Talk:2009 swine flu pandemic by country summary

^{< Talk:2009 swine flu outbreak}

See also: File talk:H1N1 map.svg, Talk:2009 swine flu outbreak by country, Talk:2009 swine flu outbreak timeline, and File talk:H1N1 map by confirmed cases.svg

Mexico

Please update the number of confirmed cases to 727 and the number of confirmed deaths to 26. Secretaría de Salud - Estadísticas

 Done Flipper9 (talk) 00:15, 5 May 2009 (UTC)

{{editsemiprotected}} Please update the number of confirmed cases to 866. Secretaría de Salud - Estadísticas Please add a note to the number of suspected cases saying: "Mexican health administration had stopped updating the number of suspected cases a couple of days ago."

Y - Confirmed cases updated; such a note already exists regarding suspect cases. --ThaddeusB (talk) 20:34, 5 May 2009 (UTC)

confirmed cases on Secretaría de Salud - Estadísticas now 1204 and death 44. Please update. —Preceding unsigned comment added by Icekoer (talk • contribs) 17:11, 7 May 2009 (UTC)

Please correct the fatalities to 48 as they are reported in the ECDC source cited. --FHessel (talk) 07:26, 12 May 2009 (UTC) There is a new ECDC source (update 12: http://ecdc.europa.eu/en/files/pdf/Health_topics/Situation_Report_090512_0800hrs.pdf) reporting 2059 confirmed cases and 56 fatalities. --FHessel (talk) 07:51, 12 May 2009 (UTC)

Should suspect cases (sub)column be dropped?

Note: This refers only to the cases column, not the deaths one (which should definitely stay)
Note: Please do not make any (format) changes to the table until this discussion has run at least 24 hours AND has a CLEAR consensus for change.

We are, perhaps, at the point where the suspected cases column is being meaningless. The primary reason I say this is because the Mexican gov't has chosen not to make their # public anymore and they account for more than half the total. Per this story "The latest figures from Mexico suggest the virus may be less lethal and infectious than originally feared. Only 38 percent of suspected cases have turned out to be swine flu..." If you do the math, that means they have tested roughly 2500 cases, so only "new" suspect cases remain to be tested. Of course, we have no idea what that number is, but it is definitely not zero. Thus blanking just Mexico's suspect cases is not a good option.

Also, some of the countries that have only "suspect cases" might never have a clear story that all cases have been cleared.

On the other hand, eliminating suspect cases would remove ~20 countries from the chart with could have very dramatic and unexpected effects. Perhaps we could put a blurb like "Other countries with suspected, but no confirmed cases include: blah, blah, blah."

Thus, I have no strong opinion about what is the best way to proceed. The one thing I do feel very strongly about is that this would be a major change and absolutely should not be done without CLEAR consensus. --ThaddeusB (talk) 05:33, 5 May 2009 (UTC)

My 2c:

• The distinction between probable and suspected cases for the swine flu is defined by CDC, and it is not clear that the other countries (and non-US media) are following the US-guideline for using these terms (why would they ?!). So I think we should, anyway, not make a possibly OR distinction between these numbers and just relabel the column "probable or suspected cases". For countries for which the usage of the terminology is clear, we can of course continue to use these sub-categories in text and in the sub-articles.
• Instead of using "-" to signify an unknown (but non-zero) number of probable/suspect cases, why not simply use "?" ? This is a more universally understood notation, and is not confusable with the "minus" sign (even though I have been following this article and discussions more closesly than 99% of our readers, I admit to being perplexed for a moment when I saw the "-(2500±)^§[3]" entry earlier today. )

Abecedare (talk) 06:08, 5 May 2009 (UTC)

I agree "?" is a much better notation than "-" for unknown values. Does anyone object to changing this straightaway?

Canada has also not been reporting suspected cases since April. Neither Mexico nor Canada have published a count of probable cases, as far as I know, although they're required to report this to the WHO. I'm not sure if this last column should be deleted completely, but we should at least relegate the most dubious entries (Mexico, Canada) to a footnote where we can explain the problem briefly. -- Avenue (talk) 07:25, 5 May 2009 (UTC)

In the main article discussion, this was already talked about. It was suggested that "-" was improper and that another symbol or group of symbols be used. I'd agree with "?". However, from a statistical standpoint and reliable sources standpoint, the column should be removed as these statistics are not universally reported or updated, there is confusion over definitions, and the numbers have a finite lifetime that would require extreme vigilance to maintain any accuracy or relevance with daily updates for every country. If we just want to list those probable cases, that would be okay...but at least the "Total" summation value at the top should be removed due to these issues. Flipper9 (talk) 13:44, 5 May 2009 (UTC)

I went ahead and changed "-" to "??" as I do not consider that a major change. --ThaddeusB (talk) 13:52, 5 May 2009 (UTC)

Looks much better! Flipper9 (talk) 14:55, 5 May 2009 (UTC)

In the meantime, I changed ?? to ?. The former drew too much attention and disturbed the flow. hmwithτ 15:01, 5 May 2009 (UTC)

This is my opinion: We should drop the suspected cases column, but only in countries with confirmed cases. We should keep the suspected cases in countries that do not have a confirmed case. The reason for the table is to track not just the size of the outbreak but also the spread. However, it seems that once countries have significant numbers of suspected cases the number becomes less meaningful. Also, why is there a confirmed death column in the countries with no confirmed cases? If they have a confirmed death isn't that automatically a confirmed case? --62.69.130.82 (talk) 11:18, 5 May 2009 (UTC)

I don't believe we should remove the suspected cases column but we should adapt it to the reality of the information that is being provided. Mexico and Canada for example, should have their numbers removed with a note saying that these countries are not providing this data any longer. The ones which are providing the data should be updated accordingly. I personally don't understand the need to use stale data when the situation is clearly evolving. Mexico and Canada would not have 0 cases, there would be a footnote to indicate that the number is not being provided any longer. GaussianCopula (talk) 23:11, 7 May 2009 (UTC)

There seems be some consensus that the figures for Mexico and Canada should be removed, so I will go ahead with this. (The update to Mexico's figure is not really an update; it comes from a retrospective piece, and may even refer to an earlier period than the figure it replaced. So that doesn't improve the situation.) -- Avenue (talk) 23:50, 8 May 2009 (UTC)

Well I am not sure I agree, but I'm sure I disagree either. It's a complicated decision because removing Mexico does have a dramatic effect on the total, which makes the number less realistic overall. That said, we don't have a good source of information so we don't have much leeway. Really this is Mexico's fault for being all "political" on us and refusing to make the info public.

In short, I don't object to the change but I'm not sure it is actually an improvement (although I do realie it definitely had to be done at some point). --ThaddeusB (talk) 00:15, 9 May 2009 (UTC)

I think that we should remove the "Total" calculation at the top as it is misleading the reader into thinking that we have sources to back up the idea that there are only that many suspected/probable cases world-wide. We can leave the individual suspected/probable cases for each individual country that we have numbers to reference, but the total serves no purpose at this time. Flipper9 (talk) 13:32, 9 May 2009 (UTC)

I would drop the column because it can be misleading. "Suspected cases" doesn't mean that they are confirmed and by thus they don't give any information how spead the virus really is. Please remove to avoid confusion.--201.153.19.149 (talk) 13:56, 10 May 2009 (UTC)

Suspected cases column is confusing because not all countries have suspected cases but the footnotes are in parenthesis. So the footnotes look like suspected cases data. —Preceding unsigned comment added by 72.150.154.235 (talk) 20:45, 11 May 2009 (UTC)

Reviewing the Suspected Cases per the FACTBOX Reuters Link

Looks like the Factbox provided by Reuters doesn't track suspected cases anymore. The reference to countries like Benin, Norway, Finland and others do not support the data any longer. I have tried to update all references I could to official government sites but I can only do that for German, English and Spanish. What do you all recommend we do with these others? GaussianCopula (talk) 00:21, 9 May 2009 (UTC)

I will delete all data referencing suspected cases that refer to the Factbox by Reuters within 24hrs unless someone can provide sourced data. I think that is the only choice. GaussianCopula (talk) 01:21, 9 May 2009 (UTC)

Relinking them to the original reference (yesterday's fact box) will fix it. Doing it now... --ThaddeusB (talk) 04:34, 9 May 2009 (UTC)

I don't think you can do that outside of google cache. It was an up to date data list that used WHO/CDC/National Health Ministries sources, which we can now see, some have stopped simply providing the information. I am only hoping that some who know Finnish or Norwegian and so forth can start updating the data with regards to their government official information. But as far as this link is concerned, I believe it is dead. That means that in the next 12 hours that data must be removed. GaussianCopula (talk) 04:48, 9 May 2009 (UTC)

I must say that I completely disagree with using a May 7 static data for suspected cases. It will never move. The current Factbox has done away with that data. I don't see the point. Although, to be honest, you proved me wrong with that google cache thing. GaussianCopula (talk) 04:53, 9 May 2009 (UTC)

I would just like to indicate the divergence from that data on the 7th to current data on the 8th, just a day old. Country Factbox Current USA 845 224 Spain 49 40 Britain 390 426 France 0 28 NZ 107 62 El Salv 2 5 Colombia 141 162 Switzerl 19 8 CR 7 81 AUS 41 19 Venezue 9 1 Chile 7 3 Peru 7 0 South Af 2 1

I do disagree very much in using this table as a static suspected cases table for the countries which we don't have updated government data. GaussianCopula (talk) 05:08, 9 May 2009 (UTC)

I respect Thaddeus decision to provide a source for the suspected cases and do want others to please comment. My position can be summarized as follows: The current FACTBOX numbers are not being updated anymore with suspected cases. The WHO, Factbox and several government health ministries are not providing suspected case information anymore. That makes a mutable number, like suspected cases, irrelevant for those countries without a source not linked to the outdated Factbox table.

I will make a move in 12h to remove that data but I would appreciate feedback beforehand. GaussianCopula (talk) 05:26, 9 May 2009 (UTC)

I'll note how things happend on the US table and hope it gives you an example how things might be able to be done here. The CDC has never reported Probable or Suspect cases, only Confirmed though daily updates. Most state health departments have reported confrimed and probable cases. A few states continue to report suspected cases. Because of an easy to access media market (lots of news reports on cases) we intially had a lot of reports on suspected numbers. However, as the event has matured and states now update their reports daily, we began removing old suspect cases even if we didn't have a source stating "0 suspect cases" (my personal rule is 1 week). The logic is that because we have evidence of a robust and active testing enviroment, week old cases most likely have been tested by now.

The same logic applies to probable cases. As CDC testing kits have been distributed to states, some states have stopped reporting probable cases and now only report confirmed cases. My personal rule is if a state now only reports confirmed cases and has a significant jump in confirmed cases since the last 'probable statement', I can set probable cases to 0. I don't consider either of these to be OR, just logical thought given a unique reporting enviroment. Other editors have made similar edit summaries.

Note this may be a unique situation to the United States and may not be applicable to other nations where there are significant, recent media reports of suspected cases far in excess of any goverment statements, or infrequent government reporting --PigFlu Oink (talk) 05:44, 9 May 2009 (UTC)

I'll wait a until end of the week then to remove the information linked to that report. There is no rush in trying to remove countries from the list. I will see if I can get some government data from those countries. GaussianCopula (talk) 18:15, 9 May 2009 (UTC)

We're really working outside of all normal Wikipedia policy here, since as far as the encyclopedia is concerned we could put up a table describing one particular day during the early outbreak and more than cover the topic. We're intentionally putting aside WP:NOTNEWS in favor of WP:IAR, because everyone does want to know the current situation, and also because the revision history of this chart will be a goldmine for people researching future articles. But because we're going beyond normal policy, we might need to be cautious about letting things get out of hand.

To begin with, I think that these numbers for "suspected" and "probable" cases should at least be added into a single figure for each country. The fact that 4 of 430 cases in the UK are "probable" rather than "suspected" at this moment is of very little significance. Once it is established that there is no reliable source for probable+suspected cases, we can replace this by N/A; and once most of the countries are N/A, we can abolish the column entirely in favor of a footnote. The figure for "probable deaths" in Mexico should also be relegated to a footnote.

Later in the outbreak, it is possible that some countries will stop providing data for confirmed cases and confirmed deaths, at which point we should provide estimated figures in parentheses for those countries. Mike Serfas (talk) 19:52, 9 May 2009 (UTC)

First of all, I recommend all parties take a deep breath here. :) There is nothing wrong with a reference that is *gasp* 36 hours old - and no it isn't pointing to a Google cache. If there is no update in a week, obviously the numbers will be dropped. Second, while I agree that the suspected/probable cases column will eventually disappear, now is not the time to do so. Third, the suspected death reference is perfectly valid and normal - for an disease outbreak there will always be some cases that are suspected as being related to the disease but never actually confirmed. --ThaddeusB (talk) 20:00, 9 May 2009 (UTC)

P.S. most of our references are to "static data" that will never change. It is our job to replace things with more recent/up-to-date sources when needed, not our sources job to update themselves. :) --ThaddeusB (talk) 20:06, 9 May 2009 (UTC)

(edit conflict) I disagree with combining suspected and probable figures. The NZ probable cases figure, for instance, has been relatively stable, with the few changes often reflecting a case shifting from probable to confirmed. The suspected cases figure, on the other hand, frequently goes up or down by 10 or 20 cases. The UK figures are unusual (Colombia is the only other country with a similar ratio of suspected to probable cases), so they don't seem like a good basis for a decision.

I agree with GaussianCopula that the Reuters factbox figures are very dated, given how variable the suspected cases figures are. But this also applies to other sources, such as those for Taiwan and the Czech Republic (from the 4th and 5th of May respectively). Given the lack of up to date information on suspected cases for many countries, I think we should remove this sub-column. I don't think we should wait any longer; these figures are embarrassingly old already. (This obviously does not apply to the 100 suspected deaths in Mexico, as that number is unlikely to change.) The lower part of the table (covering countries with no confirmed cases) could perhaps be replaced with a list of countries, rather than omitting them completely. -- Avenue (talk) 20:29, 9 May 2009 (UTC)

Another two days have passed, with no action on this. Does anyone object to replacing the factbox figures for suspected cases from 7 May with a "?", on the basis that they are now too old to be meaningful? -- Avenue (talk) 01:00, 12 May 2009 (UTC)

I agree, have been thinking this for a while. Will change them now if no new sources. |→ Spaully^₪† 12:13, 13 May 2009 (GMT)

Case Thailand

There has been an update based on the following report:

http://nationmultimedia.com/2009/05/11/national/national_30102352.php

The following report does not clear any case of H1N1. It states: "which proves that what the person most likely had was the common human flu, Public Health Minister Wittaya Kaewparadai said yesterday."

As well it says: "Lab results due here on Wednesday from the US Centres for Disease Control and Prevention will verify that the unidentified person was not suffering from the swine flu, he said.

"By then, Thailand will be free of the disease, and so far there's been only one suspicious case, pending CDC confirmation indicating that it is only ordinary flu infecting the person," he said."

That is no reason to affirm that the Thai case is negative. As such, I will revert the edit. GaussianCopula (talk) 21:23, 10 May 2009 (UTC)

El Salvador: update

According to the last report from The Ministry of Health of El Salvador there are now 4 confirmed, 2 probable and 5 suspected cases. http://www.laprensagrafica.com/el-salvador/social/32318-salud-confirma-dos-casos-mas-de-gripe-a-h1n1-.html --66.201.170.6 (talk) 02:15, 11 May 2009 (UTC)

Updated GaussianCopula (talk) 03:04, 11 May 2009 (UTC)

China is confirmed

Just saw it off from CCTV... Let me see if I could find updates on a English website —Preceding unsigned comment added by Dakbonsa (talk • contribs) 04:32, 11 May 2009 (UTC)

Reiters: China reports first confirmed H1N1 case [1]Игорь277 (talk) 04:53, 11 May 2009 (UTC)

Suggest links to both indexes of reports and latest reports combined

A number of countries have a stable flu website with links to frequent situation reports, and the situation reports themselves which are constantly being added to. It seems useful to give BOTH of these sites as references for figures; this allows the reader to go to the index page to check for possible later updates and general information, and also to link straight to the update which was latest when the article was updated. I've done this for Argentina; in case this has been changed since the time of writing, the template after my update for Argentina is here. If this idea is popular, others may care to do a similar job for other countries. I will insert the reference here both to demonstrate it and so that it can be edited, copied, and pasted for other countries. As usual, subsequent references use the reference name.

References

Pol098 (talk) 12:13, 11 May 2009 (UTC)

Ordering of the table

What is the criteria ordering the table? I think it started as number of cases, but now the US has more, and Costa Rica arrived in the middle... Should the table be ordered by confirmed cases? 195.217.138.194 (talk) 12:38, 11 May 2009 (UTC)

It is ordered by number of deaths first, then confirmed cases, then suspect cases. This is done per previous consensus. --ThaddeusB (talk) 12:44, 11 May 2009 (UTC)

In that case, you might want to make the rows sortable or put the death row in the first position to avoid confusion. -- 77.176.77.180 (talk) 16:07, 11 May 2009 (UTC)

I think putting the deaths column first makes sense. They are the most important, and it will help readers understand the sort order as discussed above. -- Avenue (talk) 01:11, 12 May 2009 (UTC)

Template protected

This template is used only on three articles. I see no reason it should be protected any more than those articles. We do have a policy of not protecting without good cause. The cause cited was high visibility. In the guideline page it says that "permanent visibility" is required. Apart from that it is my humble opinion that this template should not be protected (even semi-protection) because we have to show that we are serious about allowing articles (including templates) to be edited by anyone. Debresser (talk) 13:37, 11 May 2009 (UTC)

The template was protected since April; 27. The initial protection is here. I reprotected it, because the software showed that log when editing, which is a little confusing. I could make the reason more in line with the original protection, but I have personally no objection to remove protection now. I would advise to keep the move-protection, as it's been moved a good number of times, it should keep in line with the article name, and this is a controversial subject. Cenarium (talk) 13:56, 11 May 2009 (UTC)

Sounds fair enough to me. Debresser (talk) 14:30, 11 May 2009 (UTC)

One of the three pages using this template is still edit-protected. Still, I think this template is less at risk than those articles. In addition, I often see that a lot of vandalism is tolerated before the decision to protect is made. Debresser (talk) 14:33, 11 May 2009 (UTC)

I see you changed the protection level. Now we'll see. Debresser (talk) 15:59, 12 May 2009 (UTC)

U.S.

A note should probably be added that some places in the U.S. have stopped testing probable cases and are only testing severe cases.[2] This affects comparability of the data between countries. Rmhermen (talk) 14:07, 11 May 2009 (UTC)

Interesting. I imagine similar decisions have been made elsewhere, when the number of cases was large enough to strain the testing facilities, but those places may be less ready to admit it. It doesn't just affect comparability between countries, but also the meaning of the numbers themselves. I agree we should include a suitable caveat in the table. -- Avenue (talk) 00:53, 12 May 2009 (UTC)

Mexico suspected

There is a May 6 report by the CDC http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5817a1.htm that gives the total number of suspected cases as of May 5 in Mexico to be 11,932.

That same report says probable cases in the US is 845. The CDC has been providing current updates to this number and it now stands at 433.

I don't agree with using a May 5 number for suspected cases in Mexico and would like further discussion about it before any changes are made. GaussianCopula (talk) 20:24, 11 May 2009 (UTC)

Thanks for adding this section. I feel that the number of deaths is misleading compared to the confirmed cases without this information. When I read the CDC article my first thought was that this table needs the information. This information gives it some context and adds to the table. The difference in US cases is understandable because the large backlog has been aggressively worked on in the US over the past few days but there is no indications that 50% of the suspected Mexican cases has been rejected in the same time period. I think it should stay but I too would like to see some more conversation on this subject and will understand of others disagree. Daveonwiki (talk) 20:34, 11 May 2009 (UTC)

Here is a quick view of what is in the article -

As of May 5, using an updated case definition of fever plus cough or sore throat for a suspected case and real-time reverse transcription--polymerase chain reaction (rRT-PCR) or viral culture for a laboratory-confirmed case, Mexico had identified 11,932 suspected cases and 949 cases of laboratory-confirmed novel influenza A (H1N1) virus infection, including 42 patients who died.

The figure above shows the 822 confirmed and 11,356 suspected cases of novel influenza A (H1N1) virus infection in Mexico with dates of onset from March 11 through May 3, 2009. Both confirmed and suspected cases rose sharply from April 19 to April 26, then decreased sharply.

This chart with caption may be of more value to the article than the number in the suspected cases column. Daveonwiki (talk) 21:38, 11 May 2009 (UTC)

That's a pretty cool little chart. It probably belongs on Wikipedia, but on 2009 swine flu outbreak in Mexico rather than the main page since it only deals with Mexico. --ThaddeusB (talk) 22:55, 11 May 2009 (UTC)

It's a nice chart, but not trivial to interpret due to lags in processing cases and assigning them dates. If it does get added somewhere, we should probably include caveats like those provided by PHAC with their similar graph here. -- Avenue (talk) 00:40, 12 May 2009 (UTC)

Well actually it is organized by the date of symptom onset, not the test date, so presumably lag isn't an issue except for the last few days. --ThaddeusB (talk) 02:11, 12 May 2009 (UTC)

And that is precisely where it is misleading. The passage immediately following the chart may thus be a misinterpretation. Also, for anyone else reading this, the chart is now being discussed more fully at Talk:2009_swine_flu_outbreak#CDC_-_Mexican_Cases. -- Avenue (talk) 02:21, 12 May 2009 (UTC)

Here is a relevant quote from a recent CDC media conference transcript:

"… our MMWRs this past week showed epidemic curves for the U.S. and Mexico, and I think that "New England Journal" article earlier this week also showed an epidemic curve for the U.S. It's important to say that those curves do not incorporate a lag time that additional cases may be happening, and are not yet showing on the curves, so we think that the trends that you see, a big peak and then a decline are not an accurate assessment of the ongoing occurrence of disease. We do think here in the U.S. it's possible that increased cases are occurring, and it's too soon for us to say whether it's some communities things are getting better. Our indicators suggest there are things still be accelerating, we're still seeing an increase in cases. The issue in Mexico, there's a little bit more data to suggest they may be in some parts of the country seeing a decline, in other parts perhaps not yet, ..."

(all in capitals in original)[3] -- Avenue (talk) 02:41, 12 May 2009 (UTC)

I agree with GaussianCopula that the May 5 figure for suspected cases should not go in the table - it's just too old. -- Avenue (talk) 00:42, 12 May 2009 (UTC)

References

It says on this page This reference list does not appear in the article but when I look at the main article the references are included. Is it just me or does this sentence need to be removed? Smartse (talk) 00:43, 12 May 2009 (UTC)

I think it is intended to indicate that the reference list doesn't appear next to the table. I agree it could be worded better. -- Avenue (talk) 00:46, 12 May 2009 (UTC)

This has been discussed at length in Category_talk:Wikipedia_pages_with_broken_references#Template_cite_error_message and after all was said and done, nobody came up with something better and this is what we use nowadays through {{Templaterefsection}}. Debresser (talk) 01:04, 12 May 2009 (UTC)

Would this be any better?

== References ==

These notes will be transcluded as references in articles using this template, and can be displayed there by adding either <references/> or {{Reflist}} (if neither of these are already in place). They will not generally be displayed next to the other content generated by this template, as shown here.

It's a bit wordy, but I think it addresses the current ambiguity. -- Avenue (talk) 02:59, 12 May 2009 (UTC)

A bit wordy?? Debresser (talk) 13:30, 12 May 2009 (UTC)

CDC: H1N1 Flu Numbers Represent a "Very Great Underestimate"

Below is a footnote for insertion at an appropriate place in the article, to avoid creating the misleading impression that the reported cases indicate the actual cases. In fact, in the U.S. and Mexico, only spot testing is being done any more, unless the illness is severe enough to raise suspicions of a mutation.

(need to add ref & endref)

CDC: H1N1 Flu Numbers Represent a "Very Great Underestimate"

Emma Hitt, PhD

May 9, 2009 — The confirmed numbers of influenza A (H1N1) cases reported are likely to fall far short of the actual numbers occurring within the community, according to a US Centers for Disease Control and Prevention (CDC) spokesperson.

"The individual numbers are likely a very great underestimate of how much virus is circulating or how many people are becoming ill from it," said Anne Schuchat, MD, the Director of the CDC's National Center for Immunization and Respiratory Diseases, speaking at a media briefing today.

...

http://www.medscape.com/viewarticle/702607</ref> —Preceding unsigned comment added by 67.101.142.160 (talk) 01:51, 12 May 2009 (UTC)

Germany: update

Please update the confirmed cases for Germany to 12 (source may remain, because the source itself has been updated to 12) --FHessel (talk) 07:33, 12 May 2009 (UTC)

Finland and Thailand

Finland has 2 confirmed cases. [4] Thailand confirms first H1N1 flu case. [5] Please update. Игорь277 (talk) 08:09, 12 May 2009 (UTC)

Thailand has updated and your source from Reuters contradicts with the original reference here. It seems that the old reference points out that the 2nd "confirmed" case is not tested by the CDC, so the other case would only be "probable". -Xavier Fung (talk) 08:59, 12 May 2009 (UTC)

Got this from Reuters and there are 2 cases now. -Xavier Fung (talk) 09:15, 12 May 2009 (UTC)

Suspected ==> under investigation

In most cases anybody with flu-like symptoms is tested for this strain, and in most countries the vast majority are found not to have it. Consequently a better term than "suspected" is "under investigation". Confirmed, probable, and under investigation make a reasonable set of categories. Pol098 (talk) 12:14, 12 May 2009 (UTC)

Suspected is a reserved definition by CDC standards (I don't know about WHO or anyone else). under investigation could mean anything. Changing defintions without authortative sourced reasons could lead to numbers that are based on differing editor intrepretations of media and government reports. --PigFlu Oink (talk) 12:26, 12 May 2009 (UTC)

I wasn't aware of that, but a term which has a precise definition in only one country is hardly appropriate. I don't know how many countries use this; if it's very widespread then it's obviously suitable. "Under investigation" is very precise, without need for a formal definition; it is a case that has been reported to a testing laboratory for investigation. "Suspected" implies a pre-judgment. Possibly countries with a great many cases (US) would tend to "suspect" that people with symptoms in an affected area were likely to be infected, while countries with few cases "investigate" a great many cases routinely. A form of words that has been used is "Cases under investigation have not yet been either confirmed or ruled out as being due to this strain of influenza by laboratory tests, although some other strains may have been ruled out." Pol098 (talk) 12:37, 12 May 2009 (UTC)

New column: # cases caused by secondary transmission within the country (tells more than suspected cases)

I think it is time to add a column with the secondary transmissions within the country. Since this is becoming the most important fact to see if the pandemic is spreading in other areas than mexico - US and canada - I tried to find information regarding this and the best what I have found so far is:

http://www.invs.sante.fr/display/?doc=surveillance/grippe_dossier/points_h1n1/grippe_A_h1n1_120509/index.html It is published by the french - I am sure there must be a better link - but I could not find it. For the european countries there is some information on the ecdc website: http://ecdc.europa.eu/en/files/pdf/Health_topics/Situation_Report_090512_0800hrs.pdf.

Do peopl know sources for this important number? And should we add this column? (194.74.226.190 (talk) 16:54, 12 May 2009 (UTC))

(I made a couple of format changes above, for readability Kavri (talk) 17:16, 12 May 2009 (UTC))

I think this is a good idea, but a bit too soon. Once there is more information coming out about secondary transmissions then I do think it would make a good addition. Or, if not to the table itself, at least a section of the article discussing it. --- Kavri (talk) 17:16, 12 May 2009 (UTC)

I already started a list of these countries in the text of the 2009 swine flu article (chapter "International cases"), with sources at least for the existence of in-country transmissions. Actually, I was just going to suggest to move it to the table. So go ahead adding the column! --Roentgenium111 (talk) 19:01, 12 May 2009 (UTC)

Adding such column is a good idea but it would be better to add later since there are no clear definition of community transmission cases. I think we could wait for more epidemiological analysis of these secondary cases to justify the needs of adding a new column. -Xavier Fung (talk) 07:16, 13 May 2009 (UTC)

When/How should 'confirmed human cases' section of table be shortened?

I noticed the table is getting a bit long, so I wondered:

Do we need to shorten the list (as was done by putting the 'suspected only' countries into their own list)? If so, how? and when?

I don't think we need to do it quite yet, but probably soon, and there are a couple of ways it could be done, so I thought maybe discussing them now beforehand would be a good idea.

--Since it is unusual for an influenza virus to manifest first in the americas, and some have already refered to it as American flu, then, we could break things out by continent:

North/Central/South America for one list, then either by 'rest of world' OR by 'continent/groups of continents'

--As people are also 'number watching' it might be a case for breaking it out be number totals:

'under 10', '10-50', '51-99', '100-999', 'over 1000' ... or some similar break down.

-- and my least liked option, alphabetical: 'a-e', 'f-j', 'k-o', 'p-t', 'u-z' ... or similar.

--Some other way that is logical and that I didn't think of.

Of course, the usual ordering would continue within the categories (death/confirm/suspect).

My preference is for continent or group of continents. Ideas? Input? --- Kavri (talk) 17:04, 12 May 2009 (UTC)

If you're gonna do it, go with continent, no reason to scroll halfway down a page to find mexico. --PigFlu Oink (talk) 17:11, 12 May 2009 (UTC)

With continent or 'over 1000' Mexico would still be at the top of the page, with US and Canada following, since it would still be ordered by 'deaths/confirmed/suspect'. That would also be the case if North/Central/South America made up one group. Going with number categories would appeal to those that want to see which countries have the most/least cases...but if it wasn't a country with the very most or least, it might be more of a bother for someone that wanted to look at a specific country (ie randomly, would someone know offhand where to find Greece? and when Greece's numbers change one would have to look through the categories to find it again. Also, it would be more work to keep up to date. That said, though I still like continents/groups of continents best, there might be other/better ideas out there. By the way, do you think it should be done now? later? I don't know what the majority would consider 'too long' for a table. --- Kavri (talk) 17:26, 12 May 2009 (UTC)

Another option would be to group countries by degree of spread. Put countries with sustained community-level human-to-human transmission together (Mexico, the U.S., and Canada at present), then countries with isolated in-country transmission, then countries with confirmed imported cases, then counties with suspected cases only. I saw this done in one report a few days ago - can't remember where though. -- Avenue (talk) 12:14, 13 May 2009 (UTC)

We could make a grouping of "Under 10" or whatever number, that might be easiest for now.

I like the suggestion of grouping by within country spread. This is becoming a more published figure, though will always be lacking in smaller Central American countries. It would be good if you could find the source as this would be our guide to reference. Any other suggestions? |→ Spaully^₪† 14:20, 13 May 2009 (GMT)

Bring back the logarithmic chart

It is a very good depiction indeed. The justification for deleting it, that it needs explanation in text, seems completely spurious to me as it is very clear. In any case this extremely useful graph is now unlinked. By all means move it elsewhere on the page but don't just delete it!!! Hawthorn (talk) 00:13, 13 May 2009 (UTC)

Having updated the chart I have added it to the main article and added a link below the normal scale on the table. While more informative overall it is not immediately obvious, especially to those without a mathematics background, what it is depicting so I think the other chart should be the main image for the table.

Remember you can always add the image to articles yourself if you think it should go somewhere, be bold if you feel strongly about something. |→ Spaully^₪† 12:49, 13 May 2009 (GMT)

I disagree strongly that you would need to have a maths background to understand this chart. (Actually, isn't it a graph, not a chart?) It has large, clear numbers on its vertical axis, and you don't need to have heard of logarithms to read them. It is absolutely clear what it is depicting. I actually think this graph is easier to understand than the main one, which is confusing on first sight because of its two different y-axes. It gives more information, too. It's a shame that it's been relegated to an obscure little link, which people will only click on if (A) they happen to notice it, (B) they know what "semi-logarithmic means (what does semi-logarithmic mean, by the way?) and (C) they appreciate the advantages of these types of graphs. This is an excellent, very informative, easily understood graph. I think it should be given equal prominence to the main one, or, at the very least, put back into a collapsible section of the table. I'd like to hear some more opinions before being "bold" about this, though. GrahamN (talk) 19:42, 13 May 2009 (UTC)

I disagree with including it until it is fixed. First, this chart or graph (http://www.thefreedictionary.com/chart chart: 4. See graph) is not adequately portraying a true logarithmic progression. Here you can see courtesy of yahoo what a real logarithmic chart should look like (http://finance.yahoo.com/q/bc?t=2y&s=GM&l=on&z=m&q=l&c=yhoo&c=%5EGSPC). If you notice the scale, there is an actual natural progression which OUTSIZES the increases in comparison to the initial y-axis value progression. This is not being adequately represented on this chart.

I would also like to point out that I am against the main chart as well because the y-axis 2 has an non-proportionate representation in the chart in comparison to y-axis 1. Not only do death cases track confirmed cases up to almost 3/4 of the chart, they serve no purpose to represent them in such a way that fails to actually track how the number of suspected cases has increased significantly whereas the number of deaths has increased at a significantly slower pace.

I am for removing both graphs until a true logarithmic chart can be created. Otherwise, they are simply of a sensational nature. GaussianCopula (talk) 20:36, 13 May 2009 (UTC)

It IS a logarithmic graph. I teach this stuff at university. GaussianCopula you are wrong. The whole OUTSIZES comment is just totally whack. You'd get marks off for that in my class for sure. The other chart you link to is bizarre. I don't know what sort of chart it is but it sure isn't a logarithmic graph. It seems to have some kind of distorted percentage on one axis. Percentages are not the kind of thing you should be plotting on a logarithmic graph. They are most useful when applied to unbounded quantities. Maybe in some weird corner of finance they call these things logarithmic graphs - but no mathematician would recognise them as such.

The logarithmic chart is now so superior to the ordinary graph that we should be using this one instead as the primary graph for the article. It shows what is going in in a way that the ordinary graph doesn't. The ordinary graph is nothing more than pretty wallpaper which tells you that the number of cases is going up - Duh! You can't see in what manner. You can't see anything much really. It also attempts to display a false relationship between the numbers of deaths and cases. On the logarithmic graph we would expect to see the number of cases and the number of deaths tending towards two parallel lines. The vertical separation will then tell us the fatality rate. In the early parts of the epidemic we don't see this - the lines are far from parallel, probably because of missing data, which means any attempt to measure the fatality rate on this part of the graph is premature. The fitted correspondence between deaths and cases in the ordinary graph is a false comparison. Notice that as time goes by it is looking more and more broken and sick, whereas on the logarithmic graph the lines for cases and deaths are getting closer and closer to parallel.Hawthorn (talk) 21:09, 13 May 2009 (UTC)

Dropping the probable/suspected column

As the N/A's ramp up, as WHO, Reuters and individual countries stop providing suspected cases updates, this column is starting to stand out like a sore thumb. I am not saying that we need to drop it now but I think we should start having the discussion with regards to removing it. BTW, Seychelles and Benin are linked to a week old report. GaussianCopula (talk) 19:13, 13 May 2009 (UTC)

I'll start the discussion with the following points. Almost half the cases on the upper graph are 0 or N/A. The lower graph has also half of it's information tracking one cases plus a 2 case. I don't see the relevance of this after the almost 3 week ordeal where we had all expected some sort of ramp up in the information. I am for removing the suspected cases since outside of the US, Costa Rica, UK and Colombia; the information is proving to be of minor relevance. GaussianCopula (talk) 19:21, 13 May 2009 (UTC)

Nobody seems to be responding to you, so here's my tuppenceworth. If you are suggesting that the table should only show laboratory-confirmed cases, I'm not sure that sounds like a good idea. Wouldn't it produce bias towards countries that are more zealous in their testing and/or who have a smaller head of population per laboratory, and against countries with less effective health services and/or whose laboratories are overstretched? I confess I haven't grasped the distinction between "probable" and "suspected". The reference in the column heading doesn't help - it defines "probable", but not "suspected". Whatever the distinction is, is it so significant? Couldn't we re-name the column "probable or suspected", and just take the larger of the two figures in each case (or, if the categories are mutually exclusive, the sum of them?) GrahamN (talk) 20:47, 13 May 2009 (UTC) GrahamN (talk) 20:47, 13 May 2009 (UTC)

As the epidemic ramps up, we are going to reach a stage where the extremely costly and difficult genetic analysis to confirm each individual case becomes impossible because there are simply too many cases. In the midst of a raging pandemic it does not make sense to genetically type each individual case of the flu. At that point we are going to have to start using probable numbers to track the course of the epidemic. My vote is to soldier on with this column because I think we are going to need it soon. Hawthorn (talk) 21:23, 13 May 2009 (UTC)