Approaches to Measuring Trends in Interdisciplinary Research Publications at One Academic Medical Center

Purpose: To determine if interdisciplinary research has increased between 2005 and 2015, based on an analysis of journal articles containing at least 1 author from Johns Hopkins University, and to compare different methods for determining the disciplinarity of research articles. Method: In 2017-2018, 100 peer-reviewed biomedical science articles were randomly selected from years 2005, 2010, and 2015 and classified as unidisciplinary or interdisciplinary based on Scopus author affiliation data (method 1). The corresponding authors of the 2010 and 2015 articles were sent a survey asking them to describe the disciplines involved in their research (method 2) and to define their research as unidisciplinary or interdisciplinary based on provided definitions (method 3). Results: There was a statistically significant increase in the proportion of interdisciplinary articles in 2015 compared with both 2005 and 2010 (P =.02). Comparison of the 3 methods indicated that 45% of the articles were classified as interdisciplinary based on author affiliation data (method 1), 40% based on the corresponding author's description of the disciplines involved in their research (method 2), and 71% based on the corresponding author's definition of their article's disciplinarity (method 3). There was a statistically significant difference in the proportion of articles classified as interdisciplinary between methods 1 and 3 (P <.001) and between methods 2 and 3 (P <.001). Conclusions: This study found that interdisciplinary research increased at Johns Hopkins University over the past decade and highlights the difference between corresponding authors' views of their own research and other methods for determining interdisciplinarity.