A preliminary study on the difference between the citation counts of issued patents and their corresponding pre-grant publications

This study tries to address a basic question: do we miscount the patent citations? The citation count (i.e., the number of forward citations) of a patent is often considered an indication to the value or quality of the patent. However patents, specifically utility patents, are usually published 18 months after their applications are filed and before they are issued subsequently. These so-called pre-grant publications and the corresponding patents disclose the same inventions, and are both citable as relevant prior art by the applicants or examiners of subsequent patent applications. Most patent analysts however consider only the citations to the patents and ignore those to their pre-grant publications. This omission may lead to erroneous analytic result as a pre-grant publication has its own citations in parallel with its corresponding patent. This study assesses the impact of such omission by using empirical data from United States patent database. The result shows that citations to the pre-grant publications can be significantly more than those to the patents, and an analyst should not ignore the citations to the pre-grant publications when evaluating patents or conducting patent citation analysis.