Type 2 Diabetes Research Yield, 1951-2012: Bibliometrics Analysis and Density-Equalizing Mapping

The objective of this paper is to provide a detailed evaluation of type 2 diabetes mellitus research output from 1951-2012, using large-scale data analysis, bibliometric indicators and density-equalizing mapping. Data were retrieved from the Science Citation Index Expanded database, one of the seven curated databases within Web of Science. Using Boolean operators "OR", "AND" and "NOT", a search strategy was developed to estimate the total number of published items. Only studies with an English abstract were eligible. Type 1 diabetes and gestational diabetes items were excluded. Specific software developed for the database analysed the data. Information including titles, authors' affiliations and publication years were extracted from all files and exported to excel. Density-equalizing mapping was conducted as described by Groenberg-Kloft et al, 2008. A total of 24,783 items were published and cited 476,002 times. The greatest number of outputs were published in 2010 (n=2,139). The United States contributed 28.8% to the overall output, followed by the United Kingdom (8.2%) and Japan (7.7%). Bilateral cooperation was most common between the United States and United Kingdom (n=237). Harvard University produced 2% of all publications, followed by the University of California (1.1%). The leading journals were Diabetes, Diabetologia and Diabetes Care and they contributed 9.3%, 7.3% and 4.0% of the research yield, respectively. In conclusion, the volume of research is rising in parallel with the increasing global burden of disease due to type 2 diabetes mellitus. Bibliometrics analysis provides useful information to scientists and funding agencies involved in the development and implementation of research strategies to address global health issues.