Talk:Benzodiazepine/Archive 3

Paton 2002
A recent edit (with edit summary "reverted misleading. the review provides evidence both for 1% and 20% frequency of disinhibition") removed this text:
 * "The frequency of paradoxical reactions to benzodiazepines is uncertain. Although large studies and systematic reviews have generally found paradoxical reactions to be rare in the general population, with incidence rates below 1% and similar to placebo, some small studies have reported high incidence rates in well-defined groups such as individuals with borderline personality disorder. Most reports of disinhibition involve high doses of high-potency benzodiazepines such as alprazolam, clonazepam or triazolam."

and restored the following text, which contains some information that is incorrect, and some that is misleading or obsolescent:
 * "A review of 45 controlled trials found no difference in their incidence between patients given triazolam, flurazepam and placebo. Another review estimated that the incidence of aggressive reactions to the administration of benzodiazepines is similar to placebo and below 1%. On the other hand, in controlled trials and case series for alprazolam, paradoxical reactions were observed in 10–20% of the cases. In several trials oxazepam demonstrated a lower rate of disinhibition reactions than other benzodiazepines."

There are several problems with this edit:
 * The first two sentences of the restored text simply lists results from earlier studies, with no summary or analysis from the cited source, Paton 2002. But a review like Paton 2002 should not be cited merely to give individual summaries of the underyling sources (the original sources should be cited for that). Reviews should be cited for the synthesis work that they do.
 * The edit restored the claim that "in controlled trials and case series for alprazolam, paradoxical reactions were observed in 10–20% of the cases". This is incorrect, as the Paton gives the 10–20% figure only for case series, not for controlled trials.
 * The edit removed the topic sentence "The frequency of paradoxical reactions to benzodiazepines is uncertain". This sentence summarizes the rest of the passage, and is well-supported by the cited source, Paton 2002, which says "Although the first reports of paradoxical reactions to benzodiazepines date back over 40 years (Boyle & Tobin, 1961), the incidence of this adverse effect remains uncertain."
 * The edit removed the important point that smaller studies in well-defined groups have found paradoxical reactions to be more common, whereas large studies and systematic reviews over the general population have generally found paradoxical reactions to be rare. This is also well-supported by Paton, who writes "Some small studies in well-defined homogeneous patient groups report high rates of paradoxical reactions.... While these studies, numerous case reports and published case series suggest that paradoxical reactions are common, large studies and systematic reviews have generally found them to be rare."
 * The edit summary said "review provides evidence both for 1% and 20% frequency of disinhibition". The removed text mentioned the 1% figure, so presumably that is not at issue. The 10–20% figure is mentioned by the review for case series, but this number is of much lower quality, as it is just for a case series. The review also mentions other high figures, such as 13.7% and 58%. None of these figures are randomized controlled trials for the general population, however. It does not seem wise to mention the 10–20% figure (which is not scientific data), while at the same time not mentioning the 13.7% and 58% figures (which are from double-blind, placebo-controlled randomized trials). As all of these figures are poorly supported, it is better not to mention any of them, and simply to say "high incidence rates".
 * The edit replaced the more-informative:
 * "Most reports of disinhibition involve high doses of high-potency benzodiazepines such as alprazolam, clonazepam or triazolam." (supported by Paton 2002)
 * with the less-informative:
 * "In several trials oxazepam demonstrated a lower rate of disinhibition reactions than other benzodiazepines." (supported by a 1981 review that must predate alprazolam).

Given all these problems, I have reinstalled the new text, adding the phrase "published case series suggest that they [paradoxical reactions] are common" in order to allay any concerns of POV here. Eubulides (talk) 07:36, 14 June 2009 (UTC)

Point-by-point answer:


 * a review like Paton 2002 should not be cited merely to give individual summaries of the underyling sources... First, a review like Paton 2002 CAN BE cited merely to give individual summaries of the underlying sources. That is what the reviews are for. They give authoritative summaries of the sources, so I can avoid OR. Secondly, it is necessary to state for which benzodiazepines (triazolam, flurazepam) the paradoxical side reactions are similar to placebo. It is a misinterpretation to generalize that result to other benzodiazepines.


 * the Paton gives the 10–20% figure only for case series, not for controlled trials. Wrong, see Paton "In a placebo-controlled study of alprazolam in the treatment of panic disorder, 13.7% of patients randomised to alprazolam experienced paradoxical reactions compared with none given placebo." It is a controlled study. 13.7% is in the middle of 10-20% range. Panic disorder is a primary indication for alprazolam, and so panic disorder patents are fairly representative of the "general population" of those treated with alprazolam.


 * The frequency of paradoxical reactions to benzodiazepines is uncertain. It is a good introductory sentence. I agree that I should have not removed that.


 * Some small studies in well-defined homogeneous patient groups report high rates of paradoxical reactions.... While these studies, numerous case reports and published case series suggest that paradoxical reactions are common, large studies and systematic reviews have generally found them to be rare. Paton does say that. However she states that only in the context of a) meta-analysis was conducted only for triazolam and flurazepam. b) controlled study of alprazolam indicating 14% paradoxical reactions. If we are to include the controversial statements, we should not remove the context.


 * The edit summary said... Let's not discuss the edit summary which is, by necessity, a stripped down justification for changes.


 * The edit replaced the more-informative:... These two sentences do not contradict each other. Suggest including both.

The Sceptical Chymist (talk) 12:14, 14 June 2009 (UTC)

 Additional information. al.[40] could find no difference in the incidence of disinhibition between triazolam, flurazepam and placebo. However, the fact that no unusual or excessive adverse reactions were reported may mean that such events were not witnessed or recorded, rather than they did not occur." The Sceptical Chymist (talk) 12:31, 14 June 2009 (UTC)
 * A review by Bond (1998) doubts the conclusions of the old Greenblatt (1984) metaanalysis of 45 studies: "In a review of 45 double-blind, controlled trials, Greenblatt et


 * In the light of paucity of newer controlled studies on paradoxical reactions, the conclusion from Hall (1981) remains appropriate: "Thus, the question of frequency, severity, quality and significance of benzodiazepine-induced hostility remains unsettled." That is why we should be cautious of the over-generalizing the results of few trials of few benzodiazepines to the whole group. That is why we should mention which BDs showed what in controlled trials. The Sceptical Chymist (talk) 12:53, 14 June 2009 (UTC)


 * It is somewhat of an accepted wisdom among the professionals that oxazepam has a more benign profile of side effects than other BDs. It is a very old drug so it is difficult to come up with newer review references than already mentioned in the article. Another review supporting this is : "Next, the controversy regarding benzodiazepines and aggression is examined. Oxazepam appears exceptional here in that it is not associated with paradoxical release of aggression in patients." Later studies seem to further support this conclusion. See : "A double-blind, controlled clinical trial of chlordiazepoxide, oxazepam and placebo was conducted in 65 outpatients with past histories of temper outbursts, assaultive behaviour and impulsiveness associated with anxiety, irritability and hostility. Of those tests showing statistically significant results, there was a tendency for oxazepam to be somewhat more effective in the reduction of anxiety than chlordiazepoxide. Oxazepam was also superior to the latter on 1 subscale of tests used to measure hostility." Also see  "Oxazepam and lorazepam had very similar subjective effects, but the higher dose of lorazepam increased aggressive responding on the task more than any other treatment." The Sceptical Chymist (talk) 13:21, 14 June 2009 (UTC)

Regarding oxazepam, oxazepam still behaves like other benzodiazepines. The reeason that it has a better side effect profile is because simply it is less potent. If one was to compare 8 mg of codeine to 5 mg of diamorphine (heroin) or fentanyl one would find that diamorphine and fentanyl would have a higher incidence of side effects such as euphoria, hallucinations confusion etc because they are more potent. As oxazepam at standard doses has less severe side effects it is also a relatively weak anxiolytic, has weak hypnotic effects and weak anticonvulsant properties. There is nothing different about oxazepam from other benzodiazepines other than the dose form that it is marketed in is "low potency" and it crosses the blood brain barrier slowly. If one was to take maybe 4 - 5 times say the therapeutic dose of oxazepam you probably would find that it has the same incidence of side effects as many other benzodiazepines and is as potent.-- Literature geek |  T@1k?  13:49, 14 June 2009 (UTC)

I think that other at risk groups such as elderly, children and those with neurological or learning impairments should also be noted in the article as these are all sizable population groups and thus ommiting these groups from the articles with just an example of one group borderline personality disorder does a diservice to the reader. Sometimes just giving one example and trying to shorten a sentence adversely effects article quality in my opinion.

This is the conclusion of Paton, "The overall incidence of disinhibitory reactions is small, but those with impulse control problems, neurological disorders, learning disabilities, the under 18s and the over 65s are at significant risk."-- Literature geek |  T@1k?  16:52, 14 June 2009 (UTC)


 * "a review like Paton 2002 CAN BE cited merely to give individual summaries of the underlying sources." I'm afraid we'll have to continue to disagree here. This is not good style, unless the review is providing perspective on those sources, which was not the case here.
 * "it is necessary to state for which benzodiazepines (triazolam, flurazepam) the paradoxical side reactions are similar to placebo. It is a misinterpretation to generalize that result to other benzodiazepines." The current text is not generalizing that result. It is relying on the cited source, Paton 2002, which says "The majority of case reports of behavioural disinhibition are in patients treated with high doses of high-potency benzodiazepines ...".
 * "Wrong, see Paton "In a placebo-controlled study of alprazolam in the treatment of panic disorder, 13.7% of patients randomised to alprazolam experienced paradoxical reactions compared with none given placebo." It is a controlled study. 13.7% is in the middle of 10-20% range." This sort of calculation is original research, and we can't rely on it. Paton mentions the 10–20% figure for a different context, one that has nothing to do with controlled studies. The article can't silently combine that with the 13.7% figure for one controlled study, and then omit the 58% figures for another controlled study, and then just say "10–20%": that would be misleading.
 * "Panic disorder is a primary indication for alprazolam, and so panic disorder patents are fairly representative of the "general population" of those treated with alprazolam." As I understand it, "general population" does not mean "those treated with alprazolam"; it means everybody in general.
 * "Paton does say that. However she states that only in the context of ..." Paton is summarizing existing research. I see no evidence that her summary is controversial, or that there's any need to list each and every study that Paton refers to. Encyclopedias are supposed to summarize what reliable sources say, rather than flood the reader with unnecessary detail.
 * "These two sentences do not contradict each other. Suggest including both." I'm leery of citing a 1981 (!) review that (obviously) does not cover newer benzos. Also, the average reader does not need this much detail, and it'd be better to remove all mention of specific benzos, resulting in simply "Most reports of disinhibition involve high doses of high-potency benzodiazepines." I did that in my edit (noted below).
 * "Additional information" Those comments cite Hall 1981, Ladler 1978, Liohn 1979, and Bond & Lader 1988. These are old sources and they don't disagree with the newer review. I don't see a need to go into so much (ancient) detail in the article.
 * "I think that other at risk groups such as elderly, children and those with neurological or learning impairments should also be noted" Good point. This is noted in Paton's abstract. I added "In these groups, impulse control problems are perhaps the most important risk factor for disinhibition; learning disabilities, neurological disorders, and ages under 18 or over 65 are also significant risks". This edit also removes the unnecessary detail noted above.
 * Eubulides (talk) 09:46, 15 June 2009 (UTC)


 * Most of Eubulides objections are centered around the age of the reviews. This issue cannot be helped because most of these drugs are ancient. Most of the research in pharmaceuticals is driven by pharma companies and often stops when the drug goes off patent. Another problem with newer reviews is that they tend to parrot the conclusions of the older reviews. The telling example is the "1% of the general population" bit. It is repeated over and over again in different "new" reviews. But you cannot tell what the authors of the original 1988 review meant under this without reading the old original review and citing it. Was it general population of those who were prescribed BDs, that is mostly people with anxiety disorders? Was it general population, that is mostly healthy people? Although, why would you prescribe BDs to healthy people? The Sceptical Chymist (talk) 11:47, 15 June 2009 (UTC)
 * Please see  below. Eubulides (talk) 08:31, 17 June 2009 (UTC)

I personally think that the Paton review (perhaps unintentionally) plays down the paradoxical adverse effects. Conflict of Interest in clinical trials and selective publishing are a problem. However, as wikipedia works via reliable sources unless we find a different secondary source then we should just go with this citation and what it says. I have no doubt that severe paradoxical reactions are relatively rare but I am not convinced that mild and moderate paradoxical effects are also rare and "similar to placebo" as Paton seems to suggest.-- Literature geek |  T@1k?  12:11, 15 June 2009 (UTC)


 * Conflict of interest cuts both ways. Big pharma would rather have physicians prescribe newer, more-expensive, patented drugs, instead of the older, cheaper, out-of-patent benzodiazepines. Big pharma is therefore happy to see studies emphasizing the adverse effects of benzodiazepines, as exemplified by Paton's review. But all this is not supposed to matter. We have to follow what reliable sources say, even if big pharma is influencing mainstream opinion. Unfortunately, the article is currently significantly distorting Paton's review in a negative way. For more, please see  below. Eubulides (talk) 08:31, 17 June 2009 (UTC)

Depends, I have read a couple of interesting points by people who would be classed as "antipsychiatry" but I don't cite those individuals because their research is biased. I actually think there were some issues with how the Paton paper was cited and summarised. I have made some changes which I really hope will resolve the dispute and will satisfy yourself as well as Sceptical. It is difficult sometimes to satisfy opposing views amongst editors so hope you both are happy with the changes. See subsection below.-- Literature geek |  T@1k?  16:12, 17 June 2009 (UTC)

I guess my point is I try to seek out the best sources and avoid sources which I believe after reading them severely distort evidence such as antipsychiatry or pharmaceutical company based publications peer reviewed secondary sources or not. I am not categorising Paton as this, Paton paper is fine and am happy with it. I am just clarifying my editing beliefs is all. I still do cite pharmaceutical company driven papers as they can have high quality data in them, I am just more cautious I guess.-- Literature geek |  T@1k?  16:18, 17 June 2009 (UTC)

Current Paton-related problems
A recent edit reintroduced some of the problems noted above. That's one too many reverts for my taste. Rather than edit war over this, I am tagging the relevant section with POV-section and Summarize section and noting the problems below. My edit is also introducing citations to studies that the disputed section discussed without citing.

Here are some problems with the edit:
 * It removed the point that large studies and systematic reviews have generally found paradoxical reactions to be rare in the general population.
 * It removed the point that disagreement comes from published case series and some small studies.
 * It removed the point that the disagreement comes from studies of well-defined groups, notably of people with impulse control problems.

More generally, the edit has reintroduced problems that Colin objected to in the FA review:
 * "There are areas where the important facts are lost among the lesser." That's definitely a problem with this edit.
 * "There are far too many cases where three, four and even five citations are strung together, sometimes just to source a single sentence." Also a problem.
 * "primary research papers cited without good reason" As mentioned above, the edit discusses several sources without citing them. I have fixed this, but this means the section is citing ancient primary sources and very old reviews, which is a negative we've been trying to fix.

Worse, this edit has made the section POV. The overall effects of the edit is to emphasize this adverse effect of benzodiazepines, far out of proportion to what the source says.

In response to some of the comments above:


 * "Most of the research in pharmaceuticals is driven by pharma companies" It is not our job to filter out parts of reviews that we don't like, simply because research is driven by pharma companies. We are here to summarize the mainstream opinion, not the opinion that we think would be mainstream if big pharma didn't exist.
 * `Another problem with newer reviews is that they tend to parrot the conclusions of the older reviews. The telling example is the "1% of the general population" bit.` This is backwards. If several reviews mention the 1% figure, this article can (and should) too.
 * "Was it general population" That's what Paton said, yes. If it's important to define what "general population" is, then we should define it in the text. However, it's clearly not that important (the term is well-understood by the general reader); so we needn't define the term in the text or cite the ancient study it's derived from.
 * "Most of Eubulides objections are centered around the age of the reviews." Not so. Most of my objections are based on two things:
 * The text about paradoxical reactions in Benzodiazepines significantly mischaracterizes what is is said by the review (Paton 2002,.
 * Paton starts off by saying that in most cases benzos have a calming effect, whereas paradoxical reactions of increased anxiety etc. happen only in a minority of cases. In contrast, the text starts off by making it sound that paradoxical reactions are common, with strong language talking about "numerous cases" of "aggression ... and suicidal behavior", giving the naive reader a misimpression of how common the problem is. Only buried later in the paragraph do we discover that this happens only in a minority of cases (whether <1% or higher we don't know).
 * Paton makes a primary point that large studies and systematic reviews have generally found paradoxical reactions to be rare in the general population, whereas smaller studies and case report series have found it to be relatively common in well-defined subgroups. This important point is absent in the text: instead, it simply lists studies or reviews, and their numbers. The naive reader cannot be expected to understand this point without its being said.
 * The text swamps the reader with way too much detail about old studies, without presenting the big picture that an encyclopedia should.
 * There is no need to present the results of any single primary study here, regardless of whether that study is cited or not. This area is well reviewed.
 * There is no need to give overly exact percentages like 14% (in a study with only 70 patients!) or 10–20% (not from any controlled study at all!). Not only are these percentages unnecessary detail, they are misleading: they give the naive reader the mistaken impression that there's general agreement that 10–20% of people experience paradoxical reactions with alprazolam, whereas in reality there is no such agreement. Paton cites Gardner & Cowdrey 1985 as a controlled study reporting 58% of patients with paradoxical reactions to alprazolam, for example.

Eubulides (talk) 08:31, 17 June 2009 (UTC)


 * Is what E. doing -- WP:POINT? Or maybe WP:PUSH? And for sure, it is WP:TLDR big time! Whatever I was able to read from E.'s long-winded answer was mostly incorrect. For example, E. called old reviews - primary sources. He found POV in the text, which states that frequency of the side effects is unclear and illustrates both points of view.


 * I am reverting to his version. Let him have his way. IMHO, this way of handling issues is inexcusable for an experienced editor. E. should be ashamed. The Sceptical Chymist (talk) 10:45, 17 June 2009 (UTC)


 * That is unfair. Please stick to discussing the article and the sources, and bear in mind that a talk page isn't as easy a communication forum as if we were talking in a room.
 * We are trying to make our text faithful to the sources, and this includes giving appropriate weight to aspects of the topic. For each section, we should ask ourselves: "What are the main points to get across" and ensure those come across loud and clear. It is really tempting when reading about fascinating research studies to think the reader shares our fascination. They don't; it just clouds things.
 * I note that currently the text still doesn't satisfy some of Eubulides issues: particularly that it opens with "Numerous cases" when we later claim the rate may be "below 1% and similar to placebo". Currently, the main point the reader gets is that paradoxical reactions are "numerous" and include "aggression, violence, impulsivity, irritability and suicidal behavior". Colin°Talk 12:55, 17 June 2009 (UTC)

Numerous is fine because it is true, it has been widely reported in the peer reviewed literature. Wikipedia primarily concerns itself with reporting facts. In this case it is not misleading as it is clarified in the paragraph. Please note that the less than 1% statistic is for "healthy patients" in clinical trials. The paton paper states that in certain groups, elderly (which make up a sizable number of the general population), the young, high doses or high potency benzos, recreational users, learning impaired etc are at significant risk of paradoxical reactions. My personal opinion is that everyone here is taking the Paton and other reviews out of context.-- Literature geek |  T@1k?  14:15, 17 June 2009 (UTC)

I am going to do afew changes to try and improve that section.-- Literature geek |  T@1k?  14:17, 17 June 2009 (UTC)


 * FWIW, I can add from clinical practice that paradoxical reactions as such are pretty rare really. They also don't much rate a mention when one is prescribing benzos except as a footnote that we're all aware of but rarely see. Casliber (talk · contribs) 14:24, 17 June 2009 (UTC)

I don't dispute your clinical experience Casliber. What I am saying is that the elderly [particularly 75 and over] for example it is not rare (think of delerium, confusion resulting in agitation, anxiety etc) or children becoming hyperactive or aggressive or learning impaired and also borderline personality disorder it is not similar to placebo. I have done a rewrite of the section, removed redunancy/repeating and I have also removed the "numerous reports" to try and resolve dispute.-- Literature geek |  T@1k?  16:04, 17 June 2009 (UTC)

The main problems with benzodiazepines are tolerance and withdrawal (including the protracted nature of withdrawal) and the adverse effects of long term use in my opinion. Acute side effects upon commensing benzodiazepines are not a significant concern for the general population except certain groups which we have acknowledged now.-- Literature geek |  T@1k?  16:08, 17 June 2009 (UTC)


 * Thanks, Literaturegeek, I think your changes have helped. I've tweaked it a little to remove all trace of references to literature or studies. The paragraph now concentrates on the facts, as well as we can determine them, which is what an encyclopaedia should focus on. I think that resolves this issue for me. Colin°Talk 16:27, 17 June 2009 (UTC)

You are welcome Colin. Glad this issue iss resolved. Your tweaks looked good to me. I spoke to a geriatrician who said in his experience about 10 percent of patients 75 years or older would have a paradoxical reaction to benzodiazepines such as anxiety, agitation usually as a result of benzo induced confusion etc. He said in the case of opiates eg morphine it would be more like 30 percent who would become anxious or agitated as a result of confusion so benzos ain't as bad as the opiates for these acute side effects in the elderly. I also read in a ref today somewhere that clinical trials of benzos in the elderly are rare so I really don't think we will get good statistical data on this from say a systematic review or meta-analysis for the elderly but at least we have it reffed that it is rare in general population but more common in certain groups of people. I am happy with that as accurate as we are gonna get. Anyway as this issue I believe is resolved we should move on to other contructive criticisms or suggestions or else vote to support FA. :)-- Literature geek |  T@1k?  02:14, 18 June 2009 (UTC)
 * Thanks to both of you for a masterful rewrite of the paragraph. Eubulides (talk) 06:51, 18 June 2009 (UTC)

Characterisation of benzodiazepine structure
Wouldn't it be more accurate to mention the phenyl group somewhere in the Chemistry section? As is, it makes it sounds as if merely a benzodiazepine base is necessary, but AFAICT the overwhelming majority (though not quite all) of comemrcial benzodiazepines are in phenylbenzodiazepines. (specifically 5-phenylbenzodiazepines in the case of 1,X-benzodiazepines). I hope I'm not messing up with the chemical nomenclature and this makes sense. Circeus (talk) 18:53, 15 June 2009 (UTC)
 * I agree. The phenyl group was included in a previous version, but was overwritten by this edit.  I have tried to combine elements of the original version which IMHO are more relevant to medicinal chemistry (and therefore by definition the specific subject matter of this article) and organic nomenclature (which is relevant to the more general subject of the benzodiazepine ring system).  Boghog2 (talk) 20:24, 15 June 2009 (UTC)

No, in that particular section, it would be inaccurate to include a phenyl group. Physchim62 (talk) 20:50, 15 June 2009 (UTC)
 * Both the entire article and that particular section within this article pertain to pharmacologically active benzodiazepines which possess a phenyl or a bioisosteric equivalent at position-5. In the new version of the  graphic I have attempted to find a compromise which depicts both the core benzodiazepine ring system which is the basis for the benzodiazepine name as well as the minimum pharmacologically active core. As discussed here, perhaps I should also add an R group to position-1 and a X group to position-7 (done).  Boghog2 (talk) 21:29, 15 June 2009 (UTC)

The problem is that the structure you want to include is (1) a substituted benzodiazepine, rather than a simple benzodiazepine in the chemical sense, and (2) almost identical to the structure shown in the top right-hand corner of the article, next to the lead. I'm very sorry for wanting to tell people where the name comes from, and to illustrate the basic structural similarity. The small image is in the "technical" part of the article – improve it if you wish, I'm having trouble with my new version of ChemSketch. Physchim62 (talk) 23:25, 15 June 2009 (UTC)
 * I'm very sorry for wanting to tell people where the name comes from. What are you apologizing for? A completely agree with you and that is why I merged our two figures.  At the same time, it is important to note that this article and the chemistry section within this article concern pharmacologically active benzodiazepines and not the benzodiazepine heterocyclic ring system.  I agree that it is important to make clear where the name came from (the left hand side of the graphic) but it is equally important to make clear what is pharmacologically active (the right hand side of the figure).  In addition, the right hand figure also includes the ring numbering and make reference to diazepam.  Furthermore, I don't see the problem with a little repetition, especially considering the two figures in question are far removed from each other. Boghog2 (talk) 23:45, 15 June 2009 (UTC)

Nonbenzodiazepines
I removed the following statement from the article: "Benzodiazepines are structurally similar to several groups of drugs, some of which share similar pharmacological properties, including the quinazolinones, hydantoins, succinimides, oxazolidinediones, barbiturates and glutarimides.[86]" First of all, it is nonsense. If you look at the structures, benzodiazepines look nothing like hydantoins, oxazolidinediones etc. Second, the statement misrepresents the article, which only points at a single common structural feature "a carboxamide group as a constituent part of a five-, six- or seven-membered heterocyclic ring structure". Third, the presence of such a common structural feature in benzodiazepines is not notable because benzodiazepines share it with hundreds of thousand other compounds.The Sceptical Chymist (talk) 01:51, 16 June 2009 (UTC)


 * Thanks for pointing out the problem. I agree that the statements that you deleted do not make any sense. Perhaps what should be included instead is a statement that:


 * Nonbenzodiazepines also bind to the benzodiazepine binding site on the GABAA receptor and share similar pharmacological properties. While the nonbenzodiazepines are by definition structurally unrelated to the benzodiazepines, both classes of drugs share a common pharmacophore which explains their binding to a common receptor site..


 * A graphic of the pharmacophore (similar to Figure 18.14 on page 822 of the cited book) could also be inserted. Boghog2 (talk) 06:00, 16 June 2009 (UTC)
 * I do not have the cited book but a picture would be useful because common pharmacophore is not evident (at least to me). The Sceptical Chymist (talk) 10:26, 16 June 2009 (UTC)
 * OK, I will generate a new public domain figure which will take a sometime since I am busy with other things at the moment.


 * To the right is a proposed figure for the Benzodiazepine section. Before adding it to the article, I thought I would first bring up the proposal here.  Is the figure and caption too technical for a FAC?  Suggestions are welcome.Boghog2 (talk) 21:44, 20 June 2009 (UTC)
 * Since I have heard no comments, I went ahead and added the pharmacophore graphic to the article. Boghog2 (talk) 04:18, 23 June 2009 (UTC)

Access to the Cochrane Library
Colin recently reverted my removal of the url for this Cochrane paper saying that it's freely-available. It's certainly not freely-available to me in the United States, as I noted with a poorly autogenerated reference a while ago. As the Cochrane Library page states, many countries have free access to Cochrane, but significant other ones don't. I don't think the URL should be included. II | (t - c) 22:39, 15 June 2009 (UTC)
 * I wasn't aware that Cochrane restricted access based on the client PC's country. I guess you'll need to move to Wyoming or the UK :-). I'll leave it up to someone else to decide if this is freely available enough to justify a URL link. How about removing the standard url link and adding a suffix: "Text freely available here in some countries." Colin°Talk 20:27, 16 June 2009 (UTC)
 * In other articles when this issue (access to Cochrane0 has come up, I have omitted the URL. A source that is "freely readable" only (say) on the campus of Johns Hopkins is clearly not free; on the other hand a source that is freely readable everywhere except (say) inside the Great Firewall of China is clearly free. In this case it's more of a gray area, but when most typical Wikipedia readers can't read a source, I'd say it's not free, so I removed the URL. Eubulides (talk) 22:35, 16 June 2009 (UTC)

More sourcing comments

 * Lead:

The third paragraph in the lead has four points: These are all sourced to three papers at the end. I'd prefer really that each major point had its own citation(s). One is a comment on a review of their effectiveness/efficacy for GAD, which concludes they aren't effective. One is possibly focussing only on sedation and it deals with "paradoxical reactions" rather than all side effects. I don't see anything on them being "major drugs of abuse" but I don't have access to the full papers.
 * 1) Benzodiazepines are generally safe and effective in the short term
 * 2) Cognitive impairments or paradoxical effects such as aggression or behavioral disinhibition occasionally occur.
 * 3) Their use in the longer term is not recommended due to their propensity to cause tolerance, physical dependence, addiction and a withdrawal syndrome upon cessation of use.
 * 4) They are also major drugs of abuse.


 * Therapeutic uses, Anxiety, panic and agitation:

What are the main points we are trying to get across here? There are three aspects that our sources testify to:
 * 1) Clinical guidelines, hopefully based on evidence (... are recommended (indicated) for the treatment of ...)
 * 2) The results of studies and the analysis of those studies (... have been found to be efficacious in treating ...)
 * 3) Actual clinical practice, whether supported by guidelines and evidence or not (... are used for the treatment of ...)

The article mostly cites aspects 2 and 3. The article often says "can be useful for", which to me implies there is evidence of usefulness, but only sometimes. IMO the "can" is too vague and allows us to mention usage that is not even first or second-line recommended. I would argue that the section title "therapeutic uses" implies "usage that has been proven to have a therapeutic effect in clinical practice". Usage that goes against the evidence and advice isn't "therapeutic", even if it is widespread.

I've discovered recently, there is a difference between "efficacy" and "effectiveness" (see this article). One paper you cite is actually just a comment on a single systematic review paper  that concludes "This systematic review did not find convincing evidence of the short-term effectiveness of the benzodiazepines in the treatment of GAD. On the other hand, for the outcome of efficacy, this review found robust evidence in favour of benzodiazepines." We need to be careful in our choice of words.

Looking at the prescribing guidelines I see:


 * BNF: Benzodiazepines are indicated for the short-term relief of severe anxiety; long-term use should be avoided
 * BNF: In panic disorders (with or without agoraphobia) resistant to antidepressant therapy
 * a benzodiazepine (lorazepam 3–5 mg daily or clonazepam 1–2 mg daily [both unlicensed]) may be used
 * alternatively, a benzodiazepine may be used as short-term adjunctive therapy at the start of antidepressant treatment to prevent the initial worsening of symptoms.


 * BNF: Diazepam or lorazepam are very occasionally administered intravenously for the control of panic attacks. This route is the most rapid but the procedure is not without risk (facilities for reversing respiratory depression with mechanical ventilation must be at hand) and should be used only when alternative measures have failed.
 * BNF: Benzodiazepines may be helpful in the initial stages of treatment for mania until lithium achieves its full effect; they should not be used for long periods because of the risk of dependence.
 * CSM: Benzodiazepines are indicated for the short-term relief (two to four weeks only) of anxiety that is severe, disabling, or subjecting the individual to unacceptable distress, occurring alone or in association with insomnia or short-term psychosomatic, organic, or psychotic illness.
 * CSM: The use of benzodiazepines to treat short-term ‘mild’ anxiety is inappropriate and unsuitable.
 * NICE (CG22): Benzodiazepines are associated with a less good outcome in the long term and should not be prescribed for the treatment of individuals with panic disorder. (compare this with the BNF)
 * NICE (CG22): In Generalised Anxiety Disorder, benzodiazepines may be considered if immediate management is required, but should not usually be used beyond 2–4 weeks.

Some of this agrees with the article text but the emphasis isn't always right. For example, the main indications get a brief sentence but lesser ones get several sentences. Sometimes the article text describes clinical practice that is at odds with prescribing guidelines. The biggest examples here would be long term use and use for panic disorder, both of which appear to be absolutely forbidden by the guidelines yet routinely ignored. Perhaps such usages should appear in the "Drug misuse" section?

The final paragraph says benzos "can be very useful" for acute mania. This indication is noted in the BNF (though much reduced to "may be helpful") but the Cochrane review cited is decidedly lukewarm on the evidence. If we're going to cite Cochrane, shouldn't our text agree with it to some degree?

I think some edits have broken some sentences:
 * Broken text:
 * "Although their effectiveness is not recommended as a..."
 * "Some benzodiazepines are prescribed for the short-term management of severe or debilitating insomnia. They have strong hypnotic effects, are typically the most rapid-acting and have strong receptor affinity."

Colin°Talk 22:57, 15 June 2009 (UTC)

I made a stab at fixing the main problems that you raised. Forbidden is not the right word but unlicensed and not recommended or evidence based. breaking the guidelines means if something goes wrong eg dependence adverse long term effects severe withdrawal symptoms and the ppatient sues the doctor has to be able defend his "clinical judgement or lack of" but other than that a clinician is free to prescribe outside via exercising his clinical judgement.-- Literature geek |  T@1k?  12:00, 16 June 2009 (UTC)


 * Ultimately, benzo use is a vexed issue in amny cases and I think it depends alot on the views of the prescribing authority on addiction potential, which is responsible for the variation in views on their use in anxiety. Clonazepam was often used in mania a decade or more ago but you don't see it used now and I think most of in psychiatry accept that the benefit was non-specific. Anyway, there is a question on how much detail and how well cited this could be. I haven't looked at the page for a bit. Casliber (talk · contribs) 02:21, 16 June 2009 (UTC)

The article says lorazepam is most commonly prescribed but clonazepam is sometimes prescribed, so it was changed but perhaps you think that it should be deleted (clonazepam) from the article or is sometimes an appropriate enough word? I believe clonazepam is more often used in the USA than in other Western countries.-- Literature geek |  T@1k?  12:00, 16 June 2009 (UTC)


 * Thanks, Casliber. I'm an outsider with no expertise in this field, so I'm rather reluctant to propose alternative text or choose quality sources. I do feel this section blurs the three aspects I mentioned above, making it hard to for the reader to know what the main indications are, what is rare due to rarity, what is rare due to it being a last resort, and what is sadly common but not at all recommended. I think the foundation of an "indications" section in WP should be evidence-based clinical guidelines. I don't know where to look for US or Australian versions (a professional body?), but the UK ones are freely available. Notable deviance from those guidelines (such as routine long term use, or use in unsuitable populations groups) should be noted and discussed (and the are plenty first-class articles on that subject). Primary sources have been largely eliminated from the article, which is good. But it looks like one can find a review to support almost any opinion wrt usage in this field. And fashions change, so old papers may now be irrelevant. This is where I think the clinical guidelines from a professional body, insurance group or government body can help to establish the current consensus opinion of experts.
 * Look at what we say about panic disorder: "and occasionally prescribed for panic disorder." and "Limited data from longitudinal studies have suggested benefit from long term use in panic disorder.". Should we be noting research that merely "suggests" benefit? Surely it only gets encyclopaedic when this becomes a formal, widely respected or practised indication. We say they are "occasionally prescribed" but that fact is partly sourced to a 19-year-old US paper. Nowhere do we say that such usage is unlicensed (in the UK) and goes against NICE guidelines. Nor do we say that this is short-term use for panic disorder.
 * I will try to revisit the FAC later. Got bogged down in this section last night. Colin°Talk 08:17, 16 June 2009 (UTC)

I have added a NICE citation Colin for the anxiety section. I agree that national prescribing guidelines are the best sources for the indications sections. I actually think in this situation we should be adding a review for the evidence base of prescribing in panic disorder of benzos (mainly in the USA) but also in other countries is significant enough to warrant it and it explains to the reader the "controversy" in the literature for this offlabel indication. I have added that it is an unlicensed indication. Benzodiazepines are not licensed for long term use in any Western country (as far as I know). In USA FDA accepts evidence for efficacy of up to 8 or 9 weeks for alprazolam in panic disorder but not beyond that. Thanks for your additional suggestions. Have tried to resolve them but have to go out now, will try and work a bit more on it later.-- Literature geek |  T@1k?  12:00, 16 June 2009 (UTC)


 * Here is one national guideline:
 * These Canadian guidelines are freely available from the Pulsus Group but are behind a screwy web interface that we can't directly link through.
 * Eubulides (talk) 23:02, 16 June 2009 (UTC)

Those are not official Canadian national guidelines, it is not a policy document. That is just an ordinary review published in a journal paper written up by a group of senior pharmacists who happened to call the title guidelines.-- Literature geek  |  T@1k?  20:55, 17 June 2009 (UTC)


 * Ah, I had thought from the PubMed entry that it was a document of the The Canadian Society for Clinical Pharmacology (now the Canadian Society of Pharmacology and Therapeutics). This is not a government body, but then again the American College of Physicians is not a government body either, and yet the ACP guidelines would be an excellent source if they had a guideline on benzos (which they don't). I now see that the document itself does not state that it is an official guideline of the society, so perhaps you're right that it's just a couple of experts. Still, couldn't it be a reasonable source for guideline-related info? Eubulides (talk) 22:52, 17 June 2009 (UTC)

I would say they are experts (or senior) pharmacists but I can tell that they are not experts in the topic that they are writing on. There was a lot of good/accurate points in it but a lot of other points that they got wrong in my opinion. It is not a systematic review but I would say a review by cherry picking of primary sources by non-experts in the subject matter. Therefore I don't think that we should use it to either supplement or challenge more authoritative sources such as the National Institute of Excellence or to supplement agree with or challenge systematic reviews or official prescribing indications and so forth.

A more pressing issue I feel is that you claimed that the article has had increasing amounts of primary sources and older reviews added when I believe that you know the opposite is true. I would like you to retract that comment on the featured article review page as it is inaccurate and misleading to other reviewers who have not been following the article's development intensely and thus could mess up all of our work.-- Literature geek |  T@1k?  23:16, 17 June 2009 (UTC)


 * That comment was true when made, but it's been fixed now, so I'll go strike it. As can bee seen from  above, the sources are in pretty good shape now (thanks), and I'll mention that on the FAC page. So the only remaining point from my point of view is prose quality, which we can turn to next. Eubulides (talk) 06:51, 18 June 2009 (UTC)

I don't think we need to find new reviews or refs for the indications section unless something factual is left out of it which needs added. One or two refs per fact should be sifficient.-- Literature geek |  T@1k?  23:21, 17 June 2009 (UTC)

I think that we need to decide is the article ready to be supported for FA or if not what problems remain. I am getting burnt out from the editing but appreciate that big improvements have been made. I really think it is above the standard of other pharmacology FA articles.-- Literature geek |  T@1k?  23:25, 17 June 2009 (UTC)

Withdrawal section
I have done some major rewriting of the withdrawal section. There really was very little on tolerance, its mechanisms, why it happens, how fast it happens to various therapeutic effects etc etc so found a ref and used it. I also added in certain important aspects which I felt were left out. I now feel that the section is comprehensive and informative with all major points addressed. I felt it needed more work done on it because it is one of the most important areas of benzodiazepines. Someone had also used a source where in the opinion of a doctor "some people may need benzos long term for anxiety" and then a 2nd source which stated tolerance to anxiety effects has been demonstrated in animals to say tolerance does not happen to the anxiety effects. The sources were either not striong enough "opinion of an individual doctor and then a paper which contradicted it with animal studies. It seemed a mixture of poor sourcing with a bit of original research thrown in as well not to mention conflicting sources. Opinions are welcome.-- Literature geek |  T@1k?  02:21, 18 June 2009 (UTC)

Do not want an edit war. Sceptical keeps deleting anything related to tolerance to the anxiolytic effects of benzodiazepines. This disputes the National Institute of Clinical Excellence which says that long term treatment of panic disorder with benzodiazepines "does not have a good outcome". Disputes this review which says "there is little evidence to indicate that benzodiazepines retain their efficacy after four to six months of regular use." It also disputes this body from a national advisorary body to NHS physicians which says "Tolerance to the anxiolytic effects is slower and appears over a few months of use." The headquarters of Roche Pharmaceuticals in Switzerland published this paper which stated that "Potential development of tolerance to and dependence on benzodiazepine tranquilizers often limit their use for long-term treatment of epilepsy, anxiety and insomnia". The world council of anxiety stated that benzodiazepines are not recommended long term for generalised anxiety because of tolerance and other adverse effects. It also disputes the Committee on the safety of Medicines. Also any reference to improvements in health and wellbeing following benzodiazepine withdrawal keeps getting deleted despite being referenced to good sources and no sources provided to dispute improvements in cognition and physical and mental health. The only reason that the refs to improved mental and physical well being post withdrawal that I can see is due to someone thinking "I don't agree with that". The reality is sedative hypnotics (barbiturates, alcohol and benzodiazepines) are well known from long term use for causing profound changes in brain chemistry due to tolerance and dependence and are associated with at least in some users (not all) neuropsychiatric and cognitive impairments. I would like to discuss this evidence base and find consensus.-- Literature geek |  T@1k?  15:22, 18 June 2009 (UTC)


 * What does "does not have a good outcome" mean? That it does not work, or that it has a harmful outcome, or what? Regards, &mdash; Mattisse  (Talk) 21:04, 19 June 2009 (UTC)

NICE are referring to effectiveness of treatment, i.e tolerance, that is what the document implied to me when I read it. :) Thanks for all of the work you have done today and last night to this article. One spelling question though, should pre medication not be "premedication"?-- Literature geek |  T@1k?  21:28, 19 June 2009 (UTC)


 * Yes, you are right. &mdash; Mattisse  (Talk) 21:37, 19 June 2009 (UTC)

Please address issues on talk page Sceptical
I have requested content disagreements are discussed on talk page but we seem to be on the verge of edit warring.

Problem 1.

This edit was used to dispute text. Can I just point out that CT scans can only detect only structural brain damage (brain shrinkage, tumors, loss of brain tissue) not neuronal damage or dysfunction. You would need a PET, spect or perhaps an MRI scan to find those types of findings. These studies have never been done.

Problem 2

You seem to have totally misrepresented the findings of the review and meta-analysis.-- Literature geek |  T@1k?  17:24, 20 June 2009 (UTC)

Problem 3

The quote marks are back in the article e.g. "involvement with other issues" and the reference with a long quotation from the article does not seem to contain the phrase "involvement with other issues" so it is not a direct quote and should not be in quotes. Further, is there justification for such a long quote in the references from a copy righted work (ref 15), as it does not seem to add necessary information to the article? Regards, &mdash; Mattisse (Talk) 17:47, 20 June 2009 (UTC)


 * RE Problem 3. You (Matisse) changed the correct "problems" to incorrect "issues" when removing the quotation, diff here []. I did not notice it when I restored the quotation marks. Now fixed... As for the long citation, I thought you asked for it... Please pare it as you see needed.The Sceptical Chymist (talk) 19:10, 20 June 2009 (UTC)


 * RE: Problem 1. On one side there are outdated speculations by Ashton (1991) that there may be neurological damage. On another side, a clear negative result from a CT study cited in a recent meta-analysis. IMHO, the experimental study (despite its limitations) is sufficient to remove mere speculations. The Sceptical Chymist (talk) 19:13, 20 June 2009 (UTC)
 * Obviously source was a non-systematic review of the literature and thus just the opinion of the author using a biased sampling of the literature. See, and, . There are other studies done in high benzo abusers as well. Anyway you have not responded to me pointing out that CT scans do not detect neurona damage/dysfunction. So you are using an irrelevant ref to dispute it. That is my point. If you want to dispute a ref based on the "age of the ref", then fine lets discuss that.
 * Further, who says it is outdated? Have you got a relevant ref to dispute it?-- Literature geek |  T@1k?  19:41, 20 June 2009 (UTC)
 * I don't mean any disrespect but do you understand the difference between a neuron and a brain structure?-- Literature geek |  T@1k?  19:45, 20 June 2009 (UTC)
 * Ashton (1991) is outdated because it is 18 years old. The possibility of neural damage is her own speculation because she does not back it up by citations. This outdated speculation about neurons goes against recent negative experimental results regarding brain structures. That is how a recent review treats it :
 * Are Benzodiazepines Associated With Physiologic Changes in the Brain?
 * Given the degree of cognitive changes reported with
 * benzodiazepine treatment, several researchers have proposed
 * that anatomic or physiologic changes in the brain
 * should be demonstrable in patients with cognitive changes
 * due to benzodiazepine use. Positron emission tomography
 * research is sparse but thus far seems to indicate that
 * changes are measurable only during benzodiazepine use
 * and then disappear shortly after dose administration without
 * having a measurable effect on function. For example,
 * although cerebral blood flow to the prefrontal cortex was
 * found to be lower in the presence of midazolam, there was
 * no difference in cerebral blood flow to functioning regions
 * of the brain used during memory tasks between unmedicated
 * controls and patients who took midazolam.16 Therefore,
 * despite benzodiazepine-induced changes in the prefrontal
 * cortex, the brain was still able to perform cognitive
 * functions. Computed tomographic (CT) scans of patients
 * taking benzodiazepines long term were compared with
 * those of age- and sex-matched controls by Busto et al.,17
 * but the authors found no difference in brain atrophy in the
 * 2 groups and concluded that long-term benzodiazepine use
 * does not appear to be associated with brain abnormalities
 * as assessed by CT. The Sceptical Chymist (talk) 20:12, 20 June 2009 (UTC)


 * But the conclusion in the abstract of the paper you cite above is: "CONCLUSION: Moderate-to-large weighted effect sizes were found for all cognitive domains suggesting that long-term benzodiazepine users were significantly impaired, compared with controls, in all of the areas that were assessed. However, this study has several limitations, one being that it includes a relatively small number of studies. Further studies need to be conducted; ..." I do not have access to the complete pager. You appear to be quoting from the literature review of the paper, which is not a reliable source for this sort of conclusion. A paper's literature review is not to be used to draw conclusions. &mdash; Mattisse  (Talk) 20:24, 20 June 2009 (UTC)
 * I apologize for giving wrong PMID. I just corrected it. The Sceptical Chymist (talk) 20:33, 20 June 2009 (UTC)

Midazolam study sounds like an acute benzo dose study. It is using a short-term study to state long-term effects. I already showed that the literature is conflicting regarding CT scans. These studies are NOT on protracted withdrawal and thus are not relevant.-- Literature geek |  T@1k?  20:19, 20 June 2009 (UTC)
 * What I cite is a much newer review than Ashton's. The Sceptical Chymist (talk) 20:35, 20 June 2009 (UTC)
 * Why will you not address my points? I stated midazolam was a short term study assessing acute effects. I want to use the best quality literature, you seem to be determined to use the most dubious quality oif literature even after its faults have been pointed out. I want this to be a top quality article. What do you want?-- Literature geek |  T@1k?  20:47, 20 June 2009 (UTC)
 * The issue is what review Ashton (1991) or Stewart (2005) we prefer. Ashton's is old and does not cite any evidence. Stewart is new and cites several studies. The Sceptical Chymist (talk) 21:18, 20 June 2009 (UTC)

You ignor all my points, please address them.-- Literature geek |  T@1k?  21:26, 20 June 2009 (UTC)


 * RE: Problem 2.


 * My edit However, researchers often hold contrary opinions regarding the effects of long-term administration. : "Among the most controversial of these effects are cognitive effects...The literature is divided, however, on the persistence of cognitive effects in patients taking benzodiazepines long term." : "The literature that is available is difficult to interpret due to conflicting results as well as a variety of methodological flaws."


 * My edit: One view is that many of the short-term effects continue into the long-term and may even worsen, and are not resolved after quitting benzodiazepines. : "Long-term treatment with benzodiazepines has been described as causing impairment in several cognitive domains... The Barker et al. group followed up their first metaanalysis by addressing a question that naturally ensues: if long-term benzodiazepine use affects cognition, are these cognitive changes reversible upon withdrawal of benzodiazepines"


 * My edit: Another view maintains that cognitive deficits in chronic benzodiazepine users occur only for a short period after the dose, or that the anxiety disorders is the cause of these deficits. : "Alternatively, long-term benzodiazepine use has been reported as being unassociated with cognitive dysfunction...Often, even when cognitive effects have been observed in patients taking benzodiazepines, the effects are attributed to sedation or impaired attention; additionally, several studies report memory changes only when benzodiazepines have reached their peak plasma level, suggesting that specific cognitive changes are temporary and linked to time since last dose... Complicating the issue of the cognitive effects of longterm benzodiazepine treatment are cognitive changes associated with anxiety itself... dysfunction. Anxiety disorders have been shown to impair attention and concentration, 11 so arguably relief of anxiety may improve cognition." : "In contrast, some researchers claim little or no memory effect caused by long-term benzodiazepine use. Golombok et al. found no evidence of memory impairment in 50 patients who had used benzodiazepines for >1 year. These authors argued that there was a strong relationship between the sedative and amnesic effects of the drugs, suggesting that as patients become tolerant to the sedative effects of the drugs, memory deficits were no longer apparent. Similarly, Lucki et al. found that any impairment evident in their group of 43 long-term benzodiazepine users appeared to diminish with time after the last dose was increased. These results suggest that memory impairments, if they do occur, may be due to the acute effects of the drug and do not support the hypothesis that long-term benzodiazepine use leads to permanent memory impairment." The Sceptical Chymist (talk) 20:02, 20 June 2009 (UTC)


 * But you are cherry picking mentions of primary sources in review articles to dispute conclusions of a meta-analysis. See conclusion. "CONCLUSION: Moderate-to-large weighted effect sizes were found for all cognitive domains suggesting that long-term benzodiazepine users were significantly impaired, compared with controls, in all of the areas that were assessed."


 * I would liuke to make a third attempt to request you to address my point of using a CT scan to remove a statement on neurons.-- Literature geek |  T@1k?  20:08, 20 June 2009 (UTC)


 * Same with this rebiew, here are their conclusions. "to settle this debate, meta-analyses of peer-reviewed studies were conducted and found that cognitive dysfunction did in fact occur in patients treated long term with benzodiazepines, and although cognitive dysfunction improved after benzodiazepines were withdrawn, patients did not return to levels of functioning that matched benzodiazepine-free controls."


 * You seem to be using mentions of primary sources to challenge the conclusions of the authors, like doing your own review of their review which is original research.-- Literature geek |  T@1k?  20:10, 20 June 2009 (UTC)


 * How is it that I am cherry-picking? I am giving whole paragraphs of quotations. You wish to present only one side of the issue. However, if you read both articles you will find that they review the state of the art. Both state that there is a controversy and give arguments of both sides in details. The meta-analysis weighs overall balance somewhat in favor of reality of cognitive defects. That is what I did also by giving it more room. I can expand on the meta-analysis, if you wish. The Sceptical Chymist (talk) 20:24, 20 June 2009 (UTC)


 * The Sceptical Chymist, what you are doing is considered original research and is not allowed on Wikipedia. Please read reliable sources for medicine-related articles. &mdash; Mattisse  (Talk) 20:30, 20 June 2009 (UTC)
 * No I am not doing OR. I am using reviews, citing them almost exactly. The Sceptical Chymist (talk) 20:41, 20 June 2009 (UTC)


 * I would like to support Literature geek  |  T@1k?  on this issue, as I read the very thorough paper on the meta-analyses, and this is indeed the conclusion it reached. It also concluded that  the cognitive dysfunction might not be noticeable in everyday life, but it did conclude that it existed.  &mdash; Mattisse  (Talk) 20:17, 20 June 2009 (UTC)
 * Yes, that is exactly what I wrote. That "the former view [many of the short-term effects continue into the long-term and may even worsen, and are not resolved after quitting benzodiazepines] recently received support from a meta-analysis of 13 small studies." The Sceptical Chymist (talk) 20:41, 20 June 2009 (UTC)

You are doing your own review on a review article by using primary sources cited in the review to cast doubt on the conclusions of the authors so it is original research of the worst kind. It wouldn't be so bad if it was OR which was in keeping with the line the sources were on but you are casting doubt on the conclusions by doing your own review like making a minority opinion in the literature equal to the conclusion, i.e.e you are distorting the paper. Please quit edit warring over this!-- Literature geek |  T@1k?  21:12, 20 June 2009 (UTC)


 * RE: Problem 2. (contd.)
 * My edit: While the definitive studies are lacking... (the meta-analysis itself): From the introduction -- "Clearly, well controlled, methodologically sound studies are required which involve heterogeneous groups of subjects and multiple measures of cognitive functioning. The feasibility of such large scale studies may be limited." From the results-- "An analysis of heterogeneity was considered inappropriate due to the small number of studies that met the inclusion criteria in this meta-analysis, and the limited information provided on relevant characteristics... The small number of studies included in the meta-analysis also resulted in insufficient data to conduct a thorough investigation of the contribution of moderator variables." From the Conclusion -- "In order to fully investigate the nature of impairment after long-term use of benzodiazepines larger-scale studies, which examine many areas of memory are needed. Clearly, this is not feasible and a more likely scenario is one that involves conducting many smaller, well designed studies that thoroughly investigate certain areas of cognitive functioning and present data in such a way so as to be amenable to inclusion in a meta-analysis. Incorporating this information into a larger meta-analysis would allow for a more thorough and statistically sound investigation of the effects of moderator variables – an obvious shortcoming of the current investigation associated with the dearth of literature available." Review  discussing the meta-analysis: "In their introduction to the first metaanalysis, the authors acknowledged the small number of studies and addressed the limitations of comparing studies with different methodologies. The most obvious problems in comparing studies include variable definitions of longterm use with a wide range of doses and duration of use represented, poorly defined coexisting drug and alcohol use, and the heterogeneity of psychiatric diagnoses in both subjects and controls. Also, some studies do not define the length of time from benzodiazepine dose to cognitive testing, which may create problems differentiating acute from chronic effects. Among these limitations, heterogeneity of psychiatric diagnosis is perhaps the most significant in evaluating the meaning of results. Subjects are often selected for benzodiazepine use, not necessarily for psychiatric diagnosis, and since benzodiazepines are prescribed for a wide variety of conditions, it may be that patients with different disorders will also vary with regard to side effects and risk:benefit ratio. Furthermore, Barker et al.9 noted that subjects are frequently recruited from withdrawal clinics, which may create a sampling bias because such subjects typically have heightened concern about problems attributable to benzodiazepine use." The Sceptical Chymist (talk) 21:07, 20 June 2009 (UTC)


 * My edit: ...the former view [many of the short-term effects continue into the long-term and may even worsen, and are not resolved after quitting benzodiazepines] recently received support from a meta-analysis of 13 small studies. (the meta-analysis itself): "Of the 13 independent studies used in the meta-analysis,all were published in peer-reviewed journals. The overall mean number of patients who were benzodiazepine users was 33.5 (SD ± 28.9; range 10–96; median 21) and the mean number of controls was 27.9 (SD ± 19.6; range 10–56; median 20)." The Sceptical Chymist (talk) 21:16, 20 June 2009 (UTC)

Please stop edit warring over your original research Sceptical. We do not need to find consensus regarding original research, it is not allowed. Take debate to WP:MEDRS and try to persuade them to allow OR in FA articles.-- Literature geek |  T@1k?  21:12, 20 June 2009 (UTC)


 * Since when summarizing reviews is OR? According to WP:OR, "Summarizing or rephrasing source material without changing its meaning is not synthesis — it is good editing. Best practice is to write Wikipedia articles by taking material from different reliable sources on the topic and putting those claims on the page in your own words, with each claim attributable to a source that explicitly makes that claim." On the other hand, picking out only the sentences you like is POV since WP:NPOV policy "requires that where multiple or conflicting perspectives exist within a topic each should be presented fairly." The Sceptical Chymist (talk) 21:26, 20 June 2009 (UTC)

"without changing its meaning".-- Literature geek |  T@1k?  21:50, 20 June 2009 (UTC)

Anxiolytic tolerance
O'Brien paper used to debunk authoritative bodies etc. Sceptical claims it is "latest evidence but O'Brien study cites an opinion piece from the early 1970's from Marks J a doctor from Hoffman La Roche". Tolerance and dependence was not even accepted back in the 70's and was denied until the 1980's. The next study cited in O'Brien non-systemic review is a primary source where long term users on benzos were giev a small dose increase, no increased sedation occured but feelings of tranquility were felt when dose was temporarily increased. First off tolerance is actually defined by "having to increase the dose to achieve the same effect" so their findings are worthless. Secondaly they made a big issue out of how sleep inducing effects were not seen from dose increase but anxiolytic effects were seen. This can be explained simply by the fact (which the authors seemed ignorant of) that hypnotic effects require a larger dose of benzos!!! Sounds like a paper by authors who are not specialits in the area that they are researching! This weak review paper has been used to delete the reviews by experts in the fiel eg Ashton and disagree with expert panels, (See above).-- Literature geek |  T@1k?  20:44, 20 June 2009 (UTC)


 * It is not a good practice to represent only one view in a clearly controversial topic. Many psychiatrists in the US believe that the UK point of view represented by NICE is extreme. This found reflection in multiple articles and textbooks edited by the US-based scientists. The Sceptical Chymist (talk) 21:34, 20 June 2009 (UTC)

Where on earth did you get this statement from? It certainly wasn't in the references. "The development of tolerance has not been demonstrated in controlled studies, while anecdotal evidence goes both ways."

Only anecdotal evidence exists? Not in ref. Stop intentionally faking data and doing original research and then fighting about it. I will never compromise with faked data. Science works on evidence, please find me strong evidence, not "a doctor" who cherry picked one or 2 sources and came to dubious conclusions which you grossly worsened by adding fake facts. please find a systematic review of the peer reviewed literature and stop using weaker opinion pieces by individual doctors. I have no problem if you want to cite that some doctors disagree with this but you are faking refs, doing original research on some mission to insert what you think is "the truth".-- Literature geek |  T@1k?  21:47, 20 June 2009 (UTC)


 * The development of tolerance has not been demonstrated in controlled studies : "Tolerance. While there is controversy about whether or not patients develop tolerance to the anxiolytic effects of benzodiazepines, tolerance to some of the other effects, including the sedation generally brought on by benzodiazepines, does develop...This disapproval of long-term use was supported with the hypothesis that patients develop a tolerance to the antianxiety effects of benzodiazepines, which, if true, would mean that benzodiazepines would be medically useful only for short-term treatment, and patients would not need to use benzodiazepines long enough to develop a dependence. However, clinical evidence does not support the development of this tolerance"


 * while anecdotal evidence goes both ways. Anecdotal evidence against tolerance to anxiolytic effects -- : "in fact, anecdotal evidence from prescribing psychiatrists often indicates long-term anxiolytic effectiveness of benzodiazepines." Anecdotal evidence in favor of tolerance to anxiolytic effects, see . The Sceptical Chymist (talk) 22:14, 20 June 2009 (UTC)

He said nowhere about annecdotal evidence. He cited a single study which was dubious (see above) and then cited a doctor from Roche Pharmaceuticals. You made it say the opposite, oh my goodness. It is anecdotal the long term effecttiveness of benzos from psychiatrists. You then used it to say the opposite, hence misrepresenting refs to make them say the opposite.-- Literature geek |  T@1k?  22:24, 20 June 2009 (UTC)

Anyway i don't like fighting, I have made added what I would say is a good quality review of use of benzos for anxiety which suggested continued efficacy and showed the other side of the controversy showing tolerance evidence suggesting benzos worsen anxiety. Perhaps both views are right and some patients benefit whereas others are harmed. We are all genetically and biologically different so why not just cite both viewpoints and allow the reader to decide. I believe my edits have brought neutrality. Please review them and lets stop falling out over refs. I don't mind seeking NPOV with good quality robust refs.-- Literature geek |  T@1k?  22:24, 20 June 2009 (UTC)

POV tag
A POV tag was dded because two editors myself and Mattisse regarded edits to be original research and agreed that review articles should not be rereviewed and have doubts cast on conclusions and that we should stick to the conclusions of the author. Consensus can never be reached because original research is not allowed on wikipedia especially FA articles.-- Literature geek |  T@1k?  21:56, 20 June 2009 (UTC)

I have now cited both sides of the debate using good quality sources, citing that long term use of benzos may help anxiety (took ref from anxiety section of therapeutic uses) and the review saying benzos may worrsen anxiety. As i said in ref summary, who knows perhaps both arguments are correct, perhaps some people get continued benefits on benzos and some get "ill" on benzos, long term. I have no problem bringing balance to the article. I just simply do not like refs being faked and reviews being rereviewed, i.e. original research.-- Literature geek |  T@1k?  22:16, 20 June 2009 (UTC)


 * I suggest you read WP:NPOVD: "Sometimes people have edit wars over the NPOV dispute tag, or have an extended debate about whether there is a NPOV dispute or not. In general, if you find yourself having an ongoing dispute about whether a dispute exists, there's a good chance one does, and you should therefore leave the NPOV tag up until there is a consensus that it should be removed." The Sceptical Chymist (talk) 22:21, 20 June 2009 (UTC)

Alrighty if you insist, put it up but I will not agree with faked data and original research. Listen, I like trying to find consensus and resolving disputes. I just made an honest attempt to try and resolve the dispute. I have cited both sides of the debate now in tolerance section. Please tell me that you are happy now.-- Literature geek |  T@1k?  22:26, 20 June 2009 (UTC)

Regarding side effects section, all I did was summarise the conclusion, why are you opposed to citing the conclusion of the meta-analysis? Why do we need to then cast doubt on the conclusion by citing primary sources mentioned by the meta-analysis.-- Literature geek |  T@1k?  22:28, 20 June 2009 (UTC)


 * I disagree with both of your edits - on tolerance and on cognitive effects. They both overstate the case against benzodiazepines and understate the controversy. The literature appears to be evenly split between tho opposing views. I suggest you put two or three sentences for benzodiazepine tolerance, and I put two or three sentences against it. The same about long term effects. Then we combine them and include them into the body. The Sceptical Chymist (talk) 22:37, 20 June 2009 (UTC)

But they are a perfect interpretation of the conclusions of the meta-analysis. You may disagree with them but I can't do anything about that because those are the conclusions of the papers. Justt because you disagree doesn't mean that you can then do a personal review of the review or a review of the meta-analysis by using primary sources cited within that study to cast doubt on their own conclusions. That is original research. We can only cast doubt by using conclusions of good quality secondary sources. I cannot agree to original research. If you find a good quality source which comes to the same conclusion that you do then fine.-- Literature geek |  T@1k?  22:40, 20 June 2009 (UTC)


 * You took one sentence out of the meta-analysis ("Moderate-to-large weighted effect sizes were found for all cognitive domains") while ignoring the rest of the conclusion ("this study has several limitations, one being that it includes a relatively small number of studies. Further studies need to be conducted; ideally, well designed, controlled studies that thoroughly investigate certain areas of cognitive functioning and present data in such a way so as to be amenable to inclusion in a meta-analysis. Incorporating the information from these studies into a larger meta-analysis would allow for a more thorough and statistically sound investigation of the effects of moderator variables."). You also ignored other limitations of this meta-analysis noted in a review . That is a sure sign of POV. On the other hand, I correctly summarized the whole mess by saying While the definitive studies are lacking, the former view [many of the short-term effects continue into the long-term and may even worsen, and are not resolved after quitting benzodiazepines] recently received support from a meta-analysis of 13 small studies.


 * The same problem is with the tolerance section where the sentence you inserted "Tolerance to anti-anxiety effects develops more slowly with little evidence of continued effectiveness beyond four to six months of continued use." is POV since there is similarly little evidence in support of the tolerance to BDs' antianxiety effect. The Sceptical Chymist (talk) 10:57, 21 June 2009 (UTC)

Actually it was because the paragraph was too big as it is and I didn't want to make the section a big break down of the study listing all the limitations. Most of my edits don't include a break down of limitations or methods of a study as it can make articles untidy and unencyclopedic.-- Literature geek |  T@1k?  14:17, 21 June 2009 (UTC)

Please cite a systematic review of the peer reviewed literature saying that there is no evidence for the tolerance to the anxiolytic effects of benzodiazepines. I already cited the Committee on safety of Medicines. If you cannot find a systematic review of the literature stating your claim please retract your statement.-- Literature geek |  T@1k?  14:20, 21 June 2009 (UTC)

That is aa totally accurate representation of what the paper said. It is a cited fact, not a POV.-- Literature geek |  T@1k?  14:21, 21 June 2009 (UTC)

History
It is a minor point, but could the manufacturer of Valium be clearer in the History section. Snowman (talk) 23:03, 18 June 2009 (UTC)

Someone added that Roche Pharm patented and manufactured Valium. :)-- Literature geek |  T@1k?  21:33, 19 June 2009 (UTC)

Little issue

 * "Alcohol is also cross tolerant with benzodiazepines and more toxic and thus caution is needed to avoid replacing one dependence with another."
 * This sentence seems a little strange. It is not clear who should utilize caution. Of course the patient should, but don't physicians generally say do not mix alchol with these drugs?


 * In the "Withdrawal symptoms and management" section, the two last paragraphs begin with "Withdrawal". It would be better to vary the wording so that two paragraphs in a row do not start with the same word.

&mdash; Mattisse (Talk) 23:21, 20 June 2009 (UTC)

Hmmm, maybe it needs to be made more clear. It is about people withdrawing off of benzos including those totally off of benzos. I thought with it being in the withdrawal section people realise that it was referring to people withdrawing or withdrawn and recovering from benzos. It is not about the risks of mixing benzos and alcohol and enhanced side effects but is talking about replacing one cross tolerant drug with another cross tolerant drug, they work similarly in the brain so basically drinking alcohol is like staying on benzos and worse because alcohol is more toxic. Like a cocaine addict would need to be careful with say taking prescription ampetamines, not quite the same as stimulants don't cause strong physical dependence but you catch my drift? Thing is how do we reword it so other people understand. :) Thank you for pointing this out Mattisse.-- Literature geek |  T@1k?  00:57, 21 June 2009 (UTC)