We conducted a national, descriptive, observational study that analyzed OCs from 5 SEMICYUC congresses (2013, 2015, 2017, 2019, 2021). Alternate years were selected to reduce the volume of data while maintaining representativeness. All OCs from these five congresses were included.

Congresses held after 2021 were excluded because insufficient time had elapsed for projects to be published. Literature searches were performed in July–August 2024, and results were analyzed in September 2024.

From all projects submitted to SEMICYUC, the scientific committee selects the best for presentation as OCs, oral posters, or electronic posters. The committee reviews all submissions anonymously and in pairs, assigning scores using established criteria. Projects selected as OCs have higher scores and accordingly are considered to be of higher quality, greater relevance, or more promising.

We analyzed only OCs, excluding posters, because OCs have more chances of being published.1

All abstracts were retrieved from the online version of Medicina Intensiva.

The OCs were categorized according to the thematic areas of intensive care established at the congresses: Cardiovascular (including hemodynamics, arrhythmias, cardiac surgery, pacemakers), Donation/Transplantation, Ethics, Hematology, Infection (including sepsis, antibiotics), Metabolism/Nutrition, Miscellaneous (including emergency medicine, toxicology, simulation, new technologies), Nephrology (including renal replacement therapies), Neurointensive Care (including critical neurological pathology, traumatic brain injury, trauma), Organization (including management and quality), Respiratory (including mechanical ventilation), and Sedation/Analgesia (including delirium).

OCs accepted at SEMICYUC congresses in 2013, 2015, 2017, 2019, and 2021.

Oral or electronic posters and OCs not eligible for publication (e.g., opinion pieces).

For each OC we recorded title, authors, year, institution, country, autonomous community, intensive care area, study design (single- or multicenter), methodology, funding (independent/undeclared, industry, or public/competitive grants), association with a doctoral thesis and publication status.

For each DP we recorded title, authors, year, journal, IF and quartile obtained from Journal Citation Reports (Clarivate™).

A publication was considered derived from an OC if it had the same objective (answering the same question in whole or in part), the same design, the same or very similar recruitment period (in the case of preliminary results), the same or very similar patient number and at least one common author.

We used Google Scholar, which indexes PubMed, Elsevier, scientific societies, Dialnet and journal portals.

The search did not begin with a hypothesis or a PICO-formatted question, as is widely seen in the scientific field, but rather from the titles of OCs. To ensure that the search was as systematic, methodical and reproducible as possible, a strategy was designed beginning with a search for the full OC title in Spanish (the first 50 results were reviewed). If no DP was identified, the full title together with the first author was then searched (first 50 results). If no DP was identified at this stage, keywords from the title in English were then searched (first 50 results). The search concluded when a DP was identified or, if none was found, after these steps werecompleted. The search was conducted by a single investigator.

Quantitative variables were summarized with descriptive statistics; qualitative variables were presented as counts and percentages, including missing values. Qualitative data were compared using Fisher’s exact test. Statistical significance was set at p<0.05.

Analyses were conducted with IBM SPSS Statistics for MacOS, version 29, following International Conference on Harmonisation E9 guidelines (CPMP/ICH/363/96).

In line with former studies, most publications occurred within two years of the congress presentation.11

In the study by J. Miquel et al.,4 with a seven-year follow-up, 95.5% of DPs were published within the first four years after presentation.

In our study, the congress with the fewest DPs was held in 2021 (28%). This may be explained by the fact that insufficient time had elapsed to allow for publication of the work. Another hypothesis is that, due to the COVID-19 pandemic throughout 2020 and the associated clinical workload faced by intensivists, many projects were halted and the quality of research may also have been affected.

The “Infection” area took the lead both in the number of OCs presented and in DPs. This is likely related to the availability of the ENVIN-HELICS registry, which collects data on device-related infections occurring in ICUs and has enabled many of these studies. Certain areas, such as ethics, nephrology or those classified as miscellaneous, may show lower publication rates due to fewer submissions, lower editorial interest, the difficulty of designing high-level evidence studies or the lack of specialized journals prioritizing these topics in intensive care medicine.

The journal with the most DPs was Medicina Intensiva (28.3%), published by Elsevier Spain. It is the official journal of SEMICYUC and has become the reference journal in Spanish for the specialty. It currently has an IF of 2.7 and is ranked in the 2nd quartile (2023 data).

In our study, the mean IF of journals in which the DPs were published was 5.8 (range, 1.2–44.4). Similar studies have reported lower IFs, such as 1.51 in congresses of the Spanish Society of Shoulder and Elbow Surgery, 1.96 in those of Pharmacology and 2.32 at the European Society for Surgery of the Shoulder and Elbow Congress.3,4,10 However, differences in subject make direct comparison difficult.

We observed that studies with a higher probability of publication were multicenter, randomized clinical trials and quasi-experimental designs.

Similar results were reported by Schulte et al.,12 in congresses of the German Society of Orthopedics and Trauma Surgery, where randomized and prospective trials were published more frequently than nonrandomized and retrospective ones. Similarly, in the 2018 systematic review by Scherer et al.,1 randomized, prospective, and multicenter trials were published more often than other study types. These findings may be attributed to the fact that such characteristics increase the scientific quality of projects.

We also observed that funded projects had a higher probability of publication, both those sponsored by industry and those supported by grants. This is expected, as funded studies have undergone more rigorous evaluation processes, which may improve quality and increase the likelihood of publication.

Similar findings were described in the 2018 systematic review, in which studies with funding of any type were more frequently published.1

However, it is of note that 76% of the abstracts in our study did not declare or specify funding, limiting inferences about its effect on DPs.

Finally, it is important to consider the fact that the congress selection process prioritizes the highest-quality abstracts for OCs. This does not imply inherent high quality, but rather relative high-quality when compared to the other submissions, which may partly explain the resulting publication rate.

This study has certain limitations that should be considered.

First, only five congresses from alternate years from 2013 through 2021 were analyzed. We do not know what results would have been obtained had all congresses during this period been included.

Second, the search was not performed in parallel; it was conducted by a single investigator, and DPs may have been underestimated. For future studies, we recommend paired searches to increase accuracy.

Third, although the search was performed on a platform covering many databases, some publications may not have been detected if they were not published in indexed journals or outside the reach of Google Scholar.

With regard to the identification of DPs, the publication title may have changed significantly vs the OC title, making identification difficult. In addition, the first author may differ between the OC and the DP. Translating titles into English may introduce bias, as the search depends on the choice of appropriate terms. All these sources of bias could have been reduced by incorporating a second reviewer or by expanding the search strategy.

As a hypothesis for future research, it would be relevant to analyze factors influencing lower publication rates in certain areas of intensive care and to explore strategies to increase their visibility and publication.