About the Authors

  • The Authors and Contributors of "Patent Docs" are patent attorneys and agents, many of whom hold doctorates in a diverse array of disciplines.
2018 Juristant Badge - MBHB_165
Juristat #4 Overall Rank

E-mail Newsletter

  • Enter your e-mail address below to receive the "Patent Docs" e-mail newsletter.

Contact the Docs


  • "Patent Docs" does not contain any legal advice whatsoever. This weblog is for informational purposes only, and its publication does not create an attorney-client relationship. In addition, nothing on "Patent Docs" constitutes a solicitation for business. This weblog is intended primarily for other attorneys. Moreover, "Patent Docs" is the personal weblog of the Authors; it is not edited by the Authors' employers or clients and, as such, no part of this weblog may be so attributed. All posts on "Patent Docs" should be double-checked for their accuracy and current applicability.
Juristat #8 Overall Rank


« Views on Venue -- District of Delaware Provides Some Guidance on Venue in ANDA Cases Post-TC Heartland | Main | In re Cray Inc. (Fed. Cir. 2017) »

September 20, 2017


I think that the "death panels" comment was more accurate than not at the time, but people keep throwing this quote around even though conditions "on the ground" have changed quite a bit.

Cast yourself back to the days when IPRs first came online. I can remember visiting clients to pitch our services in IPRs, and time and again the response was that they were concerned about the estoppel provisions and did not want to involve themselves with these new, unfamiliar procedures. Better to stay with the old, familiar DJ proceedings that they knew so well.

That meant that the earliest IPR petitions dealt only with the most transparently infirm claims. You only filed an IPR if you thought "this claim is *so* clearly invalid that there is no other possible outcome than a declaration of unpatentability." If the claims were less clearly invalid (or, to put it another way, more clearly borderline), challengers were going with DJ actions instead.

The result, of course, was that an overwhelming majority of petitions were both granted and gave rise to holdings of unpatentability. If you only petition against the most clearly invalid claims, then the only results that will be possible will be cancellations. This result is a selection bias phenomenon, not any sort of intrinsic feature of the PTAB.

The next step in this progression was quite predictable: once would-be challengers saw the steep rate of invalidity holdings, they said to themselves "these PTAB trials are death-panels. Let's file our challenges there!" There was, then, a corresponding rush of petitions against stronger claims, with a resulting increase in the rate at which petitions were denied and claims upheld.

We have now reached an equilibrium in which the rate of invalidation is much less than it was at the time of the "death panels" remark. It is time to acknowledge that this characterization is no longer accurate.

So 60% are computer, e.g., how to transfer telephone calls and store data, or CBM, e.g., how to locate a store by zip code, and less than 20% are chemical/biotech, and likely most are secondary life-management pharma patents, e.g., changing dose from 0.1 to 0.05%, and saying wow it still has some efficacy, that was unexpected! And then transferring ownership to the Mohawk-St Regis. What a joke!

Get in the lab or to the bench top and invent. The PTO is doing its job!

Using your numbers, these are hardly "death panels." To prevail as a petitioner, you need to have both IPR instituted (63% chance) AND prevail at PTAB trial (win on all claims - 65% chance). Multiplying the probabilities (because a loss at either stage is a loss) and petitioner has ~ 41% chance of success. Patent owner success rate is also in the 40s% when you look at applications that are denied outright at institution stage and cases where PO wins at trial. The rest of the cases are split decisions.

One does not employ product recalls (in and of themselves)as any form of manufacturing quality improvement.

ANYONE stating that what amounts to recalls as being effective in fixing the application examination process simply has no credibility.

Anonymous at 11:48 - your supposition (and hence your math) is not correct.

You do not need ALL in each instance, thus your attempt at multiplying probabilities is plain error.

The comments to this entry are closed.

July 2024

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31