Since the foundation of the Section of Prenatal Biology and Neonatology within the Asociación Española de Pediatría (Spanish Association of Pediatrics) in 1967, the precursor of what is known today as the Sociedad Española de Neonatología (SENeo, Spanish Society of Neonatology), 29 congresses of the SENeo have been held.1,2 As is the case in congresses of other specialities, in recent years there has been an increase in the number of communications submitted to the congress, the number of participating hospital and even the submission of international works. The quality of the research has increased, with a growing proportion of multicentre studies, clinical trials, studies with larger samples, etc.3 However, we still do not know whether this increase in the number of works presented in the congress subsequently translates to an increase in relevant publications in the scientific literature.

The knowledge obtained through clinical studies is disseminated by means of communications in scientific congresses or publications in scientific journals. The former can take place in the form of oral communications (OCs), which present the findings of the most relevant studies, or posters, used to present clinical cases or less salient research.4,5 Presentation in congresses allows an initial reporting of research findings that is less demanding, quicker and more approachable, but does not have the universal reach nor require the rigour or methodology of articles published in scientific journals, which are considered the gold standard for the dissemination of scientific knowledge.4–6 There is evidence that a lower proportion of studies based on surveys or with an observational design are published compared to experimental studies and clinical trials, and that studies with a higher level of evidence are more likely to be published.7,8

An increasingly large number of studies need to be screened for publication, a task carried out by reviewers of congress communications and editors of scientific journals, who select the works that contribute something original, unknown and clinically relevant, so the existence and application of quality criteria for assessment of research output is of the essence.9 Usually, the publication of scientific output in an indexed journal goes a step further, as this is a stricter filter that screens out a significant percentage of the works.10 Thus, it would be interesting to find out the number of works presented in a congress that are subsequently published, as it may be indicative of the scientific quality of the congress.

The aim of our study was to determine the impact of the studies presented at the biennial congresses of the SENeo in the scientific literature, both in absolute terms and through bibliometric indices for the published works, an aspect that not been studied to date in to neonatology conferences.

The normality of the distribution for the studied variables was assessed by means of the Kolmogorov-Smirnov test. We expressed variables as mean and standard deviation if they followed a normal distribution and otherwise as median and interquartile range (IQR). We calculated the cumulative impact factor for each congress by adding the impact factors of the journals for each publication corresponding to works presented in each of the congresses. We used the χ2 test and the odds ratio to explore the variables that could be associated with the probability of publication following OC. Subsequently, we performed a multivariate analysis using binomial logistic regression including the variables with a statistically significant association in the univariate analysis to predict the proportion of publication through the calculation of odds ratios, 95% confidence intervals and the Wald test statistic to assess significance. The dependent variable in the multivariate analysis was the publication of the OC. The level of significance was set at 0.05. We performed the statistical analyses with the software package SPSS version 22 (IBM SPSS Statistics; Armonk, NY, USA).

Of a total of 525 OCs, 212 (40.38%) were associated with a published scientific article. Table 1 presents the characteristics of the presented OCs, the percentage of OCs that were published and the metrics of the publications for the 3 congresses under study. The hospitals that presented the greatest number of OCs were La Fe (38), Gregorio Marañón (37), 12 de Octubre (31), Sant Joan de Déu (28), La Paz (26), Vall d’Hebron (19), Río Hortega (18), Complejo Insular de Las Palmas de Gran Canaria (18), Puerta del Mar (14) and Miguel Servet (12). The hospitals with the greatest percentages of OCs associated with publication (out of those that presented more than 5 OCs over the 3 congresses) were San Cecilio (100%), Clínic de Barcelona (77.7%), Hospital Puerta del Mar (71.4%), Hospital del Mar (66.7%), Complejo Hospitalario Universitario A Coruña (63.6%), Complejo Insular de Las Palmas de Gran Canaria (61.1%), Hospital Puerta de Hierro (60%), La Fe (57.9%), Sant Joan de Déu (57.1%), Complejo Asistencial Universitario de León (55.5%), Hospital Gregorio Marañón (54.0%) and 12 de Octubre (51.6%). Considering the publications in journals in the first quartile, the hospitals with the greatest number of presentations were: La Fe (12), Complejo Insular de Las Palmas de Gran Canaria (8), Hospital Puerta del Mar (7), Gregorio Marañón (5) and Sant Joan de Déu (5). Fig. 1 presents the data on the number of presented OCs and published OCs by hospital and OCs published in first-quartile journals by hospital ordered by decreasing proportion of publication.

Figure 1.

Graph of the number of oral communications (OCs) and published OCs by hospital and OCs published in first-quartile journals ordered by decreasing proportion of publication (calculated as published OCs/OCs) for the 2017, 2019 and 2021 SENeo congresses. The graph only shows hospitals that presented more than 5 OCs over the three congresses. It includes the 12 hospitals with the highest proportion of publication. The values of the right axis reflect the absolute number of OCs or of publications, and the values on the left axis the percentage of the OCs with subsequent publication.

C. I.: Complejo Insular.

(0.25MB).

Table 2 presents the names of the authors with the largest number of publications associated with OCs presented in SENeo congresses along with the publication rate. We did not add further authors as following those cited there were up to 10 authors with 8 publications.

A total of 167 articles (78.8%) were published in international journals. Fig. 2 presents the journals that published the most articles associated with the congresses. Among the published works, 144 (67.9%) were published in specialised paediatrics/neonatology journals and 68 (32.1%) in general medicine journals.

Figure 2.

Pie chart of the journals used to publish research. The numbers show the absolute frequency of publications in each journal. The legend to the right lists the names of the journals ordered by decreasing number of publications.

(0.25MB).

When it came to the bibliometric characteristics of the publications related to OCs, the median and IQR for the impact factor, journal quartile and number of citations were 2.86 (IQR, 1.96–3.98); 2 (IQR, 1−3) and 3 (IQR, 0−7), respectively. Fig. 3 shows the temporal trend in the cumulative IF and the ratio of the cumulative impact factor and the total number of publications for each of the 3 congresses.

Figure 3.

(A) Cumulative impact factor of the publications derived from OCs. The cumulative IF for each congress was calculated by adding the IF for each publication. (B) Cumulative IF divided by the number of publications for each congress (n).

IF, impact factor; n, number of publications for the congress.

(0.11MB).

The median number of authors in the published articles was 6 (IQR, 5–8). In 122 publications (57.5%), the first author was also the first author listed for the OC. The median time elapsed between presentation in the congress and publication was 10 months (IQR, 1–23), while the median time elapsed between presentation in the congress and submission to the journal that published the article was 6 meses (IQR, −5 to 17.5). Fifty-one manuscripts (24.1%) were submitted for publication before the congress, and 44 (20.7%) were published before the corresponding congress.

Sixty-eight OCs (12.9%) presented findings from multicentre studies, 44 of which (64.7%) were subsequently published. There were 115 OCs presented with application for full membership in the SENeo, of which 40 (34.7%) were of studies that were published. When it came to the sex of the authors, 423 (80.5%) of OCs were presented by a female first author and 102 (19.5%) had a male first author; while the last author was female in 281 OCs (53.5%) and male in 244 OCs (46.4%).

Table 3 presents data on the variables potentially associated with the publication rate, explored first through a univariate analysis and then through a multivariate analysis that included those with a significant association in the previous step. The dependent variable in the model was the publication of the OC, while the variables under study were OCs derived from multicentre studies, OCs with application for full membership, sex of first and last author of the OC, subject of the OC and OC candidate for award in 2017 congress.

This study analysed data regarding OCs presented in the SENeo congress, finding a publication rate of 40%, slightly above the percentage of publication in scientific journals observed for other congresses, both of paediatrics and other medical specialities. To date, this is the first article analysing data for a neonatology congress.

The OC publication rate of 40% is a remarkable figure in light of the statistics for comparable congresses. The closest in our region was the Annual Congress of the Sociedad Española de Urgencias Pediátricas (Spanish Society of Paediatric Emergency Medicine), with a reported full publication rate of 9.4% of OCs.11 In the field of paediatrics at the international level, one of the most important studies on the subject was conducted by Basu et al., who analysed the impact and proportion of publication of paediatric critical care medicine abstracts presented at American Academy of Pediatrics, Pediatric Academic Societies and Society of Critical Care Medicine national meetings, finding a publication proportion of 41%, practically the same as the one found in our study, with a time elapsed to publication of 22 months, longer compared to the SENeo congress.7 A similar rate of 41.5% has been reported for the congresses of the Sociedad Argentina de Pediatría (Argentine Society of Pediatrics).12 Lower proportions have been reported for the American Society of Echocardiography meeting (37.6%) or the European Paediatric Orthopaedic Society annual meetings (36.7%).8,13 On the other hand, higher publication rates have been reported for the meetings of the International Society for Pediatric Neurosurgery (46%), the American Society of Pediatric Hematology and Oncology(62%) or the European Society for Pediatric Urology (47%).14–16 There are two Cochrane reviews with meta-analyses on the subject; the first, from 2007, found a full publication rate of 44.5%, and the second, from 2018, a rate of 37.3% after analysing 425 reports that included 307 028 abstracts presented in congresses.17,18 We ought to highlight the excellent publication proportion for OCs candidates to award in 2017, which far exceeded the proportions reported in the literature.

The most frequent reasons given not submitting for publication the findings presented in meetings as OCs are lack of time to write the manuscript, that the OC presented preliminary findings of studies still underway, insufficient sample size, lack of agreement with other authors regarding publication and not considering the findings relevant enough to merit publication.11,19,20 Another aspect to consider is rejection of submitted manuscripts by journal editors; Sprague et al. analysed it in relation to annual orthopaedic meetings and found a proportion of 16%.19

The impact of the publications derived from SENeo congresses was higher compared to the mean reported in the literature for other paediatric congresses or other congresses in Spain, as the impact factor of the journals that published the studies was nearly 1 point greater compared to the congresses of the SEUP (1.5) or paediatric orthopaedic societies (1.73) (1.98)9,13; and even exceeded the figures for meetings as renowned as the meeting of the American Academy of Pediatrics (1.92) or North American paediatric societies (2.64).10 Another noteworthy finding was the progressive increase in the impact factor, both in terms of the median and the cumulative value, over successive congresses, which reflects the sustained progress of the speciality in Spain, as has been the case of other specialities.21 One of the clues to understanding the reach achieved in the literature is the substantial increase in recent years of the impact factor of the journal Anales de Pediatría, the journal most frequently used for publication of OCs in the SENeo congress, which was 2.1 in year 2022, placing it in the third quartile for the speciality of paediatrics (66/130) and very close to the second quartile, based on data obtained from Journal Citation Reports.22,23

Analysing other secondary outcomes, we found that the works presented in SENeo congresses were published after a shorter time interval compared to other scientific meetings for which the time elapsed from presentation to publication ranged from 14 months to 2 years.10,13–15 This can be explained by the not insignificant percentage of manuscripts submitted for publication or even published in the months preceding the congress; the originality of the works was not explicitly required in the submission rules for the past 2 SENeo congresses, but it is a prerequisite for acceptance in other meetings, an aspect for which we found no specific data in the literature. There is also little evidence on the concordance of the first author of the OC and the published article, and we found a lower percentage in our study compared to the North American study on the field of paediatric critical care.9 In the analysis based on the sex of the author, we found a lower proportion of publication when the last author was female, consistent with previous similar studies, but in our case the difference did not continue to be significant in the multivariate analysis, all of which must be interpreted taking into account the limitations of attributing sex based on the author’s name.24–27

The main limitation of the study was the unequal follow-up period for the oral communications in each congress, for while we analysed communications presented in the 2017 congress 6 years later, in 2023, the follow-up period for communications in the 2021 congress was only of 2 years, and it would be reasonable to assume that more articles associated with the 2021 congress will be published in upcoming months. However, based on the time elapsed between presentation in previous congresses and publication, we find that the total number of publications after the 2021 SENeo congress is consistent with the temporal trends observed for the 2017 and 2019 congresses. Other studies have analysed potential publication biases based on study results (studies with positive conclusions were published more frequently than studies with negative conclusions) and the methodology used (a higher proportion of studies with a prospective design are published compared to retrospective studies), but in this study we did not have access to the abstracts of the studies presented at the 2017 congress, as only the title and author name were available for the communications of this congress.18,28

In conclusion, we can assert that a high percentage of OCs presented in SENeo congresses are subsequently published in indexed journals, which is similar to other congresses of international renown, and the high quality of these studies was evinced by the analysis of the impact factor of the journals where they were published. The rate of subsequent publication was greater for OCs derived from multicentre studies or studies concerning respiratory disease subsequently published from multicentre studies and those concerning respiratory disease.

This research did not receive any external funding.

The authors have no conflicts of interest to declare.