Archivos de Bronconeumología has once again been listed in the Journal Citation Report of this year,1 thus reinstating its impact factor (IF) with a self-citation rate of 20%. Citation of articles for documentation purposes is a common and necessary practice.2 By including relevant and appropriate citations, authors can situate their work within the current state of the art in that specific area. Notwithstanding, both the principles of communication of knowledge and publishing ethics make it unacceptable for an article to be based primarily on self-citation. Arguments for and against self-citation are numerous.3 The problem emerges when we try to assign value to citations, and the use of citations as the yardstick by which the scientific community measures and evaluates the work of its members has only served to aggravate the situation. It is logical that self-citations do not have the same value as external citations, i.e., citations from other authors.3 In this respect, a distinction must be made between self-citation by journals and publishers and self-citation by authors and co-authors who reference their own previously published papers.

Attempts by publishers to force authors to increase the number of self-citations in their journals (Coercive Journal Self Citation) are frowned upon, but it nevertheless remains a common practice in some circles. There are numerous examples, not only of direct requests to the author for self-citation, but also of arrangements between journals to cite each other's publications.4 Impact indexes that exclude self-citation are already available: 2 YIF (2-yearly IF), the Eigenfactor Score, the Article Influence Score5 and the Crown Indicator,6 and self-citation can also be excluded from the 5 YIF (5-yearly IF). Perhaps these are better tools for measuring impact. Self-citation by authors is in itself neither good nor bad. Self-citation is known to peak rapidly in the first years after a publication, more so for authors than for co-authors, while external citation comes later. Articles with more citations and a greater IF tend to generate fewer self-citations, and self-citation does not appear to contribute decisively to achieving a higher IF.7 Collaboration, whether national or international, between authors leads to a marked increase in external citations, but the effect on self-citation is moderate.7 Fowler and Aksnes8 found that more self-citation led over time to a greater percentage of external citations; each additional self-citation generated one additional external citation in the following year, 3 in the following 4 years and 4 in the following 10 years. Thus, they conclude that self-citation should not only be eliminated from impact calculations, it should be penalized.

Needless to say, self-citation requests from publishers fueled by non-scientific interests should continue to be penalized in order to stamp out improper conduct. However, when it comes to self-citation by the author/co-authors, we believe that this should be excluded from the IF calculations, and only adjusted indexes should be used for calculating impact.