GAO Makes Seven Recommendations for Improving Prior art Searches

Last month, we summarized and analyzed one of three reports issued by the Government Accountability Office (GAO) this summer, Patent Office Should Define Quality, Reassess Incentives, and Improve Clarity. This month, we summarize and analyze a second report from the GAO, Patent Office Should Strengthen Search Capabilities and Better Monitor Examiners' Work.

Purpose of the Report

Acting under a request from the House of Representatives Judiciary Committee, the GAO looked at how the USPTO goes about the process of examining patent applications to ensure that inventions meet two key requirements for a U.S. Patent: They are novel and not obvious. Patent examiners accomplish this by comparing applications to “prior art” (existing patents and applications in the United States and abroad as well as any non-patent literature such as scientific articles). Thorough prior art searches help ensure the validity of granted patents.

The GAO was asked to identify ways to improve patent quality through use of the best available prior art. The GAO included in its report:

  • The challenges examiners face in identifying relevant prior art
  • How selected foreign patent offices address challenges in identifying relevant prior art
  • The extent to which the USPTO has taken steps to address challenges in identifying relevant prior art.

Survey of Patent Examiners

In addition to surveying a cross section of USPTO patent examiners, the GAO interviewed experts active in the field such as patentees, attorneys, academics, and officials from both the USPTO and similarly sized foreign patent offices.

The survey conducted by the GAO was of 3,336 patent examiners. It asked the patent examiners for their perceptions of the Patent Office’s approach to patent quality, the challenges the agency faces in finding relevant prior art, and how the USPTO might improve patent quality and its prior art search capabilities. The GAO received responses from 2,669 patent examiners, an 80 percent response rate.

The survey was conducted across a random sample of patent examiners from eight of the technology-based subject matter groups into which patent examiners are divided, and patent examiners who had been with the Patent Office for less than one year were excluded.

One of the most telling questions was No. 3. It asked patent examiners approximately how many hours they spent per patent application on prior art searches over the past quarter in the preparation of the original First Action on the Merits, the notice patent examiners send to applicants about the patentability of their inventions. Here are the results for the eight technology areas.

Technology Area Average Hours
Biotechnology and Organic Chemistry 5.2
Chemical and Materials Engineering 6.7
Computer Architecture, Software, and Information Security 10.9
Computer Networks, Multiplex Communication, Video Distribution, and Security 8.8
Communications 8.9
Semiconductors, Electrical and Optical Systems and Components 7.4
Transportation, Construction, Electronic Commerce, Agriculture, National Security,and License and Review
Business Methods 7.3
Not Business Methods 7.1
Mechanical Engineering, Manufacturing, and Products 7.9
Overall Average 7.8

 
Foreign Patent Office Input

The GAO found that the European Patent Office and the Japan Patent Office face similar challenges to the USPTO when it comes to identifying prior art, and these offices use various approaches to improve prior art searches such as leveraging work of other patent offices on related patent applications, and integrating non-patent literature into their search tools. In some cases, these approaches are coordinated with – or are very similar to – what the U.S. Patent Office does.

GAO Recommendations and USPTO Responses

A draft of this report was given to the U.S. Patent and Trademark Office prior to the report’s official release, so we include here both the GAO’s seven recommendations, and the Patent Office’s response to each.

Recommendation No. 1: To ensure that USPTO's collaborative efforts on classification help examiners find relevant prior art, the USPTO should work with the European Patent Office (EPO) to identify a target level of consistency of Cooperative Patent Classification (CPC) decisions between USPTO and EPO, and develop a plan to monitor consistency to achieve the target.

USPTO Response: The USPTO concurs with this recommendation. Efforts are currently underway between the USPTO and the EPO to develop objective, consistent target levels called "objective metrics" for CPC. These objective metrics will allow for the determination of CPC symbol allocation by both the USPTO and the EPO in an effort to identify areas of classification practice convergence and divergence. Divergent areas will then be targeted for further investigation and corrected through bi-lateral training, revision, and reclassification. The USPTO has also been working with other partner patent offices, the Republic of Korea, in particular, to develop consistent target levels. Objective metrics are important to ensure confidence in the CPC system and confidence of examiners in their search. The USPTO's objective metrics have been well received by Korea in helping its office produce work sharing.

Recommendation No. 2: To ensure that the USPTO is able to take full advantage of its investment in new information technology tools and capabilities, the USPTO should develop and periodically update a documented strategy to identify key sources of non-patent literature for individual technology centers and to assess the optimal means of providing access to these sources, such as including them in the USPTO's search system.

USPTO Response: The USPTO concurs with this recommendation. The USPTO's Scientific and Technical Information Center (STIC) currently conducts analyses on existing and new non-patent literature (NPL) sources to be made available to individual technology centers as well as assesses usability for incorporation into the USPTO's search system. The USPTO currently has a documented monthly process whereby it measures and reports on NPL utilized by individual technology centers and adjusts NPL sources accordingly. As part of the analysis and implementation of the elements of this recommendation, the USPTO will investigate the development of a documented strategy to identify key sources of non-patent literature for individual technology centers and assess the optimal means of providing access to these sources.

Recommendation No. 3: To improve its monitoring of prior art searches and provide the USPTO with the ability to examine and address trends in prior art search quality at the technology center level, the USPTO should develop written guidance on what constitutes a thorough prior art search within each technology field, technology center, art area, or art unit – as appropriate – and establish goals and indicators for improving prior art searches.

USPTO Response: The USPTO concurs with this recommendation. The USPTO will develop technology-based search training guidance and establish enhanced goals and indicators for improving prior art searches.

Recommendation No. 4: To improve its monitoring of prior art searches and provide the USPTO with the ability to examine and address trends in prior art search quality at the technology center level, USPTO should ensure that sufficient information is collected in reviews of prior art searches to assess the quality of searches at the technology center level, including how often examiners search for U.S. patents, foreign patents, and non-patent literature.

USPTO Response: The USPTO concurs with this recommendation. A great deal of information is collected and recorded by examiners during a search of prior art. In addition, the Office of Patent Quality Assurance also makes efforts to sample examiner search data during its review of issued Office Actions. These efforts are in addition to the review of search strategy and results carried out at the technology center level by training primary examiners, supervisory patent examiners, and quality assurance specialists. These reviews are utilized by the examiners and supervisors to make the search process more effective and more efficient. Responsive to this recommendation, the USPTO will work to ensure that adequate search data are collected to assess the quality of searches at the technology center level.

Recommendation No. 5: To improve its monitoring of prior art searches and provide the USPTO with the ability to examine and address trends in prior art search quality at the technology center level, USPTO should use the audits and supervisory reviews to monitor the thoroughness of examiners' prior art searches and improvements over time.

USPTO Response: The USPTO concurs with this recommendation and will investigate utilizing audits and various reviews to monitor the thoroughness of patent examiners' prior art searches and improvements over time.

Recommendation No. 6: To ensure that examiners have sufficient time to conduct a thorough prior art search, the USPTO should – in conjunction with implementing the recommendation from GAO's patent quality report to analyze the time examiners need to perform a thorough examination – specifically assess the time examiners need to perform a thorough prior art search for different technologies.

USPTO Response: The USPTO concurs with this recommendation. The USPTO intends to further investigate the time examiners need to perform a thorough prior art search, including an assessment of the time examiners need to perform a thorough prior art search for different technologies. The USPTO has already completed an initial investigation of the time examiners need to perform a thorough patent examination prior art search. This initial investigation occurred as a result of feedback received during the transition from the United States Patent Classification System (US PC) to CPC. The investigation found that additional time for searching was warranted in certain technologies. Due to the findings of this initial investigation, the USPTO provided approximately 1,000 examiners with an adjustment of an additional 2.7 hours to their current GS-12 expectancy for certain classes/subclasses.

Recommendation No. 7: To ensure that examiners have the technical competence needed to complete thorough prior art searches, the USPTO should assess whether the technical competence of examiners in each technology center match those necessary, and develop strategies to address any gaps identified – such as a technical training strategy – and establish measures to monitor progress toward closing any gaps.

USPTO Response: The USPTO concurs with this recommendation and will assess whether the technical competence of examiners in each technology center matches those necessary and develop strategies to address any gaps identified.

We find Recommendation No. 6 the most telling. As a result, the USPTO patent examiners will be spending an additional 2.7 hours on prior art searches, an increase over the current average of 35 percent!