Country of origin

In the present study, the vast majority of publications were submitted from the US, followed by Western Europe and Japan (Table 1). These data are in concordance with Li et al.’s study, analyzing journals between 2000 and 2009 [12]. The top five countries in the present investigation were among the top seven countries in the study by Li et al. [12]. Figueredo and colleagues also found that the USA, the UK, Japan, Germany, Canada, and France published the most articles in anesthesia journals between 1997 and 2001 [13]. In their study, the largest increase in scientific production in anesthesiology was observed in Germany [13]. However, they concluded that the geographical distribution of scientific production must not only be analyzed in absolute terms but should also be evaluated relative to other variables, such as financial spending on research and development, as well as population growth [13].

Articles per year

Altogether, 42,549 articles were published in the five journals within the time frame analyzed. In Anesthesiology, the mean number of articles published per year was 498.8±67.6. Compared to 1991, the number of articles per year decreased by 5.0% until 2010. In Anaesthesia, a decrease of 45.1% in the mean number of articles published was observed in the same period. In order to attain higher quality in journal articles and consequently a higher IF, several prerequisites for article acceptance have been developed in recent years. Currently, clinical trial registration, as well as plagiarism checks and ethical committee approvals, is a standard practice. Although this practice ensures higher-quality articles, it also could result in a decrease in the number of accepted publications.

Li et al. [14] confirmed these results in 18 different anesthesiology journals. The authors postulated that this decrease might be associated with an increasing IF. Feneck and colleagues also found a decrease in research publications from the UK in anesthesia journals from 1997 to 2006 [15]. In our study, the journals Br J Anaesth, J Neurosurg Anesthesiol, and Anesth Analg showed significant increases in the numbers of publications per year (1991 to 2010). Another study published by Li et al. [14] showed that these journals have decreasing numbers of articles while analyzing the period of 2000-2009 only. However, it is accepted that the absolute number of scientific publications is less important due to the difficulties that can occur in quality assessments of studies [16].

Impact factor analysis

Among the many surrogate parameters, the IF from the Institute for Scientific Information [2] has achieved the greatest popularity, especially in government research and medical schools [3, 4]. It is a common, but fictive, instrument for quantifying the importance of a medical paper based on citations [17] and it is used to rate the quality of a (medical) journal [4, 5]. Today, the IF also has enormous significance for research department funding [3, 18]. The IF was originally designed and sold as a product for advertisers, i.e., it was sold to advertisers, at very high costs for the purpose of helping them target their advertisements to the most widely read journals within the set of journals that were read by potential buyers of their products [19]. Although the IF is not an absolute measurement of the quality of a journal [3], it can quantify the influence of a journal in the medical field [4]. Because the IF influences medical research and funding [18], changes in the IF can significantly affect publication practice [6]. During the time frame analyzed, the IF of all journals in the present study increased (2.249 in 1991 vs. 3.639 in 2010; i.e., +61.81%). This phenomenon is not new and has been observed for several years also by other authors [20]. The median and highest IF in the present study have increased since 2005 [3]. This increase in IF has been observed for years [3, 20]. One main reason for the IF boom is the growth of research production [4, 20, 21]. More researchers are writing more contributions in more journals [22, 23] with more references [3]. In contrast, several journals have decreasing IFs, especially those in non-English languages and those that are difficult to access.

However, IFs can be manipulated by many strategies: self-citation; increasing editorials and reviews (more likely to be cited); and a reduction of the number of case reports (less likely to be cited) [4, 16, 18]. Therefore, the importance of the IF remains controversial [24] and there is no established current alternative available to rate the quality of scientific publications [4].

As the main factor that could lead to overestimate the IF is the amount of self-citations within a journal, in 2010 Landoni et al. [25] have suggested to apply a new metric called “new IF 20%”. This instrument, accounting the weight of self-citation with the aim of limiting their influence on the final result, is employed when the self-citation percentage is more than 20% of the total. Other authors present innovative journal metrics as IF substitute, e.g. the CAPCI factor (Citation Average per Citable Item) by E.P. Diamandis [26]. In addition, the International Committee of Medical Journal Editors (ICMJE) has outlined guidelines for the authorship of scientific papers [27], but it remains unclear whether authors comply with these prerequisites when submitting manuscripts. Therefore, the contributions of each author should not only be judged by the authors themselves but also by the editors to prevent misconduct by researchers and inadequate authorship.

Article type

Although the absolute numbers of comparative studies and RCTs were comparable between 1991 and 2010, there were significant variations per year. In contrast, the absolute number of case reports decreased significantly by 45.8% between 2003 and 2010 and by 54.7% between 1991 and 2010. Li et al. [14] also found a decreasing number of articles per year (both clinical studies and RCTs) when analyzing several different journals. Because case reports are cited less frequently in the years after their publication, publishing them can be counterproductive for a journal striving to obtain a higher IF. Therefore, publishing fewer case reports to enhance the IF might have motivated the major anesthesiology journals that were analyzed in the present study.

Pages per article

The mean number of pages per article was 4.3±1.4 and did not change between 1991 and 2010 (Figure 4; p>0.05). However, only in the Br J Anaesth the pages per article remained stable during the time frame analyzed (p>0.05). For Anesthesiology, Anesth Analg, and Anaesthesia the pages per article increased but decreased significantly for J Neurosurg Anesthesiol. However, due to lower publications costs and the availability of electronic publication [28], restrictions on the length of articles might not have been as relevant as such restrictions previously were.

Number of authors per article

The mean number of authors per article was 3.6±0.7 in the present study, and it increased significantly in all journals analyzed between 1991 and 2010 (3.0 vs. 4.3; relative increase of 43.3%). This increase was the lowest in Anesthesiology (+24.8%) and was significantly higher in the other journals (+62.5%). The increase in the number of authors per articles has also been a commonly observed phenomenon in other medical journals and in different countries [29, 30]. This phenomenon has been observed even in the most highly ranked journals (N Engl J Med, J Am Med Assoc, and The Lancet) [29, 30]. Kumar et al. [30] reported a gradual increase in the average number of authors over the past three decades [30]. Drenth et al. also found an increasing number of authors in the Br Med J [31]. These findings are congruent with the results of the present study, although different medical fields were analyzed. A number of reasons could be responsible for this phenomenon [30]. There has been an increasing trend toward cooperation among researchers in multiple disciplines and an increase in multi-center studies, as well as an increase in the complexity of research projects, which could facilitate collaboration and result in an increase in the number of authors per article [30, 32]. Additionally, it has been shown that studies with many authors are cited more often than studies with fewer authors, thereby contributing to the IF of the journal [23].