Very few studies have made use of new electronic resources such as Lex Machina, which provides full case dockets and other relevant data for intellectual property law cases. While the Lex Machina data is based on existing data from PACER, the federal court system’s official electronic filing database, Lex Machina has performed a significant amount of error correction and tagging on top of PACER’s docket reports. For instance, some cases have been corrected to account for errors in coding by the courts, case types have been resolved of false positives and negatives, and a vast amount of meta-data has been added to each case. Most importantly, the Lex Machina data includes important features such as case outcome; litigants; patents-in-suit; and tags for case events like summary judgments, trials, and appeals (which the PACER data does not include).”
To find my specific subset of patent cases, I use Lex Machina’s robust patents in-suit data (which lists all of the patents disputed in a single case) and match those patents with the FDA’s Orange Book data.
Thus, searching Lex Machina for all cases where any one of these patents are in dispute should provide a complete list of all potential ANDA cases.
I then limit my dataset to only cases filed between January 1, 2006 and August 1, 2011. While the Lex Machina data is excellent for all available years (2000 to present), it is particularly good after 2004, when most federal courts completed efforts to modernize their court filing systems.
Xiangnong Wang, Understanding Current Trends and Outcomes in Generic Drug Patent Litigation: An Empirical Investigation,Stanford University Public Policy Program, (2012)