Advances in the study of childhood diseases and their social and economic impact require a continued support of research, including the publication of research findings in scientific journals and open-access journals. A better understanding of the trends in the performance of Anales de Pediatría in different bibliometric indicators and an increased awareness of the complexities and challenges faced by the journal may contribute to improving its future development.

Previous studies have analyzed the scientific production in the field of pediatrics in Spain.1–7 However, little is known about the evolution of different bibliometric indicators of the performance of Anales de Pediatría in the 2 international databases that contribute data and metrics regarding citation and impact in which Anales is indexed, Scopus and the Science Citation Index Expanded of the Web of Science Core Collection (SCIE).

The coverage of Anales in these databases has been inconsistent and irregular, both in terms of the annual number of indexed publications and in the frequency distribution of publications by type of document. For instance, the journal has been indexed in Scopus between 1973 and 2002 under the title Anales Españoles de Pediatría and from 2003 to present under the title Anales de Pediatría,8 whereas it has only been indexed regularly in the SCIE since 2009. These differences in coverage may give rise to numerous discrepancies in bibliometric indicators depending on whether the data are extracted from one or the other database. For instance, the total number of citations of documents in the journal in the past 9 years (2009–2017) is 4638 in Scopus, compared to 2799 in the SCIE—1839 fewer citations (data extracted on September 12, 2018). There are even variations in other indicators that are less dependent on the source of the data, such as the average number of citations per publication, which is 2.27 in Scopus and only 1.5 in the SCIE.

Taking this context into account, our objective is threefold:

• a)

  To analyze the evolution of the main bibliometric indicators of the scientific output, collaborations and impact of Anales de Pediatría in the SCIE and Scopus databases in the years that they indexed the journal.

• b)

  To offer accurate data on the actual coverage in these sources and to evince how differences in the editorial policies of the databases regarding how records are indexed may contribute to the variability in indicators, which is source-dependent and results from the discrepancies in the data contributed by the databases that process records and citations.

• c)

  To identify the indicators for which Anales de Pediatría exhibits an increasing, decreasing or uneven evolution, aiming at establishing possible trends.

The methodology involved several phases: 1st, download of the records of databases; 2nd, selection of strictly scientific types of documents (original articles, editorials, scientific letters and review articles); 3rd, normalization of missing or incorrect data; 4th (a) extraction of output indicators (number of articles per type of article category and year of publication; (b) calculation of collaboration indices (year-by-year trends in the author collaboration index; year-by-year trends in international collaboration); (c) Evolution of citation-based indicators (number of citations; impact factor; 5-year impact factor without self-citation; Scimago Journal Rank [SJR]; h-index; most cited documents or “hot papers”; annual citation rate of the most cited documents; worldwide ranking of the journal for the impact indicators); 5th, identification of the indicators in which Anales de Pediatría exhibited an increasing trend, a decreasing trend and an uneven trend.

To obtain these data and these metrics, we used 4 sources: the SCIE of the Web of Science Core Collection, Journal Citation Reports (JCR), Scopus and SCImago Journal & Country Rank.

All these indicators are defined in the sources specified above and have been analyzed in previous studies,9–13 except the citation rate of the most cited documents, calculated by dividing the total number of citations by the years elapsed from the publication of the work.

Our science metrics research focused on the analysis of the journal Anales de Pediatría to quantify its coverage in the SCIE and Scopus and the evolution of its most relevant bibliometric indicators.

The number of works indexed in Scopus in the past 44 years has been inconsistent, with peaks and troughs that can affect trends in indicators. In the 9 years of indexing in the SCIE, we found more consistent indexing that was similar to indexing patterns in Scopus, except for 2014, for which Scopus erroneously indexed works published in different years. Ideally, Scopus will correct these errors, some of which have already affected the scientific literature,14 for instance in the management of so-called “online-first articles”, in the connection of cited works with the citing authors or the duplication of records, among other phenomena.15–17 These circumstances, which are unrelated to the effectiveness of the editorial strategies of the journal or the quality of its published contents, can seriously affect indicators, inflating or deflating them.

On the other hand, we noticed that these databases do not always adhere to the classification of documents made by journal by including them in specific sections. We found that some of the documents indexed as articles in Scopus were classified as letters or editorials in the SCIE. In a previous study, Abad et al.6 found that the proportion of letters in the Spanish pediatrics output was of 19.2%, a proportion that vastly exceeded the global average (7.3%) and that probably resulted from the influence of the letters published in Anales de Pediatría, a circumstance that was also noted in 2011 by González Alcaide et al.4

The author collaboration index has increased throughout the trajectory of Anales de Pediatría. The value of this indicator is influenced by several factors, including the limitation imposed by the journal on the number of signing authors, which have led to a sustained value of this index in contrast with the marked increasing trend that had been observed in previous years.

As we already mentioned, the average number of citations per document is lower in the SCIE (2.27 in Scopus vs 1.5 in the SCIE). This is due to the higher journal coverage of Scopus. Taking into account the area of pediatrics alone, this database indexes 6 Spanish journals. Another study that compared results across several databases (Scopus, Web of Science and Google Scholar) also highlighted the variations in citation counts between databases.18

The impact factor and its variants have increased progressively in the SCIE, as have, to a lesser extent, the SJR and the ranking of the journal among journals in its subject area. Should this gradual increase be sustained in upcoming years, the journal would continue climbing the rankings and even enter the second quartile. When it comes to the h-index, since the “life” of Anales de Pediatría has been much longer in Scopus compared to the SCIE, it was only to be expected that the h-index is higher in the former database compared to the latter, as this is an index that increases as journals persist in databases.

The most cited articles of Anales de Pediatría provide a perspective of historical developments and insight on the important advances that have occurred in pediatrics, as it is fair to assume that highly cited articles are those read most times and which are most important within this specialty. The subjects of the most cited articles are varied and usually involve scores or indices, recommendations, consensus documents and childhood diseases. We did not find what Chhapola et al.19 called “sleeping beauties”, that is, articles that go unnoticed (sleep) for a long time and then, seemingly out of the blue, attract considerable attention in the form of citations, which manifests a delay in the recognition of vital work. We also found no evidence of the citation bias toward recent publications postulated by some authors20 and that could be explained by how easily and quickly these publications may be found using web search engines.

In considering the limitations of the study, we ought to keep in mind that all bibliometric analyses have inherent strengths and weaknesses. The databases considered in our study have great international prestige and provide detailed citation metrics, so they are the sources used most frequently in bibliometric studies, and today, the Web of Science is considered the gold standard in this type of research.19 Some of the limitations involve the inconsistency or inaccuracy of the records stored in these databases and regarding citations. We also ought to highlight that citation metrics do not necessarily reflect the quality of studies and is not the only method available for the assessment of the impact of research.

To conclude, we ought to underscore that most of the main bibliometric indicators of Anales de Pediatría have exhibited an increase, especially the impact factor and its variants (Table 4). We must also highlight the sustained ascension of Anales in the ranking of pediatrics journals of the JCR, moving from the fourth to the third quartile in an increasing trend that could bring it to the second quartile in upcoming years. On the other hand, we should strive to increase participation by foreign researchers and institutions, as well as the number of citations. The open-access and bilingual Spanish-English publishing policies promoted from the Editorial Committee may contribute to achieving these goals, as would a diffusion strategy that took into account the visibility of the journal in social networks and other digital platforms. The data retrieved in our study confirmed that Anales de Pediatría has consolidated its leadership as the main vehicle for the diffusion of Spanish pediatrics.

As an interesting project for the future, we propose the establishment of a surveillance body to track the evolution of bibliometric indicators that would guide strategic decision making for the purpose of correcting detected weaknesses or reinforcing existing strengths. Another useful measure would be to analyze the flow of citations from Anales to other Spanish and international journals, and from them to Anales de Pediatría.

The authors declare that they have no conflicts of interest.