A Historical Perspective of Pediatric Publications: A Bibliometric Analysis

The world of medical research is rapidly proliferating. Since 1945 there have been more than 40 million publications referenced on the Thompson Reuters Web of Science (WoS). Pediatric academia has been in existence since 1471 with Bagallarder's Little Book on Disease in Children.1 However, pediatrics as a specialized field of medicine only developed in the mid-19th century. Archives in Pediatrics was the first pediatric journal and was established in 1884. The very first pediatric publication was “Convulsions in Children” by William T. Plant.2 Since then, the number of publications on the subject of pediatrics has dramatically expanded. There are now 191 journals solely dedicated to the specialty of pediatrics and in 2010 alone, more than 22 800 pediatric articles were published.Bibliometric methods, including citation analysis, have been used to evaluate trends in publications within specialties including neonatal surgery.3 They have also been used to examine highly cited articles within a specialty,4 providing insight into what types of article are of importance and interest to the readership. In this article, we identify and analyze the top 100 most frequently cited articles published in journals dedicated to pediatrics between 1945 and 2010. We identify and interpret the characteristics that mark these most-cited articles.A comprehensive search of the WoS database was performed with the category “pediatrics” selected. Data were retrieved for the reference period 1945 to 2010. All searches were conducted between March 2012 and August 2012. Search results from WoS included entries from the “science citations index expanded,” “social sciences citation index,” and “arts and humanities citation index” databases. Data downloaded from the WoS was assessed and analyzed by using Microsoft Excel spreadsheet software.There were 497 240 articles published in 191 pediatric journals between 1945 and 2010. Just under 50% of these articles were published in the past 16 years (Fig 1).The 100 top cited publications are listed in Table 1. The mean number of citations from this cohort was 782.3 (range 457–2284). Of the top 100 cited articles published, 94 were original articles with 2 of these proceeding papers. Only 2 of the articles were laboratory-based studies. There were 28 articles on endocrinology. The next most published category was neonatology, with 22 articles, followed by psychiatry with 15. Respiratory and immunology had 7 and 6 publications respectively. English was the written language in all but 1 publication. The countries where these articles originated are shown in Fig 2.The 100 top-cited publications are to be found in only 16 journals (Table 2). There were 89 of these articles published in just 5 journals: Pediatrics (40), Journal of Pediatrics (22), American Academy of Child and Adolescent Psychiatry (10), Archives of Disease in Childhood (9), and JAMA Pediatrics (8).Analysis of the top 100 cited articles in pediatrics provides a diverse yet compelling read. One of the most striking features of the list is the strong bias toward recently published articles (Fig 1). Although it could be argued that this is because of newer medical technology or advances in science, only 2 of the articles are laboratory based. The bias toward recent articles could be explained by the rapidly expanding role of the Internet in medical publishing. The almost global progression of medical journals to online publishing has made it easier and quicker to find an article. In addition, the development of multiple integrated search engines enables articles to be found using just keywords. Recent articles will be more likely to be published online and have keywords that can be found using a search engine. They are therefore easier to access, which creates a bias for articles published in the Internet era.Although the bias toward recent articles exists, there are articles on the list written many years ago that are highly cited. Interestingly, most of these articles are not clinical trials providing a scientific breakthrough, as one might assume. These articles are of a descriptive nature, providing, for example, a clinical observation or a syndrome. Bruton’s article on agammaglobulinemia published in 1952 and Menke’s description of a gender-linked degenerative disorder with abnormal hair are 2 such examples. Kawasaki’s article is another example of the descriptive theme seen in the older studies that feature in the top 100 list. It is fascinating to consider that most of the older articles do not represent major breakthroughs in basic science. The only such example in the top 15 is Avery’s article on hyaline membrane disease. Despite this, the older articles continue to be cited, particularly because their findings were novel, yet pertain to pediatricians in current medical practice.The lack of clinical trials in the top 100 cited articles is not exclusive to the older articles. Although the more recent articles do not report new syndromes or clinical observations, they are still descriptive in nature. Most of the studies in the list have similar characteristics in that they describe criteria or definitions that can be used as tools by pediatricians: Tanner staging, Dubowitz examination for gestational age, and the 4 grades of intraventricular hemorrhage. All but 1 of the top 10 in the list belong to this category. That there is a lack of clinical trials in the top 100 citations may be attributed to the ethical constraints of conducting research in pediatrics. Notwithstanding, medical research is important to ensure optimal evidence-based care in children.It is not surprising that the top-cited articles were published in higher-impact journals. In our cohort, these journals include Pediatrics (40), Journal of Pediatrics (22), and American Academy of Child and Adolescent Psychiatry (10).This top-cited list has limitations, including the retrospective nature of the search. In addition, when the initial search was performed, only dedicated pediatric journals were included. Although this could not have been avoided at the time, articles from prominent journals, such as New England Journal of Medicine and The Lancet among others, may have been excluded.This article provides a historical perspective on the publications in pediatrics between 1945 and 2010. Citation analysis of pediatric publications reveals a sustained interest in articles that are based around criteria or guidelines rather than clinical trials. Older articles describing important syndromes or observations are still popular but there is a strong bias toward recent articles. This can be attributed in part to the ever rapidly expanding role of the Internet in evidence-based medicine, which provides a manuscript for directing patient care, exploring new ideas, and guiding the path for future research.