Originally Published IVDT June 2009
Regulations & Standards
The evolution of the 510(k) submission: Part 1 of 2
Medical devices were originally unregulated, then regulated as drugs. FDA’s oversight of devices has changed and grown during the past four decades.
Marlene A. Hanna
 |
Marlene A. Hanna, RAC, is a manager in the Department of Regulatory Affairs at Ortho-Clinical Diagnostics Inc., a Johnson & Johnson Co. (Rochester, NY). She can be reached at mhanna1@its.jnj.com.
|
In the late nineteenth and early twentieth centuries, drugs, not devices, were the target of government regulation—and for good reason. Their impact was dramatic, their potential uses in the treatment or cure of diseases and other medical conditions were matched or exceeded only by their concomitant risks.
In contrast, medical devices were few in number up until the 1970s, and their effectiveness in treating or diagnosing medical conditions was considered modest. Most early medical devices were viewed as “quack machines,” were harmless for the most part, and were ignored by reputable physicians and government alike. The Pure Food and Drugs Act of 1906 did not apply to medical devices, and the subsequent Federal Food, Drug, and Cosmetic Act of 1938 (known hereafter as “the Act”) carried little authority beyond the drug industry.1 The Act simply gave definition to the term medical device and banned the adulteration and misbranding of such products. It also provided for rudimentary safety testing.2 Beyond that, the counsel of “buyer beware” remained the tool most often used to protect the public from potentially harmful medical devices.
1960s: Medical Devices Increase in Number
As the 1950s ended, society changed swiftly, and with the period of dynamism in postwar Europe and the United States came a rapid rise of medical technology. Drugs, of course, remained the flag bearer of medical innovation. But machines, devices, apparatuses, and diagnostic tests began to play an increasingly vital role in hospitals, doctors’ offices, and clinical laboratories across the nation. Contact lenses, devices that tied blood vessels, and intrauterine birth-control devices were just a few of the medical products that made their way into the marketplace before FDA took significant steps to regulate beyond drugs.3
Scrambling to assure safety in a changing healthcare industry, FDA turned to the only tool it had: drug regulations. FDA classified these medical devices as new drugs and maintained that they were subject to the existing review and approval process designed for pharmaceuticals. The courts tended to agree. For example, in United States v. Bacto-Unidisc, FDA alleged that an antibiotic sensitivity disc that never even touched the patient’s body was a “new drug” falling under the umbrella of FDA authority.4 The Supreme Court concurred. Thus, the FDA’s reach into the world of medical device regulation was validated—but changes were needed.
Outgrowing Drug Regulations
FDA soon recognized that existing drug regulations were simply ineffective to govern the explosion of medical technology and the surge in medical device manufacturers seeking entry to the marketplace for their new products. A string of highly publicized medical device failures in the 1970s resulted in numerous physical injuries and dozens of deaths. The most prominent of the faulty devices was the Dalkon Shield intrauterine device, used by approximately 2.2 million women in the United States between 1970 and 1974.5 Aggressively promoted as a safe and effective form of birth control, the Dalkon Shield was linked to 16 deaths and 25 miscarriages as of June 1975.6 By early 1976, more than 500 lawsuits seeking compensatory and punitive damages totaling more than $400 million had been filed.7 This situation prompted new regulatory changes, beginning with the Medical Device Amendments of 1976 (MDA).8
The MDA did a number of things, not the least of which was to resolve once and for all the issue of FDA jurisdiction to regulate medical devices. The amendments increased the scope of the definition of a device to include devices intended for use in diagnosis of conditions other than disease, such as pregnancy, and in vitro diagnostic products, including those previously regulated as drugs;9 however, the question remained how best to use this newly gained authority. The market had already become congested with medical devices that had skirted FDA review, and now with FDA jurisdiction firmly established, many more manufacturers would be approaching FDA to seek clearance or approval of their devices. Many questions began to emerge; for example, what should be done about products already in clinical laboratories and doctors’ offices? How should new medical device innovations be evaluated? Would different standards be employed to evaluate higher- and lower-risk devices?
The latter question was resolved rather easily, or at least seemingly so. FDA developed a class system whereby devices were evaluated based on the degree of risks and benefits the device presented. With recommendations from classification advisory panels, FDA placed each device into one of three distinct categories: Class I devices were deemed low risk, requiring only general controls designed primarily to prevent adulteration and misbranding. 10 Among the general controls were requirements for establishment registration and device listing (21 CFR 807), compliance with good manufacturing practices (21 CFR 820 and, later, the quality system regulation), and labeling in accordance with 21 CFR 820 (or, for in vitro diagnostic devices, 21 CFR 809). Except where exempted by law, 510(k) premarket notifications were required as well.
As products classified as posing elevated concerns, Class II devices were subject not only to those same general controls, but also to special controls such as performance standards, human clinical trials, and postmarket surveillance.11 Class III devices, which were considered high risk, were burdened not only with general and special controls, but also with gaining premarket approval. The additional scrutiny was warranted for this group of products normally charged with supporting or sustaining human life, or otherwise posing an unreasonable risk of adverse effects.11
The answer to the former question regarding devices already on the market was not easily defined by regulation. Any device introduced before May 1976 would be allowed by FDA to remain on the market. However, all would be reviewed and classified by panels of experts.12 This proved to be a Herculean task; it was not completed until 1988.13
There were other new issues the agency needed to confront. For example, what about products that were designed and manufactured after 1976 but similar to one or more products already on the market, particularly those not then burdened by Class II performance standards? Equity demanded that FDA grant market access to sponsors of these products. In turn, FDA demanded that manufacturers indicate that their proposed device was “substantially equivalent” to one already being sold. And thus was born the 510(k); i.e., Section 510(k) of the MDA. Premarket notification provisions were intended to serve as a screening mechanism to allow for marketing of medical devices with reasonable assurance of safety and effectiveness.14
510(k) Notification
510(k) notification seemed very reasonable to industry at the time. By simply notifying FDA at least 90 days in advance of the first product sale, a device manufacturer could enter the marketplace. Companies seeking Class II designation for their products needed only to show that the product had the same intended use as a device already on the market (a predicate device), and, at the time, that it had the same technological characteristics as that predicate device.15 For those companies whose devices reflected significant technological improvements, the notification requirement was only slightly different. Those device makers needed to advise that, while the intended use was the same, a different technology was employed—one that raised no new safety concerns and was at least as safe and effective as the previous technology.15
And that’s how it worked for a while, more or less. FDA had 90 days to issue a determination either accepting or challenging the applicant’s finding of substantial equivalency. After that, the manufacturer was legally entitled to proceed to the marketplace.15 (Exercising caution, many applicants waited beyond 90 days for formal FDA clearance anyway.) FDA clearance was virtually a formality—FDA had its hands full trying to classify the devices already on the market. Moreover, while it now had a new set of regulations bestowing upon it legitimate jurisdiction over these devices, FDA personnel did not necessarily have all of the expertise needed to review many of the newer technologies and more-sophisticated devices since, at the time, FDA staff were more strongly oriented to, and trained in, drug-product technology.
Safe Medical Devices Act of 1990: More FDA Power, Less FDA Work
Perhaps because of the agency’s strong support from Congress, FDA scored major points with the 1990 amendments to the 1976 regulations. Its 15-year endeavor to create performance standards for all Class II devices was declared finished—or, more precisely, FDA was told by Congress that it didn’t have to finish it.16
Instead, FDA was now expected to monitor the postmarket performance of many devices and create post-market performance standards for some of the higher-risk products.16
The Safe Medical Devices Act (SMDA) of 1990 formally codified the 510(k) process and provided rigidity to the notion of substantial equivalency. Most important, however, was a new requirement that a manufacturer obtain written clearance from FDA before introducing a medical device into interstate commerce.16 With the passage of SMDA, mere notification was officially a thing of the past—written FDA consent was now a necessity.
The response to SMDA was predictable. Industry complained that the government was expanding, both in terms of size and scope of control. Critics further predicted that requiring government clearance of 510(k) submissions would deprive business of the flexibility it needed to compete in a rapidly changing global healthcare industry. On the other hand, FDA advocates maintained that companies shouldn’t decide for themselves whether their own products are entitled to bypass the scrutiny of regulators hired to protect public safety. Indeed, if FDA were to be nothing more than a repository for FDA notifications, it would really serve no purpose at all.
The government view prevailed; however, FDA soon struggled to meet its responsibilities in the face of rapidly advancing technologies. As a result, a backlog of medical devices was kept from the marketplace while FDA pondered whether those products were substantially equivalent to products already being sold. FDA made approximately 6000 clearance decisions per year on 510(k) submissions in 1989 and 1990, when the industry was technically making the decisions on substantial equivalency. That number fell to 4862 in 1992, after FDA began making the call under the new rules implemented via SMDA in 1990. As a result, the number of aging submissions awaiting FDA action climbed. There were 2291 submissions awaiting FDA review at the end of 1991. That figure ballooned to 5157 at the end of 1993.17
The delays, and intense criticism of FDA, caught the attention of Congress.
Part 2 of this column, to be published in the July/August issue of IVD Technology, will examine ensuing congressional action designed to reduce the logjam at FDA and evaluate the results.
References
1. Pure Food & Drugs Act, 34 Stat. 770 (1906).
2. FD&C Act, sect. 505 (1938).
3. RA Merrill, “Regulation of Drugs and Devices: An Evolution,” Health Affairs, Summer (1994).
4. United States v. Bacto-Unidisk, 394 U.S. 784 (1969).
5. In re Northern Dist. of Cal., Dalkon Shield IUD Prods. Liability Litigation, 693 F. 2d 847, 848 (CA9 1982); ante, at 1–2.
6. H.R. Rep. No. 94–853, 8 (1976).
7. The Dalkon Shield was ultimately linked to “thousands of serious injuries to otherwise healthy women.” DC Vladeck, “Preemption and Regulatory Failure,” Pepperdine Law Review 33, no. 95, 103 (2005). By October 1984, the manufacturer had settled or litigated approximately 7700 Dalkon Shield cases. R Sobol, Bending the Law: The Story of the Dalkon Shield Bankruptcy (Chicago: University of Chicago Press, 1991), 23.
8. 21 U.S.C. ss360c-1, Medical Device Amendments.
9. 2005 Fundamentals of US Regulatory Affairs, (Rockville, MD: Regulatory Affairs Professional Society, 2005), 133.
10. FD&C Act, sect. 513 and 520.
11. 2005 Fundamentals of US Regulatory Affairs (Rockville, MD: Regulatory Affairs Professional Society, 2005) 136.
12. FD&C Act, sect. 513 and 520.
13. 53 Federal Register 23856 (1988).
14. “Guidance for the Preparation of 510(k) Submissions,” HHS Publication FDA 97-4224 (Rockville, MD: FDA, 1997).
15. FD&C Act, sect. 510 (1976).
16. FD&C Act, 104 Stat. 4511 (1990).
17. R Higgs, “Wrecking Ball: FDA Regulation of Medical Devices,” Policy Analysis, no. 235 (1995).
Copyright ©2009 IVD Technology
