Beyond outputs: new measures of biomedical research impact.

Biomedical research evaluation has traditionally been based on analysis of outputs and their citations by other papers. However we should try to map the routes by which research actually improves patient care and reduces illness, and develop indicators for them. We must allow for the lengthy time-scales involved and the importance of researchers being physically close to healthcare professionals, whose practice can be improved through international and governmental regulations and through approved guidelines. Each of these will depend on a body of research evidence. We must also evaluate the effects of research on policy makers and the public, who often learn about it through the World Wide Web and through the mass media, particularly newspapers. The latter provide a major bibliometric resource but one that needs to be tapped in individual countries using common standards in order to provide internationally-comparable indicators. [ABSTRACT FROM AUTHOR]