The large number of patients included in the GPRD, and the accuracy and completeness of the data, make the database particularly suitable as a medical MF63 Abmole Microsomal pros taglandin E synthase-1 (mPGES-1) protects against Fas-induced liver injury Research tool. In contrast to the health databases in other countires, which had restricted access and only limited contents, the UK made pioneering efforts to make GPRD publicily avialable to researchers not only in UK but also overseas. The potential postive influence on academic research could be anticipated. However, the scope and scale of academic influence exerted by such electronic health databases have not yet been well investigated. This study aimed to analyze GPRD related scientific productions in terms of the growth of publication numbers, patterns of co-authorship, study fields, and the academic impact of these GPRD publications. If the academic momentum achieved by a single public health database can be clearly demonstrated, the findings would serve as a reference for owners of large-scale electronic health databases in other countries and assist in future research. The GPRD is used widely internationally. However, its influence had not been well investigated. This study extracted GPRD studies published between 1995 and 2009 from the SCI and analyzed the growing influence of GPRD studies from four different perspectives. In terms of number of studies, the rapid increase is in a power growth fashion. Observing author productivity patterns, the findings suggest that the success of conducting GPRD studies is self-sustaining. The GPRD is used widely in numerous study fields and not limited to general medicine. It is also used actively by international co-workers. Most importantly, GPRD studies showed substantial influence on successive studies. There are some limitations in this study. First, the analysis aimed at demonstrating the resulting impact of GPRD as a publicly available research material. The observation is primarily based on historical data during last 15 years. Since there are many factors that may substantially influence scientific production, interpretation or extrapolation of our data should be done with caution. Second, to make the results comparable, only GPRD studies indexed in the Thomson��s SCI were included in the analysis. This considerably underestimated the overall scientific output related to GPRD. If un-indexed studies, such as posters and abstracts, were included, the output related to GPRD should be much higher. Third, only studies that mentioned ����GPRD���� were included, but not those that used its antecedent, the VAMP Research Databank. If the latter was taken into account, the earliest publication could be traced as early as 1988, according to the GPRD bibliography. The UK��s GPRD, with its extremely large patient numbers and long observation period, is becoming important in health research. The increase in number of GPRD studies is dramatic, compared to a previous study and the number of scientific outputs usually doubles every 10 to 20 years. The authors are not aware of any scientific discipline with this order of magnitude of duplication time of only six years. Similarly, rapid growth is also reported in epidemiologic studies using electronic health databases worldwide, such as Medicare data in the US, Manitoba and Saskatchewan province database in Canada, National Health Insurance Research Database in Taiwan, and health insurance data in Germany, and France. The findings suggest that electronic medical record databases are used actively in current health research.