Talk:Ethosuximide

Affordability

This book discusses affordability in prose form of anti epileptic medications in LMIC.[1] As such it fulfills the RfC requirements IMO. Doc James (talk · contribs · email) 01:57, 30 March 2020 (UTC)

This does not meet the RFC and WP:NOTPRICES requirement: "secondary sources discuss[ing] pricing extensively" [for that drug] or as policy states it "mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention". This is very much "passing mention" of ethosuximide and a single figure in a table in a 900 page specialist textbook on Epilepsy. That is not mainstream media. Further, the price in the source, is for Italy in 2001 (not 2011 -- the book was only published in 2008). A twenty-year-old Italian retail price has zero weight or relevance in an English encyclopaedia. The article text here juxtaposes wholesale prices in 2014/2019 for a month supply of a given daily dose, with a retail price in 2001 for a year's supply of approximately half that daily dose. This baffling mix of incompatible numbers is unacceptable. The RFC conclusion was clear. All these prices are unacceptable. -- Colin°^Talk 09:02, 1 April 2020 (UTC)

Disputed content

Per the broad discussion here about Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices, and specifically the issues raised about the use of 2001 data from one country to support price statements at Ethosuximide here, the pricing content here is disputed. SandyGeorgia (Talk) 00:35, 31 March 2020 (UTC)

RfC

We have the sentence "As of 2008 it was generally affordable in many areas of the world."

Based on a 2008 textbook from Wiley on Epilepsy that states

In 28 countries, which contain over 40% of the world population, the per capita annual gross nation product barely suffices to buy a year's supply of carbamazepine or valproate for one or two patients... For most patients living in these countries, only phenobarbital, phenytoin, and ethosuximdie may be avaliable at prices affordable by the general population, and many of the newer drugs may not be avaliable at all.

Doc James (talk · contribs · email) 00:37, 31 March 2020 (UTC)

Revised RfC Statement: [2] 00:51, 5 April 2020 by Doc James

• Should the lead of ethoxuximide summarize the section on cost as "As of 2008 it was generally affordable in many areas of the world."

Support

• Support we have textbook by a well respect medical publisher. I have provided an exact quote of the text above. We have other sources that also describe it as inexpensive.(2005 textbook,2019 textbook) Doc James (talk · contribs · email) 00:38, 31 March 2020 (UTC)
• I support based on sources, just to clarify as I never write this much it should be noted that references speak for themselves, I hope this matter can be solved, thank you--Ozzie10aaaa (talk) 17:51, 3 April 2020 (UTC)*
• Support having been away from the discussions for the last month and trying to catch-up here is how I perceive the situation. Saying that the medicine is 'generally affordable' does not violate the spirit of NOPRICE which (to me) is intended to prevent companies from advertising their wares on Wikipedia like a catalog. Even without that impression I think the rationale for inclusion is sound. Wrt the wording, I like that it is relativistic as I can understand the concern with specific costing (although I don't agree with the opinion). As a infrequent editor, I can find no consensus about when the affordability of a medication can be mentioned and when it can be in the lead (despite spending an hour reading thru related RfCs) so I feel its reasonable to look at how the mention is made in this article. Imo, this is completely reasonable, well placed, and reasonably sourced. Its also appears to be well intentioned (e.g. the editor doesn't appear to be a socket puppet for a supplier of Ethosuximide). Summary from me; given that no consensus exists Wikipedia wide I turn to the specific issue and I think its reasonable here. Ian Furst (talk) 17:52, 7 April 2020 (UTC)

Oppose

First, I oppose the !voting rather than discussing format here. Second, this is yet another RFC [3] to dispute the already held RFC (which is not a productive way to proceed IMO). Third, the problems per the discussion about Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices, specifically the issues raised about the use of 2001 data from one country to support price statements at Ethosuximide here, make this source obviously unsuitable for this text. A thorough read of the source shows that the data for this drug is based on 2001 prices in one country only, Italy. As the data is two decades old, based on a decade-old source, WP:WEIGHT is not met for inclusion, per WP:NOPRICE. None of the cost data in this article is policy-compliant. SandyGeorgia (Talk) 00:54, 31 March 2020 (UTC)

I also dislike this "please vote" style. Can't we just talk about it? Or have a ==Poll== section that doesn't treat supporters and opposers as the only possible responses, and as parties that need to be carefully segregated from each other? My own view doesn't fall neatly into "Support" or "Oppose", but the (very) few editors who use this style tend to disregard comments in the "Discussion" section when they're trying to figure out what the conclusion is, so it doesn't even feel like it would be worth it to say what my view is. WhatamIdoing (talk) 01:21, 31 March 2020 (UTC)

• Strongly oppose: When the content in the article body is worked out, we can readdress. Ignoring those problems, the proposed addition to the lede is grossly undue, as well as misleading. We need a discussion about WP:MOSLEAD in general, focusing on MOS:LEADNO, rather than repeat such RfCs. --Hipal/Ronz (talk) 16:57, 7 April 2020 (UTC)

Discussion

I object to creating poll-RFCs for every issue. This seems to be a coronavirus infection on Wikipedia. The textbook is 12 years old and their data is based on a price in Italy from 2001, so nearly twenty years old. If you read this article, and the last source offered above, you will see that Ethosuximide has been dropped by many manufacturers and is unavailable in many markets. This is why James had to cite a 2014 MSH record for a supplier who delivers by their own fleet of trucks to the north of the Democratic Republic of Congo for the "Developing world" price, rather than use the 2015 record -- because there isn't one: there are no developing world suppliers at all that MSH list.

To add to that, the textbook rather weirdly bases their "affordable" conclusion by taking the per capita annual gross national product of a nation and comparing this with the retail price in Italy that was seven-years out-of-date when the book was published in 2008. Even if we accept their €133 price (the price in Italy, not in the developing nation pharmacy), that is $145, or $12 a month. The WHO/HAI reckon that a drug is unaffordable if it costs more than a days wages to buy a month's supply. There are no African nations where the the standard wage of a government employee reaches anything like $12 a day. Many don't even reach $1.2 a day. It is rather important to get economic facts from economic experts rather than neurologists. Don't you think?

Any affordability statement (low cost, inexpensive) is likely to have only regional applicability for the market/readership of the book or article, and also be subject to chronological irrelevance with time. A North American textbook or a UK textbook, will be concerned with such matters in their domestic market. Such statements clearly cannot be extended to nations with different economic levels or drug price policy and availability: such statements are not international in scope. -- Colin°^Talk 09:18, 1 April 2020 (UTC)

I don't see the point of responding to these biased RfC's. They're clearly efforts at IDHT and CONLOCAL. The ArbCom case should sort these behavioral problems out. --Hipal/Ronz (talk) 20:13, 1 April 2020 (UTC)

(per WAID above), if discussion is needed then lets discuss--Ozzie10aaaa (talk) 18:14, 3 April 2020 (UTC)

Please do. Start with why policy and a massive RfC specifically to address such disputes are being completely ignored. --Hipal/Ronz (talk) 19:02, 3 April 2020 (UTC)

are they being ignored?...are you certain?--Ozzie10aaaa (talk) 19:38, 3 April 2020 (UTC)

This is not a proposal to include numerical prices. It is just a rough cost and I do not think the prior RfC ruled that out. Doc James (talk · contribs · email) 22:53, 3 April 2020 (UTC)

Completely irrelevant. You've been asked multiple times how anyone could assume that the RfC, policies, consensus applies solely to "numerical prices". Those questions have been ignored so far.

So, Ozzie10aaaa, there's a nice example for you right there. But if you want to demonstrate that you are taking the RfC, policies, and consensus into account, please do so. --Hipal/Ronz (talk) 16:19, 4 April 2020 (UTC)

Im certain with time this will work out, thank you--Ozzie10aaaa (talk) 01:04, 5 April 2020 (UTC)

Then what's the point of this RfC, if editors refuse to engage? Votes don't create consensus, collaboration does. --Hipal/Ronz (talk) 16:55, 5 April 2020 (UTC)

Sure so what source and what text do you want to have summarize cost? Doc James (talk · contribs · email) 17:03, 5 April 2020 (UTC)

I like the clarification of the RfC statement.[4] Apologies if my attempt to make it clear that the RfC was changed adds confusion to the situation. --Hipal/Ronz (talk) 17:08, 5 April 2020 (UTC)

Such an old reference makes it seem like it's not so noteworthy. Can we find something more recent? --Hipal/Ronz (talk) 17:09, 5 April 2020 (UTC)

Other

• "The very favorable response to ethosuximide-an inexpensive AED" 2005 book
• "It is relatively inexpensive." 2011 textbook
• Document form the American Academy of Pediatrics "inexpensive" 2014
• "Low cost" 2019 textbook published by Elsevier. Also comments on the fact that availability may be poor.
• "Less expensive" 2017 textbook by Elsevier
• "The results of the earlier phase of the study concluded that ethosuximide — the oldest and least expensive of the three medications"

Doc James (talk · contribs · email) 21:31, 5 April 2020 (UTC)

Thanks. I'd like to hear what others think of using some of these to demonstrate stronger due weight.

Thank you for bringinging up availability. I've been wanting to, but have been waiting until after we get these pricing guidelines worked out. From my perspective we need some lengthy discussions on what drug availability information is due in articles. Pricing and prices are just a part of total availability. Supply and accessibility are also important parts. I've not even looked yet for what articles we have about these subjects. --Hipal/Ronz (talk) 21:46, 5 April 2020 (UTC)

Availability often depends on the drug approval process and not so much the price. In a lot of countries though price does play a role as in Canada if a company want to set the price to high the regulators will not allowed it to be sold. Doc James (talk · contribs · email) 21:51, 5 April 2020 (UTC)

User:Hipal with these additional sources is their anything you would support as a summary of cost in the lead? Doc James (talk · contribs · email) 21:55, 5 April 2020 (UTC)

I'd like to hear from others, as I said. I'm unimpressed with those sources as supporting mention in the lede. There's also the bigger issue, discussed in the RfC and it's preparation, of how vague and transitory information like this probably doesn't belong in the lede. --Hipal/Ronz (talk) 22:59, 5 April 2020 (UTC)

The good news is that we are looking at and talking about sources in the way that is supposed to happen for all Wikipedia content. (I asked, for weeks during the RFC formulation, for examples of drug pricing for the 530 articles that didn't depend on databases, and none surfaced ... so for the first time, we are talking about the kinds of things we should be talking about).

James, would you mind numbering your bullet points above for ease of discussing the sources (as No. 1, No. 2, etc)?

1. The first source is a 2005 book, which seems irrelevant for pricing 15 years later.
2. The second, 2011 textbook, I am unable to pull up that text, so can't see what is said, but again, ten-year-old data isn't going to be very useful on pricing.
3. The third source, American Academy of Pediatrics is recent and reliable, but tells us only pricing in US I suspect.
4. Says low cost, but also says taken off the market in many countries.
5. Can't see that, don't know if it's me or my browser lately ? @Colin:
6. Sixth, medcitynews ? Least expensive of the three (in the US ?) isn't entirely useful.

It seems to me that (partial response because I can't see all the sources) we may have sources that mention the cost in the US, that may be usable in the body of the article to discuss US drug price being low, but I am not seeing an indication of WEIGHT to satisfy including mention of price in the LEAD. I will change my opinion if someone provides more in-depth on the sources I can't access, but I see nothing that indicates pricing on this particular medication is a LEAD summary WEIGHT issue, but don't mind if we mention relatively low cost in the body, but that it has been pulled from the market in many countries. This doesn't appear to be a situation where pricing of this particular drug has generated considerable coverage, as for example compared to epipen, insulin, pyrimethamine, where pricing is controversial and has been covered by many reliable sources. SandyGeorgia (Talk) 00:19, 6 April 2020 (UTC)

Update: I see that concerns of price fixing has been added to the lead since I last looked at the article,[5] which changes the story about price relevance in the lead. SandyGeorgia (Talk) 00:56, 6 April 2020 (UTC)

You mean the relevance is now present? Doc James (talk · contribs · email) 01:04, 6 April 2020 (UTC)

We edit conflicted. IMO, the lawsuits are worthy of LEAD mention, but want to hear from others. But ... contradiction. The lead says the drug is affordable in most areas of the world, but the body says:

In the United States the wholesale cost of this amount is about US$96 per month as of 2019.[23] With discounts this amount may be purchased for around US$57 in the United States.

In the United States, that is not by any means an "affordable drug". Inexpensive drugs in the US are on the Walmart $4 per 30-day, or $10 for 90-day lists. That is not the case here. GoodRX says with a coupon, you can get it for $50, and its $63 at Costco, where drugs are discounted. That is not "affordable" in general or relative to other drugs. Why are we saying in the lead that it is "affordable"? This is the problem with using outdated sources. We can solve this by doing what is always done with V WEIGHT, LEAD and source-to-text integrity; discuss whatever prices are mentioned in the sources that cover the lawsuits, rather than digging around for old price data. SandyGeorgia (Talk) 01:11, 6 April 2020 (UTC)

I believe that, in the WHO/HAI model, any drug that costs less than a day's gross income at minimum wage (i.e., currently US$58) would be considered "affordable", so the claim isn't patently unreasonable. Perhaps this explains what the sources mean by "relatively low": the price isn't necessarily impossible, but it's not actually low, either. WhatamIdoing (talk) 14:36, 6 April 2020 (UTC)

WhatamIdoing, I haven't seen it put that way round. They use one day's wages as a ballpark for judging if a drug is generally considered unaffordable. That doesn't mean that prices below that are "affordable". Stretching to afford a drug is not a binary thing for anyone. Further, WHO/HAI often comment that the family income may have to pay for more than one drug. I think that as you and Sandy show, the price is on the borderline for being classed unaffordable in the US (unless your insurance pays). Also, the minimum wage or unskilled government worker metric varies to where it sits in a country. In many African reports, they suggest it is roughly a median figure, meaning a lot of the population earn much less. In the UK I suspect it is a much lower percentile. It depends how much of your workforce are unemployed, employed though casual labour, self-employed in low-paid work, subsistence farming, etc, etc. -- Colin°^Talk 15:29, 6 April 2020 (UTC)

(edit conflict) I would think that a drug described as an inexpensive anti-epilepsy drug in 2005 would remain so, since we don't seem to see other sources remarking on it becoming expensive. Per WP:NPOV We should be reporting all of the significant views in the literature, but we're not obliged to discount one view simply because we can't evidence of a contrary one. Per WP:ASF, we should state the conclusion that it's an inexpensive drug in Wikipedia's voice, without inline attribution in the absence of contrary sources. All of that relates to the body of the article, of course. I much prefer to see the article text settled before I try to determine how or if a particular section is summarised in the lead (but I accept that's just my personal preference).

I'd like to see if there is more to be said around issues of cost, particularly how it has compared to other popular AEDs over time and geographically, if (and only if) we can find secondary sources that discuss that. Additionally, there does seem to be a controversy about alleged price fixing: we could usefully tease out some more relevant content from current sources 7 and 8 to put into the body of the article. I suspect it's too NEWSy for the lead at present. --RexxS (talk) 01:13, 6 April 2020 (UTC)

There are several things here I am not yet understanding, RexxS. First, is the price disparity, because the $4 lists are relatively new things (compared to the dates of these sources), changing industry-wide pricing and leading to significantly lower drug prices, and to physicians prescribing more of those that are on the $4 lists. (This is a US thing-- Colin or you can comment on UK.) So it's not clear that what was described as inexpensive in 2005 remains the case, if it is not on the $4 list. (At the Free Clinic where I interpret in Spanish, we are almost always able to prescribe on the $4 list; I suspect this is new phenom since the older sources, and I have only once in years encountered a case where we couldn't prescribe from the $4 list.) My other concern about the price fixing is, does that belong in the lead of each drug, in the body of each drug, or in the company of the drug article being sued? As you note, NOTNEWS is an issue. Agree on dealing with body first, then summarizing to lead-- always good practice, and here we are dealing with lead and body as if they were separate. SandyGeorgia (Talk) 01:28, 6 April 2020 (UTC)

Sandy, I am rather shocked to see any WP:MED editor cite a online news article that is clearly churnalism: a press release from 2010. These sources are all either dated or US-specific, or both. I got the book you couldn't read. What it contains is a table with a column describing features of older anti-epileptic medicines and a column for newer anti-epileptic medicines. Ethosuximide is classified as among the older of seven drugs they mention, vs 13 newer drugs. Some of the pro/cons are simply drug manufacturer spin such as "safer in pregnancy" or "less complex drug interactions" or "better tolerated". The inclusion of Felbamate among the new drugs suggest to me these descriptions should be taken with a shovel of salt as it is bad for drug interactions and was largely withdrawn after deaths reported. One of the claims is the older drugs are "less expensive" and the newer drugs are "more expensive". For anyone who knows anything about drugs, this is a "do bears shit in the woods" level of obvious generalisation. The source is not specifically claiming ethosuximide is "less expensive" than other drugs, merely that the authors have assumed it must be because it is old. In fact, as our article notes, the supplier issues mean that its price has not followed the generic=>inexpensive model. But the most interesting feature of the sources is what is revealed by the URL. James has searched Google for 'Ethosuximide "inexpensive" OR "low cost"'. This is not how we write neutral articles on Wikipedia. -- Colin°^Talk 10:09, 6 April 2020 (UTC)

According to GoodRX: Ethosuximide (Zarontin) is an expensive drug used to control seizures in certain types of epilepsy. This drug is less popular than comparable drugs. It is available in brand and generic form. Generic zarontin is covered by most Medicare and insurance plans, but some pharmacy coupons or cash prices may be lower. The lowest GoodRx price for the most common version of ethosuximide is around $47.30, 80% off the average retail price of $246.98. (my bold). I see we are still quoting the price in Italy in 2001 and still wrongly saying that is the price for 2011. We are also giving the UK price for 2011, which is nine years ago, and quite irrelevant. The article says "£0.68 per 250 mg tablet versus £0.11 per 250 mg of liquid". But today's BNF says the price is £3.09 per 250mg tablet and £4.32 per 250mg (5ml) of liquid. So not only are our prices wrong today by a factor of 40x, but the claim the syrup is cheaper than the tablet is also wrong. The main competitor drugs for absense epilepsy are valproate and lamotrigine, both of which are much cheaper, though I have not found sources making a price comparison when deciding which to use. This is likely because all three are affordable in rich countries, and the choice is a clinical one. Per WP:WEIGHT we must conclude the body of sources do not find the relative or absolute price of ethosuximide to be notable. This further emphasises that by desperately searching for any source supporting a "low cost" "inexpensive" price statement about ethosuximide, and claiming it to be low cost, we in fact totally mislead our readers. For childhood absence epilepsy (its primary indication) ethosuximide is a clinically good first choice, but it is currently the dearest option. At the 1250mg daily dose James used earlier, this equates to £464 per months for tablets and £648.75 per month for syrup. Valproate would roughly cost about £30 for a month, so it is 15-20x more expensive. -- Colin°^Talk 11:07, 6 April 2020 (UTC)

Two other sources. Concise guide to the currently available drug options in epilepsy shows that even in 2012 in the UK ethosuximide was "the only costly older AED". And this story in The Sunday Post says Children’s epilepsy experts are also worried over the soaring price of anticonvulsant drugs. A commonly used one, Ethosuximide, has soared from 9p per 250mg capsule in 2006 to £1.89 per capsule in 2016/17. Epilepsy Scotland said: “Neurologists have raised issues. They say that they are managing to keep prescribing it to children but fear they may be asked to reconsider in future. The cost has skyrocketed in price. It is not easy to change a child’s drugs because new ones may not control their seizures.”. Now, The Sunday Post is a newspaper my gran used to read for some light entertainment at the weekend, so I'm not claiming this meets WP:V, but it does at least indicate that the increase has registered with epilepsy charities and doctors. But WP:WEIGHT demands we are familiar with the body of literature on an article topic, and weigh what we write accordingly. And the literature, albeit with a rich western bias, really does not generally comment on the price or relative price of ethosuximide. That is why the sources that do mention it even in passing are so old and poor. Nobody is talking about this, and neither should we. -- Colin°^Talk 12:48, 6 April 2020 (UTC)

Since there seems to be in this discussion some confusion about which pieces of text are either database-generated, UNDUE, or too dated to be useful-- as have been demonstrated with an analysis of reliable sources on this page-- I have tagged the disputed portions here to hopefully lend some clarity. The question of whether we should even have a cost section for this drug seems to come down to how or where we decide to handle the NEWS-y price fixing lawsuit. SandyGeorgia (Talk) 22:51, 6 April 2020 (UTC)

Price in one country misrepresented as the price in the developing world

A sentence that was discussed in great detail, and universally condemned, in Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices is still present in this article.

I see that:

• User:Colin removed it recently in this edit
• User:Doc James restored it in this edit a couple of hours later.

If you look at Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices#Comments about neutrality and due weight for Doc James' comments on 3 April, you will see him say that "we can adjust that one to that one country." Regardless of whether you believe that the RFC authorizes use of this database at all, I thought we had achieved, if nothing else, an agreement that pretending that the Democratic Republic of the Congo is "the developing world" in this specific article was a bad idea. James, did you maybe just forget about those conversations? Or have you changed your mind? WhatamIdoing (talk) 20:09, 5 April 2020 (UTC)

Okay adjusted to "The wholesale cost from an NGO in the Democratic Republic of Congo is about US$28 per month for the defined daily dose of 1,250 mg as of 2014."[6]

What wording do you propose User:WhatamIdoing? Doc James (talk · contribs · email) 20:18, 5 April 2020 (UTC)

James, I don't think that we should be using that source at all, given that I understand the RFC outcome to say that using that source at all violates WP:OR. (That is, the RFC's closing statement said that using that primary source is "not straightforward", and WP:PRIMARY says that "A primary source may be used on Wikipedia only to make straightforward, descriptive statements". That adds up to every single use of that drug pricing database being a violation of WP:OR.)

Given that using that source at all appears (to me, and apparently to many other editors) to be a violation of WP:OR and in defiance of the RFC's stated conclusion (as confirmed by closing admin User:Ymblanter, e.g., "We believe that the RfC established that sourcing prices only from a database involves original research."), I'm not sure that there's any point in the ideal wording for content from an unusable source. I think you should remove that sentence entirely. Did you maybe not realize that the RFC said that the IMPPG database can't be used as the sole source for any fact or any sentence in any article, not even if other sentences in the article contain information about prices and are sourced to other, acceptable sources? You can't do One sentence.^{[good source]} Separate sentence.^{[bad source]}; the good source doesn't make up for the bad source. Per the RFC, the only acceptable way to use that database if it's an 'unnecessary' second citation, e.g., One sentence.^{[good source]} Separate sentence.^{[another good source][redundant bad source]}. If this is not your understanding of the RFC outcome, then maybe you should directly ask one of the closing admins' on their talk pages whether it's okay to use only the MSH's IMPPG database as the sole source for a drug price (NB: not the sole source in the section – the sole source for the individual price). WhatamIdoing (talk) 20:49, 5 April 2020 (UTC)

I agree. My understanding is that price guide should not be used.

Could we just use the Medicaid ref? --Hipal/Ronz (talk) 20:55, 5 April 2020 (UTC)

Yes I do not interpret their conclusions the same as yourself.

They also say "database which covers some ground and does not require interpretation... should be discussed at the talk page"

We could write "One NGO in the Democratic Republic of the Congo supplied ethosuximide at the wholesale price of $US0.18 per 250 mg tablet in 2014. The daily defined dose is listed as 1.25 grams."

That requires no interpretation as requested.

Why would we want only US pricing information by only including medicaid? That would be in breach of knowledge parity.Doc James (talk · contribs · email) 21:01, 5 April 2020 (UTC)

That doesn't appear remotely encyclopedic, nor due mention. It's just a bit of information without context. All this was covered in the RfC and it's preparation. --Hipal/Ronz (talk) 21:17, 5 April 2020 (UTC)

(edit conflict)

Nobody says that we want only US prices. I'm saying that, as far as I can tell, the RFC closed with a consensus against using that database as the only source for any price.

I'm really surprised and dismayed that you left out so much of the sentence you quote there. James, that's "selective" to the point of seeming dishonest. User:Ymblanter's whole sentence says "If there are some issues (like, I do not know, one database which covers some ground and does not require interpretation) which have not been discussed during this RfC then they should be discussed at the talk page". And it comes immediately after the sentence that says "We believe that the RfC established that sourcing prices only from the database involves original research." "The database" == the source you're trying to cite for this sentence. The sentence you (mis-)quote is about hypothetical other databases, not about MSH's IMPPG database. And it doesn't say that these other databases are acceptable; it only says that there's a possibility that some hypothetical and heretofore undiscussed database might not have the same fatal problems. This means that if you find a database that has "issues" that "have not been discussed during this RfC" – a condition that, after 40,000 words, I think firmly excludes the source you're trying to use here – then those other databases should be discussed first on article talk pages.

Seriously, James: If you still believe that the RFC permits you to use MSH's IMPPG database as the sole source at the end of any sentence, despite lots of editors telling you that they believe you're wrong, would you please just go ask one of the closing admins? Just drop a note on their talk pages saying something like "I read the RFC's closing statement. Is it still okay if I put <this sentence> in Ethosuximide, with MSH's IMPPG drug price database as the sole source for it?" If User:Wugapodes or Ymblanter respond with a cheerful "Yes, that's perfectly fine!", then I really do think everyone else will leave you alone about it. But until then, you seem to be the only editor who still thinks it's perfectly fine, and I do think you should double-check that your interpretation is the correct one. WhatamIdoing (talk) 21:22, 5 April 2020 (UTC)

I am not "sourcing prices only from the database". We have 5 6 sources used for prices here.

User:Hipal states "Could we just use the Medicaid ref?" Ie US prices. Doc James (talk · contribs · email) 21:36, 5 April 2020 (UTC)

Yes the RfC concluded that numerical prices should generally not go in the lead. With exceptions for medications that have picked up the notice of the popular press such as for being expensive in the developed world such as pyrimethamine and onasemnogene abeparvovec.

And yes the RfC concluded interpretation of the IMPPG database should not be carried out and not be used as the only source when prices are discussed.

I do not read it as saying the IMPPG database as published by the World Health Organization can never be used just that it needs to be done without interpretation and along with other high quality secondary sources.

WP:NPOV states we are to "represent fairly, proportionately, and, as far as possible, without editorial bias, all the significant views that have been published by reliable sources on a topic."

There is support for different examples of prices per WP:RSUW which states "Wikipedia aims to present competing views in proportion to their representation among experts on the subject, or among the concerned parties. This applies not only to article text, but to images, external links, categories, and all other material as well." - and that includes medications prices of course.

Yes would appreciate the comments of the closing admins. User:Ymblanter and User:Wugapodes... Doc James (talk · contribs · email) 21:47, 5 April 2020 (UTC)

Editors in the RfC pointed to this sentence as an example of OR and source interpretation that was unacceptable. I don't see how I can read Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices#Ethosuximide as anything other than consensus for this sentences exclusion, especially given the wider discussion of principles. — Wug·a·po·des​ 03:23, 6 April 2020 (UTC)

Sure and the sentence discussing in the RfC is no longer in the article. Doc James (talk · contribs · email) 15:02, 6 April 2020 (UTC)

But, it does still seem to be in the article? You've slightly changed it, but I find that

The wholesale cost from an NGO in the Democratic Republic of Congo is about US$28 per month for the defined daily dose of 1,250 mg as of 2014.

is still in the article, still using the same source. Wugapodes, could you please officially confirm for us whether the RFC concluded that the MSH IMPPG source should be used to support any content about drug prices in any sentence (i.e., no matter how it got re-written) in this article? WhatamIdoing (talk) 20:19, 6 April 2020 (UTC)

Sources
References

Yes if the conclusion was "MSH IMPPG" as published by WHO cannot be used for anything ever, it would be useful to say that. If you meant that the source can be used in the body but no interpretation of the source is permitted than that would be useful too. Doc James (talk · contribs · email) 20:30, 6 April 2020 (UTC)

James, see my question and response from Ymblanter. I'm really not sure why you think the original research and weight problems of a database-sourced price only apply to the lead. You aren't going to get any luckier with Drugs.com, NADAC, BNF, GoodRX, etc either. Their fundamentals are the same as MSH IMPPG. The use of a product price database to source a drug price on Wikipedia is no longer allowed. -- Colin°^Talk 21:24, 6 April 2020 (UTC)

(edit conflict) I don't think consensus at the RfC was sufficient for a blanket ban on the MSH IMPPG database, but a number of editors indicated that it is better to not include price information than to use database information as the only source for claims. To be clear, the RfC should be taken to cover that database and any substantially similar database regardless of publisher; citing a photocopy of a letter reproduced in a history book is still citing a primary source because you are citing the primary material itself not the secondary analysis of it, even if the primary source object is contained in what is generally a secondary source. To the specific sentence now in the article, revisions like these were addressed as hypotheticals in the RfC and generally disfavored. This was covered in the closing statement: In addition, there are concerns that proper explanation of the situation for the indicated price would give the price undue weight. While the sentence now in this article resolves interpretation and verifiability concerns, it raises questions of due weight. Is a single wholesale transaction in the DRC important enough for this level of specificity when weighed against all the other information available about this drug? Being pretty ignorant of the pharmacology field, I don't have much of an opinion on the specifics, but in general editors foresaw situations like the current text and were suspicious of them. — Wug·a·po·des​ 21:43, 6 April 2020 (UTC)

Okay so there is not a blanket ban on this source. Thank you. Basically we need no make sure there is no interpretation. We are a global encyclopedia with a global readership. We already struggle with being US/EU centric. With lots of efforts to improve coverage specifically in Africa. Africa does represent 15% of the global population yet I doubt it even comes close to reaching 15% of our coverage. The population of the DRC is also 81 million (which is more than the UK). Doc James (talk · contribs · email) 23:40, 6 April 2020 (UTC)

James, you interspersed and outdented a comment not in chrono order:[7] I re-indented it [8] so others can continue the discussion without interrupted flow. Could you please take the time to read the rest of Wugapodes statement, and not just the first sentence, and respond to commentary in order with the correct formatting? SandyGeorgia (Talk) 00:50, 7 April 2020 (UTC)

Wugapodes, thanks for clarifying. Let's be clear, Wikipedia does not allow editors to make novel conclusions and claims based on primary literature that are not themselves repeated and discussed in the secondary literature. None of the 500+ novel price claims in our drug articles are repeated by the secondary literature. It is not just that a price from a supplier in a north east region of DRC is a ridiculous source, but also that our secondary literature do not think that a $28 monthly price based on a 1,250 mg daily dose using 250mg tablets is notable and worth commenting on. And this repeats even more so for the many drugs with far more varied dosing regimes, treatment indications and formulations. The idea that secondary literature can be ignored, both in terms of WP:V and WP:WEIGHT is untenable. All the secondary sources James has offered so far for Ethosuximide mention cost no more than in "passing mention" (e.g. the Italian price and the UK syrup vs tablet price), and some even less so (e.g. the source saying older drugs are generally cheaper than newer ones). We are a long way here from the "extensive discussion/commentary" let alone "mainstream media" required by policy. -- Colin°^Talk 22:10, 6 April 2020 (UTC)

Yes the RfC concluded that numerical prices should generally not go in the lead. With exceptions for medications that have picked up the notice of the popular press such as for being expensive in the developed world such as pyrimethamine and onasemnogene abeparvovec. ... Doc James (talk · contribs · email) 21:47, 5 April 2020 (UTC)

James, I have seen you state several times that the RFC concluded that numerical prices should not go in the lead. I want to understand where you got this numerical notion, because there is no such thing in the RFC. The RFC included no such distinction. (I might also mention that you just pinged both Ymblanter and Wugapodes, who were both already pinged just above by WAID, in case they might want to consider your repetitive pinging as "harassment"; we've seen you make that claim in these discussions.) SandyGeorgia (Talk) 00:28, 6 April 2020 (UTC)

I cannot believe that the wording of the RfC close "Editors are generally opposed to inclusion of prices in the lede." can be perceived as anything other than numerical prices. First, prices are inherently numerical. Secondly, the next sentence "While basic calculations are not considered original research, the pricing statements ..." talks about calculations, which are also inherently numerical. That presumes that calculations are relevant to prices, which only makes sense if they are numerical. Thirdly, the question in the RfC only deals with the numerical prices found in a database, and the conclusions of an RfC should not be extrapolated beyond what the participants were asked to consider. They did not comment on any aspect of pricing other than the numerical prices found in databases, because the questions were narrowly framed to address just that. Finally, any relevant content that is extensively discussed in multiple reliable secondary sources should be included in the article because of WP:NPOV. Where that content forms a significant topic within the article, it should be summarised in the lead, per WP:LEAD. What the RfC made clear is that consensus is against using a raw figure from a primary source as that summary. I certainly have no problem with that. --RexxS (talk) 01:33, 6 April 2020 (UTC)

RexxS (do you prefer to be pinged, or not?), if you view WP:WEIGHT and WP:LEAD in broad context, do you not see that what is included in a lead is unrelated to whether the content is a precise number or not? Understanding how LEADs are written, what I can't understand is how we can perceive that the word numerical would change the overall conclusion. The two sentences you mention above are about two different things. Have you read through all of the discussions leading up to the RFC, and the full RFC? Because the closing admins did ... how we can come up with a "numerical" construct here is, IMO, not in the RFC discussions, not in the RFC, and not in any logical reading of how Leads are constructed, summarizing the most important parts of an article to the lead. Whether a price is numerical or generalized, its relevance to the LEAD guideline does not change, and it would take some sort of maneuvering to get that out of LEAD.

Let's take a hypothetical example relative to MEDRS we may all understand better. If we had text discussing incidence, prevalence, and so on of a given condition in the body of an article, we an summarize that text to the lead in general terms (common condition, more common in men than women, etc) precisely because that content is worthy of inclusion in the LEAD per the guideline. Our decision does not hinge around whether we include the exact prevalence or incidence or gender ratio: it hinges around whether the content in any shape (numerical or not) belongs in the lead. How we choose to summarize it to the lead (precise numbers or general concept) is a whole 'nother thing than whether the concept belongs in the lead at all. Take this example; what is pregnancy doing in the lead? There is nothing in epidemiology which gives a WEIGHT indication for why it's in the lead. It appears we have forgotten how to write leads in medical articles, and in this case, if 50% of pregnant woman had Hepatitis E, I could understand it being in the lead, and then it wouldn't matter if we added the number or not. SandyGeorgia (Talk) 01:52, 6 April 2020 (UTC)

At least we have a response to the question on why some editors believe the RfC applies only to "numerical prices". That's progress. --Hipal/Ronz (talk) 02:13, 6 April 2020 (UTC)

(edit conflict) @Sandy:, I don't mind if you ping me or not. Please feel free to use whatever is habitual for you; I honestly don't mid either way. Reflexively, please let me know if you prefer not to be pinged on each of my posts. I understand you have the page watchlisted.

Yes, of course I see that viewing WP:WEIGHT and WP:LEAD in broad context means that we summarise the important bits of the article in the lead to provide a concise overview. From the perspective of those two policies/guidelines, it wouldn't matter what format the summary was in. Many leads contain nary a number, but Technetium, for example, is chock-full of digits. However, we just had an RfC that looked at three articles that had numerical prices in the lead as examples. Editors were clear that drug pricing is something that may need interpretation and that a single primary database of prices was unlikely to be sufficient sourcing to discuss the drug's cost in general. So, despite WEIGHT and LEAD being completely agnostic about the format of the summary that should be there, we have been told that in the case of drugs, we are not to use prices (and by that I mean the numerical prices as found in primary sources) to write that summary. I suppose one could argue that if a drug had a very notable scandal about specific pricing, and if the scandal formed a large part of the article text, you might be justified in mentioning that specific price in the lead. But that's going to be a rarity and would need considerable discussion.

I'm sorry to be contrary, but I'm pretty certain that in "Editors are generally opposed to inclusion of prices in the lede. While basic calculations are not considered original research, the pricing statements in the examples require interpretation of primary sources that may not be straightforward." The two sentences are talking about exactly the same thing: the use of numerical prices in the lead, as exemplified by the examples.

I have read through all of the discussions leading up to the RfC and the full RfC. Nobody who knows me would ever think otherwise. If nothing else, I am thorough in my research. The RfC and its preliminaries considered numerical prices like "The wholesale cost in the developing world is about US$27.77 per month as of 2014." The reasons the closers gave for discouraging prices in the lead are: (i) the issues surrounding "interpretation of primary sources that may not be straightforward", (ii) the difficulty of verification (especially where prices are not widely discussed), and (iii) concerns about explaining the indicated price would require an undue amount of space in the lead. All of that is very reasonable, but it's a step too far to extend that to forbidding summaries like "it is a relatively inexpensive drug", when there are multiple reliable secondary sources telling us that as a fact, and they are well-represented in the body of the article.

To take your hypothetical analogy: you are quite right. LEAD and DUE are silent on whether we summarise content numerically or textually, so we agree that they do not distinguish between those formats. But what if somebody complains that several articles are quoting an incidence of 1 per million in the population; and that figure is taken from a single database that gives the figure for one country in one year; and then we have an RfC where editors make clear that the problems with specific figures are so great that we shouldn't put them in the lead. Would you then support removing "this conditions is quite rare" from the lead of an article that has multiple secondary sources discussing the incidence in exactly those terms?

If an RfC specifically disallows or discourages something specific that is allowed by LEAD and DUE, we should be extra cautious not to extend that prohibition beyond what the RfC discussed. --RexxS (talk) 02:45, 6 April 2020 (UTC)

So I think we are agreed that:

1. This sentence has only one source. (The five or six other sources are for other sentences.)
2. The source you want to use for this sentence is the one that the big RFC focused on.
3. Including a sentence about the price in just one country (e.g., the US) can be a WP:DUE problem.

Where we disagree is:

1. You believe that putting in five or six other sources at the end of other sentences means that this source becomes suitable for this sentence. I do not believe this.
2. I believe that the RFC concluded that using this database always requires a level of interpretation that constitutes a WP:OR violation. You do not believe this.
3. I believe that the RFC concluded that including a sentence about the price from one seller to one African country is usually a WP:DUE problem. You do not believe this.

Are there any other points of agreement or disagreement that it could be added to the list? WhatamIdoing (talk) 23:30, 5 April 2020 (UTC)

And some editors believe the MEDMOS RfC applies to "numerical prices" only. --Hipal/Ronz (talk) 02:13, 6 April 2020 (UTC)

Not at all. The RfC concludes that editors don't think that numerical prices in the lead are a good idea. I agree with them. But it also has not changed the current consensus on drug prices in the rest of the article. It affirms that pricing information should be sourced to reliable secondary sources. Nor does it change the position on including cost information somewhere in an article, which is still governed by WP:NPOV and our prior consensus. --RexxS (talk) 03:01, 6 April 2020 (UTC)

FWIW, Hipal/Ronz, I think the main focus of that RFC was dollar/cent explicit prices. Some editors, myself included, suggested that there were in fact other ways of describing the cost of medicine, such as adjectives ("low cost") or relative ("the cheapest statin"). Some of those ways have their own difficulties, for example "low cost" is very much regional and also a judgement wrt a class of medicine/indication. So issues surrounding those probably require further careful discussion, but certainly not in this vote-RFC format.

None of this prevents us using existing policy, which for example requires that the lead summarise the article body, both of which must take into account the WP:WEIGHT of the point being made. So the practice documented here in 500+ articles of adding price information only to the lead must stop. On the other points, I agree with WAID, and pricing based on databases such as MSH must stop. That's a firm conclusion of the RFC, and in fact, as a closing admin made clear, this article's use of MSH price database was held up as the very worst example. -- Colin°^Talk 09:26, 6 April 2020 (UTC)

Good spotting, User:Hipal. User:Doc James appears to believe (and he's welcome to correct me if I'm wrong) that the first sentence in the RFC close, "Editors are generally opposed to inclusion of prices in the lede", means that we can't write "It costs $10" in most article's introductions, but that we can write "It's affordable/expensive/whatever" in all the introductions (assuming that statement is described later in more detail). Most editors seem to think that nothing about prices or affordability, in any form, belongs at the beginning of most articles (with exceptions for famously expensive drugs, etc.). So that's another specific point that we need to clear up: Does the word "price" in the first sentence mean "numerical price" or "any reference to prices whatsoever"? WhatamIdoing (talk) 14:55, 6 April 2020 (UTC)

I think that, aside from the arguments over RFCs and policies, we need to apply a degree of common sense. As my mums says, God gave you a brain in your head, so use it. The MSH price from 2014 that James is using is an anomaly. In the 25 years that MSH have been recording drug prices, they have only had a single supplier record in 1999 and another in 2014, but none in any other year, including 2015. In a rational discussion, I hope we'd all agree that is embarrassingly weak data, suggesting more than this drug is not available in the developing world, and not confirming anything about its price. My guess is that a doctor in a private clinic in DRC asked the NGO if they could source some ethosuximide for a patient who wasn't responding to valproate. They got a price for the doc, who tried it and it didn't help, or the price was too high. Either way, the price appeared in that year's price list and then vanished.

What we do know is that over the last decade the price of ethosuximide has been affected by major manufacturers, including the originator brand, pulling out of markets. This should make us concerned about using dated sources, or even using recent sources if their source data is dated. It may also strongly suggest that the price could vary widely depending on which market there is a good supplier or supplier-competition. Wikipedia claiming a drug is generally affordable when one can't in fact purchase it at all in most parts of the world, is daft.

The article currently juxtaposes the dated (2001) and economically dubious claim that the drug is generally affordable "in most areas of the world as of 2008" with an original-research price of $28 a month, citing one supplier in the North Kivu region of DRC. Now it is widely accepted by WHO/HAI that affordability can be judged on the wages of the lowest paid government worker and if it costs more than a day's wages to buy a month's supply of a drug, it isn't affordable. What sources I can find suggest the minimum wage in the DRC is $1.83 per day. That's a long way short of $28, which is a wholesale list price, and we know in the developing world, retail prices can be many many multiples of this. And there are many other countries in Africa and the developing world where wages are not even a tenth of $28. So both these claims, juxtaposed here, cannot possible be true. Neither claim belongs on Wikipedia. -- Colin°^Talk 09:26, 6 April 2020 (UTC)

I think that, given Wugapodes' statement at 03:23, 6 April 2020 above, that this sentence will be removed as soon as Doc James has had a moment to catch up on his watchlist and ask any questions about it that he might still have remaining. I don't think we need to debate the facts again. WhatamIdoing (talk) 14:56, 6 April 2020 (UTC)

WAID I think given Wugapodes' statement at 21:43, 6 April 2020 their is not justification to remove this sentence. Doc James (talk · contribs · email) 23:57, 6 April 2020 (UTC)

So there we are; we have multiple people seeing the statements say something that James does not see and is not able to explain to our satisfaction. SandyGeorgia (Talk) 00:55, 7 April 2020 (UTC)

Moving forward

Trying to look at the bigger picture, my suggestion for moving forward would involve agreement on several seemingly disputed points, some general, some specific:

General

1. Do we have to agree on the body of the article before we try to determine how it is summarised in the lead?
2. Do we agree that drug articles can have a section on Cost?
3. Do we agree that in this context, the word 'price' should be understood per the Cambridge English Dictionary definition as "the amount of money for which something is sold or offered for sale"?
4. Do we agree that drug articles can reference primary sources such as prices, if and only if there is substantial discussion of cost in reliable secondary sources?
5. Do we agree that drug articles can quote (but not interpret) specific prices, when relevant to the discussion on cost, in the body of the article?
6. Do we agree that drug articles cannot quote specific prices as part of the article summary in the lead?

Specific

1. Should this article have a section on Cost?
2. The present section is tagged as having its factual accuracy disputed. Can we agree on which facts are accurate, and which are not? (this will require specific examples)
3. Are there other problems with the section? If so what are they specifically?

Anything else? --RexxS (talk) 17:13, 6 April 2020 (UTC)

Thanks for summarizing with numbers, for ease of response. I will circle back later and catch up with rest of page.

1. Although working from body first, to lead last is always my preferred approach, in this particular case, it appears that there are some changes that can be made to the lead now, while deferring other changes. Specifically, the disputed statement (now shown conclusively to be inaccurate) should be removed, and we can separately decide how and where to deal with the price fixing case, relative to NOTNEWS, WEIGHT, etc.
2. Drug articles can have a section on cost when sufficient relevant secondary sources (not databases) mention them and we an write meaningful text without breaching V, WEIGHT, and OR. Basic WP:NOTPRICE has not been rejected by the RFC. All database-source (and indeed, all poorly sourced content, as Colin's posts above delineate) should be removed. That could be done now. We need a much more careful examination of sources to help avoid walls of text to explain our sourcing policies and uphold NPOV.
3. Unsure why that matters ... we are guided by sources, WEIGHT and LEAD, regardless of semantics, use of "numerical", etc ... we should take care not to wikilawyer wording and respect the spirit of policy and guideline. Perhaps I am not understanding the question?
4. I believe we have rejected primary sources substantially, for the specific case of drug prices, for all of the reasons and problems laid out in the RFC.
5. Same as 2 above. When policy and sourcing allows, subject to consensus and the usual examination of sources (such as the in-depth examination by Colin above). That is, we need to take much greater care with source-to-text integrity, reliability of sources, WEIGHT, etc.
6. I do not believe we have found this broadly, via the RFC. We are guided by policy and guideline. When sufficient WEIGHT issues and sources discuss specific prices in a specific controversy (think Epipen), we might find consensus to add them to the lead, according to WP:LEAD. These cases will be few and far between, but we shouldn't exclude that they exist.

Then:

1. So far, we don't have an indication that this article should have a Cost section, but we need to decide how to handle the price fixing. Is that best left altogether out per NOTNEWS, added to the manufacturer's articles, added to drug pricing general article, or added to every drug article affected?
2. See my previous answers. Anything sourced to a primary source database, anything shown to be inaccurate for discussion above, and anything which deals with one country only, hence broadly irrelevant and UNDUE. I think that means there's nothing left in this case.
3. Just that we need a general discussion about where/how to handle the price fixing.

On the "anything else", just that we need to all go over to WT:MED and have a discussion about pinging, how to keep track of, honor and respect multiple differences, so we can keep up with each other without spurious charges of harassment when using a common feature of the website. MY feelings on pinging are recorded at the top of my talk page: Every Click Physically Hurts. We should have a broader discussion at WT:MED. SandyGeorgia (Talk) 18:00, 6 April 2020 (UTC)

RexxS, I don't think it is helpful to contemplate further areas of agreement/disagreement on drug prices in this Ethosuximide talk page. This is an article surely that until a few days ago was likely on a couple of people's watchlist. The idea that a secondary source making passing mention of the cost of ethosuximide (for that is all that has ever been shown for this drug) means that we can use primary sources to (a) make claims that require original research or (b) offer a price that no other source offers, is pretty ridiculous and already rejected explicitly by Ymblanter at WP:MED. This is really like saying because lots of sources think cancer drugs are useful in treating cancer, then I can use my primary source case study to suggest that Ribena is a universal and effective treatment. At WP:MED I asked to confirm: "In particular, it is not sufficient to find secondary sources mentioning the the cost of a drug (whether in dollars or abstractly) and then include a different price sourced to a primary database of product prices. The prices themselves must be sourced to secondary sources who make extended commentary on that price" and Ymblanter replied "Yes, this is what we concluded", with a caveat that there may be a magical database that doesn't have MSH's flaws (there isn't).

RexxS, earlier you wrote "I would think that a drug described as an inexpensive anti-epilepsy drug in 2005 would remain so, since we don't seem to see other sources remarking on it becoming expensive". Given what I posted today, that the price in the UK had increased 40x since 2011, and this had been noted by an epilepsy charity and by neurologists, and given the restrictive aspects of price James was actually Googling for, you might want to accept here you were wrong. This is part of consensus forming.

This article is still quoting a retail price in Italy from 2001 (and falsely claiming it is from 2011, three years after the source was published). This Italian price is for about half the dosage as the preceding two prices. Why? Why a 20-year-old half-the-dose price in Italy on the English Encyclopaedia? Anyone? And it juxtaposes this with a UK wholesale price that we all now know is wrong by 40x along with a comment on syrup vs tablet costs that is now inverted from reality in 2020. Are we really on this project in a stage where we must all wait for James to consent to the removal of that nonsense. That basic facts we know are not just false but in fact the opposite to reality are retained. -- Colin°^Talk 21:18, 6 April 2020 (UTC)

@Colin: I disagree with your assertion that it's unhelpful to try to identify the areas where most of the editors participating here agree or disagree. If you are genuinely interested in finding areas of compromise, rather than insisting that your own view should prevail, then I would have hoped you would have welcomed attempts to isolate the disagreements.

If we have secondary sources that discuss a forty-fold increase in the price of ethosuximide, why wouldn't we use those? Moreover, if we are discussing controversies such as price fixing and artificially limited availability, why shouldn't we include statements in the Cost section such as "In the UK, April 2020, the BNF gave the indicative price and tariff price of 56 capsules of ethosuximide 250 mg as £173 from six suppliers"? There is consensus to add pricing "where the sources note the significance of the pricing". WP:PRIMARY is not violated by a mere statement, without interpretation.

There is clearly a disjoint between your research and my observations. The NICE costing statement from January 2012 (3.2.5) states a price of £0.27 per mg, while the BNF (2020) I quoted above indicates a price of £0.012 per mg. I'm not seeing a forty-fold increase. No doubt that sort of discrepancy is exactly the sort of thing that our secondary sources would discuss, and should be reported our article. I'm quite happy to acknowledge if my assumption that no sources have shown a dramatic price increase turns out to be wrong. But you don't seem to have done us the courtesy of indicating the sources that support your assertion, and that really isn't helpful.

Because this is an international encyclopedia, we should strive to give information on a global basis. A 20-year-old price from Italy is no better or worse than a 20-year-old price from the USA. If you were to try to find some common ground, you would find that over-egging a reasonable argument (20-year-old prices) merely serves to diminish it.

If you wish to remove the entire section on cost (as you've already done once), then make that proposal. Other editors will have no problem in finding sufficient secondary sources discussing the cost of ethosuximide or zarontin to rebut that.

Alternatively if you want to improve the article and wish to re-write the section to report the secondary sources that you're already aware of, please propose a suitable starting wording, and I'll be happy to contribute to a solution that all participants can live with.

@Doc James: for a list of as many secondary sources relating to the cost of ethosuximide/zarontin as you're aware of, when you find time. --RexxS (talk) 23:11, 6 April 2020 (UTC)

RexxS, I may have misunderstood, but it seems that Colin said that we know there is a 40-fold increase, but we don't have reliable sources for that (that is, we have our own original research, which demonstrates why we shouldn't be using old data, but we have nothing reliable to replace it the old data with). Did I get that wrong, Colin? My understanding at this point is that we have no usable secondary reliable sources to even discuss, based on everything above, except those which discuss the price fixing. And that our next decision is where the price fixing text belongs: here or in the manufacturer article or nowhere per NOTNEWS. SandyGeorgia (Talk) 23:22, 6 April 2020 (UTC)

RexxS, I would appreciate if you would re-read my first sentence, particularly the last five words, and consider to strike your first paragraph. Thank you. We don't have reliable secondary sources discussing the price increase. We have a tabloid newspaper. And I have already indicated the sources that justify my assertion. The 2011 price comes from a NICE guideline which very much as an aside noted that in November 2011 the tablet price was 6.75x the syrup price and providers should bear this in mind. The 2020 price comes from a primary source of product prices (BNF -- not accessible outside of UK) that the RFC has essentially forbidden for price statements. We can discuss and note the price increase on this talk page, and we can (hopefully) come to an agreement that if ethosuximide was ever reasonably considered low cost, it is not in 2020 in the UK nor the US, and never really was available anywhere in the developing world. We can't include that in the article, and neither should we include the old data, particularly now we know it is not merely old but inverted from reality in 2020. You seem unconcerned that the Italian price is likely to be wrong too. I cannot think of any other medical information where this project would be happy with 20-year-old data like this, or strive to claim that a 20-year-old Italian retail price has any weight, or that juxtaposing different dosages and wholesale/retail is doing anything but bamboozle our readers.

The requirement for noting the price in a drug article remains and has always been extensive discussion/commentary in mainstream media, and WP:NOTPRICE explicitly disallows "passing mention". Every single one of the sources on this talk page or article page is either a primary product price database or passing mention in a non-mainstream-media publication. This was covered in the RFC.

As for ethosuximide prices, it turns out the NICE document gets both syrup and capsule prices wrong by a factor of 100 too high: their price is really in pennies rather than pounds. It doesn't change their argument that the tablets are 6.75x more expensive than the syrup in November 2011. But if you look at the footnote, you'll see the actual prices for the pack of 56 250mg capsules (£38.23) and for 200ml of 250mg/5ml syrup (£4.22). The equivalent prices today for the same pack or bottle are both £173. That's an increase of 4.5x for the tablets (which were dearer to begin with) and 41x for the syrup. -- Colin°^Talk 07:54, 7 April 2020 (UTC)

@SandyGeorgia: Colin said "the price in the UK had increased 40x since 2011, and this had been noted by an epilepsy charity and by neurologists. If that is the case, then surely he is quoting some published secondary sources (the epilepsy charity and neurologists). I didn't think he was reporting his own conversations with them. Wouldn't we consider an epilepsy charity and a neurologist reliable sources for commentary on an AED? My original research (fine for a talk page) showed a cost decrease, so I was genuinely interested in reading the sources Colin refers to. I'm aware of a couple of sources from 2007 and 2011 that speak of the cost effectiveness of ethosuximide in comparison with other AEDs during that period, so I think there is room for some discussion in the article of how that position has apparently shifted over the last decade. Let's see if editors can track down some decent sources that discuss it.

@Colin: I re-read the whole of your first sentence and still don't find it constructive. You still haven't given your sources and I'd like to to read what the epilepsy charity and neurologists said.

Again, you ascribe sentiments to me that are not true. The problem with 20-year-old Italian data is that it is 20 years old, not that it's Italian, and that was the point I already make quite clearly. Would you have a problem with using a 1-year-old Italian source to discuss a subject concerning Italy?

You contend that "The requirement for noting the price in a drug article remains and has always been extensive discussion/commentary in mainstream media", but that is unfounded. We have consensus to include pricing in the body of the article where sources note its significance, and your extra conditions do not have to be met. Unless, of course, you are only considering the question of numerical prices, where the feeling, if not the consensus, favours rejecting databases as the sole source, and I certainly don't have a problem with that.

So the NICE document is quoting £0.27 per mg for the capsules when it should be 0.27p per mg. Fair enough. That's a £173/£38.23 = 4.5x increase without adjusting for the (low) rate of inflation. The syrup currently costs 1.73p per mg in the BNF as opposed to the 0.042p per mg from the NICE document, the 40x increase. We're on the same page. Perhaps the only remaining disagreement is whether we should reject as a source the NICE document, of which about half is devoted to examining the cost-effectiveness of AEDs, and which makes recommendations based on those, including the whole of page 7 discussing ethosuximide. It seems rather more than a passing mention to me.

Given that we have consensus to include pricing in the body of the article where sources note its significance, should we be writing something along the lines of

• In 2011 in the UK, ethosuximide was over six times more expensive in tablet form than in syrup form, and NICE noted the potential cost impact of switching.

in the Cost section? --RexxS (talk) 12:17, 7 April 2020 (UTC)

See my comment at User talk:Barkeep49#Discretionary sanctions. The sources are there in full and always have been. Nobody mentioned anything about personal conversations.

Before I quote policy and the closing admins, let's clear up the pricing/price confusion. This was covered in the RFC Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices see "Pricing/prices" box. Drug pricing is the practice of determining a price to charge or agree to pay for a drug. How drugs are priced (internal reference pricing, external reference pricing, controlled and regulated vs free market), whether this is fair, whether some companies are exploiting the market, the politician and economic consequences, how this translates to the population getting the medicine they need, etc. That's all encyclopaedic and we have whole articles on it and could write more. Drug prices are absolute numbers like $34.45 a pill. Further there are treatment costs such as $25 a month, which again were discussed during the RFC and a number of original-research and weight and missing-information issues were discussed. The routine inclusion of these is an area of conflict discussed during the RFC. Then there are other cost comments such as adjectives like "low cost" or relative like "cheapest statin". There are probably others. The main focus of the RFC was to look at deriving an absolute price (whether per pill or treatment cost) in numbers from a product price database such as MSH's. In practice, editors, including the closing admins, interchange price and pricing, but in both cases they were generally talking about the price of a drug rather than how drugs are priced.

The important thing to remember is that a comment may indicate some necessary preconditions (for including a drug price say) but not necessarily the sufficient conditions. In the closing comments, two extremes were described. One was a drug "discussed extensively" in the secondary literature (e.g. insulin) and another with "little discussion" in the secondary literature. At the one extreme "that information [what the secondary sources say] may be worth including in the article". Note the "may". At the other extreme "it generally should not be included". Note the "generally". This much reflects the long-held policy at WP:NOTPRICES. We all agree that insulin's price (in the US at least) is discussed extensively by mainstream media sources. At the opposite extreme really is ethosuximide, who's price, even though it has increased sharply in a decade, is not found notable by anyone except The Sunday Post, a tabloid, and who only ever earns passing mention in the secondary literature. I have looked, RexxS. I explicitly asked Ymblanter to clarify. Here is my question and response. There is no doubt that the full requirements of WP:NOTPRICES are upheld. You need as a minimum extensive discussion in mainstream media sources and explicitly that "passing mention" is insufficient. Further the admins clarified that "Where pricing information is included, claims should be sourced to reliable, secondary sources and not solely primary source data from price databases". This means the price itself must come from the secondary source that discusses that price. Again that's in the diff with Ymblanter. Wugapodes has further clarified the ethosuximide MSH price should go and again clarified that the rewrite to avoid interpretation was also rejected by the RFC and closing remarks. James appears to have seized upon "I don't think consensus at the RfC was sufficient for a blanket ban on the MSH IMPPG database" as a declaration of the near opposite statement that the database is absolutely fine. Nothing could be further from the truth. Since none of the prices derived from the MSH IMPPG, Drugs.com, NADAC, BNF, etc, appear in any of the secondary literature (because they are all cherry picked and original research derived by a Wikipedian) it is very hard at this point to see how the MSH IMPPG, or similar database, can be used as the "sole source" of a drug price. The candard that secondary sources mentioning some aspect of cost meant primary source can be used to give/derive a price is busted. All the reliable secondary sources on this page do no more than passing mention of the price of ethosuximide. If there were many newspaper articles like the one in The Sunday Post then we could consider including a price, if that helps our readers, and the price would have to come from the secondary literature that discusses the price, not from a product price database.

I am sorry the above is a bit wordy. I am trying to explain as simply as I can, for the benefit of all readers of this page, as we seem to be at various stages of catch-up, denial, grief, anger, acceptance, etc. -- Colin°^Talk 13:43, 7 April 2020 (UTC)

As for the syrup vs tablet point from 2011, my mind really boggles that a minor "by the way" comment about a dated price discrepancy between syrup/tablet should feature on Wikipedia in 2020. We all here know that the discrepancy has gone, though there are no secondary sources to confirm that. Retaining this will only cause our readers to wonder and assume that the discrepancy remains. We are not required to include every factoid that turns up on a google search for "ethosuximide cost". I can guess that the capsule had recently had a price hike (because normally syrup formulations of drugs are a bit more expensive than tablets) and we know that since then the syrup played catchup. What this discrepancy should further remind us is that There Is No One Price For A Drug and that prices are quite volatile. -- Colin°^Talk 13:55, 7 April 2020 (UTC)

@Colin: When you stated "the price in the UK had increased 40x since 2011, and this had been noted by an epilepsy charity and by neurologists" without naming your sources, I think I'm entitled to wonder what these were – "surely he is quoting some published secondary sources ... I didn't think he was reporting his own conversations with them."

Thank you for taking the time to lay out at length your interpretation of the policies and practices surrounding the issues of drug prices and pricing in Wikipedia. First of I take it that you agree with my suggestion at the start of this section that the word 'price' should be understood as an amount of money. Unfortunately I find it difficult to reconcile that with your assertion later on about "other cost comments". I reject your assertion that the closing admins interchanged 'price' and 'pricing' as if they were unaware that it had been extensively examined in the RfC.

Turning to policy then, we have WP:NPOV "representing fairly, proportionately, and, as far as possible, without editorial bias, all the significant views that have been published by reliable sources on a topic." Let's see how that applies to the issue of cost and cost-effectiveness. I believe that the view of NICE in 2012 on the cost-effectiveness of ethosuximide is significant. This is an important public body giving recommendations to practitioners in the UK. They make a point of the price differential between syrup and tablet formulation at that time, but you want to reject it on the grounds that your own original research leads you to believe that it is no longer the case. You really can't have it both ways: it seems that you're happy to use an OR argument to exclude raw database prices; but you also want to use your OR to exclude a significant view from a major national institution, dismissing it as a minor "by the way" comment. It wasn't. It was part of an extensive survey of effectiveness and cost by NICE in order to give guidance to practitioners, and is as encyclopedic as any other view expressed in a reliable secondary source of that nature.

The relevant policy on WP:NOTPRICES needs to be read and understood. It states

Sales catalogues. An article should not include product pricing or availability information unless there is an independent source and a justified reason for the mention. Encyclopedic significance may be indicated if mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention.

There is a requirement that any inclusion of pricing needs sourcing independent of the manufacturer and discussion of the significance of pricing in sources. Where you have misunderstood is you mistake the statement that "Encyclopedic significance may be indicated if mainstream media sources provide commentary on these details" for a necessary condition for inclusion, rather than a sufficient one. We agree that "If the mainstream media makes more than a passing mention on pricing, it may indicate encyclopedic significance." But you have inferred a incorrect contrapositive: "A non-mainstream medium making more than a passing mention on pricing does not indicate encyclopedic significance." It is not necessary for the reliable source to be "mainstream media", and you cannot discount the significant views of 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. Per WP:MEDRS, those are ideal sources.

I can't debate other editors with you. Their views and actions are their own. If you want to work toward consensus with me on the article content here, you should try to unambiguously address the simple questions I ask at the start of this section. I think you've indicated agreement with the understanding of the word 'price', but not much else that I can see. The most pressing point at present is whether you accept that pricing information can be part of the main part of an article under the same conditions as any other piece of content, or whether you want to impose a higher bar. --RexxS (talk) 15:18, 7 April 2020 (UTC)

RexxS, you raise many points, most of which involve misunderstanding what I wrote or what the closing admins wrote or misrepresenting longstanding policy. I've made my position as clear as I can and cannot make it clearer without repetition. Some of the arguments you are making were discussed at the RFC, where you chose not to participate. You are currently raising more points of misunderstanding with each post than you are clarifying existing ones.

You seem to think that guideline-required sourcing is required both in order to say something in an article and equally also to choose not to say something you know is simply wrong or hopelessly dated or misleading. You say that I "want to reject it on the grounds that [my] own original research leads [me] to believe that it is no longer the case." So, RexxS, is the current UK NHS price of Ethosuximide syrup significantly cheaper than the price of a capsules for the same dose? This isn't a question of religious belief or politics or opinion on how best to treat coronavirus. This is something you can work out for yourself with basic primary school level maths. -- Colin°^Talk 16:10, 7 April 2020 (UTC)

Colin your entire first paragraph is simply a personal attack on me. I misunderstood nothing. You have chosen not to address the points I made on NPOV and NOTPRICES, and I think that speaks volumes to the shaky foundations that you base you arguments on.

Don't put words into my mouth: "guideline-required sourcing is required"? Seriously, what are you trying to make me say? I do believe that content has to be based on good quality reliable sources. I believe that content related to cost, cost-effectiveness and availability of a drug belong in the drug's article, within the bounds of WEIGHT, when we have good quality reliable sources that cover those issues. I believe we have consensus to add drug pricing to articles where the sources note the significance of the pricing. Which of those do you disagree with?

In the absence of reliable sources that discuss it, it doesn't matter what you or I think the price of ethosuximide is in 2020. I already did those calculations in my prior post, I have a degree in maths, and taught it for 25 years, so I don't need you to patronise me about "basic primary school level maths". The question is simply whether you can reject encyclopedic content describing pricing in 2011 based on a good source, because your calculations show that the pricing is different now.

Now, can you please say whether you still maintain that ethosuximide was never considered low cost? Can you tell me whether you believe the RfC disallowed pricing information anywhere in an article? Do you still maintain that "You need as a minimum extensive discussion in mainstream media sources"? --RexxS (talk) 18:05, 7 April 2020 (UTC)

RexxS, claiming that IMO you misunderstand several things is not a "personal attack". I don't think this conversation is working because the degree to which we disagree and misunderstand each other seems to be escalating like some pandemic virus. It isn't helped by your open declaration of bad faith and accusation that I'm deliberately making it hard for you to follow. Perhaps User:WhatamIdoing can clarify the rules on editors choosing not to say something that is clearly in 2020 wrong and misleading. I am truly perplexed by your "You really can't have it both ways" argument that if someone says something in the past that was true then, that we need an equivalent source saying the opposite or saying "You know that thing we said eight years ago, it isn't the case any more" to choose not to include that on WP today. Sourcing doesn't work that way. I said what I wanted to say about "low cost" at 09:18, 1 April 2020 and further investigation has only shown the claims by James about the sources to be misleading or clearly outdated by events. The absence of good quality secondary sources discussing a matter is more an indication that neither should we mention it, than that any old crap from decades ago that Google finds has to be included here. -- Colin°^Talk 18:33, 7 April 2020 (UTC)

• "Any old crap from decades ago: noun. For example, the retail cost of treatment with a drug for an atypical period of time at lower than usual dose in a random non-English-speaking European country from twenty years ago". -- Colin°^Talk 18:38, 7 April 2020 (UTC)

Wrt my "choice" not to address some of your points "speak[ing] volumes about [my arguments] shaky foundations". I am more than aware of your repeated accusations of blugeoning, and your continued comments about how much I have written on the subject of drug prices. So I haven't responded to all your points per the exponential pandemic point I made earlier. There is only so much wrongness one can deal with in one go. If I write something, you attack what I have written. If I don't write something, you attack what I have not written. -- Colin°^Talk 19:28, 7 April 2020 (UTC)

Colin, WP:MEDDATE says "These instructions are appropriate for actively researched areas with many primary sources and several reviews and may need to be relaxed in areas where little progress is being made or where few reviews are published" and "In many topics, a review that was conducted more than five or so years ago will have been superseded by more up-to-date ones, and editors should try to find those newer sources". Would you agree that the area of cost/cost-effectiveness/availability of AEDs is one where few reviews are published, and that therefore an eight-year old paper from NICE fits the requirements for MEDDATE?

Given that I proposed a wording that was based on an eight-year old respected UK source, I don't see how complaining about "the retail cost of treatment with a drug for an atypical period of time at lower than usual dose in a random non-English-speaking European country from twenty years ago" helps the discussion.

Why do I have to go searching through your contributions to find something you could have linked. Is 09:18, 1 April 2020 in EST, GMT, BST, CET? What? Simple basic courtesy to other editors is needed if you're going to avoid accusations of deliberate obstructiveness. Can you please say whether you still maintain that ethosuximide was never considered low cost?

Are you willing to state whether you believe the RfC disallowed pricing information anywhere in an article?

Do you still maintain that we need as a minimum extensive discussion in mainstream media sources to meet NOTPRICE?

"If I write something, you attack what I have written. If I don't write something, you attack what I have not written." You do realise that I can justifiably level exactly the same complaint against you. --RexxS (talk) 21:45, 7 April 2020 (UTC)

Pings

OK, here's how I work with pings. Processing through them hurts, as it just gives my fingers more clicks to deal with (see the top of User talk:SandyGeorgia)-- early in the day and I am already headed for the Tylenol bottle. When I'm following a page, I don't need pings, but I hope I would never be so arrogant as to demand or expect other people to remember that, or to remember every editor's preference with respect to a feature of the website. IMO, if someone doesn't want to be pinged, it is their responsibility to alter their user preferences.

My other practice is to let the talk page know when I am unwatching, and request that THEN I please be pinged if my feedback is needed.

I am unwatching this page; please ping me when AGF returns. It is apparent that no matter how clearly Colin types, if his posts aren't being read, we won't make progress towards consensus.

Colin, Ronz, SandyGeorgia and WAID are all clear on the V, WEIGHT, NOT and LEAD issues here, and yet this page has turned into walls of misunderstanding, misreading and misinterpreting requiring lengthy explanation. Asking for clarification would be so much shorter. It looks like this: "Colin, I see you said X, but I see Y: could you clarify?" Response, "When I said X, I meant this; is it more helpful if I lay out the full math?"

It appears that the text will remain tagged disputed, as most of us don't routinely edit war, and there is no way to resolve this, although consensus is darn clear that most of the text in the article now should be removed. Ping me back if people want to honestly start reading and digesting posts with the good of the Project in mind. Bye for now. SandyGeorgia (Talk) 15:29, 7 April 2020 (UTC)