The relation of patent description and examination with validity: an empirical study

Abstract

This paper describes an empirical study to identify factors of patent description and examination that affect the outcome of later challenges to patent validity. Using ternary dummy variables based on the change in the scope of protection before and after Japanese invalidation trials as the dependent variable, the study compares models estimated by ordered and binary logistic regression. The standardized length of the first part of the description explaining the prior art, and the second part explaining the details of the invention, were found to relate to the "maximum" and "minimum" survival of patents, respectively. Other factors were also identified. Different modes of survival are discussed. Models on the likelihood of a challenge to patent validity were also estimated. The variables related to patent validity were compared to those related to the likelihood of a challenge. This revealed factors affected by selection biases based on the opponent's decision to make the challenge. Factors including the first and second parts of the description were not affected by selection biases and thus confirmed to be related to patent validity. Factors that not only affect patent validity but also affect an opponent's decision to challenge patent validity are also discussed.