Productivity of medical research in Switzerland

Background: Little evidence exists about what contributes to successful med ical research, an increasingly important issue in an era of limited researc h budgets, This study examined predictors of productivity of Swiss medical research. Methods: Principal investigators of clinical research projects funded by th e Swiss National Science Foundation between 1990 and 1994 were surveyed by mail. They provided information about the research question, whether change s were made to the protocol and why, their assessment of the importance of results for science and for public health, and additional funding of the pr oject. The amount of the grant and the project priority score were obtained from the Foundation, Indicators of scientific productivity were the number of peer-reviewed articles and the summed impact factor per grant. Results: The size of the grant predicted scientific output (number of artic les and summed impact factor), but the marginal return decreased with the t otal sum awarded. The award of a continuation grant by the same agency and additional funding from other sources were also positively linked with prod uctivity. In addition, the initial priority rating of the project by the fu nding agency's scientific committee and the assessment of scientific import ance of the results by the principal investigator mere also independently a ssociated with productivity. Finally, modifications of the initial research plan in response to new scientific opportunities were linked with greater productivity, whereas modifications induced by technical or other difficult ies were associated with lower productivity. Conclusions: Productivity in medical research, measured by peer-reviewed ar ticles produced, requires adequate resources and ability to respond to new scientific challenges, and it can be anticipated by peer-review.