phenomenal" what? What percentage of pneumonia have fever at presentation? We have this in the text. We are qualifying it in the text right were it should be qualified. We need to avoid trying to put everything in tables.

The CDC is a better source than a primary source in JAMA from Feb 2020. Doc James (talk · contribs · email) 23:17, 24 April 2020 (UTC)

This reference is from Apr 6th.[20] This one is from Mar 16th to 24.[21] IMO it is better to go with the newer source as we learn more about this disease. Doc James (talk · contribs · email) 08:22, 25 April 2020 (UTC)

How about this:

Interpretation and Uses of Medical Statistics, Leslie Daly, Geoffrey J Bourke, Blackwell Science, 2000, page 63:

”A fault that is fairly common is to attempt to show too much material in a graph. In general, it should avoid excessive information and detail and yet be self-contained, in the sense that it should present the essential points without the reader having to search the text for explanations.”

To me, this is saying the graph should achieve both goals.

And please note that I started with CDC as an authoritative secondary source and then dove deeper. In fact, I like to think if I were in a residency program and you one of the attendings, well, that same quality over a much, much bigger workload might warrant an A, or at least a B+, right? FriendlyRiverOtter (talk) 00:03, 26 April 2020 (UTC)

The current graph lists the common symptoms and the percentages. We can have an indepth discussion about when each symptom occurs in the course of the disease. With respect to due weight IMO this should go in the body of the text. Doc James (talk · contribs · email) 03:28, 26 April 2020 (UTC)

@Doc James:, you ask how often there’s fever when pneumonia presents? Other than occasionally hearing the phrase “walking pneumonia,” I have no idea. And I’d say that’s a big part of the point. If a lay person as interested in Coronavirus as I am doesn’t know, I bet a bunch of others don’t either. (even given that pneumonia is a separate but overlapping topic)

You know, I talked up your work back in October on the Fermi Paradox talk page in Aliens might detect our electromagnetic leakage, though we cannot. I think you do great work. Doesn’t mean we have to agree on this topic here.

I’d only suggest that you do a lot of technical articles. And this one draws in a lot of interested lay persons. We don’t need to talk down to them (which we don’t), and in fact which is almost always a mistake. We just matter-of-factly tell them what they don’t know. Plus, I’d personally say that we accept that a significant percentage will just skim and look at graphs and tables and photo captions. FriendlyRiverOtter (talk) 22:48, 30 April 2020 (UTC)

I am supportive of us including when the fever occurs, just not sure it is needed in the table. Lets see what others think and go with that. Doc James (talk · contribs · email) 06:18, 1 May 2020 (UTC)

@Doc James and @FriendlyRiverOtter - I appreciate the thought that you two have already put into this, regarding both the abridged symptoms in the infobox, and the information in the symptoms section. It just seems to me that the information in the symptoms section is leaving out the experience of so many who have had more mild experiences of the virus, for example, without any fever, or any shortness of breath, etc. but who did indeed have some of these other symptoms. Is there some reason this is omitted? I understand if published information surveying incidence of symptoms in mild cases might be lacking as of yet, but shouldn't there be at least a disclaimer along with that table, just at least mentioning that many are believed to have mild "versions" of the virus, or even remain asymptomatic. From all I have been reading for these few months, 83-99% of people with this virus - who will indeed be otherwise symptomatic - do not all get a fever. I think this specifically is the one number that could be most misleading to many who end up relying on this information to inform their observations. I can try to help find something suitable as a source for such a disclaimer. I'm trying not to be biased by my own and the experiences of about 20 people I know, but it certainly motivates me to add into this discussion here.

To this point, isn't the new CDC information you're citing referencing only studies done only on hospitalized cases, which will be biased towards more severe cases? And those studies are all in China, when we know they're are different strains in the world. Could these strains be causing a different effect? Or different populations be more susceptible to fever as a symptom? I mean this merely as justification for my above idea of a mild/asymptomatic case disclaimer.

Also, it was me who removed the word "early" from the description of loss of smell as a possible symptom because: 1) I did not find that word or its implication in any of the cited sources, 2) I have only read about experiences to the contrary (reddit), and 3) my experience and everyone else I knew that lost smell due to this found it to be the opposite - starting mid symptoms with sense of smell only returning as other symptoms were gone or leaving. Yes 2) is anecdotal, 3) is original research, but 1) is not. Spettro9 (talk) 07:28, 26 April 2020 (UTC)

@Spettro9: I would be okay with a disclaimer at the end of the table, or better yet, a good source which gives a percentage estimate.

And good catch on removing “early” from loss of smell. If none of our cited sources say it, we shouldn’t either.

PS It sounds like you do it exactly right. You might get interested in a topic because of the personal experiences of people you know, but then you jump in and do the research and find sources. FriendlyRiverOtter (talk) 19:55, 27 April 2020 (UTC)

@Spettro9: From MOS:INFOBOXPURPOSE:

When considering any aspect of infobox design, keep in mind the purpose of an infobox: to summarize (and not supplant) key facts that appear in the article (an article should remain complete with its summary infobox ignored). The less information it contains, the more effectively it serves that purpose, allowing readers to identify key facts at a glance. Of necessity, some infoboxes contain more than just a few fields; however, wherever possible, present information in short form, and exclude any unnecessary content.

Any piece of information that requires explanation is not suitable for inclusion in an infobox, and keeping the number of items in an infobox as small as possible is a benefit to the reader. The symptoms field in the infobox cannot include all the possible information about symptoms, and it's not its job to do so. --RexxS (talk) 11:47, 26 April 2020 (UTC)

@RexxS: I think it was Einstein who said, As simple as possible, but no simpler. And would you generally agree with this?

I view this particular Infobox as right on the borderline. With CDC’s “Interim Clinical Guidance”, April 6, both giving the range for fever as 83–99%, but also making major qualifications as far as fever often presenting as a delayed symptom. Qualifications which the interested layperson will neither know or assume. So, a person comes to the article with certain misconceptions, does a fairly good job reading and skimming, which I think as we both strongly suspect means more attention to the photos and graphics than to long text, and then pretty much leaves with the same misconceptions. No, I don’t think that’s the best work we’re capable of. FriendlyRiverOtter (talk) 21:19, 27 April 2020 (UTC)

@FriendlyRiverOtter: no, I disagree with almost everything Einstein said. He was a theoretician, not an engineer or a social scientist, and knew nothing of the construction or use of infoboxes. When you have been through the same struggles as I went through in the Infobox Wars, you'll be a lot less cavalier about trying to cram unsuitably large and nuanced factoids into the tiny space available in a properly designed and deployed infobox. MOS:INFOBOXPURPOSE isn't difficult. For readers who want an "at-a-glance" summary of symptoms, "fever, cough, and shortness of breath", as suggested by James in the very first post are the best we're going to get in an infobox. In the body of the text, we can talk about sources showing that lack of appetite is more common than shortness of breath, and why lack of appetite is such a non-specific symptom that it isn't particularly helpful for diagnosis. That's where we can discuss whether anosmia presents early or late (or either) in the progression of the disease, and just how common it is. And so on. But the infobox is for the simplest possible, straightforward presentation of non-contentious facts that are the most useful to the readership. --RexxS (talk) 21:50, 27 April 2020 (UTC)

@RexxS: (and also FriendlyRiverOtter) - I know I'm writing here under "Symptoms in the infobox" section of the Talk page, but I really only meant adding some sort of disclaimer (about mild and asymptomatic cases) to the table in the symptoms section of the article listing symptoms and their incidence. I can understand the infobox being more concise, eg just fever, cough, shortness of breath. (I see the the loss of smell is gone from the symptoms section table but added to the infobox now - I leave that to you and others.) I have 2 questions: 1) Can we just add to that table a "Note - as many as x% of cases could be asymptomatic or mild"? 2) I should know my wiki-etiqutte better, but are we trying to keep these references to journals? Are news outlets acceptable? (Because there are a LOT of estimates to put in that X %) Thank you Spettro9 (talk) 08:40, 29 April 2020 (UTC)

The infobox lists "none" as one of the symptoms. I think that is a disclaimer. The person who is interested and reads that can read more in the body of the article. This keeps the infobox brief as intended I believe. MartinezMD (talk) 13:18, 29 April 2020 (UTC)

@Spettro9: The best advice is in WP:MEDRS, where it describes how to identify the best sources. The highest quality sources are always preferred, so if you have good journal articles, they are preferred over newspaper reports. Newspapers are not considered secondary sources because they are not competent to analyse primary sources, only report them; they add nothing to the original source. For statements that are not biomedical claims, we don't insist on secondary sources, but you still can use primary sources only with care. You can't aggregate figures from sources, for example; the best you can do is to attribute each figure to its source, which is messy when you have a lot of equally good sources. Just quoting a range is also fraught with dangers, as one end or the other might be a complete outlier, thus misrepresenting the true spread of values found in sources. We are not allowed to perform our own statistical analysis on the figures; that has to be done in a secondary source. I think in this case, you're going to find it hard to justify where that "x% of cases" comes from. --RexxS (talk) 13:25, 29 April 2020 (UTC)

Semi-protected edit request on 29 April 2020

A review shows that several vitmain (e.g. Vitamin D, C) and trace elements (e.g. Zn and Se)are useful in both prevention and treatment of COVID-19 [22]. Ranil7 (talk) 20:52, 29 April 2020 (UTC)

@Ranil7: The Highlights states "Supplementation of Vitamin A, D and Zn and selenium may be beneficial for both prevention and treatment of viral infections". Where do you think this information belongs in the article? GoingBatty (talk) 21:32, 29 April 2020 (UTC)

(edit conflict) @Ranil7: Unfortunately that's not what the source reviewed or concluded. It was a review of studies on the means of enhancing immunity in previous viral infections. The intention, of course, was to suggest what might be useful to prevent and treat COVID-19, but it wan't capable of demonstrating effects on COVID-19 itself. Looking at section 4 'Discussion' in the source, it elaborates on what previous research has shown for some vitamins, trace elements, nutraceuticals and probiotics, but it is speculation to assume that what applies to other viruses also applies to COVID-19. It incidentally shows that what we expect isn't always what happens as it reports that vitamin E, which we might expect to improve the immune response, actually has adverse effects in most of the studies.

This paper is almost certainly of value in suggesting to researchers what would be fruitful for future studies on prevention and treatment of COVID-19, but doesn't reach any usable conclusion for us. --RexxS (talk) 21:52, 29 April 2020 (UTC)

I'm not even sure it has that value. Vitamin C has been a treatment in search of a disease for decades ever since good nutrition reduced cases of scurvy. MartinezMD (talk) 03:56, 30 April 2020 (UTC)

~~ At the moment, no studies have conducted for COVID-19 on the nutrutitional aspects. Many clinical settings have started supplimenting vitamin D, Zinc etc. Ranil7 (talk) 07:20, 1 May 2020 (UTC)Comments from non-clinicians are not practical in this kind of epidemic, when you get a solid evidence on vitmain supplimentation on COVID-19, many thousond has to pay the priceRanil7 (talk) 07:20, 1 May 2020 (UTC) And there is virtually no adverse effect of given dosesRanil7 (talk) 07:20, 1 May 2020 (UTC)

References

COVID-19 possibly linked to strokes

A WebMD article published on 10 April 2020 referenced JAMA Neurology which stated that

[a] study out of China finds that strokes, altered consciousness and other neurological issues are relatively common in more serious cases of COVID-19.

Looking at 214 cases of severe coronavirus illness treated in Wuhan city during the early phase of the global pandemic, doctors reported that 36.4% of patients displayed neurological symptoms.

Sometimes these symptoms appeared in the relative absence of "typical" symptoms of COVID-19 -- fever, cough, diarrhea -- the team said.

More recently, publications like CNN have noted that the virus "appears to be causing sudden strokes in adults in their 30s and 40s who are not otherwise terribly ill". Has anyone found similar sources reporting as such? —Tenryuu 🐲 ( 💬 • 📝 ) 23:41, 23 April 2020 (UTC)

Well we should not use the popular press. We should likely wait for a decent review article on statement by a major medical organization. Doc James (talk · contribs · email) 05:24, 24 April 2020 (UTC)

• I think this absolutely must be included, along with effects on kidney, gut and brain. See, for example, "Clinicians trace a ferocious rampage through the body, from brain to toes" by Science (journal). This is a perfectly valid review source per WP:MEDRS. The statements by medical organizations, governments and politicians are a lot less relevant than WP:RS. It tells, for example,

A 25 March paper in JAMA Cardiology documented heart damage in nearly 20% of patients out of 416 hospitalized for COVID-19 in Wuhan, China. In another Wuhan study, 44% of 36 patients admitted to the ICU had arrhythmias. The disruption seems to extend to the blood itself. Among 184 COVID-19 patients in a Dutch ICU, 38% had blood that clotted abnormally, and almost one-third already had clots, according to a 10 April paper in Thrombosis Research.

My very best wishes (talk) 15:22, 24 April 2020 (UTC)

Doc James, your thoughts on using JAMA Cardiology that's been cited second-hand? —Tenryuu 🐲 ( 💬 • 📝 ) 18:48, 25 April 2020 (UTC)

We have reviews in high quality journals that cover this topic.[23][24]

These are what we should be using, not primary sources and commentary on primary sources. Doc James (talk · contribs · email) 00:39, 26 April 2020 (UTC)

@Tenryuu and My very best wishes: The first source you're looking at can be found at "Explanation for COVID-19 Infection Neurological Damage and Reactivations". It's a letter reporting a single study. We really must not be using that as a source for biomedical claims. However, as James says, there are some better sources available.

"Neurological complications of coronavirus and COVID-19" is a review stating

Neurological symptoms have been reported in patients affected by COVID-19, such as headache, dizziness, myalgia and anosmia, as well as cases of encephalopathy, encephalitis, necrotising haemorrhagic encephalopathy, stroke, epileptic seizures, rhabdomyolysis and Guillain-Barre syndrome, associated with SARS-CoV-2 infection.

Future epidemiological studies and case records should elucidate the real incidence of these neurological complications, their pathogenic mechanisms and their therapeutic options.

So we have a qualitative confirmation of a linkage to stroke (among a number of other neurological complications), but no insight into the incidence of stroke - it could be one in a million for all we know.

"Cerebrovascular disease is associated with an increased disease severity in patients with Coronavirus Disease 2019 (COVID-19): A pooled analysis of published literature." is a pooled analysis of four studies. It states

We pooled studies from published literature to assess the association of a history of stroke with outcomes in patients with COVID-19.

A pooled analysis ... showed a ∼2.5-fold increase in odds of severe COVID-19. While a trend was observed, there was no statistically significant association of stroke with mortality in patients with COVID-19 infection.

That tells us that people with a history of stroke are more likely to experience severe COVID-19 symptoms, but are no more likely to die from it. Probably. It's really early days to be making definitive statements, and as usual, I advise waiting for more comprehensive reviews to come online. --RexxS (talk) 01:18, 26 April 2020 (UTC)

We should be very cautious with all those odd rare complications with COVID-19, as a symptom WITH covid-19 doesn't really mean FROM covid-19. If those early antibody tests are half right, that would suggest that a lot of people would go to the hospital having covid-19 for something completely unrelated (and potentially much more deadly). Iluvalar (talk) 01:43, 26 April 2020 (UTC)

"all those odd rare complications with COVID-19, as a symptom WITH covid-19 doesn't really mean FROM covid-19" and some other comments above... Actually, such comments are pretty much WP:OR. Actually, the article in Science (good secondary RS per WP:MEDRS) claims very clearly that all of those are not just symptoms, but result of direct action of COVID-19, it tells about mechanisms, etc. It also tells, with data and references, such things are very common. If more secondary RS would be needed, they are easy to find. That was widely covered. Therefore, I still believe this info absolutely must be included. My very best wishes (talk) 03:01, 26 April 2020 (UTC)

@My very best wishes: Here's what the article in Science actually "claims very clearly":

... pathologists are struggling to understand the damage wrought by the coronavirus ... a clear picture is elusive ... Without larger, prospective controlled studies that are only now being launched, scientists must pull information from small studies and case reports, often published at warp speed and not yet peer reviewed ... Some COVID-19 patients have strokes, seizures, confusion, and brain inflammation. Doctors are trying to understand which are directly caused by the virus ... We’re still at the beginning ... We really don’t understand who is vulnerable, why some people are affected so severely, why it comes on so rapidly … and why it is so hard [for some] to recover.

That's what the article is really telling you: we don't know yet. If you think that secondary MEDRS sources linking COVID-19 to strokes are so easy to find, why haven't you produced them? James has included what little there is to report so far. --RexxS (talk) 04:53, 26 April 2020 (UTC)

Of course more studies are needed. No one denies it. But that has already published in multiple RS (and therefor must be included to the page). One should read whole article in Science. Even the title of it (How does coronavirus kill? Clinicians trace a ferocious rampage through the body, from brain to toes) tells by itself. Well, I do not have time to really contribute here, so it was only a suggestion, but an important one. Look, even "First known coronavirus death in the US was due to a "heart rupture" (see here ). Even that alone justifies the inclusion. My very best wishes (talk) 18:31, 26 April 2020 (UTC)

Good, at least we have agreement that more sources are needed. As for that has already published in multiple RS (and therefor must be included to the page), that is pure nonsense. WP:MEDRS determines whether a source is usable to support a biomedical claim. Mere publication in an RS isn't sufficient, and contributors to medical articles need to be aware of the policies and guidelines that apply. I disagree that the "pop science" article in Science makes any case for a link between Covid-19 and stroke beyond what we already know from the more scholarly sources that I drew to your attention above. --RexxS (talk) 20:10, 26 April 2020 (UTC)

Summary based on reviews

User:Moksha88 you removed "A number of neurological symptoms has been reported including seizures, stroke, encephalitis, and Guillain–Barré syndrome. Cardiovascular-related complications may include heart failure, irregular electrical activity, blood clots, and heart inflammation." Wondering why? Doc James (talk · contribs · email) 13:39, 30 April 2020 (UTC)

@Moksha88: (as James cocked up the first notification) please have another look at Special:Diff/953952119 and consider self-reverting. I don't think your edit summary is accurate. --RexxS (talk) 22:26, 30 April 2020 (UTC)

@RexxS and Doc James: My apologies for not clarifying. Technically speaking, seizures, strokes, and all the other conditions listed are not signs or symptoms; they are complications that arise as the disease progress and would be better incorporated under 'Pathophysiology.' If you look, they are already mentioned there, and I've requested out for some of those papers to do a better rewrite. I pasted those two sentences in my sandbox in the mean time here. Moksha88 (talk) 02:33, 1 May 2020 (UTC)

Never mind, let me merge for now, but 'Pathophysiology' will need a cleanup. I'm hoping to tackle it this weekend once I get the necessary sources. Moksha88 (talk) 02:54, 1 May 2020 (UTC)

We can have a section in signs and symptoms call complications maybe. Doc James (talk · contribs · email) 06:14, 1 May 2020 (UTC)

@Doc James: I was just about to say deja vu when I recalled two prior discussions on this point (1,2). I looked at other highly rated articles again, namely Influenza and HIV/AIDS, and it seems 'Prognosis' is where this type of material landed. Moksha88 (talk) 00:27, 2 May 2020 (UTC)

Yes sometimes it is put in the prognosis and other times it is put in a complications section. I do not have a strong feeling either way. Probably one of these is best. You have thoughts on which we should use? Doc James (talk · contribs · email) 06:29, 2 May 2020 (UTC)

@Doc James: This article is complicated as it is, so let's just incorporate under 'Prognosis.' Moksha88 (talk) 14:14, 2 May 2020 (UTC)

References

Leaked from lab

Discussion here Talk:2019–20_coronavirus_pandemic#Leaked_from_lab Doc James (talk · contribs · email) 19:10, 17 April 2020 (UTC)

The history section is too short. Shouldn't there be, among other things, a reference to the Wuhan Lab as a possilbe source -- or not -- of covid-19? Even the see-also section doesn't seem to lead to more history. For the lab, see, e.g., https://www.japantimes.co.jp/news/2020/04/18/asia-pacific/wuhan-lab-china-coronavirus-controversy/#.XpwZPv0zbIU. Kdammers (talk) 09:33, 19 April 2020 (UTC)

I support the inclusion. Even BBC and CNN have discussed it, although for the moment they labeling it as speculation --Forich (talk) 14:45, 20 April 2020 (UTC)

I don't support the inclusion in this article because it is WP:UNDUE. This is an encyclopedia, not a newspaper, and we reflect the established mainstream view on topics. The correct place to report on this sort of unsubstantiated speculation is Misinformation related to the 2019–20 coronavirus pandemic, where a section already exists. --RexxS (talk) 15:18, 20 April 2020 (UTC)

RexxS, I agree on following WP:UNDUE to give this hypothesis its due weight in the entry. Please note that the policy does not --Forich (talk) 15:55, 20 April 2020 (UTC)say that you omit all mentions of minority views; instead, it says "Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in the published, reliable sources." Since the proportion of space that sources have devoted to the hypothesis is tiny, I suggest we only include one phrase mentioning the accidental-leakage hypothesis

@Forich: But WP:FRINGE is more precisely the guideline on how to deal with "an idea that is not broadly supported by scholarship in its field". As an example, the treatment of Flat Earth theories in the article on the Earth is a single sentence, which makes clear their lack of prominence in relation to the mainstream view. if you are looking for a form of words for this article, I suggest that the section Coronavirus disease 2019 #Misinformation already contains sufficient DUE information on "the origin, scale, prevention, treatment and other aspects" as well as a link to Misinformation related to the 2019–20 coronavirus pandemic for those looking for more detail. --RexxS (talk) 17:07, 20 April 2020 (UTC)

RexxS, are you saying that the accidental-leakage hypothesis is fringe and belongs on a "misinformation" entry? Are you sure we are refering to the same thing? You seem to be confusing this with the man-made hipothesis, in which case I agree that it is fringy. The hypothesis I am referring to, its composed of a) zoonotic origin following the scientific consensus + b) a debunking of the bat-in-the-seafood market jumping point argument + c) mentioning how there is circumstantial evidence suggesting that the Wuhan lab could have obtained a copy of the virus (after nature produced it) and did not prevent its accidental leak. These assertions are not mine, but they have been put together in several reputable news outlets, including BBC and CNN in the past week (April); please do not confuse this with the bogus man-made hypothesis that circulated in February.--Forich (talk) 01:54, 21 April 2020 (UTC)

@Forich: yes, I'm saying exactly that. Fringe theories belong in articles about the fringe theory. The newer the theory, the less likely it is to have become a mainstream view. --RexxS (talk) 07:37, 21 April 2020 (UTC)

I'm with RexxS here. I haven't read CNN, but the BBC has indeed reported on the claims, though not as a factual assertion. They simply reported that some were suggesting it. The BBC (and CNN) reporting on a fringe theory doesn't make it any less a fringe theory. The BBC even reported on David Icke (hack, spit) promoting the 5G mobile phone nonsense - it doesn't make it not fringe. Boing! said Zebedee (talk) 08:26, 21 April 2020 (UTC)

https://thebulletin.org/2020/03/experts-know-the-new-coronavirus-is-not-a-bioweapon-they-disagree-on-whether-it-could-have-leaked-from-a-research-lab/ --Espoo (talk) 22:06, 28 April 2020 (UTC)

AIDS was first studied because of outbreaks amongst the male gays of New York in the early 1980s. It was blamed on these gay people, and it was immediately assumed that HIV originated there. Only after 30 years of scientific study did we know that HIV likely came from Africa from as far back as the 1920s, and it came from apes (SIV). Aren't we a bit early in blaming Wuhan and China for the origin of this virus COVID-19. Even if the virus came from bats, can we be sure it was not from African bats or Australian bats, given that bats are also consumed by Africans and Australians. 2A00:23C5:C102:9E00:A98F:74CF:E799:43D5 (talk) 03:14, 2 May 2020 (UTC)

Special:Contributions/2A00:23C5:C102:9E00:A98F:74CF:E799:43D5|2A00:23C5:C102:9E00:A98F:74CF:E799:43D5, great suggestion. Only problem is, there is no "we", because the opinion of Wikipedians does not matter. It's against Wikipedia:OR. We limit ourselves to report what other sources say on the subject. Hopefully, the good sources will follow your logic and abstain from rushing to blame China for this.--Forich (talk) 17:09, 2 May 2020 (UTC)

Issue of patients avoiding hospitals even with heart attack symptoms?

Metzler, Bernhard; Siostrzonek, Peter; Binder, Ronald; et al. (16 April 2020). "Decline of acute coronary syndrome admissions in Austria since the outbreak of COVID-19: the pandemic response causes cardiac collateral damage". European Heart Journal. doi:10.1093/eurheartj/ehaa314.

” . . Comparing the first and last calendar week [ for the month of March], there was a relative reduction of 39.4% in admissions for ACS. . ”

” . . According to these assumptions, 275 patients were not treated in March 2020. Based on data showing that the cardiovascular mortality of untreated ACS patients might be as high as 40% (as it was in the 1950s),[2] we can theoretically estimate 110 ACS deaths during this time frame. The number of deaths associated with this unintentional undersupply of guideline-directed ACS management is very alarming, particularly when considering that the official number of COVID-related deaths in Austria was 86 on 29 March. . ”

ACS = Acute Cornary Syndrome

—————-

Garcia, Santiago; Albaghdadi, Mazen; Meraj, Perwaiz; et al. (April 2020). "PRE-PRINT: Reduction in ST-Segment Elevation Cardiac Catheterization Laboratory Activations in the United States during COVID-19 Pandemic". Journal of the American College of Cardiology (JACC). doi:10.1016/j.jacc.2020.04.011.

Page 3 (page 4 in PDF):

“Our preliminary analysis during the early phase of the COVID pandemic shows an estimated 38% reduction in US cardiac catheterization laboratory STEMI activations, similar to the 40% reduction noticed in Spain.”

I think it’s okay to use a pre-print as long as the reference clearly indentifies it as such, and in the text itself such as the phrase “a preliminary study.” FriendlyRiverOtter (talk) 16:11, 27 April 2020 (UTC) FriendlyRiverOtter (talk) 22:56, 2 May 2020 (UTC)

————————-

‘Where are all our patients?’: Covid phobia is keeping people with serious heart symptoms away from ERs, Stat News, Usha Lee McFarling, April 23, 2020.

”The same is true for appendicitis and stroke. Clinicians say patients with these life-threatening conditions have also stopped seeking treatment in large numbers. ‘My worry is some of these people are dying at home because they’re too scared to go to the hospital,’ Gulati said [Martha Gulati, chief of cardiology at the University of Arizona].”

——————

Amid the Coronavirus Crisis, Heart and Stroke Patients Go Missing, New York Times, Gina Kolata, April 25, 2020.

———————

And for the more cultural aspects of Coronavirus (which also have big medical consequences of course!), I think it’s fine to use journalistic sources. FriendlyRiverOtter (talk) 17:58, 27 April 2020 (UTC)

Infection fatality rate

This section contains the sentence:

• Similarly, as of April 25 in New York City, with a population if 8.4 million, there were 16,673 (0.20%) deaths confirmed from COVID-19, and 20,900 (0.25%) excess deaths.

Because the bald percentages in that sentence do not represent the IFR, the topic of the section, I believed the first was misleading and the second confusing. I therefore amended it to read:

• Similarly, as of April 25 in New York City, with a population if 8.4 million, there were 16,673 deaths confirmed from COVID-19 (a PFR of 0.20%), and 20,900 excess deaths.

Jmv2009 reverted my edit a minute later, with the edit summary Confusion is on purpose, as excess deaths may also need to be considered. https://www.bloomberg.com/opinion/articles/2020-04-24/is-coronavirus-worse-than-the-flu-blood-studies-say-yes-by-far. Purposeful confusion is not the goal of an encyclopedia, and shouldn't be allowed to remain. We should never be positing raw percentages but always carefully distinguishing between IFR and PFR. We certainly shouldn't be calculating excess deaths as a percentage of population in the same context, as that figure doesn't represent any metric used. Of course we can report excess deaths, but the raw figure is all that needs to be presented to achieve that. Jmv2009's reversion should be undone. --RexxS (talk) 16:25, 28 April 2020 (UTC)

Thank you. Was confused about what the confusion was about. Think I fixed it. Jmv2009 (talk) 17:14, 28 April 2020 (UTC)

I like the excess death figures, but we need to be careful with them too ! The graphs on NYT are misleading. Yes countries of the world are pushing the count of excess deaths in the recent past and are announcing their finding all at once. It create those peak in "factor of 2" but in reality, those deaths were slowly ramping up during the previous months. I have faith in the mitigation impact, but the virus didn't went from a growth rate of 60 to 0.1 in 1 week just because a few of us stayed home. Iluvalar (talk) 17:27, 28 April 2020 (UTC)

I've been struggling the last hour to find the exact delay in the death certificate for perspective, the best found so far "This delay can range from 1 week to 8 weeks or more" [25] in a note under a table. Not a satisfactory source, but it illustrate my point. Iluvalar (talk) 19:38, 28 April 2020 (UTC)

In the "Case fatality rates" table, for some countries instead of the fatality rate by ages wrongly recorded deaths distribution by ages. Please, correct it. — Preceding unsigned comment added by 95.35.206.174 (talk) 01:12, 3 May 2020 (UTC)

Transmission - speech

Aerosol emission and superemission during human speech increase with voice loudness: [26] Sciencia58 (talk) 18:41, 1 May 2020 (UTC)

This linked page is in German. This is the English Wikipedia. Do you have a suggestion for the actual page? Also, superemission? TylerDurden8823 (talk) 19:47, 1 May 2020 (UTC)

I imagine the same applies for coughing, one coughs harder and it spreads further. Doc James (talk · contribs · email) 01:44, 3 May 2020 (UTC)

It doesn't really matter that this is the English Wikipedia. Per WP:NONENG, while English sources are preferred if they are of equal quality and relevance, non English sources are perfectly acceptable when they are not. Even if the transmission increase is when speaking German, that still is unlikely to matter. However, considering the massive worldwide scale of the pandemic if this is a well accepted detail that is significant, there should be English sources for it. Note that a few English sources aren't enough to establish that, but their absence is strong evidence against it. Since I cannot understand German, I have not evaluated the source, however my feeling is similar to Doc James. Such a finding is not most likely particularly surprising or illustrative. Yet despite that, it probably also lacks good evidence. Nil Einne (talk) 03:23, 3 May 2020 (UTC)

Protests are relevant to the pandemic, not the disease

This article is about the disease, not the pandemic (and the social reaction to the pandemic), so the protests section should really be removed and put in that article. — Preceding unsigned comment added by 37.170.62.208 (talk) 21:25, 29 April 2020 (UTC)

I agree. --RexxS (talk) 21:57, 29 April 2020 (UTC)

+1 Doc James (talk · contribs · email) 13:37, 30 April 2020 (UTC)

Concur Nil Einne (talk) 03:24, 3 May 2020 (UTC)

Add information about Aarogya Setu app under Technology, contact tracing.

https://en.wikipedia.org/wiki/Aarogya_Setu

https://en.wikipedia.org/wiki/Coronavirus_disease_2019#Information_technology

AdithyaKL (talk) 09:20, 2 May 2020 (UTC)

Maybe we should just mention COVID-19 apps here and than have all the specifics on that subpage? Doc James (talk · contribs · email) 01:38, 3 May 2020 (UTC)

Ah yes @Doc James: , that seems correct. The "Also see" list needs to updated only. AdithyaKL (talk) 12:20, 3 May 2020 (UTC)

Reinfection

See also: Coronavirus_disease_2019 § Reinfection, fr:Maladie à coronavirus 2019 § Etude d'une infection à base d'une souche de coronavirus bénigne, and Antibody-dependent enhancement

There has been a lot of speculation in the lay media about possible reinfection. Are there any known viruses that regularly reinfect (non-immunocompromised} people after the primary immune response is complete, and cause clinically significant illness? Thanks, 2600:1702:2670:B530:D09A:C706:C10:EC5E (talk) 05:41, 24 April 2020 (UTC)

• Influenza. Doc James (talk · contribs · email) 06:35, 26 April 2020 (UTC)
  • No, one cannot get reinfected by the exact same strain of influenza again. This is stated in our own article, Flu season. By the way, I am actually a registered editor of Wikipedia with tens of thousands of edits, so please don't be dismissive as if I was an IP user. The section needs to reflect that if no immunity developed after infection, this would be unlike any known virus in any known organism. 2600:1702:2670:B530:443D:8BF1:2898:435A (talk) 07:05, 27 April 2020 (UTC)
    • You can look up the sources.. [27][28] Doc James (talk · contribs · email) 08:30, 27 April 2020 (UTC)
      • In my book, a vaccine will offer less protection than a previous infection. It's still a reasonable bet that reinfection is unlikely or less severe, at least for some time. Anyway, plenty of people will get re-exposed after having been tested positive, so we'll find out pretty soon. Jmv2009 (talk) 05:15, 30 April 2020 (UTC)
        • We have no vaccine for the common cold and we dont have immunity to the common cold. They're often coronaviruses. We do develop IgM and IgG antibodies to COVID-19, I think we should put that in, if Roche says we do, but I can't find their study to criticise any help? I reckon we need to be on the lookout for reinfection in the next few months, and it is completly unclear to me why certain cities get hit and certain dont. Many theories about. Maybe temperature. These public health measures will need to be in place for a very very long time. --Almaty (talk) 14:28, 3 May 2020 (UTC)

Coronavirus_disease_2019#Medications

Have restored "Per the World Health Organization, as of April 2020, there is no specific treatment for COVID‑19." per https://www.who.int/news-room/q-a-detail/q-a-coronaviruses

Emergency authorization in the US, well appropriate to mention, needs to be put into context. Still a lot of questions around the medication. Doc James (talk · contribs · email) 08:54, 2 May 2020 (UTC)

Thanks User:Doc James. Yes, but the WHO isn't the correct source to comment on medicines, rather national medicine regulatory authorities. Remdesivir is approved and has some effect, and it doesn't have any other indication, so it is an approved specific treatment. We both have concerns about context, but we can solve that by "Remdesivir has not shown to be clearly effective on mortality in RCTs" similar to what happens with oseltamivir --Almaty (talk) 09:19, 3 May 2020 (UTC)

The WHO actually is a good source to comment on the efficacy of medicines. Per WP:MEDRS, we should not be including biomedical claims based on a single study - particularly when another study reaches the opposite conclusion.

Ideal sources for biomedical information include: review articles (especially systematic reviews) published in reputable medical journals; academic and professional books written by experts in the relevant fields and from respected publishers; and guidelines or position statements from national or international expert bodies. Primary sources should generally not be used for medical content – as such sources often include unreliable or preliminary information

It's fine to note that one country has approved the use of remdesivir in an attempt to treat COVID-19, but in the absence of any MEDRS evidence that it has any effect at all, it remains inappropriate to claim the efficacy of the drug in Wikipedia's voice. --RexxS (talk) 00:27, 4 May 2020 (UTC)

Discussion on how to describe Hubei in the lead here and at COVID-19 pandemic

 You are invited to join the discussion at Talk:COVID-19 pandemic#Hubei description in the lead. {{u|Sdkb}} ^talk 04:51, 4 May 2020 (UTC)

Sorry for the repeat; the anchor broke last time. We could still use some more input on this — we're pretty split so far! {{u|Sdkb}} ^talk 04:52, 4 May 2020 (UTC)

NYTs

Is not a very good source for medical content per this text:

"Low oxygen levels detected using a pulse oximeter are a low cost quick diagnosis method for those with asymptomatic pneumonia relative to CT scans. Patients display acute or silent hypoxia where the oxygen level in blood cells and tissue can drop without any initial warning, even though the individual's chest x-ray shows diffuse pneumonia with an oxygen level below normal. Doctors report cases of silent hypoxia with COVID-19 patients who did not experience shortness of breath or coughing until their oxygen levels had plummeted to such a degree that the patients risked acute respiratory distress (ARDS) and organ failure. In a New York Times opinion piece (20 April 2020), emergency room doctor Richard Levitan reports "a vast majority of Covid pneumonia patients I met had remarkably low oxygen saturations at triage — seemingly incompatible with life — but they were using their cellphones as we put them on monitors.""

For those who climb / study high altitude medicine we all know that these degrees of low oxygen are completely compatible with life. Doc James (talk · contribs · email) 07:44, 4 May 2020 (UTC)

References

Infectious disease or respiratory disease?

In the first line it says that it's an "infectious disease" which is a link which directs to the article for "infection." However, my understanding is that - just like SARS - COVID-19 is a respiratory disease. The following article says, "two strains of [this species of coronavirus] have caused outbreaks of severe respiratory diseases in humans: severe acute respiratory syndrome coronavirus (SARS-CoV or SARS-CoV-1), which caused the 2002–2004 outbreak of severe acute respiratory syndrome (SARS), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is causing the 2019–20 pandemic of coronavirus disease 2019 (COVID-19)" https://en.wikipedia.org/wiki/Severe_acute_respiratory_syndrome-related_coronavirus — Preceding unsigned comment added by 110.142.94.82 (talk) 09:37, 3 May 2020 (UTC)

I would say "infectious disease" as it is not just a "respiratory disease" Doc James (talk · contribs · email) 07:50, 4 May 2020 (UTC)