Abstract to publication rate: Do all the papers presented in conferences see the light of being a full publication?

BACKGROUND: Every year the scientific sessions of Annual National Conference of Indian Psychiatric Society (ANCIPS) are marked by presentation of free papers, posters, and award paper sessions, which are usually meant for presentation of new research which is not yet published. Hence, it is expected that these papers will be published in near future so that the scientific literature is distributed and shared with wider audience. AIM: This paper aims to evaluate the abstract to publication rate of papers presented during ANCIPS in the years 2012-2014.
MATERIALS AND METHODS: For this study, all the free papers, posters, and award papers presented during the ANCIPS of 2012-2014 were listed, and electronic searches were carried out to search for published articles. In addition, one of the authors of papers not found in the electronic searches were contacted through E-mail.
RESULTS: A total of 1081 papers were presented during the ANCIPS in the 3 year period under study. Of these, 64 were award papers, 622 were free papers, and 395 were posters. Majority (n = 807; 74.6%) of these could be categorized as research data-based presentations; this was followed by case reports/series (203; 18.8%), review of literature (n = 35; 3.3%), and others (n = 36; 3.3%). Overall, only 27% of the papers were published after at least 5 years of the presentation. Of all the award papers, 69.6% of papers were published, whereas only 26.8% of free oral papers and 22.5% of free posters were published. About half (45.6%) of the papers were published in national journals. In terms of indexing, among those which were published, 62.8% were published in Medline-indexed (PubMed-listed) Journals with a mean impact factor of 1.
CONCLUSION: The present study shows that only 27% of the abstracts presented during the ANCIPS are ultimately published as full text articles in the next 5 years. Copyright:

INTRODUCTION

The basic purpose of the scientific conferences is to share the knowledge, including sharing of unpublished research. Every year, Indian Psychiatric Society (IPS) organizes its Annual Conference, which is attended by a large number of psychiatrists and other mental health professionals. The scientific activities during the conference involve plenary lectures, award papers, symposia, workshops, free papers, and posters. The award paper session involves presentation of selected papers for the various categories. In contrast, the free paper and poster session mostly involves presentation of unpublished research.

According to the various principles of publication ethics, it is in general believed that every research activity involves lot of effort on the part of the investigator(s) and also the participants.[1] Hence, it is suggested that every research must be ultimately published. It is to be understood that as such abstracts of the conference proceedings are not indexed and are not considered as publications.[2] Hence, it is desirable that all the abstracts presented in various conferences must be converted into full manuscripts.

There are different ways to evaluate the quality of papers presented in different conferences. One of the parameter which has been used to evaluate the quality of presentations made in different conferences is abstract to publication rate, which is basically understood as the proportion of abstract presented in a conference, which are ultimately published in peer-reviewed journal in subsequent time frame. Studies from different parts of the world, involving various specialties, which have included 149–1897 abstracts, have evaluated the abstract to publication rate, and these suggest that 3.8%–78% of the abstract are published, with oral papers more likely to be published than the posters.[345678910111213141516171819202122232425262728] Occasional studies have taken abstracts of specific type of studies, such as randomized controlled trials, rather than all the abstracts presented in a particular year.[4] The authors of these studies have evaluated the publication rates, after 1–5 years of the conference, mostly by searching for the paper in the search engines such as Medline/PubMed, Google scholar, Embase, Scopus or local scientific research base (for example, Korean Medical Database). Only one study additionally have contacted the authors of the abstract to find the publication status of the papers.[21]

In terms of specialty, the assessment of abstract to publication rate has been evaluated in one or more studies in the specialties of ophthalmology, orthopedics, otorhinolaryngology, radiology, gastroenterology, emergency medicine, cardiology, urology, endocrinology, nephrology and urology, pediatrics, and internal medicine.[345678910111213141516171819202122232425262728]

Although every year, many psychiatric conferences are organized across the globe, there are limited number of studies in psychiatry on this issue. In a study from United States of America, the authors evaluated the 658 abstracts of American Academy of Child and Adolescent Psychiatry Annual Meeting of 2012 and 2013. In their PubMed search, the authors reported 46% of the papers to have been published after 4 years of the conference.[25] In another study from Turkey, the authors evaluated the abstracts of poster presentations of National Congress of Child and Adolescent Psychiatry held during the years 2005–2008. In their PubMed and Google Scholar search after 5 years, authors reported 25.2% of the papers to have been published.[22]

We could not locate any study from India, which has evaluated the abstract to publication rate in any specialty. This provided the impetus for this study. This study aimed to evaluate the abstract to publication ratio of papers presented during the Annual National Conference of IPS (ANCIPS) during the years 2012–2014.

MATERIALS AND METHODS

For this study, we took the abstracts of the ANCIPS held in the year 2012, 2013, and 2014. We limited ourselves to 2012–2014, in view of the existing literature, which suggests that a time lag of 1–5 years is required for the abstracts to be published.[345678910111213141516171819202122232425262728] For this study, we took the abstracts of all the free papers/oral papers, posters, and award papers. We limited ourselves to these because of the fact that these presentations usually involve presentation of unpublished research. As this study did not involve collection of any patient specific data, no ethical clearance was sought for this study.

First, all the abstracts were listed, using a predesigned pro forma, which included information on year of presentation, title of the paper, name of all the authors, institute of origin of the paper, type of presentation, type of data covered in the paper (i.e., case report/series, original data, and review of literature), and E-mail address of correspondence, if available.

All the data were coded/listed accordingly, and then, Internet searches were carried out by using PubMed, Google Scholar, Scopus, and Google search engines. The searches involved use of various permutations and combinations involving the title of the paper, name of all the authors, name of one of the authors, and key words involving the topic covered. Whenever a match was found, it was compared with the published abstract as part of the Indian Journal of Psychiatry Supplement, and if the data matched in both the published paper and the abstract of the full published paper, then the abstract was considered to be published. If the published paper included more or nearly equal number of patients than the abstract of the conference, the data published as part of the abstract was considered to be published. Similarly, if the published paper, included at least one of the aspects included in the abstract, then it was considered to be published. If the data included in the abstract were found to have been published as part of more than one paper, then it was considered as published, and the higher impact factor of the journal, among the various journals was considered for analysis.

In addition, we search the IPS directory to search for the E-mail address of one of the authors of the paper and listed the same. If we did not find any match for the abstract in our Internet search, the authors were contacted through E-mail, requesting them to tell us whether the scientific data included as part of the abstract has been published or not. The researchers were also asked to furnish the citation of the work, if that has been published.

Depending on the journal in which the articles were published, the articles were categorized as published in International journals (defined as those published from country other than India), Indian journals of National level (Journals of National Professional Associations), and Zonal or State journals (Journals of Zonal or state level Professional Associations). Further, depending on the Indexing status, journals were also categorized as Medline indexed (PubMed listed) and non-PubMed indexed. In addition, a note of journals indexed only in Scopus, Embase, and Copernicus but not in PubMed was also made.

All the data generated were analyzed using SPSS-14. (SPSS for Windows, Version 14.0. Chicago, SPSS Inc.). Frequencies and percentages were calculated for the discrete variables. Mean, standard deviation (SD), median, and range were calculated for the continuous variables.

RESULTS

As shown in Table 1, during the 3 years under the study, 1081 papers were presented as part of the free (oral) papers, posters, award oral papers, and award posters. Maximum numbers of papers were presented during the year 2013, and this was followed by 2014, and least number of papers were presented in 2012. In terms of type of papers, during all the 3 years, free (oral) papers were the most common type of papers. When the papers were evaluated in terms of type of data covered, majority of the papers in all the years included data emerging from prospective studies and this was followed by case reports/series.

Table 1

Distribution of various papers and type of papers

	2012 (n=277), n (%)	2013 (n=457), n (%)	2014 (n=347), n (%)	Total papers (n=1081), n (%)
Type of presentation
Free (oral) papers	233 (84.1)	217 (47.5)	172 (49.6)	622 (57.5)
Posters	28 (10.1)	227 (49.7)	140 (40.3)	395 (36.5)
Award oral papers	16 (5.8)	13 (2.8)	17 (4.9)	46 (4.3)
Award posters	-	-	18 (5.2)	18 (1.7)
Type of study
Data-based prospective study	197 (71.1)	300 (65.6)	231 (66.6)	728 (67.3)
Data-based retrospective study	17 (6.1)	43 (9.4)	19 (5.5)	79 (7.3)
Case report and series	43 (15.6)	84 (18.3)	76 (21.9)	203 (18.8)
Topic review	1 (0.4)	18 (3.9)	16 (4.6)	35 (3.3)
Not known/others	19 (6.9)	11 (2.4)	3 (0.9)	33 (3.1)

Initial search for the various publications yielded 232 abstracts being published. For the remaining 849 papers, 372 mails were sent to one of the authors. Out of the various authors contacted, 80 provided the information either in the form of articles not being published or articles being published. Overall information provided by the author's added information about 64 more published articles. Many of the researchers, who informed that the research was not published, referred to the fact that the paper was presented by their students and they have not been able to get it published.

When the abstract to publication rate was calculated, as is evident from Table 2, overall 27% of all the papers were published. In terms of type of papers, 69.6% of the award oral papers were published subsequently. This was followed by 44.4% of award posters being subsequently published. In terms of free paper, 26.8% of free (oral) papers were published and only 22.5% of the free posters were subsequently published.

Table 2

Number and types of papers published

	2012 (n=277), n (%)	2013 (n=457), n (%)	2014 (n=347), n (%)	Total papers (n=1081), n (%)
Type of presentation
Free oral papers	73 (31.3)	48 (22.1)	46 (26.7)	167 (26.8)
Free poster	8 (28.6)	53 (23.3)	28 (20.0)	89 (22.5)
Award oral paper	10 (62.5)	11 (84.6)	11 (64.7)	32 (69.6)
Award poster	-	-	8 (44.4)	8 (44.4)
Total	91 (32.85)	112 (24.5)	93 (26.8)	296 (27.0)
Type of study
Data-based prospective study	59 (29.9)	81 (27.0)	71 (30.7)	211 (29.0)
Data-based retrospective study	7 (41.2)	8 (18.6)	2 (10.5)	17 (21.5)
Case report and series	22 (51.6)	22 (26.2)	16 (21.05)	60 (29.5)
Topic review	0	1 (5.5)	3 (18.75)	4 (11.4)
Not known/others	3 (15.8)	0	1 (33.3)	4 (12.1)

When the published articles were categorized on the basis of type of journal in which these were published, it was seen that majority of the papers were published in national journals, with only two-third of the papers being published in Medline-indexed journals. The mean impact for of the Medline-indexed journals was 1.02 (SD – 3.2; median – 1.00; range – 0.04–8.44) [Table 3]. In terms of specific journals, most of the articles were published in Indian Journal of Psychiatry followed by Indian Journal of Psychological Medicine and Industrial Journal of Psychiatry [Table 4].

Table 3

Number and types of papers published

	International journals, n (%)	National journals, n (%)	Zonal journals, n (%)	State/others journals, n (%)	Indexed journals#, n (%)	Only Medline indexed journal, n (%)	Impact factor$ (n=186) (mean)
Free oral papers (n=167)	65 (38.9)	74 (44.3)	14 (8.4)	14 (8.4)	133 (79.6)	94 (56.3)	1.04
Free poster (n=89)	28 (31.5)	39 (43.8)	14 (15.7)	8 (9.0)	70 (78.7)	58 (65.2)	0.92
Award oral paper (n=32)	10 (31.3)	17 (53.1)	3 (9.4)	2 (6.3)	28 (87.5)	28 (87.5)	1.14
Award poster (n=8)	2 (25.0)	5 (62.5)	0	1 (12.5)	6 (75.0)	6 (75.0)	0.96
Total (n=296)	105 (35.5)	135 (45.6)	31 (10.5)	25 (8.4)	239 (80.7)	186 (62.8)	1.02

#Medline indexed, Embase, Scopus, Copernicus; $Impact factor - Only for Medline indexed journals (median - 1.00; SD - 3.2; range - 0.04-8.44). SD – Standard deviation

Table 4

Journalwise distribution of published papers (with at least 10 published papers)

	Journal
	Indian Journal of Psychiatry	Indian Journal of Psychological Medicine	Industrial Journal of Psychiatry	Asian Journal of Psychiatry	Delhi Journal of Psychiatry
2012 (n=91), n (%)	11 (12.1)	5 (5.5)	2 (2.2)	5 (5.5)	8 (8.8)
2013 (n=112), n (%)	13 (11.6)	16 (14.3)	3 (2.7)	6 (5.4)	2 (1.8)
2014 (n=93), n (%)	17 (18.3)	7 (7.5)	8 (8.6)	1 (1.1)	2 (2.2)
Total (n=296), n (%)	41 (13.9)	28 (9.5)	13 (4.4)	12 (4.1)	12 (4.1)

When the institute of origin of the authors for the various papers was evaluated, maximum number of papers were from Postgraduate Institute of Medical Education and Research, Chandigarh, and this was followed by papers from National Institute of Mental Health and Neurosciences, Bengaluru, Central Institute of Psychiatry, Ranchi, and Banaras Hindu University, Varanasi [Table 5]. In addition, publication rate as per the various institutes was evaluated. For this, institutes with at least 10 papers during the 3 years under the study were included in the analysis. As is evident from Table 5, the publication rates for various institutes varied from 7.7% to 60.0%, with the publication rate close to 30% for most of the institutes.

Table 5

Institutewise distribution of papers

Names of the institutes	2012 (n=277), n (%)	2013 (n=457), n (%)	2014 (n=347), n (%)	Total number of papers presented (n=1081), n (%)	Total number of papers published (n=296), n (%)
Postgraduate Institute of Medical Education and Research, Chandigarh	28 (10.1)	30 (6.6)	32 (9.2)	90 (8.3)	54 (60.0)
National Institute of Mental Health and Neurosciences, Bengaluru	11 (4.0)	21 (4.6)	14 (4.0)	46 (4.3)	13 (28.3)
Central Institute of Psychiatry, Ranchi	9 (3.2)	19 (4.2)	12 (3.5)	40 (3.7)	11 (27.5)
Banaras Hindu University, Varanasi	23 (8.3)	13 (2.8)	4 (1.2)	40 (3.7)	12 (30.0)
Kasturba Medical College, Manipal	7 (2.5)	15 (3.3)	12 (3.5)	34 (3.1)	9 (26.5)
Sawai Man Singh Medical College, Jaipur	5 (1.8)	11 (2.4)	13 (3.7)	29 (2.7)	7 (24.1)
King George’s Medical University, Lucknow	10 (3.6)	11 (2.4)	7 (2.0)	28 (2.6)	8 (28.6)
Institute of Human Behaviour and Allied Sciences, New Delhi	5 (1.8)	4 (0.9)	16 (4.6)	25 (2.3)	4 (16.0)
All India Institute of Medical Sciences, New Delhi	6 (2.2)	7 (1.5)	10 (2.9)	23 (2.1)	11 (47.8)
Seth G.S Medical College and KEM Hospital, Mumbai	5 (1.8)	4 (0.9)	9 (2.6)	18 (1.7)	5 (27.8)
Lady Hardinge Medical College, New Delhi	5 (1.8)	2 (0.4)	6 (1.7)	13 (1.2)	1 (7.7)
Government Medical College, Trivandrum	4 (1.4)	4 (0.9)	4 (1.2)	12 (1.1)	3 (25.0)
Christian Medical College, Vellore	2 (0.7)	3 (0.7)	6 (1.7)	11 (1.0)	1 (9.09)
Armed Forces Medical college, Pune	1 (0.4)	3 (0.7)	7 (2.0)	11 (1.0)	3 (27.3)
Dr. DY Patil Medical College, Hospital and Research Centre, Pune	2 (0.7)	3 (0.7)	6 (1.7)	11 (1.0)	2 (18.2)
Others	154 (55.6)	307 (67.2)	189 (54.5)	650 (60.1)	153 (23.5)

When the time lag from abstract to publication was evaluated, based on the definition of months taken from the conference to actual publication of the paper, the mean time lag was 20.8 (SD – 14. 9; median = 17; range: 2–69) months for all the publications. When the time lag for the each year was evaluated, the time lag for 2012 was 20.96 (SD – 15.48; median = 17; range: 2–69) months, 2013 was 22.1 (SD – 14.9; median: 17; range: 2–60) months, and 2014 was 17.54 (SD – 14.1; median: 12; range: 2–54) months. The time lag for publication of various award papers was 21.8 (SD – 14.41; median: 19.50; range: 2–55) months and that for free papers/posters was 20.05 (SD – 14.9; median: 15; range: 2–69) months.

DISCUSSION

The basic purpose of medical research is to generate information which can be generalizable and be used to improve the knowledge and outcome of various diseases/disorders. To achieve this basic goal, all the research must be published. A way of dissemination of research information is presenting the findings of the research in the conferences. However, the proceedings or the abstracts of the conferences are usually not indexed and are usually not available to others to refer to the conducted research. Abstracts of various research papers form the proceedings of various conferences, but these are not considered as publications in real sense.[5] Due to this, it is suggested that all the research papers presented as part of the conferences must be subsequently published as full articles.

Evaluating the abstract to publication rate for a specialty or a conference of specific national society can help in judging the quantity and quality of research in a particular specialty.[29]

The present study shows that about only one-fourth of the papers presented as part of the free (oral) papers, posters, and award papers sessions during the ANCIPS get published in the subsequent 5 years. When the findings of the present study are compared with the existing literature, which suggest that the abstract-publication rate is 3.8%–78% across various specialties, then the finding of the present study can be considered as falling in the above reported range.[345678910111213141516171819202122232425262728] However, when one attempts to have a closer look at the available studies, it is evident that most of the studies[345678101112131416171819202123252627] have reported higher abstract to publication rate than that found in the present study. Further, it is important to remember that in these studies authors have mostly themselves calculated the abstract to publication rate and have limited themselves to the Medline-indexed journals only and have not contacted the authors for checking the publication of the abstract. In contrast to these studies, in the present study, we not only included non-PubMed-listed publications but also contacted the authors to check whether the corresponding research was published or not. When one tries to compare the abstract to publication rate of the present study, with the available studies for the specialty of psychiatry, which reported publication rates of 25.2% and 46%, again the findings of the present study can be considered to be much lower, at least to one of the 2 studies.[2225] Taken together, it can be said that the abstract to publication rate is lower in the present study than most of the previous studies which have evaluated the abstract to publication rate involving various other specialties and for psychiatry from other countries.

It is important to understand the reasons for low publication rates. The various reasons for the low publication rates could be lack of availability of proper journals to disseminate the results, lack of motivation to publication, lack of mentorship from the senior authors to guide the budding researchers to write the manuscripts, and lack of any incentive for publication of research. Other issues, which could possibly act as demotivating factors, could be authorship issues, conflict of interest, conflict among the researchers, and lack of ethical clearance for the study.

The Medical Council of India has made it mandatory for the post graduate students to at least present one poster presentation, to read one paper at a national/state conference and to present one research paper which should be published/accepted for publication/sent for publication during the period of his postgraduate studies so as to make him/her eligible to appear at the postgraduate degree examination (Medical Council of India-Post Graduate Medical Education Regulations. Amendment, 2017). This suggests that in times to come, there would be more presentation from postgraduate students in various conferences. This can be considered as a step in the positive direction. However, there is a need to also look at the factors which can lead to improvement in publication of this research.

Previous studies also suggest that compared to posters, free (oral) papers are more often published.[17212427] Findings of the present study also support the same and suggest that the scientific committee of the conference should make provision for accommodating more free (oral) papers. The lag period for publication for various papers in the present study was 20.8 months with a median of 17 months. This finding is also comparable with the existing literature[12182125] and suggests that most of the papers are published within 1.5 years of the conference. Another important finding of the present study was publication of award papers and the free papers, in journals with nearly equal impact factor. There could be many reasons for the same. First, although it is not mandatory to publish the award papers in Indian Journal of Psychiatry (IJP), many authors submit their award to the IJP. At the same time, authors who present free papers and posters too publish their papers in the same journal.

What can be done at the level of Indian Psychiatric Society to improve the abstract to publication rate?

IPS as an organization has already initiated some of the efforts to address the issue of low research and publication outcome. One of the important steps include formation of a publication task force, which over the years have organized different workshops with respect to improving the publication and presentation skills of young researchers. Second, the IPS has initiated multicentric studies which have provided the opportunities to young researchers to collaborate. Third, recently, it has been decided to increase the number of issues of IJP from 4 to 6 per year. This would possibly help the authors to publish more papers in IJP. Fourth, IPS has also come up with its own first publication on research and publications.[29] Fifth, in last few years, IPS has started with poster award for young researchers, which are supposed to be based on the postgraduate thesis. This is a welcome step and possibly would motivate the young researchers to write their papers. However, the criteria for these poster awards do not make it mandatory to submit the full paper for competing for the poster award. Hence, modification of the criteria in such a way that it requires submission of full papers to compete for the award, can in one way, make the young researchers to write the full papers before the conference itself. Once the draft of the papers is ready, it is always a step forward for ultimate publication. This is also reflected by the findings of the present study, which suggest that compared to free paper and posters, chances of publication of award paper is double or more.

However, still many things require attention. Various organizations follow more rigid criteria for selection of the papers for presentation during the annual conferences. For example, All India Ophthalmological Society use 2 level of scrutiny for selection of papers for presentation during the annual conference.[30] At the initial step, papers submitted for the free paper sessions are evaluated by the scientific committee. Authors of the papers which are selected at the level 1 are advised to submit the full text of the paper for further evaluation. If the full text papers are not uploaded within the given time frame, then the paper is automatically rejected. This strategy in one way compels the authors to prepare at least a draft of the full paper of the abstract submitted before making the actual podium presentation. IPS can take a leaf from here, and the scientific committee for the annual conference can follow a similar pattern. This could possibly definitely improve the abstract to publication rate. In majority of the cases, the abstracts are submitted by the postgraduate trainees or senior residents, who want to attend the conference. Many institutes make it mandatory that residents need to have at least one presentation to attend the conference, and resultantly, they work hard for collection of data, analysis, and preparing the presentation. However, once the conference is over, depending on the stage of the career and need, the data are mostly not pursued further and it is left almost to the senior authors to convert the abstract to a full manuscript. Possibly, this contributes to lack of publication of the abstract. However, if the trainees by default have to prepare at least a draft of the paper before the conference, it is likely that they would more often be able to convert the same to a full paper. Another strategy which can be followed is that for all award sessions, submission of full text must be made mandatory. A further innovation can involve having awards for free paper and poster session with the precondition that for the same, full papers have to be submitted. For example, this strategy is followed by the IPS-North Zone, which allows all the papers presented in the free paper session of the zone to compete for the BPSS award, provided the authors submit the full text of the paper for evaluation.[31] Further, having more frequent workshops, focusing on publication ethics, and research methodology including statistics can help the young researchers in pursuing their research to logical conclusion of having a publication with their name. In addition, the publication committee can devise strategy to develop a pool of research mentors for young colleagues in writing the manuscripts. This can help the people who lack skills to write scientific papers.

The present study has certain limitations. First, all the authors contacted for the information did not respond. Hence, it is quite possible that some of the studies could have been missed. Second, this study did not evaluate the reasons for not publishing the research by the various authors who made presentations during the conference. Fourth, we did not specifically evaluate the quality of research being presented in the ANCIPS.

CONCLUSION

The present study shows that only 27% of the abstracts presented during the ANCIPS are ultimately published as full text articles in next 5 years. The abstracts of award papers sessions have more chance of being published with publication rate of 44.4% to 69.6%. Accordingly, there is a need to devise strategies to improve the conversion of abstract to full text manuscripts.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.