Impact Factors and Prediction of Popular Topics in a Journal

The impact factor (IF) for 2015 was recently released and this could be the time to once again reflect on its use as a metric of a journal. Problems and concerns regarding the IF have been addressed extensively elsewhere 1 2. The principle of the IF for a given year is that it represents the average number of citations of articles published in the journal in the two previous years.While authors frequently cite the IF as a determining factor for submission, the IF does not predict how many times individual articles will be cited. In a study from a peer-reviewed cardiovascular journal, nearly half of all published articles were poorly cited, i. e., less than five citations in five years 3. A similar percentage seems to apply to our journal. In nearly all journals we estimate that the majority of citations relate to a minority of the articles. Some articles are never cited. 13 % of the articles published in our journal from 2010 to 2013 have never been cited. Even authors of poorly cited articles benefit from the IF since many institutions use the combined impact factors of their published papers to measure research activity and this may be reflected in their research budgets.The competition for the printed pages in the six annual issues of Ultraschall in der Medizin/European Journal of Ultrasound (UiM/EJU) has resulted in high rejection rates (between 80 % and 90 %). One negative review with recommendation of major revision may therefore result in rejection. Peer-review fraud where the submitting author listed recommended reviewers with fake email addresses supplying fabricated peer reviews has recently been described in the New England Journal of Medicine 4. Some of the editors of our journal believe they have experienced this as well. Fabricating reviews in order to get a high IF for an article is to be considered fraud and is inexcusable.One aspect of using impact factors as a measure of the quality of a journal is that the IF only goes back two years. There may be differences between journals for different medical specialties since the citations in some areas seem to "burn out" within a few years while some articles continue to be cited even after several years. Therefore, a citation window that is longer than 2 years has been proposed 5.For this editorial we took a look at the 60 articles published in UiM/EJU in 2010. Half of them were no longer being cited in 2015. However, 10 articles were cited more than 5 times in 2015, and 5 of these were cited more than 10 times 6 7 8 9 10. It therefore seems that many of our articles have a long scientific life and generate more citations than indicated by the IF. Moreover, some articles have the highest number of citations after three years when they are no longer contributing to the impact factor. The most frequently cited articles from 2010 were multicenter studies, recommendations, and papers on hot topics like contrast-enhanced ultrasound (CEUS) and elastography, but it should be noted that there were also articles on the same topics that were poorly cited.The same trending topics continued into 2013 now topped by European guidelines and recommendations 11 12 13. 9 of the 10 most cited articles we published in 2014 were on CEUS or elastography 14 15 16 17 18 19 20 21 22, but the most cited article from that year so far has been on peripheral nerves 23. Surprisingly many good scientific papers on obstetrics/fetal US and musculoskeletal US have low citation rates 24 25 26. Our predictions for 2016 based on the topics of submitted articles in the last 12 months are that CEUS and elastography will continue to be popular topics.It is also worth mentioning that there can be a discrepancy between which titles are cited and which are accessed online. In addition to international guidelines, our CME articles are usually popular according to online access. CME articles are well established educational papers but they are rarely cited for the IF. Looking at the most read full-text recent articles on our journal's website shows that multicenter studies as well as recommendations backed by a national society or by the EFSUMB (European Federation of Societies for Ultrasound in Medicine and Biology) are still important 27 28 29 30 31 32 33. Upcoming important topics appear to be pediatric use of CEUS, simulation training and the introduction of ultrasound to medical students 34 35 36 37. Some of these are also backed by EFSUMB.A recent paper on the IF of radiology journals found that subspecialty radiology journals had a higher IF than general radiology journals 38. This could prove a challenge to interdisciplinary journals like ours but we take pride in continuing to cover all aspects of ultrasound in more than 15 fields.The distribution between reviews, original articles and case reports in a journal is worth addressing. An important aspect of a journal is the publication of original scientific research articles. CME articles, pictorials and letters are important for other reasons but are cited at a lower rate. The value of case reports with regard to the IF is low since they are rarely cited 39 and we have observed that some journals have abandoned the publication of case reports, thus leaving them to spin-off journals. The rationale is that keeping case reports in a journal will only increase the denominator, thereby decreasing the IF 39. At our journal we have seen a decline in case report submissions but still want to publish them and even put one case on the front cover of every issue. Case reports still hold an educational value 40 and are important to our readers.In conclusion, a healthy mix of original articles, CME articles, reviews and case reports combined with a few international guidelines and recommendations is important to UIM/EJU. Although we see popular topics like CEUS and elastography, it is not possible to predict which articles will be read or even cited based on the topic, with multicenter studies being the exception.