Citation indexing and highly cited articles in the Australian Veterinary Journal

The concept of tracking scientific manuscripts to see if they were cited in the reference lists of subsequent journal articles was first proposed by Eugene Garfield in 1955.1 His formula to establish which journals had the highest relative citation rates quickly became known as the impact factor (IF), a number that has now risen to prominence in the academic world of veterinary science, as elsewhere.2 Although the IF focuses only on those citations obtained during the previous 2-year period, the idea of counting the total number of citations received by a particular article also attracted the attention of Garfield, who first began undertaking such investigations in the 1970s. The term ‘citation classics’ was later coined by him to describe specific journal articles that had received an unusually large number of citations.3 Relatively few studies of citation classics appear to have been conducted in veterinary science, although those that do exist have previously identified avian lymphomatosis,4 feline leukaemia5 and neonatal calf diarrhea6 as topics with potentially high citation rates. In 1977, Cho published a study of citation characteristics in veterinary science, albeit with analysis limited to Advances in Veterinary Science and Comparative Medicine.7 One of the earliest citation classics from the Australian Veterinary Journal (AVJ) appears to have been identified in 1976, during a study of the most frequently cited articles in Australian and New Zealand journals.8 Only one recent study of citation analysis appears to have been conducted in the veterinary profession, however,9 which despite searching across 123 veterinary periodicals, did not appear to locate any citation classics from the AVJ. Although international veterinary research tends to be spread among the specialist veterinary journals, as well as periodicals in general medicine, local publications have a very important role to play in the global dissemination of scientific information. The AVJ was first published in 1925 as the Journal of the Australian Veterinary Association, with Max Henry as the founding Editor-in-Chief.10 It was the official voice of the Australian Veterinary Association, which had held its inaugural meeting 4 years earlier, in 1921.11 In 2006, Crawley-Low12 classified the AVJ as a ‘Zone 2’ core veterinary medicine journal. It was first indexed by the Institute for Scientific Information (ISI) (now known as Thomson Scientific) in 1968, and by September 2007, the AVJ had over 7000 articles listed in the National Library of Medicine's PubMed® database. Despite this fact, it was hitherto unknown which of these 7000+ articles were destined to become highly cited in later years. The purpose of the current study, therefore, was to identify the 10 most highly-cited articles ever published in the AVJ. An additional aim was to examine the changing nature of the journal's IF over time. The AVJ citation data was extracted from Thomson Scientific's Web of Science®database in 2007. According to this database, the most frequently cited article is a 1980 study describing anthelmintic resistance in nematodes, conducted by RK Prichard et al at the University of Sydney.13 As of September 2007, this article had been cited 185 times. The second most highly cited article was a study of bovine interferon-gamma published by JS Rothel et al14 in 1990 (179 citations), followed by a study of Salmonella typhimurium infection published by AB Lloyd et al15 in 1977 (140 citations). By proportion, these top three citation classics have garnered almost half of all citations in the list (Table 1).13–22 The data was also highly skewed, with the 10th most-cited article accumulating less than half the number of citations of the top-cited article. Early data from the 1974 Science Citation Index® shows that the IF of the AVJ at that time was 0.755,23 although it had apparently risen to 1.586 by 1980.24 According to the 2007 Journal Citation Reports®, the journal's current IF is 0.595. Regardless of one's opinion on the subject, the IF, citation rate and identification of citation classics remain contentious aspects of modern academia. On one hand, the journals of general medicine, cancer research, genetics and immunology, for example, tend to have consistently high IFs, with many regularly exceeding 10 or 20. Similarly, some of their articles accumulate citation counts in the thousands or hundreds of thousands. Despite this, the size of a research field inherently determines the absolute number of citation classics it will ever contain, an issue that also influences the IF.25 Veterinary journals are known to have relatively low scores in this regard, particularly when compared with journals in general medicine. Christopher, for example, has previously reported that 80% of dedicated veterinary journals had an IF of less than one.2 Even though the absolute citation counts tend to be lower and are thus omitted from larger investigations, smaller journals in specialist fields, such as veterinary science and veterinary medicine, do have their citation classics, as shown in this study, and important information can be obtained by looking at what articles they originally published and which of these went on to become highly cited. It takes time for any article to become a citation classic, with the most cited article of all-time for the AVJ being almost 30 years old. This delay occurs because of the time taken for a published article to be located by a new author, for the article to be cited in a new manuscript, for that new manuscript to be accepted, for the new manuscript containing the citation to be published, and finally, for the new manuscript's reference list to be examined and entered into the citation tracking system. Even so, with the inexorable march of technology, the proliferation of the internet and the widespread use of bibliographic reference management software, it appears that a greater number of references are being cited per article in recent years, and the citation/publication time-lag is probably also being shortened. The second most highly-cited article in the AVJ, for example, is less than 20 years old.14 Also, 40% of the top 10 articles involve research pertaining to cattle, suggesting that this area has a relatively high citation density in the Australian veterinary field. Interestingly, a large study of veterinary citation classics conducted by Elsinghorst reported that the broad category of ‘microbiology’ was the most common theme among the highly-cited articles.9 The relative skewness of citation data identified during the current study is also consistent with other investigations of veterinary literature. In 1977, for example, Cho reported that less than 10% of the periodical titles cited contributed 66% of the total periodical citations.7 It has been suggested that the ‘usefulness’ of scientific information is always proportional to the time, effort and resources that are needed to actually find it.26 Similarly, while citation data and IF analysis represent a convenient way of evaluating the use of scientific journals, the relationship between reading an article and actually citing it has always been, and continues to be, notoriously difficult to evaluate. Although no single database provides access to all relevant literature in veterinary science,26 IFs and highly-cited articles nevertheless provide an interesting and useful insight into which authors, articles and topics are influencing the veterinary profession over time.