Talk:Statin

Muscle pain and vitamin-D

[1] suggests that statins lower levels of vitamin-D and that this is associated with muscle pain that can in most (90%) cases be alleviated with vitamin-D suppliments. We could mention this in the side effects section ? Rod57 (talk) 04:19, 9 January 2010 (UTC)

Where is the associated peer-reviewed article? Is it methodologically worth the paper it is written on? JFW | T@lk 23:38, 12 January 2010 (UTC)

New research on statins and M.S.

http://news.biocompare.com/News/NewsStory/320941/NewsStory.html 128.138.210.150 (talk) 21:24, 20 April 2010 (UTC)

BBC and British Heart Foundation

I'm not sure why this was removed, but the sources are reliable.

A 2003 study indicated that statins may increase the risk of heart failure by depleting the body of coenzyme Q10. In response, Tim Bowker of the British Heart Foundation said the benefits of taking statins for people with heart disease nevertheless outweigh the risks.

SlimVirgin ^{talk|contribs} 01:52, 1 August 2010 (UTC)

I think you need to have a look at WP:MEDRS - a guideline that we apply when selecting sources for medical articles. This paper is a clear example of what is not a reliable medical source. It is not cited in adequate secondary sources. Of course anyone can put out a press release about something they think is scary, and it is disappointing that the BBC fell for the bait, but 8 years down the line there is still no better data on the Q10 story, people with heart failure are still receiving statins, and there is even some data that Q10 supplementation in people on statins doesn't do much in terms of modifying the risk of side effects.

Statins are widely used, because they have been shown to reduce the relative risk of heart attacks by about a third. A proportion of people don't tolerate statins. But if you put 1 million people on statins, a number of people will experience side effects, and if those people start making a lot of noise in the blogosphere then it is easily assumed that statins are bad. They're not. JFW | T@lk 07:43, 1 August 2010 (UTC)

Hi JD, MEDRS is not policy, WP:V is. This article is poorly written and somewhat POV. There are lots of concerns out there about statins that seem not to be reflected here and that I think readers would appreciate. The BBC is not the blogosphere. SlimVirgin ^{talk|contribs} 00:10, 17 August 2010 (UTC)

This is not the first time Langsjoen's work has been ridiculously hyped. Q10 is not reliably associated with heart failure, and therefore changes in Q10 are not necessarily a cause of heart failure. The problem is that Langsjoen published his paper in an insignificant journal, then started blasting off press releases thereby more or less forcing the BHF to comment. This article already spends a fair amount of space on legitimate concerns such as voiced by THINCS (the entire "safety" section, with three subsections and 18 references. Therefore I don't really see why you think this article is POV, just because we are not including the Langsjoen paper. JFW | T@lk 19:33, 17 August 2010 (UTC)

The BBC is not the blogosphere - which is presumably why the BBC article treated Langsjoen's findings with an appropriate degree of skepticism, noting that "his views have been challenged by UK experts, who say that the safety of statins has been extensively checked." Langsjoen's concern in the BBC article seems to rest on a purported "epidemic" of heart failure in the US, which he attributes to statin use. The BBC article correctly points to an alternate explanation: the success of statins in preventing heart attacks and strokes has resulted in a larger population of patients with cardiovascular disease who are alive and at risk for heart failure.

If we want to base our coverage on the BBC article, then we should probably reflect its emphases, which are responsible and make clear that Langsjoen's hypothesis was treated skeptically by experts in the field. That was back in 2002; I'm not aware that any subsequent work by other investigators in the intervening 8 years has linked statins to heart failure, but I'm open to being corrected.

There have been at least 2 major randomized trials of statins in patients with heart failure: the CORONA trial (PMID 17984166, NEJM 2007) and GISSI-HF (PMID 18757089, Lancet 2008). Neither found any benefit from statins in heart failure patients (except for a subgroup of CORONA, PMID 19892235). At the same time, both randomized controlled trials found that statins were safe in heart failure, and were not associated with any increase in adverse events - which is a rather direct contradiction of Langsjoen's hypothesis. MastCell ^Talk 21:42, 17 August 2010 (UTC)

That's fine, and we can include that, but I wish people would develop material and not keep removing it because it's not liked. My point is that there are other perspectives on statins (social, economic, "how they made me feel" perspectives) that are not included in this article, as well as other scientific perspectives, and the media counts as a reliable source for them. There's no reason within policy to keep removing them. I'll write up some more using other sources and add it later. This article can't only reflect the views of the drug companies and doctors who are happy prescribing them. SlimVirgin ^{talk|contribs} 17:10, 18 August 2010 (UTC)

It's not a matter of "not liked", it's a matter of WP:WEIGHT. A bit of gentle trawling of the internet would have revealed that Langsjoen's work was hyped. Not everything that appears in the press is notable. I see no problem with including social and economic perspective (especially if supported by academic rather than interview-based subjective newspaper journalism), but I have a lot of problems with "the way they made me feel" reports. These are utterly and utterly unverifiable, and publishing them would be a tad irresponsible, wouldn't you think? JFW | T@lk 19:10, 18 August 2010 (UTC)

Hi JD, the sourcing policy is V, and it is very clear that non-academic sources are allowed in all these articles, if high quality (my bold):

Academic and peer-reviewed publications are usually the most reliable sources where available, such as in history, medicine, and science, but they are not the only reliable sources in such areas. Material from reliable non-academic sources may also be used, particularly if it appears in respected mainstream publications. Other reliable sources include university-level textbooks, books published by respected publishing houses, magazines, journals, and mainstream newspapers. Electronic media may also be used, subject to the same criteria.

If there are points of view being published about statins in reliable newspapers or similar, they belong in this article. SlimVirgin ^{talk|contribs} 21:48, 20 August 2010 (UTC)

Sure. What high-quality non-academic sources are we talking about? It would probably be less contentious to deal with specifics. MastCell ^Talk 22:03, 20 August 2010 (UTC)

There's the BBC article and some Telegraph articles about concerns, and one from Fortune magazine about how the drug was developed, which would be useful in the history section . The Fortune article also mentions the Peter Langsjoen study; JFW may not like it, but journalists do. I'll write up what I have in mind over the next few days. Have been meaning to for some time, but I keep forgetting. SlimVirgin ^{talk|contribs} 00:38, 21 August 2010 (UTC)

I'm fine with any accurate representation of up-to-date sources. The Fortune piece is OK, although I guess I'm a bit iffy on devoting so much airtime to a single article from 7 years ago. That's fine for describing the historical aspects of statin development, but if we're talking about side effects or indications, then I suspect there are more relevant and current sources. I want to make sure that we're looking for high-quality, up-to-date sources across the board (academic or lay press), rather than picking articles that make the points we think should be made. Do you think these pieces from the Telegraph represent up-to-date, encyclopedic perspectives on statins?

I'm not totally sure that quoting the line about "rumors of muscle damage, tumors, and death" is the most responsible choice. At present, we know that statins don't increase the risk of cancer in humans (if anything, perhaps the reverse). We know they don't increase the risk of death, and in fact clearly save lives in the appropriate patient populations. And we know quite a bit about statin-related muscle damage. If we mention these rumors because they were relevant to the historical development of statins, then I guess that's fine, but I wonder a bit whether we're implying something that isn't correct. MastCell ^Talk 04:30, 21 August 2010 (UTC)

The Telegraph articles represent a perspective. What we mustn't do in these articles is allow the perspective of the pharmaceutical industry to dominate. It can be included, and it may even constitute most of the material because they pay for the research. But we can't allow them to take over. I'm concerned about the number of primary sources in the article. Are they indeed primary sources, or are they overviews? SlimVirgin ^{talk|contribs} 05:10, 21 August 2010 (UTC)

• SlimVirgin, have you carefully read the Statin#Safety section? I think that since I revised it [2], it is pretty detailed, covers most of the issues raised; a while back I started it off with a large, detailed freely-accessible 2008 review from Golomb, who JFW above refers to as the "statins are bad" lady. I'm not sure the pharmaceutical perspective currently dominates the section, although I just tweaked some sketchy poorly-supported claims. Generally I agree with MastCell on citing these newspapers in this case. I agree with you generally that sometimes the scientific literature doesn't have seem to have a critical perspective, but I don't think that is the case here. Plus, it's a bit hard to imagine that newspapers could be more critical than scientists and get things right. II | (t - c) 17:13, 21 August 2010 (UTC)

Thanks, II, will take a look. SlimVirgin ^{talk|contribs} 09:52, 22 August 2010 (UTC)

I take issue with SlimVirgin's assertion that prior to the expansion, the article was dominated by "drug company views". The article already had a long section about the important side-effects. Just because it didn't contain several paragraphs about Langsjoen's unrepresentative study in a journal with an impact factor of almost 0. (PMID 16873939 is a clear example that this is just a family affair; the man seems to believe that Q10 is a magical treatment for heart failure.) I have no idea why this got hyped, other than some cleverly targeted press releases. JFW | T@lk 23:39, 21 August 2010 (UTC)

Jfd, are the papers cited here primary sources, or are they secondary reviews, and if the former, who financed the studies, and how have we decided which ones to include? My concern is that articles should be based on secondary sources only, with primary sources used only to augment them.

I'm also concerned that the article isn't aimed at our general readership, but is very technical. I don't think many of our readers would understand it, e.g. the first sentence

Statins (or HMG-CoA reductase inhibitors) are a class of drug used to lower plasma cholesterol level. They lower cholesterol by inhibiting the enzyme HMG-CoA reductase, which is the rate-limiting enzyme of the mevalonate pathway of cholesterol synthesis. Inhibition of this enzyme in the liver results in decreased cholesterol synthesis, as well as increased synthesis of LDL receptors, resulting in an increased clearance of low-density lipoprotein (LDL) from the bloodstream.

SlimVirgin ^{talk|contribs} 09:52, 22 August 2010 (UTC)

The article is in need of improvement when it comes to citing secondary sources. I don't think it is our task to investigate who financed the primary studies (this is exactly what secondary sources should be doing) unless there is a question as to which secondary source may be most reliable. As for the intro, I will have a stab at improving that. JFW | T@lk 10:49, 22 August 2010 (UTC)

You seem to have taken ownership of this article! I don't mind that if it's complete and easy to read, but this needs to be aimed more at the general reader, offering the things they might want to see, in language they can understand. I'd appreciate therefore if you wouldn't keep reverting me. Expand my edits by all means, but please don't remove them.

As for the primary source business, that is a major issue. We can't base this on primary sources, which is OR. SlimVirgin ^{talk|contribs} 10:58, 22 August 2010 (UTC)

I'm not sure if WP:OWN is the main issue. I'm one of the few editors who is prepared to deal with discussions about this article. If I see changes that I think are to the detriment of the article I will challenge them. Again, if removal is the most straightforward solution then that is exactly what I will do. JFW | T@lk 21:01, 22 August 2010 (UTC)

Primary sources

Just to continue from above, take this sentence as an example:

Statins lower LDL cholesterol (so-called "bad cholesterol") by 1.8 mmol/l. This translates to a 60% decrease in the number of cardiac events (heart attack, sudden cardiac death), and a 17% reduced risk of stroke.

This is a very definitive statement (not "a study suggests that" or "a researcher argues that"), and it's also not clear what it means: a 60 percent decrease in what sense, and for whom? But it is based only on one source, which looks like a primary source, a study in London. [3] How have we reached the decision that this study is the definitive one? SlimVirgin ^{talk|contribs} 10:01, 22 August 2010 (UTC)

This is a meta-analysis, which is a secondary source that collates the results from a large number of primary studies, identifies probable confounders and removes studies of which the results cannot be generalised. Please see WP:MEDRS for details on the reliability of meta-analyses. A particularly good source of meta-analyses is the Cochrane Collaboration; their reviews are often regarded as the "final word" on particular treatments. JFW | T@lk 10:49, 22 August 2010 (UTC)

Can you show me where it makes clear that it's a secondary source? SlimVirgin ^{talk|contribs} 10:58, 22 August 2010 (UTC)

"Design Three meta-analyses: 164 short term randomised placebo controlled trials of six statins and LDL cholesterol reduction; 58 randomised trials of cholesterol lowering by any means and IHD events; and nine cohort studies and the same 58 trials on stoke [sic]." [4] Fvasconcellos (t·c) 17:47, 22 August 2010 (UTC)

Readership

JFW, can I ask whether you accept that this article should be aimed at a general readership? SlimVirgin ^{talk|contribs} 10:38, 22 August 2010 (UTC)

I'm not sure why you feel the need to ask me this question (I'm not the only author of this article, and I have a track record of writing readable medical articles on Wikipedia). But now you ask, I think it should be aimed at the general readership but contain broad information that might be useful to both general readers and experts without compromising ono readability. JFW | T@lk 10:59, 22 August 2010 (UTC)

The reason I'm asking you is that you're the one who's reverting me, but you're right, I apologize, you may not be the primary author. Whoever wrote this was not writing for a general audience, and was not complying with the policies, so it comes across as one person's opinion based on his own pick of primary sources. I therefore hope we can work together to produce an article that an ordinary person, with no medical knowledge, will have a reasonable chance at understanding, and one based entirely on secondary sources giving an overview, from both medical and non-medical perspectives.

I suggest for a start that we try to build on each other's edits and not remove them. That may mean that I don't like some of your edits, and you don't like some of mine, but between us we should be able to produce something that offers a three-dimensional view. SlimVirgin ^{talk|contribs} 11:05, 22 August 2010 (UTC)

I can't guarantee that I won't revert an edit that I think is not benefiting the article or even making it more unreliable, such as a long list of brand names in the "history" section, and I think WP:BRD applies here. JFW | T@lk 11:19, 22 August 2010 (UTC)

If you didn't like it in the history section, you could have moved it to the lead, and written it the way you preferred, as you did eventually, rather than engaging in wholesale reverts. The point is to build, not revert. Otherwise we will have a situation every day like the one we had today, which benefits neither of us, and doesn't help the article. SlimVirgin ^{talk|contribs} 11:24, 22 August 2010 (UTC)

My view was that the brand names were already in the right place. That was what my edit summary said, and this was not a "wholesale revert" but a revert based on an argued position. Again, I feel a need to remind you of WP:AGF. It is good that we now have achieved consensus, which is something you just as easily could have done by taking it to the talk page earlier. JFW | T@lk 11:30, 22 August 2010 (UTC)

But they were not obvious, and most readers will know these drugs by their brand names, so the brand names must be very obvious. We are not a medical journal. We're writing for a general readership, and the aim is to make this article as accessible to them as possible, and to include material they might want to read.

It was wholesale reverting, JFW. You removed the brand names that I added three times in a row. Your advice to go to talk is good, but it can't only apply to me. Improve my edits, develop them, build on them, place them in context, add better sources, fix my writing, but don't just remove them. If you feel they're so awful that they must be removed entirely, please discuss on talk first. If I've made a genuine error, you'll find me very open to hearing that. SlimVirgin ^{talk|contribs} 11:57, 22 August 2010 (UTC)

You made no serious attempt to address my concerns, so this is not "wholesale reverting". You could just as easily have gone to talk. This is my last post in this thread, so please escalate to dispute resolution if you feel there are unresolved issues. JFW | T@lk 14:23, 22 August 2010 (UTC)

What kind of dispute resolution would you suggest that we're not engaged in here? SlimVirgin ^{talk|contribs} 14:27, 22 August 2010 (UTC)

Primary sourcing

I don't think I disagree with SlimVirgin that this high-profile article should be less reliant on primary research studies. WP:MEDRS gives some clear pointers as to how we can make most use of secondary sources. I think the trouble is that there is such an immense research body on statins that few sources would dare to attempt to give a full overview of the subject. Even when searching PubMed for:

"Hydroxymethylglutaryl-CoA Reductase Inhibitors"[MeSH Major Topic]

and restricting to "review" gives you 2000 citations. Clearly we must be very selective in the kind of reviews we decide to use. JFW | T@lk 11:18, 22 August 2010 (UTC)

And we must not rely entirely on medical sources. We have to give a general overview too. SlimVirgin ^{talk|contribs} 11:19, 22 August 2010 (UTC)

In particular, we need secondary sources on the whole red yeast rice content. At the moment this is based almost entirely on primary research. JFW | T@lk 11:23, 22 August 2010 (UTC)

It would be interesting to go through the sources and identify which ones are secondary sources. I would do that myself, but as I see it most of them are primary sources. I don't know whether you're defining a primary source differently. Regarding the source I highlighted above, can you say which parts of it indicate that it's a secondary source so I can see where you're coming from? SlimVirgin ^{talk|contribs} 11:26, 22 August 2010 (UTC)

The majority are indeed secondary sources, in the sense that they do not provide new data but analyse and discuss data obtained by others. They may reach novel conclusions, but as long as the data is the same we have traditionally been regarding these as secondary sources. Again, this is outlined in detail in WP:MEDRS and has the support of many active medicine and pharmacology editors. JFW | T@lk 11:33, 22 August 2010 (UTC)

If the sources contain their own conclusions, and they are studies in their own right, they're primary sources. Secondary sources give only an overview of a subject, written by people who are not involved in it. The point is that we always want Wikipedia articles to be based on sources who are not involved in the issue. Primary sources can be used to augment those sources (so long as they are used only to make only descriptive claims), but the article must be based on secondary sources. These can be medical or non-medical secondary sources. We require this to avoid original research by Wikipedians who could pick and choose primary sources that they prefer, even though secondary sources have not highlighted them. SlimVirgin ^{talk|contribs} 11:49, 22 August 2010 (UTC)

Thanks for the lecture. This issue should be addressed on WT:MEDRS, because it pervades all medical content on Wikipedia. JFW | T@lk 14:20, 22 August 2010 (UTC)

It should be addressed here, because the policies are clear (WP:NOR) and best practice is clear. All WP articles must be based on secondary sources. Otherwise we're allowing Wikipedians to decide which primary sources to pick and choose from, but that's the job of secondary sources. There's very strong consensus about this in every area of WP, and it's not something we should have to argue separately on each and every talk page. That's why we have policies! :)

If the studies you're using as sources really are secondary sources (overviews by uninvolved people), I'm fine with them. But if they're primary sources standing alone, not augmenting secondary sources, they need to be replaced, removed, or at least need to have a secondary source added that shows they're studies that are often discussed. SlimVirgin ^{talk|contribs} 14:34, 22 August 2010 (UTC)

SV, tackling the primary vs. secondary source issue is often as simple as having a look at the study design. I agree that over-reliance on primary sources is a serious issue here (perhaps more so than JFW :), but I am more concerned on increasing reliance on non-medical sources. Unfortunately, in encouraging use of reliable mainstream media sources, WP:V does not provide for the overwhelming ineptitude of these sources in describing science—something to which I am exposed on a daily basis as a medical writer—and that's where WP:MEDRS comes in. The very purpose of Wikipedia guidelines is to provide a set of best practices for application of policy. You'd be surprised at the amount of misinformation a single NYT or Guardian article can contain.

For instance: why on Earth are reports of muscle pain and muscle cramps cited to a Reuters news piece? Any general medicine or pharmacology textbook will report this, not to mention any decent review article on statin safety. Why are the actual studies mentioned in this USA Today piece not cited, with the piece itself as the | laysummary = parameter? Fvasconcellos (t·c) 18:13, 22 August 2010 (UTC)

I have no problem with the use of secondary sources that are medical, but I want to make sure we have a broad mixture of medical and non-medical opinion in the article, and I'll be adding some more about popular concerns (but I will make clear that that's what they are). My main concern here, though, is that the article seems almost entirely based on primary sources. I don't know how to tackle that. SlimVirgin ^{talk|contribs} 18:22, 22 August 2010 (UTC)

Are the current "popular concerns" not sufficient? I agree that we must replace what we can with appropriate secondary sources. JFW | T@lk 20:03, 22 August 2010 (UTC)

No, there are popular concerns that I would like to add. SlimVirgin ^{talk|contribs} 21:14, 22 August 2010 (UTC)

I'd be most grateful if you could inform us what the nature of those "popular concerns" might be, so I can do the required background reading. JFW | T@lk 21:28, 22 August 2010 (UTC)

In fact, I think ImperfectlyInformed is the person who added a large number of primary studies, particularly in the "adverse events" section, and he has argued (e.g. on WT:MEDRS) that parts of these reports may still be reliable enough to be regarded as secondary content. I don't personally carry the blame for all that content, you know? JFW | T@lk 21:05, 22 August 2010 (UTC)

I don't know who added the primary sources - I'm guessing anonymous IPs - but I did not, although I've tweaked a few. As far as I can remember I added the Golomb review, the Silva et al meta-analysis, and a clearly secondary Doc News. However, as you said, I think that "primary" studies are often reliable. To clarify what I've argued, I think that research articles are often mostly secondary in their introduction/discussion sections - and I think you (JFW) admitted that this was probably correct a while back. II | (t - c) 01:17, 24 August 2010 (UTC)

I may have conceded that this content of primary studies may have an element of reliability (see WT:MEDRS, where this was discussed) but I also said that most of the time such content is quite selective in only pulling in the content needed to provide context for its own line of research, rather than give a broad overview as a good review would do. Therefore, this should still be avoided. JFW | T@lk 05:56, 24 August 2010 (UTC)

Interactive pathway map

How useful is this map? It takes up quite a bit of space, and overlaps considerably with the smaller (if static) pathway shown one section up. Fvasconcellos (t·c) 17:52, 22 August 2010 (UTC)

I can't see what use it would be to a general reader, so I'd be fine with removing it. SlimVirgin ^{talk|contribs} 18:24, 22 August 2010 (UTC)

I agree, but that's not exactly the reason I had in mind. This article doesn't cater enough to the general reader at the moment, but no article should ever be useful to the lay readership alone. We may not like to admit it, but Wikipedia is used as a source of scholarly (gasp!) information by medical professionals, scientists and their in-training counterparts (double gasp! triple gasp!) all the time. Fvasconcellos (t·c) 18:35, 22 August 2010 (UTC)

The article as it stands is of almost no use to the readership it's aimed at, that's the problem. I can't imagine why any medical professional would choose to use Wikipedia, but even if they do, they have easy access to better source material. The general reader doesn't have that easy access, so it would make sense to aim the article at them. SlimVirgin ^{talk|contribs} 18:46, 22 August 2010 (UTC)

The "general readership" includes medical professionals quickly wanting to double-check some facts. I would not exclude them just because some of the current content is a bit hard to understand. JFW | T@lk 21:03, 22 August 2010 (UTC)

I don't want to exclude anyone, but the way it's currently written it excludes everyone else, that's the thing. SlimVirgin ^{talk|contribs} 21:10, 22 August 2010 (UTC)

I know that this is your opinion, because I rephrased it for you. Do you or do you not acknowledge that even specialised groups of readers might need to be able to derive information from the article? (I don't actually have an opinion as to whether the diagram should stay. It probably shouldn't, in fact.) JFW | T@lk 21:28, 22 August 2010 (UTC)

"Members" table

Would look better as prose, with some actual content. Structural formulae should probably be removed—they add little in this context and the different orientations and styles make it difficult to spot the structural similarities. Brand names must be cited (Martindale probably the best source); combinations should get their own section. Fvasconcellos (t·c) 18:49, 22 August 2010 (UTC)

I have no copy of Martindale's at home. Would you be prepared to take care of this, FV? JFW | T@lk 20:05, 22 August 2010 (UTC)

Sure. Give me a couple of days and I'll whip something up and leave it at the Talk page for consideration.

SV, how do you feel about turning this section into prose? I don't think the current table is at all useful for lay readers. Fvasconcellos (t·c) 20:29, 22 August 2010 (UTC)

That would be extremely helpful if you're willing to take the time. SlimVirgin ^{talk|contribs} 21:09, 22 August 2010 (UTC)

I will try to tackle this over the next few days. Perhaps it'll give me the energy and inclination to address more serious sourcing issues as well. Fvasconcellos (t·c) 23:15, 22 August 2010 (UTC)

Source request

Could we have a page number here, please? "Still, the Coronary Primary Prevention Trial of 1984 had demonstrated that cholesterol lowering could significantly reduce the risk of heart attacks and angina."

SlimVirgin ^{talk|contribs} 18:51, 22 August 2010 (UTC)

[5] Fvasconcellos (t·c) 18:55, 22 August 2010 (UTC)

Many thanks. SlimVirgin ^{talk|contribs} 19:22, 22 August 2010 (UTC)

You're welcome. A few points regarding this edit, which I will respectfully refrain from reverting:

• As JFW, I feel we should note that Simons is a journalist (and Steinberg a doctor, etc.). Just because he's writing in FORTUNE doesn't automatically mean he is a journalist :)
• The 1984 study did demonstrate an association between lower cholesterol and lower risk.
• You removed the access date from the Simons source. Was it just an accident, or had you accessed the article at an earlier date?

—Fvasconcellos (t·c) 20:24, 22 August 2010 (UTC)

We don't need to add access dates for articles, unless the website is one we think is unstable. I don't mind if you restore that he's a journalist if you know that he is, and that the study did demonstrate it, if the source says that clearly, though I wonder why cardiologists weren't listening if the evidence was as clear as you're saying. SlimVirgin ^{talk|contribs} 21:13, 22 August 2010 (UTC)

I fixed the Simons template code that got broken as a result of removing the access date. I did change the claim that the CPPT proved an association, because statistical significance is not a matter of personal opinion (unless methodological deficiencies are subsequently demonstrated). I continue to feel that we should make it very clear that Simons is a journalist, while Steinberg is a cholesterol researcher.

Once more, I don't think there should be a paragraph break between the summary cited to Steinberg and the claims made by Simons. These are chronologically and topically connected. I would request that no paragraph break is inserted without a clear justification on this talk page. JFW | T@lk 20:56, 22 August 2010 (UTC)

It's a different thought, so a different paragraph. SlimVirgin ^{talk|contribs} 21:13, 22 August 2010 (UTC)

It's not. One follows the other conceptually. JFW | T@lk 21:28, 22 August 2010 (UTC)

3RR

JFW, you've violated 3RR today, and I'd really appreciate it if you would stop reverting my edits. A revert is any edit that undoes someone else's work, whether it's the same or different material each time, whether it's a wholesale or partial revert. Once someone has asked you to stop reverting, and points out that you've done it three times, it's provocative to move onto new material and keep on doing it, so I'm asking you again please to stop. SlimVirgin ^{talk|contribs} 21:29, 22 August 2010 (UTC)

Could you indicate where I violated 3RR? And you didn't ask me to stop reverting until afterwards. And I see pretty much no evidence that you address the legitimate concerns I raise in edit summaries. There's probably no point in redoing an edit after another has undone it; the only solution is to address the concerns raised about the edit in question. JFW | T@lk 21:33, 22 August 2010 (UTC)

It's unfortunate that you're making me post the reverts.

• Version reverted to (without brand names in history or lead): [6]

• 1st revert 23:44 Aug 21, removed brand names that I'd just added
• 2nd revert 10:36 Aug 22, removed brand names
• 3rd revert 10:50 Aug 22, removed brand names

• Version reverted to: (JFW joined up two paragraphs making the Simon material less obvious) [7]

• 4th revert 19:00 Aug 22, re-joined the paragraph
• 5th revert 20:52 Aug 22, re-joined the paragraph

There were other edits that may have amounted to reverts, but these were the obvious ones. Again, I suggest that we try to develop each other's edits, while leaving them intact as far as possible. SlimVirgin ^{talk|contribs} 21:51, 22 August 2010 (UTC)

Why thanks for taking the trouble; I don't think an allegation of 3RR should be made without diffs. I have been doing a lot of article work over the last few months and I tend not to get embroiled into this kind of things (the only thing I normally revert is obvious vandalism or other unconstructive edits). My apologies for my part in this, but I hope you will concede that we missed opportunities to discuss edits rather than simply redoing them (e.g. readding the brand names). I will attempt to develop your edits, but I request that you take my concerns seriously (rather than readding challenged content) and perhaps try to be more open about those "popular opinions" you intend to add to this article. JFW | T@lk 22:03, 22 August 2010 (UTC)

Thank you for the apology. My concern about the article is that it seems POV. It looks as though it's trying to defend statins, rather than just neutrally publishing the medical and non-medical reliable sources who discuss them. A related concern is that most of the article's readership won't understand it. So perhaps you could focus on the material from medical sources, and allow me to focus on introducing material from non-medical reliable sources. So long as we both use in-text attribution (this researcher says X, this study concluded Y, this journalist argues Z), and try to use very clear language, then the readers can judge for themselves. SlimVirgin ^{talk|contribs} 22:07, 22 August 2010 (UTC)

I think the article appears POV because it has received a lot of attention in the past from people (see previous talkpage posts and Talk:Atorvastatin) bent on disputing the scientific basis of statin use. What you are left with is attempts by other editors to balance this, and an article that has the smell of apologetics to it. I agree this is not a desirable situation, but the answer is not to add further material to "restore the balance" but closely examine whether it actually needs to sound this way.

As I said a while ago, there is a very strong evidence base for the use of statins in people with previous heart disease, and a convincing but slightly weaker one for "primary prevention". I think one would be justified is saying that this is the current position of practically every professional society with an interest in lipids. I think we need more content around this, because these guidelines actually drive what GPs and physicians do in their consultations with patients (e.g. NICE clinical guidelines 48 and 67, the parallel American and Canadian guideline documents). The "dissenters" have various arguments, ranging from concerns about data on primary prevention in women (which is limited) to a complete denial of the lipid hypothesis. There is ample source material, but it is going to be a fair amount of hard work.

With regards to "non-medical reliable sources", I think a very high degree of selectivity is required. As Fvasconcellos indicated above, journalists can be very bad at grasping and presenting scientific concepts, and horrible misunderstandings often arise out of bad reporting—which then create very uncomfortable moments in consulting rooms. JFW | T@lk 22:20, 22 August 2010 (UTC)

This is not a medical journal or any other kind of scholarly journal. Medical material is of course appropriate here, but so is other material, so long as the publications are reliable. Please don't try to argue about that, JFW, because it's not appropriate here. The policies are clear and editors shouldn't have to argue them anew on each talk page. I don't know whether you have enjoyed our editing today. I haven't. But it will be like this every day unless we cooperate, so please allow me to edit the article without reverting. I will likewise agree to leave material you add intact. SlimVirgin ^{talk|contribs} 22:30, 22 August 2010 (UTC)

SV, which are the "reliable non-medical sources that discuss them"? You seem to be under the impression that sources cannot be free of influence from the pharmaceutical industry unless they are published in the popular press. There is a wealth of high-quality information on concerns over statin safety and efficacy in the scientific literature, and I still don't see any compelling reason to use popular sources at all except in very restricted contexts and with proper qualification.

"Medical material is of course appropriate here, but so is other material, so long as the publications are reliable." No, it isn't. Not necessarily, and that is a fundamental flaw in WP:V that should be discussed elsewhere. Reliable MSM sources does not equal adequate information. Just as reputable medical journals publish junk science every now and then and a lot (and I mean a lot) of industry-supported "research" and should thus not be used indiscriminately, the cluelessness of the mainstream media when reporting study results, to mention just one of many flaws, is a source of constant frustration for people who are actually familiar with the science and capable of interpreting it, for instance. We should not perpetuate this. Fvasconcellos (t·c) 22:35, 22 August 2010 (UTC)

It may be a fundamental flaw of V, but that's the place to argue it, because that's the policy. What I feel I'm seeing here is an effort to keep out criticism. Now it could be, as JFW says, that an apologia has developed because of the tensions of editing. But even so, readers are coming here and are finding an article that they can't understand and that seems to be saying everything is fine, and everything is understood, and study results are plain facts. Wikipedians can't set themselves up as judges to this extent.

I do think an effort has to be made on this page, using whatever sources are available, to present a view that hasn't been financed by the pharmaceutical industry, which is making unimaginable profits out of statins. That needs to be borne strongly in mind when choosing our sources. SlimVirgin ^{talk|contribs} 22:42, 22 August 2010 (UTC)

You must bear in mind that the current incarnation of the article is based on subpar sourcing, as you yourself have mentioned repeatedly today. "Readers are coming here and are finding an article that they can't understand and that seems to be saying everything is fine"—I fail to see where this article says that everything is fine. Does it place enough emphasis on adverse effects, which are the main source of criticism? No, it doesn't—that must be improved. Does it use the best, most recent, reliable evidence on efficacy of statins (or lack thereof) as primary prevention? Yes, it does. Does it mention criticism of the lipid hypothesis in a fair, balanced way that respects WP:WEIGHT? Yes, it does, and that's a major accomplishment whan one considers the drama this article has seen in the past.

Yes, this article must be expanded and improved, extensively, particularly in terms of sourcing. But I see no evidence of any deliberate, concerted effort to keep out criticism, and I don't see any flagrant misinformation or bias that must be "balanced out" with countervailing opinions. Fvasconcellos (t·c) 22:53, 22 August 2010 (UTC)

I'm not aware of the history of the article, or the reasons for the drama, but was it related to the article not containing enough criticism? If so, and if I'm still getting that impression as an editor not previously involved, perhaps there's something in it. But regardless, these discussions aren't helping to improve the article. I want to add a popular concerns section, and I'm willing to call it anything you want—Silly Concerns That No One Should Listen To—but I want to make sure they're included. And again I ask in general that others be allowed to edit this article without getting this sense of OWN. SlimVirgin ^{talk|contribs} 22:57, 22 August 2010 (UTC)

OWN? Erm... OK. I don't think I'd ever edited this article before today :) I have nothing against including a broader discussion of "popular" concerns, as you call them. In fact, I think it would work well as an expansion of the Concerns section. But the impression I'm getting is one of profound distrust of any medical/scientific sources. Believe me, I've seen enough junk science (and sadly helped translate some of it) to develop a healthy distrust myself. But my distrust of MSM ability to adequately report on science—an evidence-based distrust, if you will, pun fully intended—is far, far greater, and well-founded to boot.

I hate to draw another editor into this, but this is reminding me of a brief discussion I had over at Talk:Reye's syndrome a few days ago, over a concern that sadly has yet to be addressed. This is Wikipedia, people. Please don't attribute to malice (bias, whitewashing, whatever) flaws that are most likely due to deficient article development and misinformed edits by well-meaning editors. Fvasconcellos (t·c) 23:12, 22 August 2010 (UTC)c) 23:10, 22 August 2010 (UTC)

I wasn't referring to you, F. It's not that I don't trust scientific studies, just that I don't want to rely on them, especially if they're financed by the people who are selling the drugs, and particular not if they're primary sources being chosen by Wikipedians. We wouldn't rely only on restaurant reviews that were written by the restaurants (or at all, in fact). That's the only principle I'm arguing for. SlimVirgin ^{talk|contribs} 23:35, 22 August 2010 (UTC)

All I did today was add the brand names in a more prominent place, and add two uncontroversial paras to the history section. And yet I've had to post about it something like 35 times. Something not right there. When people are asked to justify every tiny thought, and explain every relevant policy, even for edits they've not yet made, it signals OWN. SlimVirgin ^{talk|contribs} 23:42, 22 August 2010 (UTC)

• For the benefit of someone with a short attention span who values his weekend time, could anyone summarize the actual content under dispute in a few words? Instead of arguing in the abstract about "popular" vs. "scientific" concerns, it might be easiest to collect high-quality, current, relevant sources (both academic and non-academic) here, so we can have a discussion grounded on specifics. That might make it easier to find common ground. MastCell ^Talk 23:38, 22 August 2010 (UTC)

If you'd be willing to help compile some non-academic secondary sources that would be very valuable, MC. SlimVirgin ^{talk|contribs} 00:07, 23 August 2010 (UTC)

I'd be happy to work on it. I still can't say I understand the current content dispute, though - since I last looked in on Friday, there have been about a dozen new talk page threads and 20+ edits to the article. If anyone feels like distilling the subject of contention, I'd be grateful, although I can also get off my lazy ass and actually read the talk-page threads in full. :) MastCell ^Talk 04:17, 23 August 2010 (UTC)

I think it is not OWN at work here but a genuine desire to keep the article neutral. I have already conceded that the current content has problems with primary sources and readability. I think medical and scientific articles should be based on scientific sources whenever possible, as they are the prerequisite for non-scholarly writing on the subject. Only if such sources do not exist is it appropriate to resort to non-scholarly sources. When that happens, we need to be very clear about the notability and quality of those sources. I think the articles about the Langsjoen research were a clear example of hype, much like dichloroacetic acid was hyped relentlessly as a cancer treatment before genuine research was being completed.

I have previously alluded to the fact that my editing behaviour (and that of other medicine contributors) is guided by WP:MEDRS; it may not be policy but it is a guideline that has received widespread support and makes very clear how we identify the most reliable sources. Many references in this article don't comply with MEDRS, which is something that we need to address as a matter of urgency. JFW | T@lk 19:26, 23 August 2010 (UTC)

We have to stick to the policies, J. The urgent thing is to remove the primary sources. Did you see my query about one of them above? You said it was secondary, but it has some strong claims attached to it, so could you point out how you're judging it to be a secondary source? Here it is again:

On average, statins can lower LDL cholesterol (so-called "bad cholesterol") by 1.8 mmol/l (70 mg/dl), which translates to a 60% decrease in the number of cardiac events (heart attack, sudden cardiac death) and a 17% reduced risk of stroke after long-term treatment.

SlimVirgin ^{talk|contribs} 19:36, 23 August 2010 (UTC)

Meta-analyses review and attempt to synthesize existing primary literature. They are sometimes called "systematic" reviews (as opposed to "narrative" reviews, which generally reflect an expert's summary of a subject but lack a formal or rigorous methodology. Secondary sources are generally those which review, analyze, and synthesize existing primary sources. In that respect, the article in question is a secondary source - it represents a formal analysis and synthesis of the existing literature on statins, LDL cholesterol, and clinical outcomes. MastCell ^Talk 20:11, 23 August 2010 (UTC)

Okay, thank you, MC. In this particular case, is it correct to say that the authors would not have been involved in any of the studies they reviewed? SlimVirgin ^{talk|contribs} 20:16, 23 August 2010 (UTC)

In general, that's the case. In this particular instance, I don't know for sure without reviewing all of the authors' publications. I'm not sure that it matters whether the authors of the meta-analysis were involved in one or two of the 222 trials they included in their analysis. The article details the methods they used to identify and select the studies for inclusion in the review, which should address any concern that they've preferentially focused on their own results. MastCell ^Talk 20:37, 23 August 2010 (UTC)

Thank you, that's very helpful. SlimVirgin ^{talk|contribs} 20:46, 23 August 2010 (UTC)

Writing

I keep trying to fix this writing, but JFW reverts me. Is there a need to write it this way? "Writing in Fortune magazine, journalist John Simons writes that ..." SlimVirgin ^{talk|contribs} 23:58, 22 August 2010 (UTC)

How about "In a 2003 feature piece on Lipitor published in Fortune magazine, journalist John Simons writes that...". I'll always favor more background information then less, though, so you (and other editors) might prefer a shorter introduction. Here's his bio, by the way. Fvasconcellos (t·c) 00:17, 23 August 2010 (UTC)

The best thing would simply be: "Journalist John Simons writes in Fortune magazine that ..." but I think JFW reverted that twice, or at least once. There's something badly wrong on this page where people have to get permission to fix grammar and repetition, and still get reverted, then have to negotiate over straightforward English. I'm not sure I've ever encountered this. SlimVirgin ^{talk|contribs} 00:23, 23 August 2010 (UTC)

And you haven't; that was never the issue. The issue was adding or omitting "journalist" and splitting or not splitting the paragraph (latest revert). Fvasconcellos (t·c) 00:32, 23 August 2010 (UTC)

But somewhere along the way the writing kept getting messed up too. Please don't make me find the diffs again. SlimVirgin ^{talk|contribs} 00:35, 23 August 2010 (UTC)

Of course it did; that's why I'm no fan of reverting. Why don't we just move on from this?

My issue is with not mentioning that Simons is a journalist and Steinberg a clinical researcher. It is a content issue, nothing more, and shouldn't be so contentious (ba-dum ching). I'm signing off for tonight, and for what it's worth, "Journalist John Simons writes in Fortune magazine that ..." is fine by me. There are far more serious issues in need of attention here. Fvasconcellos (t·c) 00:45, 23 August 2010 (UTC)

I'm not that interested in the actual publication where Simons writes. That is in the reference. What is much more important is that we know that he is a journalist who makes a rather controversial claim, namely that a drug company disease mongered hypercholesterolaemia to make a buck. That is why I want to make it clear that he's a journalist. I'd be much happier if we had further sources that could corrobate this claim, because it is essentially contradicted by Steinberg (hence my reluctance for a paragraph break) who makes it clear that NIH had cholesterol on the agenda before Merck developed the first statin. I thought I was reasonably clear about this. JFW | T@lk 19:26, 23 August 2010 (UTC)

But Steinberg also makes clear that doctors weren't listening, so what he says doesn't contradict what Simon says. SlimVirgin ^{talk|contribs} 19:32, 23 August 2010 (UTC)

It does to an extent. The version you made, without the Steinberg quote, suggests that cholesterol as a health problem was discovered by Merck. Quod non. JFW | T@lk 19:36, 23 August 2010 (UTC)

It doesn't say that. If you read Steinberg and Simon carefully, what they say is perfectly consistent. I'd also appreciate if you wouldn't judge a writer negatively because he's a journalist. Journalists often have a better overview, and he specializes in this area. SlimVirgin ^{talk|contribs} 19:38, 23 August 2010 (UTC)

He makes a controversial claim. That's why I am especially interested who makes that claim. I would have been equally interested if he was a dustman or a Nobel Prize winning professor. And journalists are also capable of oversimplifying things because they (or their editors) want to sell newspapers, specialism or not. JFW | T@lk 19:43, 23 August 2010 (UTC)

All sources are capable of that, to sell newspapers and to sell drugs, so it's pointless to speculate on motives or skill. And you're not taking into account that what he said and what Steinberg said are the same. Doctors weren't listening. The public wasn't listening. Cardiologists were skeptical. Therefore they had to be persuaded. If you think it's controversial, do you have a source that says otherwise? SlimVirgin ^{talk|contribs} 19:46, 23 August 2010 (UTC)

To repeat what I just said, your initial edit gave the impression that cholesterol was invented by Merck. That is not so. Which I proved with Steinberg. If there is no contradiction then there is an even stronger argument to let the one argument follow the other in the same paragraph. As for stating that Simons is a journalist, this is exactly why I made sure that Steinberg is quoted as a cholesterol researcher—no immediate bias is implied but the reader can do their math. Finally, I think a claim that a drug company has discovered a disease entity to make money (and not made anywhere else AFAIK) casts aspersions and implies unethical behaviour and is therefore intrinsically controversial. JFW | T@lk 19:55, 23 August 2010 (UTC)

You are arguing that drug companies don't do what they do for money; and that they don't invent disease? :)

Look, J, you're giving the impression of OWNership, and somewhat obsessively so, where everything has to be checked and questioned, and tweaked and fixed and argued over, no matter how trivial. Please allow me to edit. I'm an experienced editor. I won't add nonsense, and if I do someone else will point it out. I may add things you don't like, but that doesn't mean they have to be removed.

If you think something is wrong or controversial, by all means add another source that shows that or post the source here. In the meantime, could you take a look at the source I posted above and clarify how it's a secondary source? SlimVirgin ^{talk|contribs} 20:04, 23 August 2010 (UTC)

With regards to WP:OWN, make an issue of it if you wish. You have been displaying preconceptions about a very well established medical treatment. I will continue to check, question, tweak and fix whatever is necessary. And I have already agreed that I will not outright remove your additions, but I'm sure that does not stop me from rephrasing things (e.g. the distictions between anonymous "researchers" vs the evidence provided by randomised trials with 60,000 patients).

I'm not sure which source you are troubled about. If it is Law et al 2003 (doi:10.1136/bmj.326.7404.1423), please see WP:MEDRS. It might not be your kind of secondary source, but it does not invent anything that had not already been concluded by the controlled trials that it is based on. JFW | T@lk 20:15, 23 August 2010 (UTC)

Look, you can continue with the arrogant responses, or you can try to work with me. The latter will be nicer for both of us. SlimVirgin ^{talk|contribs} 20:18, 23 August 2010 (UTC)

WP:NPA. I'm trying hard to work with you, but I get the ongoing impression that you are deliberately misunderstanding what I am trying to say. It would be good if you could make an effort to remove that impression. JFW | T@lk 20:24, 23 August 2010 (UTC)

• Ye Gods. What is this dispute about? Is it about quoting the Fortune article to the effect that Merck had to promote a niche for statins? That can probably be solved with additional high-quality secondary sources. Clearly, it took a while for statins to "catch on". I suppose the Fortune article focuses on Merck's role in countering that sluggishness? This article (New York Times, 1999) directly addresses the slow adoption of statins, suggesting that responsible factors included a lack of concern about heart disease (vs., say, cancer), financial incentives for physicians, the cost of statins, the conservative nature of most physicians, and safety concerns ("even though the latest data suggest they are extremely safe and produce few side effects.") If our goal is to address the slow uptake of statins, then we can probably address this dispute by incorporating these sorts of sources? MastCell ^Talk 20:28, 23 August 2010 (UTC)

Book titles

JFW, is there a reason you want book subtitles to be lower case? I fixed one of them but you reverted. [8] The publishers don't write it that way. [9] SlimVirgin ^{talk|contribs} 20:24, 23 August 2010 (UTC)

I don't think there's a policy on capitalising book titles or not. Gramatically there is no reason to. At any rate, I'm unsure why you removed the citation template; do you actually feel that adding the citation template back should count as a revert? JFW | T@lk 20:29, 23 August 2010 (UTC)

In English, book titles are usually capitalized. Some publishers choose to do otherwise, but this one has not. You didn't just add the template back, you added back another error. That's the problem with reverting that I tried to point out yesterday. SlimVirgin ^{talk|contribs} 20:48, 23 August 2010 (UTC)

But I don't regard this as an error (you certainly did not point this out in your edit summary when you changed it) and I dispute that this should be regarded as a revert. I also think you are wasting an awful lot of time ranting at me on threads like this. JFW | T@lk 20:56, 23 August 2010 (UTC)

Then stop reverting and I won't need to keep posting on talk about trivia! SlimVirgin ^{talk|contribs} 21:04, 23 August 2010 (UTC)

Can you really not take my point that I was not intending to revert but to fix the removal of the citation template? I really wish you would, because I think this would stop the proliferation of such threads, rather that me doing one thing or another. JFW | T@lk 21:06, 23 August 2010 (UTC)

Citation templates

May I ask why citation templates are being removed? Per WP:CITE:

Because templates can be contentious, editors should not change an article with a distinctive citation format to another without gaining consensus. Where no agreement can be reached, defer to the style used by the first major contributor.

Slim, I realize you may be frustrated by JFW's reversions, but it has to work both ways. Insisting on changes that suit your personal preferences is unnecessary and not constructive. Fvasconcellos (t·c) 20:59, 23 August 2010 (UTC)

I want to fix the refs, which are oddly written (all book titles in lower case, for example). When I tried to fix something yesterday, JFW complained that I had messed up a template. So today I rewrote it without a template, but he still reverted to the errors. F, the point is that all my edits here are being reverted or discussed in such detail and with such hostility that when someone like Mastcell asks what the dispute is about, I can't reply because I don't know myself. But I do know OWN when I see it, and I speak as someone who doesn't mind OWN if it's producing a great article, but this is a poor article. Perhaps one of you can ask him to stop reverting, rather than focusing on me. SlimVirgin ^{talk|contribs} 21:03, 23 August 2010 (UTC)

I'd gladly do so, but this bafflingly minor issue is one where you are going against (1) established practice and (2) editing guidelines. This article has long used citation templates; editors coming to it now should follow this practice unless consensus arises to do otherwise.

Again: why is this (and grammar, and writing, and who's a journalist and who isn't) still an issue when this article has major flaws that need work?

Did you get a chance to read MastCell's explanation of meta-analyses and systematic reviews above? That should help you clear up your "primary source" confusion. Fvasconcellos (t·c) 21:07, 23 August 2010 (UTC)

Sorry, I don't follow your point, F. The issue was not templates, but writing book titles in lower case, which I haven't seen before where the publisher writes it differently. And the issue of the journalist wasn't an issue of mine, so you'll have to ask JFW about it; I was puzzled too about the reverting over that.

I agree about the major flaws. What do you see them as? SlimVirgin ^{talk|contribs} 21:12, 23 August 2010 (UTC)

For what it's worth, my issue is with the Adverse effects section. It is unclear, repetitive, and contradictory; referencing is dismal. (I was surprised to not see any mention of the National Lipid Association Statin Safety Task Force reports from 2006, for instance, nor any medical textbooks, nor any recent review of side effects apart from Golomb.) Fvasconcellos (t·c) 21:33, 23 August 2010 (UTC)

• (Might as well put this here) Can I propose a moratorium on new talk page threads? They're being added much more quickly than I (at least) can digest and respond to them. We're not operating under a deadline, and while we all agree that the article needs improvement, I don't see anything that is so dangerously inaccurate that it needs to be addressed This Very Minute. Can we work on one or two concrete issues and try to see them through to some sort of resolution before moving on? I personally would find that much more constructive. MastCell ^Talk 21:15, 23 August 2010 (UTC)

Source query

Are the following primary sources?

Research continues into other areas where statins also appear to have a favorable effect, including dementia, lung cancer, nuclear cataracts, and hypertension.

SlimVirgin ^{talk|contribs} 21:08, 23 August 2010 (UTC)

Yes, every single one of them. Some of them could be replaced by secondary sources, but to identify these would take time. JFW | T@lk 21:13, 23 August 2010 (UTC)

Agreed that these are primary sources. They are used to identify areas of active research. In that context, I don't see the problem - they're not used to claim a benefit, only to show that people are interested in exploring these questions. We could change "appear to have a favorable effect" to "may have a favorable effect" if less definitive langauge is preferred.

In terms of secondary sources, for statins and dementia see New York Times 2005 and 2007. For statins and cancer, see New York Times 2004. I'm not suggesting that these are the best available secondary sources; only that such sources exist. MastCell ^Talk 21:23, 23 August 2010 (UTC)

I would like to request that we don't allow anything in this article to rely on primary sources alone. For every study showing A we could probably find one that shows not-A, so it's pointless. We need to find uninvolved sources who have decided these things are worth mentioning. SlimVirgin ^{talk|contribs} 21:55, 23 August 2010 (UTC)

I think this usage is well within policy. WP:PSTS states: A primary source may only be used on Wikipedia to make straightforward, descriptive statements that any educated person, with access to the source but without specialist knowledge, will be able to verify are supported by the source. We use these primary sources only to indicate that research is being performed on these questions. That is a straightforward, descriptive statement that any educated person can verify without specialist knowledge. We're not citing these studies as proof that statins are beneficial in dementia or cataract prevention, which would certainly require a secondary source. In any case, as I noted above, secondary sources do exist - I'm just not sure they're necessary for the statements we're making in this particular passage. MastCell ^Talk 22:10, 23 August 2010 (UTC)

As long as we can agree about what constitutes a "secondary source", I completely agree with you. As I indicated above, there are thousands of publications about this subject, many of which will qualify as potential secondary sources. JFW | T@lk 22:12, 23 August 2010 (UTC)

(ec to Mascell) But we are also supposed to be able to show that the material the primary sources are used to support is notable; otherwise we could start quoting from the Bible using descriptive statements that any educated person etc etc. We need secondary sources to steer the course of the article; that's standard practice. And we describe the issue the way the secondary source does, using the primary source only to augment if needed. SlimVirgin ^{talk|contribs} 22:15, 23 August 2010 (UTC)

As of secondary sources try PMID 17325332. Looking for data of young people with minimal cancer incidence is not nearly as interesting. Richiez (talk) 22:25, 23 August 2010 (UTC)

Yes, this is something that was suggested by the PROSPER trial. However, the difference is not statistically significant (OR 0.99-1.13). Many other trials, such as the Heart Protection Study, and some large meta-analyses, have confirmed that there is no significant association between statins and cancer. Indeed, some cancer types are more uncommon in people taking statins. What exactly are you suggesting? JFW | T@lk 22:46, 23 August 2010 (UTC)

(edit conflict) @Richiez: There are at least 3 meta-analyses concluding that statins don't increase the risk of cancer (PMID 16214597, PMID 16391219, PMID 17131313). The one you mention is very circumspect about the limitations of their analysis (and the accompanying editorial is entitled "Pravastatin and cancer: an unproven association", which speaks for itself).

@SlimVirgin: it seems unarguable to me that one sentence outlining current and ongoing research is appropriate for articles about drugs. It seems of obvious and common-sense interest to the general reader; we're hardly belaboring the point or axe-grinding with a single, neutral sentence; and in any case (as noted) these areas have been discussed by secondary sources. MastCell ^Talk 22:47, 23 August 2010 (UTC)

It has been suggested since the Scandinavian Simvastatin trial that statins raise the cancer risk in well known subgroups: women and the elderly population. None of PMID 16214597, PMID 16391219, PMID 17131313 does address this issues - they reuse data from trials that were carefully designed so that these particular groups were underrepresented. Richiez (talk) 23:02, 23 August 2010 (UTC)

Hmmm. I don't think the trials were "carefully designed" to underrepresent people at risk for cancer, but perhaps can point me to some substantiation of that claim? In any case, as the editorial I linked makes clear, there are methodological reasons to be a bit skeptical of the findings of PMID 17325332. It's a good study - if it weren't, CMAJ likely would not have published it - but I don't think we should accept it uncritically or grant it special status simply because it potentially reflects negatively on statins. We should be asking how experts have synthesized the evidence, because this is Wikipedia and our goal is to accurately convey the current state of human knowledge.

I'm not aware of any body of expert opinion which currently holds that statins increase cancer risk. To the contrary, the National Cancer Institute regards it as possible, although not proven, that statins may protect against specific cancers ([10], secondary source). MastCell ^Talk 23:27, 23 August 2010 (UTC)

• I personally have no major problem with the above sentence. However, I think part of the concern here is that there could be some double treatment when it comes to primary sources and speculative research directions. When the primary sources are pointing to benefits, they are OK. When they're pointing to harms, there's a tendency for an emphatic argument that such studies are not OK, which is a bit counterintuitive considering the do no harm precept, although I realize that such a precept could be viewed both for and against statins given the opportunity cost of not using statins - heart attacks. The idea that doctors might have a bias towards viewing statins favorably is supported by the fact that 50% of doctors in a survey denied even muscle adverse effects. The above sentence points to a beneficial impact on dementia - yet 2 RCTs have found negative cognitive side-effects while 2 have found no effect [11]; the BMC paper above is a secondary source in its report on PMID 15883262, a very small RCT which found beneficial cognitive effects, but overall points to a "mixed picture" - basically a secondary conclusion which puts the research into context. As far as cancer, PMID 17662392, currently reflected in the article, also points to increased cancer risk. II | (t - c) 01:17, 24 August 2010 (UTC)

We do need to discuss cancer risk in the article, and we will need to cite either a single review that pulls in all the meta-analyses (the "borderline positive" and the negative ones) to create balance. This shouldn't be difficult. I don't think there is any evidence in the literature of Richiez's allegations (that statins are not being researched in people at risk of cancer). JFW | T@lk 05:56, 24 August 2010 (UTC)

I have not done a literature search on cancer risk specifically but I did an extensive search on statins for geriatric and female patients some time ago. The conclusion for me was quite clear, when I have more time I will dig out my literature and update the article. I also have an issue with the dozens of "statins are also investigated for possible use in XXXX" snippets throughout the article. Near everyone can claim he is investigating the possible positive effect on something but in retrospect most of these claims never materialised, in some cases the contrary effect was proven yet the text snippets suggest there may be some effect. Richiez (talk) 08:04, 24 August 2010 (UTC)

Statins have been researched in all populations, but there is less data in the use in primary prevention in women (who may have a different form of vascular disease too, which is more diffuse and less stenotic). There is also no long-term data in children (yes, children with familial hypercholesterolemia require statins). If we can find a secondary source that confirms this, I see no problem with mentioning this in the article. I think your claim that people at risk of cancer are excluded from statin trials in not correct, and easily refuted. I do share your concern about "statins have been investigated in condition X", because these studies are relatively easy to do and a lot of them have methodological flaws; a secondary source listing all investigational uses for statins would be preferrable. JFW | T@lk 19:21, 24 August 2010 (UTC)

I agree re. investigational uses. My first impulse is to remove the offending statement altogether and replace it with a more well-developed paragraph with such a source—if we can find one. Dementia and colon cancer are two prime examples of "promising" applications that haven't panned out. Fvasconcellos (t·c) 20:12, 24 August 2010 (UTC)

<long multiple discussion threads partly noted> When considering primary vs secondary sources as it applies to primary studies vs metanalaysis, it is also important to remember that WP reflects current majority realworld knowledge, rather than emerging (good research) that might one day change accepted opinion. So conventional medicine currently has as a de facto consensus that statins reduce disease in secondary prevention, and also high-risk primary prevention (per National Institute for Health and Clinical Excellence. Quality and Outcomes Framework etc). Now I agree alarming that all-cause mortality in primary prevention put into doubt by 2010 metanalaysis, but NICE/QOF have not yet changed. I've left the disenting views, but their priority within the article needs rearranging (also in as much that sections such as indications is written with description of historical expansion of suggested roles, and so the 2010 metanalaysis seemed to read better in that sequence). "Safety" section seems better as "Adverse effects" in keeping with WP:MEDMOS and the lipid-hypothesis questioners then seems better at end of section (they don't after all disagree with the maority over issues of liver and muscle side effects occuring). Hence this edit seriesDavid Ruben ^Talk 23:21, 25 August 2010 (UTC)

Thanks for effecting those edits, David. People need to understand who is prescribed statins and why. The mortality benefit is only one side of the coin. Being in hospital with a heart attack is also not pleasant, and I reckon many people would be prepared to take a tablet a day to achieve this outcome alone, even if their life expectancy remains unchanged.

I totally agree that NICE and NCEP guidance is therefore a very reliable source. We still need to say something more about how Framingham Risk Scoring (and possibly QRISK2) is implemented when it comes to statin decisions in real life. And perhaps something about the NICE diabetes type 2 guideline, which suggests "prescribe and then stop measuring lipids". And perhaps something about the NICE guideline for familial hypercholesterolaemia and how it might influence statin prescribing. JFW | T@lk 19:34, 26 August 2010 (UTC)

"prescribe and then stop measuring lipids" would be simplicity itself, except QOF seeks Total<4 & LDL<2, which rather requires retesting, dosage adjustments and alternative statin selection. David Ruben ^Talk 22:31, 26 August 2010 (UTC)

Additional rare side effect not found in package insert is significantly reduced white cell count; reversible upon discontinuation of drug.

Personal experience verified by Winship Cancer Institute, Atlanta GA based on two years treatment daily Lipitor 10 mg.

Patrea Pabst —Preceding unsigned comment added by 75.203.139.22 (talk) 19:17, 9 November 2010 (UTC)

Sadly cannot be added until formally reported, see WP:NOR. JFW | T@lk 22:29, 9 November 2010 (UTC)

New report suggests statins may be overprescribed

A report by the Cochrane Collaboration [12] says "Although reductions in all-cause mortality, composite endpoints and revascularisations were found with no excess of adverse events, there was evidence of selective reporting of outcomes, failure to report adverse events and inclusion of people with cardiovascular disease. Only limited evidence showed that primary prevention with statins may be cost effective and improve patient quality of life. Caution should be taken in prescribing statins for primary prevention among people at low cardiovascular risk."

Nature's blog [13] says "But now a review by the widely respected Cochrane Collaboration suggests there is little evidence such a strategy is cost effective. In addition, the mainly industry-funded trials in this area are guilty of horrific selective reporting and an unwillingness to give up their data, according to an editorial accompanying the latest review."

Should this be reported in the article and if so, how? Poujeaux (talk) 17:01, 30 January 2011 (UTC)

A blog usually reflects someone's personal opinion, hence the use of loaded terminology such as "horrific". The fact that cholesterol lowering in low-risk patients is unrewarding is not news; the British guidelines have traditionally relied on the Framingham (and more recently QRISK2) risk score to identify those at higher risk.

I get a bit annoyed when people rant about manufacturer-sponsored trials. Obviously, if those trials could be performed by impartial bodies with funding from government or charities, that would be great. However, setting up a high-quality trial is dreadfully expensive. You can't have it both ways. JFW | T@lk 18:32, 30 January 2011 (UTC)

I agree the blog is not appropriate. I think the Cochrane article could get a brief mention in the final section of the article. "The fact that cholesterol lowering in low-risk patients is unrewarding " may not be news to you but I think it might be to the non-expert reader of the article! Poujeaux (talk) 09:03, 31 January 2011 (UTC)

In that case, we should not just cite the Cochrane review (see WP:WEIGHT and WP:RECENTISM) but also current evidence-based guidelines that have been stating the same for a number of years. JFW | T@lk 19:46, 31 January 2011 (UTC)

Vitamin D

Pro crast in a tor (talk · contribs) added a short paragraph in the "controversy and criticism" section suggesting a link between statins and vitamin D activity. This seems interesting, but the sources were being used in a fashion that seemed to indicate WP:SYNTH, and this is probably something we should only really mention when it has appeared in a high-quality secondary source. JFW | T@lk 16:27, 7 February 2011 (UTC)

Statins as possible alleviators of host response to influenza infection

In a letter to the New Yorker of March 7, 2011 (http://www.newyorker.com/magazine/letters/2011/03/07/110307mama_mail2), Dr. David S. Fedson (Former Prof of Medicine at the University of Virgina) refers to the use of statins to reduce mortality when given to patients with seasonal influenza. It seems worth exploring to expand this article's range. It would be necessary, of course, to discover the actual research and practice behind this. Jelly Catullus (talk) 15:52, 6 March 2011 (UTC)

We should only cite information that is compatible with WP:MEDRS. Citing newspaper reports is usually not sufficient, especially seeing that the correspondent makes no attemt to cite sources of his own. JFW | T@lk 20:51, 6 March 2011 (UTC)

Clarification Needed

"Endo and his team reasoned that certain microorganisms may produce inhibitors of the enzyme to defend themselves against other organisms, as mevalonate is a precursor of many substances required by organisms for the maintenance of their cell wall (ergosterol) or cytoskeleton (isoprenoids).[57]"

This sentence needs further explanation. What does mevalonate have to do with HMG-CoA reductase? —Preceding unsigned comment added by 50.4.1.182 (talk) 02:59, 17 May 2011 (UTC)

It is the enzyme that makes mevalonate. JFW | T@lk 13:38, 17 May 2011 (UTC)

potentially dangerous uncited material

At the end of the Drug Interactions section is this line: "An alternative, somewhat risky, approach is that some users take grapefruit juice to enhance the effect of lower (hence cheaper) doses of statins; this increases risk and the potential for statin toxicity." It is uncited and I believe a health risk so I am deleting it immediately. THEMlCK (talk) 09:14, 20 December 2011 (UTC)