Hi Tito,

that's a contested issue. There are many pros and cons as the inclusion of not yet 'really' published research can contribute to a bias as well as to a bias avoidance. Even the Cochrane handbook is not clear about this.

http://handbook.cochrane.org/chapter_10/10_3_2_including_unpublished_studies_in_systematic_reviews.htm

If the poster presentations have run through a peer review, it would certainly make a case for including them. However, as somebody who sits on several of such review bodies of conferences, I am sure that the review process itself is not as thorough as it is with journal papers.

Dirk

Firstly, if you have found outcomes that met YOUR inclusion criteria, then the simple answer is yes. 

However, meta-analyses commonly combine the results of studies & trawl 'the literature' implying that the studies must be 'published'.  Posters however raise all sorts of issues & defy stict definitions of 'published' in that although we put the presented info in front of perhaps 1000s of people (& perhaps also on the web), we still don't tend to regard them as published.  There has only been 1 legal case that tested this position, & it deemed that a poster is in fact a publication. Practically however, a poster will often only present the key aspects of a study, similar to an abstract 'writ-large'. Therefore it may be difficult to account for the rigor of the presented study, & also to include it as 'peer-reviewed literature' (for what its worth).

I am a great fan of posters, but in their current format as a medium of presentation, education, publication, networking etc. there is much room for improvement.  It can be simply demonstrated that you cannot 'read' the vast numbers of posters (or their abstracts) that are often on display within the alloted time. People report dissappointing levels of engagement during poster sessions. Also, posters offer a reduced amount of information & are seldom available (in full) after events.  I estimated posters to cost @ 2.2 billion USD pa, so why do we continue to produce over a million of them every year, without concern for their efficacy or 'use'?  A couple of sceptical answers spring to mind, but that is what my current research is looking into  ;-)

If your inclusion criteria included 'peer-reviewed literature' & considered the rigor & depth of results  of the presented study, then it might be difficult to include poster studies in a meta-analysis, but you can definately include them in a systematic review, on the condition that your cosen inclusion criteria are met.  The problem is, that if these criteria allow non peer-reviewd material & 'shallow' data  presentation, then some might call this into question.

Yes you can . It does depend on the previous contributors thoughts as well as how you classify the level of the work. That level is not the highest level of validity or significance, but in some topic areas, it may be where that issue is in terms of being studied.

Thank you for the responses.

The data presented in the posters were good however, we just could not determine ramdomization. I have emailed the authors and waiting for their responses.

I also talked to other authors regarding this and they all have varied insights on whether to include the study or not. Without them, I will have 7 RCTs which we believe that it is reasonable given the fact that the technique is new.

TT

Poster presentation is probably a "grey zone " between published and unpublished .

It may be mentioned in a discussion but I think one should avoid including data and statistics .

If the poster was named in a conference then yes. It would be grey literature and grey literature should be searched and included. This is an AMSTAR checklist item 4.

http://amstar.ca/Amstar_Checklist.php

I think it really boils down to what the editor and/or peer reviewers of the manuscript that you submit believe about including posters. I would not go too far down this road until you have contacted the editor of your target journal; otherwise, you may have wasted your time. 

I agree with comments above regarding the difficulty for others to  "search" all poster presentations on a topic. While meta analyses are by definition reviews of the literature, are poster presentations "literature"? Including poster presentations may make it difficult for others to replicate your work (they may not have access to the scientific organization's poster presentations). The question then becomes, have you searched ALL posters on your topic. Also, it is difficult to know if all conferences' posters were peer-reviewed; some conferences take all comers. This whole topic is related to publication bias; not all posters eventually get published or are even accessible to others.

Personally, as a peer reviewer, I would be a bit skeptical of posters, so again it depends on who reviews your paper. 

John:  Even the greatest fan of posters is going to be sceptical of them, simply because their peer-review is either impractical (1000s may be delivered in a single session), or more commonly that the peer-review process centres on checking only that the submission (normally the abstract & rarely the poster) conforms to the meetings format requirements.  As for searching 'all of the posters' ... that really is impossible.  Some are reported, some are not - some you can access, some you can not ...  @ 30% of posters are developed into a full paper - the rest just seem to get lost.  I have been researching posters since 2009 & although I can put a conservative estimate at over 1 million being produced every year, I have really no idea how we can track them all. 

At larger international conferences, you could not even read the titles of all the posters that are provided in the abstract book, unless you devoted some serious time to non-stop reading (a good reader will read @ 300 effective wpm). Given that abstracts are @ 300-500 words in length, & posters have 500-1500 words, just the math alone proves that we cannot see all that is on offer, even if we wanted to.  But we still devote vast amounts of time, effort & money to poster presentation.

That said though, there is no reason why the medium can't be developed to be more efficient (& reliable) in disseminating quality information.  We just have to re-focus on what we want them to do (other than justify a funding application to attend a certain conference)  ;-)

 Nicholas, I totally agree with you about the impracticality (if not impossibility) of searching all posters. And I would definitely be skeptical. As an author I would not include them and would  teach my junior faculty or students NOT to include them either. 

There are other issues with posters, too, which Tito alluded too--such as not being able to tell (in detail) the methods such that you might not even know if a clinical trial was randomized, or if there was intention to treat analysis, as well as a whole host of other potential flaws. The text is usually just too short. 

thanks!

John, I also agree.  I do feel though that given the potential of posters, we should do more to develop the medium.  I would like to see the professions, the research funders, as well as the conference industry, do more in this regard (it is a mutual situation).  As it stands though, we are not getting the most out of the medium (but still spending a fortune on it).

Tito Tubog estimated

The question you raise is controversial.

The posters (posters) Well-designed and manufactured, are attractive and an excellent teaching technique to publicize the results of the work done by researchers in training.

As the space allocated for this purpose in general is not much information an investigation has to be handled in abstract form, where probably more extensive and detailed part is the section on methodology, procedures and results.

The information and knowledge derived from it, can be original and valuable, but usually the selection and evaluation of the research subject or to present in this mode (poster, poster) is not as rigorous as when publishers sent the prestigious biomedical journals, which have widely, expert reviewers (pairs) and the quality of its contents, provide researchers with an adequate score (impact) that can translate into their curricula.

On the other hand, in the Meta-analyzes and Systematic Reviews Based on Meta-analysis, publications that meet quality criteria to be considered truly human experiments are included, and these publications are "Randomized Controlled Trials (often double-blind) "

If the poster is the means by which they spread the results of research work already published in a biomedical journal and its design it is that of a "Clinical randomized controlled trial" if it can be included in a Meta-analysis and Systematic review, but if you do not meet quality criteria and with this type of design, you can not use it.

Cheers

Sincerely

Dr. Jose Luis Garcia Vigil

A systematic Review should definitely include grey literature. Searching for grey literature may be a challenge, but the bigger international conferences normally publish their abstract as supplement in a journal and some of those are accessible via databases.

In metaanalysis the best solution in my opinion would be to include all available data meeting the inclusion criteria and do a sensitivity analysis excluding grey literature and low quality papers.

That depends on the paucity of data in the area you are researching. If you decide from the outset when you design the study that for what ever reason (I would assume justifiable) that you are going to include posters then this question should not arise. You say that the poster "met our inclusion criteria", I assume that the sort of literature to be included was already established and that this indeed included posters. If this did not include poster at the outset then you shouldn't change them to include the two posters as you will be introducing bias. This question differs from one of whether posters should be included at all, a question that should be asked before the literature is searched.... 

Following the Cochrane guidelines, all source of data has to be included in the main analysis of a meta-analysis. Of course the quality of an abstract will be poor, and sensitivity analyses excluding the poor quality article(s) must thus be done. If excluding the low quality studies the results does not change, no problem. If results change sensibly, you have to take into account this aspect for a correct interpretation of results.

Cheers,

Francesco

Tito, what is  your a-priori inclusion criteria for studies to be included in the SR, if it didn't include abstract, I wouldn't dilute the SR ! It may be very tempting to include outcome of interest from these poster/abstracts, but if the poster is not published in a an index journal as a peer-reviewed abstract, you will invite a lot criticism ! 

Hi Tito, In my views the systematic review should include the data from the published journal articles. If your poster has the data source and published also then, it can be included in the review. As such the review is review of all studies It doesn't say that you can'nt include the posters. But the quality of your review increases only when it has included major studies published .

Another convenient aspect to comment at this time of discussion is whether to include information and results emanating from the so-called "gray literature"

I have no doubt that there are excellent works that have not passed through the filter of peer review (pairs) of the Editorial Committee of an indexed biomedical journal (RB), but they are not the majority.

Comparing "gray quality literature" extracted from the databases of the page (Web) of the National Institutes of Health in the United States, where it is recorded and published the progress of research protocols with the same work one once published in an indexed RB, the concordance is around 60-70%, indicating that many of the works were not completed, or the type of results, the authors decided not to submit to the process of editorial review one RB.

In short, consider the results of research published in abstract form in a poster, induces a greater bias if data already published results are included in RB (in which it is fair to mention, they have also identified several types of bias that invalidate or give little validity to some research).

Sincerely

Dr. Jose Luis Garcia Vigil

It depends on your inclusion criteria and if all the information is available in the poster.

1: If you opted to accept grey literature, you should accept posters as an information source. Otherwise, you shouldn't include it.

2: Posters usually don't have all the data you have proposed to abstract. Sometimes, e-mailing the authors will solve this issue.

In a Meta-analysis and Systematic Review Based on Meta-analysis, critical to include published information requirement is that the primary source is a Randomized, Controlled Clinical Trial has been published in a Biomedical Journal indexed with group publishing and expert reviewers (pairs).

In principle, the summary of a poster as complete and quality information for inclusion in the Meta-analysis and Systematic Reviews is not accepted. for the reasons given in my previous involvement related to "gray literature"

The only exception to include a sign on Meta-analyzes and systematic reviews, is it a poster where the summary of a randomized controlled clinical trial previously published in a biomedical journal is indexed.

Data in pester or abstract presentation are not complete and the results are affect by this drawback

I used a conference poster in one of my systematic reviews, where the CSR (Clinical Study Report) is available at https://clinicaltrials.gov/.

Correct

Poster Research

As this thread concerns posters, I would like to draw you attention to some research I am doing. As we can see, there is debate over the value and validity of poster-presented information. I am conducting some interviews about people's views & experiences of conference posters. To participate, all you need is to be able to answer 'Yes' to the following question:

• Have you had direct experience of international Academic/Scientific conferences that use the poster medium ?

The interviews are conducted by exchanging questions & answers here on the Research Gate platform by private message. No one else can see your responses & you can write as much or as little as you like. No particular attention is given to language or presentation. We can either swap individual questions & answers (you can still ask things & expand like in a normal interview), or you can treat it like a survey & get all the questions at once.

More details are available in the attached invitation, or you can contact me by message on RG, or at: [email protected].

I hope you will take part & share your views on what is a very under-researched medium :-)

Regarding systematic reviews are of two types:

It is the second type of Systematic Reviews, where if you can include information and published as a poster.

In my view the data can be included if it is a complete study. And some of the conferences do publish good data in poster abstracts. 

Syed - what is your systematic review on?  Also, where are you finding this body of poster work?