Discussion
Over the last decades, the scientific literature available in football has dramatically increased both in terms of quantity and quality. However, to our knowledge, this is the first bibliometric analysis conducted in football. Here, we show that the number of citations for the 50 most frequently cited articles in medicine and science in football ranged from 251 to 869.
A bibliometric analysis is an extraction of statistics on journal articles in a research field. Its structure resembles the design of a systematic review.9 Bibliometric analyses are commonly used to measure the impact of scientific articles, but should not be considered as the standard reference of scientific quality of the investigations. Citation rankings may also provide an estimate for trends for the research undertaken worldwide in a specific research subject (eg, football). We observed that 80% of the articles presented in the current ranking were original research, but only three articles had a clear experimental design. Observational studies (eg, notational analysis of performance) have a long tradition in football research, aiming the identification of critical events (often referred as performance indicators) that are thought to be the keys to success in the sport.10 In fact, observation studies are essential for a comprehensive awareness of the distinct individual and team performance components. Notwithstanding, the lack of experimental studies within the top 50 most cited articles in football clearly underpins how far we still are from establishing the theoretical and methodological guidelines for the applied science and medicine in football.
Nine review articles, but only one meta-analysis, were included in the current ranking. Meta-analyses characterise and combine findings of prior studies in order to increase statistical power, provide quantitative summary estimates, and identify data gaps and biases.11
Some of the limitations for conducting scientific studies in football are well known: it is difficult to have access to large sample sizes of top-level players, longitudinal training studies are arduous and laborious, and studies that address and replicate appropriate real-world methodological issues are rare. However, despite the important role of narrative review articles, the debate within the community dealing with medicine and science in football might still be far from high levels of evidence.
Still, the theoretical knowledge, methodological approaches and professional practice in football are supported by comprehensive and complex contributions from a variety of sources.12 In this line, the input of individuals involved in sports medicine, both from a clinical and scientific perspective, should be highlighted. It is noteworthy that 44% of the articles presented in the current ranking were related to sports medicine. The other topics main topic areas were training and testing, performance analysis and physiology, but no study within the top 50 most-cited was devoted to biomechanics, nutrition, sport psychology, coaching or social sciences.
The articles presented here were all published between 1983 and 2011. Online databases have limitations in tracking older articles, especially for articles published in alternate media, textbooks or meeting symposia,4 which may have been missed in the current analysis. In addition, it can also be argued that old data are of limited value in the current rapidly changing sports environment. It should be noted that the impact of recent studies could not be captured with the current analysis. Any article needs some time after publication to receive enough number of citations. However, this is common in all research fields and does not negate the trend and conclusions presented here. We might expect a relative emergence of systematic research analysis and experimental studies in medicine and science in football over the last few years, underpinning the recent growing interest in those areas and, at the same time, highlighting the need for more research in the future. Still, the current study may also be limited by other research publication dynamics, which can include publication biases, as language, predominance of positive results, journal preferences when citing and self-citation.13 14
In the current manuscript, we present a network of authors’ collaborations within the scope of medicine and science in football. This network underpins the most prolific scientific collaborations within medicine and science in football. Five main clusters of network collaborations were identified. We hope the current manuscript can establish an historical, comprehensive analysis to the most influential scientific literature published within the area of medicine and science in football, helping researchers, sports scientists, clinicians and practitioners on developing new insights to the original theoretical and methodological understanding of the game of football.