Medical Device Link.

Originally Published MDDI May 2001

BOTTOM LINE

What has been the effect of the landmark legislation on the device industry? An analysis of one key sector draws some interesting conclusions.

Richard W. Treharne and Thomas L. Craig

President Gerald Ford signed the Medical Device Amendments into law in 1976.

Winston Churchill once declared that "the farther backward you can look, the farther forward you are likely to see." May 28, 2001, marks the 25th anniversary of the signing of the Medical Device Amendments law by President Gerald Ford. In light of this anniversary, it is appropriate to pause to examine the effect of this law over the last 25 years. This article presents a detailed analysis of its effect on a single medical device industry sector— orthopedics. The orthopedics sector represents approximately 10% of the entire device industry, and can be considered typical of the industry as a whole.

In 1996 the entire U.S. medical device industry was estimated to total about $40 billion in sales. Today, the domestic market is estimated at $66.5 billion annually. Depending on what products are included and when the data are examined, most analysts would estimate the size of the current orthopedic market to be about $6.5 billion. The orthopedic sector comprises such products as arthroscopy instruments, diagnostics, trauma devices, total hips and knees, spinal implants, and bone substitutes, among many others.

For new orthopedic devices to enter the market after May 28, 1976, the Medical Device Amendments required that the devices first be reviewed by FDA through one of two mechanisms. Premarket notification via the 510(k) process was required for products claiming "substantial equivalence" to a device on the market prior to passage of the law, and the investigational device exemption (IDE)/premarket approval (PMA) route was mandated for more innovative, non–substantially equivalent devices.

510(K) HISTORY

A review of the 510(k) database on FDA's Web site and a book entitled The 510(k) Register makes it possible to put together an estimate of the number of orthopedic 510(k)s over the past 25 years for about 11 companies and their consulting groups. ¹ Although these data may not be perfectly accurate—some consulting firms that performed 510(k) submissions on behalf of client companies may have been overlooked and the most recent 510(k)s are not counted—the total, 1071, does give a representative view of the number of 510(k)s for orthopedic devices that have been cleared since 1976. It should be noted that the list of companies is not comprehensive, since many other orthopedic companies exist and have not been included in this analysis. The key point is that for the 11 companies checked, the average number of 510(k)s cleared each year over the past 25 years was approximately four per company.

IDE/PMA HISTORY

Before analyzing the number of PMA applications that have been approved by the FDA's Orthopedic Advisory Panel since 1976, it is helpful to know the number of orthopedic IDEs, since only IDEs can possibly turn into PMAs. (The one possible exception would be orthopedic device PMAs granted by FDA based solely on foreign data. At least one of the orthopedic devices in the PMAs listed fits this category.)

FDA has publicly stated that, since 1976, there have been at least 31 IDEs on bone substitutes and at least 21 IDEs on pedicle screws. None of the pedicle-screw IDEs ever led to approved devices, and only two of the 31 IDEs for bone substitutes have become commercially available products. In other words, out of 52 IDEs, only two resulted in approved devices. Although some IDEs could have been for the purpose of collecting data for 510(k) applications, and some of these may have been too recent to have been converted into PMAs, according to FDA's program operations staff, as of September 2000, 249 orthopedic IDEs have been approved since May 1976. This means that, on average, FDA has approved approximately 10 orthopedic-related IDEs per year. Because it can take years for enough clinical data to be gathered to support a PMA application, we will assume that a fair number of IDEs to use for calculating a PMA conversion rate is 200.

Of the assumed 200 eligible IDEs since 1976, the advisory panel and/or FDA has only approved 43 products. A summary of these approved orthopedic devices can be found in Table I.

Of the 43 devices that have reached the market, it appears that all were available outside the United States first. At most, only about half of the 43 were (are) on the U.S. market for five or more years after FDA approval (one possible reason being because many of the products had already gone through their life cycle outside the United States). Of those on the market after five years, only a few can truly be considered to have caused dramatic or permanent shifts in orthopedic surgical practices or procedures. Furthermore, some of the PMA products were already on the market for differently labeled indications via the 510(k) process—for example, hips and knees with porous coatings—and were not completely novel. In these cases, the PMA amounted to a labeling change of an existing product.

Perhaps a more interesting analysis begins with the fact that, over the last 25 years, the average time from submission of the PMA application to final approval was approximately 2.5 years. To be added to that time frame would be the time required for preclinical IDE testing, IDE approval, IRB approvals, patient enrollment and follow-up, and compiling and reporting of results. A conservative estimate of the average time required for all these functions to be accomplished would be six years. Furthermore, apparently none of the IDEs/PMAs in the list required that human pilot studies be performed. Mandating such studies is a current FDA trend for novel technologies, and would add at least two additional years to the process.

Therefore, based on the past 25 years' experience, this analysis shows that if an investment firm were to invest in a company that has an IDE product in orthopedics, it would have roughly a 21% chance of receiving a return on its investment. If, after approximately six years, a PMA were granted, the investors would have only about an 11% chance of receiving any return for more than five years, and less than a 5% chance of receiving a significant return on the investment. These calculations show that a $6.5 billion industry comprising approximately a dozen major companies produces on average a little less than two PMA products per year.

Why the low number of PMAs? One reason is the potentially multimillion-dollar cost required to conduct a clinical study, since few orthopedic devices warrant such an expense. Another reason, which is related to the cost, is FDA's interpretation of "reasonable assurance" for product safety and efficacy. Some might argue that this phrase has often been assumed by FDA to mean "unquestionably safe and effective"—a condition difficult to prove to the agency's satisfaction without studies and data that are beyond reproach. The question to be asked is whether the practical outcome of the IDE/PMA process for this particular industry sector—which may or may not be typical for all sectors—was what Congress intended when the law was passed 25 years ago.

Are there additional reasons why so few orthopedic device IDEs are being converted into PMAs? Among the possible explanations are that patent-interference issues may have come to light during some studies, financial constraints or other business matters may have hindered final completion of other studies, or IDE conduct problems may have prevented completion of PMA reviews. Another explanation is that the market may have changed during the course of some studies, making their continuation a moot point. Of course, there is always the possibility that the investigational product may not have worked as well as planned.

THE EFFECT OF THE PASSAGE OF THE MEDICAL DEVICE AMENDMENTS

Would President Ford—who himself received a total artificial knee implant—have signed the Medical Device Amendments 25 years ago if he had known the effect the law would have on the introduction of new products into orthopedics? Former FDA chief counsel Peter Barton Hutt has stated that the 1976 Medical Device Amendments were passed 14 years behind schedule. ² What would have happened if President Kennedy had signed the same law into effect in 1962? Would President Ford have been able to benefit from the type of surgery he had? An argument could be made that he wouldn't have, since the 14-year "delay" in signing the law allowed total hips, total knees, many spinal implants, and other orthopedic devices to enter the market prior to 1976 and those devices were subsequently improved via the 510(k) process. Ironically, it was the introduction and sale of these 510(k) products that allowed the orthopedic industry to fund the IDE studies that have been performed. It is also ironic that the growth of the industry has come about without much impetus from PMA-type products.

Even with the 14-year delay, for nearly 10 years after the signing of the 1976 law its effect was quite minimal. During those early years, 510(k)s took only about 30 days to be cleared, with difficult ones perhaps requiring as long as six weeks. In 1993, the average time for 510(k) reviews was nearly a year. While in recent years approval times have become shorter, any future increase in volume or complexity of 510(k)s can only increase the delays in the introduction of new orthopedic products to the market.

Another reason for the delays is that the drafters of the Medical Device Amendments probably did not intend for the 510(k) process to go on forever. Rather, they probably envisioned a system of small-scale, simple clinical trials prior to the introduction of any new medical device in the United States. Former FDA commissioner David Kessler expressed this view at a May 1, 1996, House Commerce and Health Committee subcommittee hearing when he said, "I think we need to move away from this whole idea of substantial equivalence . . . It doesn't make any sense." ³ FDA appears to agree, since it has been increasing its criteria for orthopedic product 510(k)s to the point that the 510(k) review process resembles a mini-PMA review, with 10–15% of the filings accompanied by clinical data. Moreover, after 25 years, one would think that, for certain devices, nearly every variation of substantial equivalence that could result in a "510(k)-able" product would have been exhausted. The net effect of the passage of time, the increasing FDA scrutiny of 510(k)s, and the low rate of new orthopedic products reaching the market via the IDE/PMA route has been a gradual slowdown in product innovation relative to what occurred prior to 1976.

So what did Congress think would happen when it approved and President Ford signed the Medical Device Amendments law? It probably thought that in 25 years there would be no more 510(k)s being processed; that all new products and materials would be brought to market with clinical data showing "reasonable assurance of safety and effectiveness"; that in 2001 there would be more biomaterials on the market than in 1976; and that, by its actions, it was making the United States a safer place for the latest medical device technology without sacrificing innovation or world leadership. The final effect of the 1976 Medical Device Amendments is just now coming into full flower. And the reality is that FDA's new product review process may have evolved into something quite different from what the designers of the law intended, the net effect of which has been a measurable slowdown in the availability of truly innovative orthopedic products in the United States.

CONCLUSION

Twenty-five years after the enactment of the Medical Device Amendments law, several important questions present themselves:

• What has been the positive effect of the passage of the amendments?
• Is the American patient better off today because of the law? Are medical devices safer and more effective than they were in 1976?
• Based on this look into the past, what can we expect to see in the future? What will be the medical device approval process for the next 25 years?

  Table 1. Orthopedic PMA products approved from May 1976 to October 2000.

REFERENCES

1.FOI Services and Washington Business Information, The 510(K) Register, 1995 ed., vol 2 (Gaithersburg, MD: DIOGENES, 1995).

2."The Development of the Medical Device Amendments. An Interview with Peter Barton Hutt," Medical Device & Diagnostic Industry 18, no. 5, (1996): 50–52.

3.The Gray Sheet 22, no. 19 (May 6, 1996): 4.

Richard W. Treharne, PhD, is senior vice president of regulatory affairs at Medtronic Sofamor Danek. Tom Craig is director of regulatory affairs at Smith and Nephew Orthopedics. Both companies are located in Memphis, TN, and make medical device implants for orthopedic surgical procedures. Craig was the industrial representative to the Orthopedic and Rehabilitation Advisory Panel from 1989 to 1993 and is currently the president of the Orthopedic Surgical Manufacturers Association.

Photo courtesy of the Gerald R. Ford Library

Copyright ©2001 Medical Device & Diagnostic Industry