Federal Courts Erroneously Continue to Claim that FDA Only Spends 20 Hours On Average Reviewing 510(k)sJune 10, 2015
By Jeffrey K. Shapiro –
In 1987, an FDA official testified that 20 hours were spent on average to review a 510(k) submission. Hearings before the Subcommittee on Health and the Environment of the House Committee on Energy & Commerce, 100th Cong., 1st Sess. (Ser. No. 100-34), p. 384 (1987).
We have not been able to locate any testimony updating this figure, albeit our search was not exhaustive (we have a day job). We filed a Freedom of Information Act (FOIA) request for documents with this information and were told by FDA that there were no responsive documents.
In any event, the 20 hour figure put forward by FDA was relied upon by the U.S. Supreme Court in Medtronic v. Lohr, 518 U.S. 470, 479 (1996) (here). In that case, the Court was reviewing whether a products liability claim against a medical device cleared in 1982 was preempted under Section 521 of the Federal Food, Drug, and Cosmetic Act (FDCA). Therefore, at the time, and for purposes of that case, the 20 hour figure was reasonably accurate.
The trouble comes afterward. Since Medtronic v. Lohr was decided in 1996, by our informal count, the 20-hour figure has been regurgitated in at least 52 federal court opinions down to the present day. Last year, there were three opinions that reiterated the 20-hour figure. There has already been another one this year (see here, here, here, and here.)
Anyone associated with filing 510(k) submissions knows that this 20-hour figure is hopelessly outdated. In the 1980s, a 510(k) tended to be a short document with a narrative comparison of a proposed device to a predicate device and little or no supporting data. The length and complexity of 510(k) submissions today far exceeds 510(k) submissions in the earlier bygone era. That is a function of the greater complexity of devices today, the substantial data and information requirements now part of the 510(k) program, and the ongoing exemption of lower risk devices from 510(k) review, thereby biasing the pool of 510(k) submissions toward greater complexity. We wrote in great detail about the statutory and administrative evolution of the 510(k) program here. Virtually all of the important changes occurred after 1990. Just as today’s Honda Accord is much more powerful, feature laden and reliable than one purchased in 1982, so the 510(k) program is much more robust than the one FDA operated in the 1980s.
Just last year, our firm assisted with the preparation of a 510(k) submission that was approximately 8,000 pages. The review team at FDA did not keel over with fright, even though they doubtless knew that this review would take them a lot longer than 20 hours. In fact, after a couple of months of review, they asked for a whole lot more information.
A moment’s reflection is all it should take to realize how implausible it is that a federal government program would remain frozen in place for more than 30 years with no changes. Of course it would metastasize over time, both as to scope and burden. The 510(k) program is certainly no exception.
The only thing that has remained frozen is the federal judiciary’s description of the 510(k) program. (In addition to the 20-hour review figure, we also note that many cases repeat descriptions of the 510(k) program taken from journal articles written by law practitioners in the 1980s.) Our purpose here is not to wade into the merits of applying preemption in medical device products liability cases. Our point is simply that the federal courts keep reiterating information that is manifestly wrong. If this information is, even in part, a basis for present day decision-making about preemption in products liability cases, that is a serious embarrassment to the federal courts.