COULD BIBLIOMETRIC DATA BE USED TO PREDICT THE CLINICAL SUCCESS OF DRUGS? Available to Purchase

Bibliometric traits were used to describe the development and clinical use of levodopa as an antiparkinsonian agent. Yearly percent compositions of a complete collection of 2,335 papers, published from 1960 through 1974, which reported the administration of levodopa to humans, were obtained for: single‐author, non‐regular‐journal, DOPA‐word in title, short, O reference, English language, and contemporary papers. From this basic description a hypothesis was erected, according to which these bibliometric traits can be used for predicting the future of a developing drug. A successful drug (a drug of choice) will have a development peak and a clinical‐use peak on a graph plotting the annual slopes of the percent composition of the trait against the year, for ‘bursts’ of 3–6 years. The main premonitory event (predictor) is the switch in polarity of the slope. Single‐author papers were judged to be the best bibliometric trait for predictive use.