Talk:Aspirin/GA1

GA Review

The article is fairly close to meeting the GA criteria, but needs a little bit of work to bring it up to standards:

• There are several citation issues in the article. First, there's a 'citation needed' tag that needs to be addressed. I've also seen several sections with uncited or poorly cited material, mostly in the history section, but also in a few other places.

YCitation issues appear to have been adressed.CrazyChemGuy (talk) 22:54, 2 May 2008 (UTC)

• The prose could use a bit of a cleanup. I would recommend a good, thorough copyedit. Some material looks a bit too technical, and almost looks like some information from the abstracts of scientific articles was just copied directly into the article (e.g. 'adverse effects'). That section should have a summary at the beginning of the section, instead of just going right into subsections. Try to tie the information together instead of just listing it. Some sentences need to be corrected for grammar and rephrased.

• There's some manual of style issues, so I would recommend someone with experience in that area to copyedit the article. Look at wikilinking (only full dates should be linked, not years and not month years). Why is rat wikilinked? There doesn't seem to be context for this. The placement of reference citations in sentences should also be examined -- citations should be placed immediately after the sentence(s) being cited, immediately after the punctuation with no space between (like this. not like this. ). There's some references listed which only have a single external link with no author, title, publisher, date of publication, date URL retrieved, included in the citation.

• I would promote the mechanism of action section; this seems to be a bit more important than some of the other sections. Probably should fall right after 'therapeutic uses'.

YSection has been moved. CrazyChemGuy (talk) 19:02, 30 April 2008 (UTC)

• WP:MEDMOS does not advocate having a 'dosage' section; we have to be very careful with this as we don't want to look like we're providing actual medical advice. This section could be removed from the article.

YDosage information was removed from the section; section should now comply with WP:MEDMOSCrazyChemGuy (talk) 19:02, 30 April 2008 (UTC)

• 'Contraindications' is just a bulleted list. While generally well cited (except for one point), it should ideally be converted to prose and there should be more information to tie this section together better. As it's written, it looks like it belongs in the drug information brochure by the manufacturer, instead of in an encyclopedia.

YConverted to prose and merged with Resistance.CrazyChemGuy (talk) 19:02, 30 April 2008 (UTC)

• The 'resistance' section is very short, and could probably be combined with another section. It seems out of place in the lineup of sections here. On it's own, it doesn't comply with WP:MEDMOS.

YMerged with Contraindications.CrazyChemGuy (talk) 19:02, 30 April 2008 (UTC)

• Likewise, 'polymorphism' is also very short, and should probably be combined with another section as well. Perhaps add it to a section on 'chemical and physical properties', in a discussion on the structure.

YMerged, along with Synthesis, into section on Chemistry. CrazyChemGuy (talk) 23:12, 2 May 2008 (UTC)

• There's no information on pharmacokinetics (absorption, distribution, metabolism, and excretion) in the article.

YA section on pharmacokinetics has been added.CrazyChemGuy (talk) 19:02, 30 April 2008 (UTC)

These are the biggest issues with the article the way I see them now. I will put the article on hold for one week and review it again after that. Dr. Cash (talk) 23:46, 27 April 2008 (UTC)

There are more comments which should be addressed in the current peer review for the article (although for future reference peer review should not be conducted at the same time as a GA review). Dr. Cash (talk) 23:49, 27 April 2008 (UTC)

Dosage strengths removed as per above comments and WP:MEDMOS (vary by country, individual products and age). See changes. Patient product literature, the references given and most importantly the prescription instructions of ones own doctor should be followed... In teh case of thromboprophylaxis, some advocate 75mg, others 150mg and that's just in UK (little evidence for either of these doses - just historical usuage) and different values used in US I believe - sufficent to point out much lower strengths and once daily rather than multiple dosing when used as an analgesic. David Ruben ^Talk 11:43, 28 April 2008 (UTC)

• Oppose. Out of the first four references I checked, three were false in the sense that they stated exactly the opposite of what was written in the article. Serious (and long) work is needed to re-check all the references. I will fail the article unless somebody thinks it is unfair. Paul Gene (talk) 10:34, 29 April 2008 (UTC)

I won't oppose your opposition, but I looked at those references and I don't think the situation is as simple as you put it. It illustrates the larger problem of writing about recent medical studies, and that is that there are no definite conclusions in many cases and one can always find a study with the opposite result. For example, let's look at the prostate cancer issue you addressed in this edit. The two references provided are for individual studies that had negative results, but they acknowledge that there is an existing hypothesis and do not claim to have disproven it definitely but end with weaker statements such as "These data suggest that aspirin use may not be associated with reduced risk of prostate cancer" and "These findings do not support a protective role of regular aspirin use on prostate cancer risk". This is because it's hard to be certain from just one study; if you look for example at PMID 15998950 , that one ends with "These results support the hypothesis that long duration regular NSAID use is associated with modestly reduced risk of prostate cancer." So what is the truth? I have no idea, and I'm not a specialist in prostate cancer or aspirin, like most Wikipedia editors here, I imagine. The only thing I can say is that some studies support the hypothesis and some don't. I can also see from reading the abstracts that there are different definitions of regular aspirin use. One of the negative studies defined it as once per week for six months, while the positive study defined it as 30 per month for 5 years. That already is an obvious difference that might explain the different results.

The real question is, what should the Wikipedia article say? I think the original text was reasonably cautious, as it said "The drug may be effective..." (that is, it may or may not...). The problem is that it only cited specific negative studies. Ideally it should cite a review or meta analysis that can attempt to provide a bigger and more balanced picture. If no such reference exist, at least it should say "some studies support this hypothesis[ref] and some don't[ref]. Another option is to try to avoid reporting the bleeding edge of medical research on Wikipedia, as it is very hard to do it in a balanced way. Let's wait until the research makes it into the textbooks rather than citing very recent specific studies. I suspect this will not be the most popular option because some people will be afraid of looking outdated, but for an encyclopedia this can be better than looking like the evening news shows, which one day report "X cures cancer!" and the next day report "X causes cancer!", which only results in confusion to the audience. --Itub (talk) 11:27, 29 April 2008 (UTC)

No, you misunderstood. Of course you can cite the most recent research in WP! All I want is this — if it is written "aspirin prevents cancer", the references given should support this. So if references say "mostly likely not" the article should say "most likely not" — not "likely yes". Since this problem keeps occurring in this article, (I saw it before in the chemistry part) I conclude that a serious work is needed to check all the references, and it is premature to promote it to GA. Paul Gene (talk) 18:09, 29 April 2008 (UTC)

I don't think it's time to fail the article just yet. When the article was first placed on hold, a week was given for improvements to be made. While I do agree with you that the issue of the citations needs to be addressed, failing the article will only take away motivation for the article to be improved. Personally, with my current real life workload, I should get plenty of time to work on the article later in the week. I may not be able to fix everything that has been suggested here, but even then, the article would still be better than if it was failed immediately. I would strongly suggest keeping the article on hold to see if these issues can't be fixed. CrazyChemGuy (talk) 20:58, 29 April 2008 (UTC)

The article certainly won't be failed right now; part of the process of GA review is a general improvement phase, which includes addressing issues are they pertain to the Good Article criteria. Some articles pass easily; others take more time; hence the reason for the on hold step. It would be more helpful, though, if Paul would be more specific in the concerns regarding citations. Simply saying, "Out of the first four references I checked, three were false..." isn't really all that helpful. Which ones were "false"? Where do you see the problems. Let's try and help each other out here rather than being vague and throwing out "oppose" votes (it's not a vote). Dr. Cash (talk) 21:09, 29 April 2008 (UTC)

Read the discussion and edit history. My point is that article is very far from GA. There are major not merely cosmetic issues. Paul Gene (talk) 22:43, 29 April 2008 (UTC)

On a side note, I've added a section about pharmacokinetics. The prose could probably use some work, as I admit I'm not the best at that, and it sounds rather technical, but I did try to ensure that my sources were quite reliable - in this case, they are all from well known medical journals, such as JAMA. I think this should improve the article a lot, once the section gets smoothed out (along with the prose/copyediting needs of the rest of the article); if anyone has any suggestions about how to improve this section I'd love to hear them. CrazyChemGuy (talk) 23:50, 29 April 2008 (UTC)

I still have some issues, mainly with some of the references, and am having a hard time passing it as is.

• In the lead: "Low doses of aspirin may also be given immediately after a heart attack; these doses may inhibit the synthesis of prothrombin and therefore produce a second and different anticoagulant effect,[1] although this is not well understood." Why is the citation on the statement right after the comma, not after the sentence. If it's not well understood, says who? If there are disputes in the literature, this should be pointed out. But I wouldn't focus on something somewhat controversial in the lead; the lead should summarize the article. Focus on experimental treatments later in the article.

• The last two paragraphs of the history section have completely uncited material.

• A copyedit should still be done. I can see several instances where there should be commas. Others were there shouldn't be. Others were phrasing is awkward.

• Some of the references in the 'therapeutic uses' section just aren't used very well. For example, in this sentence: "Aspirin is one of the most frequently used drugs in the treatment of mild to moderate pain, including that of migraines,[7] and fever." why is the citation on the migraines, and why is there a comma after migraines. It should be relatively trivial to find a damn pharmacology book that backs up the whole sentence, and put a citation after the whole thing, instead of in the middle. It seems like editors here are trying to use a wikipedia reference to give more weight to someone's paper, instead of just citing this from a book -- this is really basic info here.

• The overciting of other material can be seen in the 'experimental uses' section, specifically in the sentence on reducing the risk of various cancers. Five citations after the gastrointestinal bleeding part, two after "lung", two at the end of the whole sentence!! I think we need to really look at what citations are really necessary here, and which ones are merely filler. Citations that aren't really necessary can be removed.

I'd still like to see this get to GA, but I still don't think we're quite there yet. It's already been quite awhile on hold, so try and finish this up by saturday, or I'll really have no choice but to fail it. Dr. Cash (talk)

I'm not quite sure that the material in the 'experimental uses' section is overcited. I checked each citation, and one of them was not properly placed, and I moved it, but other than that it did indeed seem that they all supported the material the claimed too. Given that the correlation between aspirin use and various types of cancer isn't simply something that is widely accepted, I think it's okay to have that many citations there - I think claims such as a correlation between aspirin and colon cancer needs several studies to be cited and all in agreement. I'm not quite sure why you feel it is a problem for this information to be "over" cited. After editing the article some more, I can't find much else in the article to improve, save for perhaps some copyediting, though I'm sure much of this is just because I've been staring at this same text for as long as I have. I think it's at the point now where it can pass as a GA, especially after I give it a good copyedit here, but the real test of this will be in the eyes of a fresh, unbiased editor, so... I suppose it's time to see what you think. CrazyChemGuy (talk) 15:30, 17 May 2008 (UTC)

Some points

I will try to do a formal GA review later, but quickly some points:

• Why is the chemistry section at the bottom? Wouldn't it make more sense to progress from chemistry to pharmacology to clinical uses to contraindications/adverse events?
• The external links seem to generally be duplicating the content of the history and pharmacology sections.
• Ref 28 looks borky. Needs proper formatting (AHA).
• Some other references are bare URLs.
• The treatment of toxicity is still a bit vague. What is actually done is the use of bicarbonate solution to acidify the urine and aid in the excretion of salicylates. This needs a good reference.

I was a bit dismayed that we allowed a citation "aspirin may increase pancreatic cancer risk" based on a single study. People read Wikipedia and do get worried about such claims. In this case I easily found a later study that disproved the association, but we really ought to be careful. JFW | T@lk 07:27, 7 May 2008 (UTC)

I think I agree with Paul Gene above that this article still needs some work before it will reach GA. JFW | T@lk 07:30, 7 May 2008 (UTC)

Like I said above, citing specific single studies is dangerous, and there are many such citations in this article. For every study, you can always find another study the opposite result. ;-) Therefore, I suggest we should try to cite only secondary literature that reviews multiple studies and puts them into perspective. --Itub (talk) 09:30, 7 May 2008 (UTC)

There is nothing wrong with citing particular studies, especially if they have made some waves, but my point here was that the "aspirin causes cancer" link had been disproven by subsequent research. Omitting the meta-analysis was IMHO irresponsible.

Which other "single studies" are you referring to? JFW | T@lk 20:50, 7 May 2008 (UTC)

For example, the sentence "Aspirin did not reduce cataract formation in diabetic patients.[11]" This is citing one study from 1992 which in fact reached this conclusion. But what about the earlier studies that reached the opposite conclusion? How about mentioning why someone might think that aspirin might reduce cataract formation in the first place? When I see a statement this on Wikipedia, based on one, seemingly cherry-picked specific study, I can think of three things: 1) the user who added it is an expert who knows that this is the scientific consensus and that the cited study is considered the "gold standard" in the field; 2) the user stumbled on this study for whatever reason, thought it was interesting, and cited it, without looking at the broader context (probably without even reading it); or 3) the user who added it has an agenda and is purposefully adding only one side of the story. I would like to hope that the right explanation is 1), but I'm afraid that most of the time it's not, and in fact Wikipedia content policies on neutrality and verifiability are based on the assumption that it's not.

Let me give you an example of how I think such a statement should be presented. These are the first few sentences from the introduction of a 2001 paper, PMID 11231774: "A possible role for aspirin in reducing risks of cataract was first suggested in studies of patients with arthritis or diabetes mellitus[1-2] and in a study of patients undergoing cataract surgery.[3] Most subsequent observational epidemiologic studies, however, have generally shown no association between self-selected aspirin use and cataract.[4-14] Results from 2 of 3 randomized trials that included an evaluation of the aspirin-cataract hypothesis also indicate no apparent benefit of aspirin in reducing risks of cataract...". OK, we don't really need to give 14+ references for this minor part of this Wikipedia article, but we could at least write the text in a similar manner and attribute it to the 2001 paper. This would also have the advantage of being 9 years more recent and based on more subsequent studies. Note that the citation would be not so much to point to the results of the 2001 study, but to the mini-review contained in the introduction of the paper (so that we don't need to cite the 14+ references ourselves). Or ideally, if there is a review we can cite that would be better.

I haven't looked at the other references in the Wikipedia article in detail, but I suspect that a similar problem will occur with some of the references in the "Experimental uses" and "Adverse effects" sections. --Itub (talk) 08:50, 8 May 2008 (UTC)

GA status update

I've checked the article and cleaned up a few minor grammatical issues. I've also checked all the references in the 'experimental uses' section, and they all appear to support what they are citing; though I rephrased the part about pancreatic cancer, since both citations came to different conclusions. Other than these issues, I don't see any other major issues with the GA criteria, and I think it can be passed. Since we've already pretty much exhausted the limit of on hold status already, a decision needs to be made. Unless there are other major objections by either Paul Gene or JFW, I will pass this article as a Good Article in 48 hours. Dr. Cash (talk) 16:28, 28 May 2008 (UTC)

Seeing no evidence of further issues from others, I have passed this article as a Good article. Congratulations! Dr. Cash (talk) 14:26, 30 May 2008 (UTC)