MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Can Product Recalls Be a Good Thing?

Medical Device & Diagnostic Industry Magazine MDDI Article Index Originally Published MDDI August 2005 WASHINGTON WRAP-UP

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

Originally Published MDDI August 2005


Can Product Recalls Be a Good Thing?

The director of the CDRH Office of In Vitro Diagnostic Device Evaluation and Safety makes waves when he claims that recalls prove a QA system is effective.

James G. Dickinson

FDA Tries Third-Party Inspections Again | FDA Guidance on Postmarket Studies | ASR Claims Circumventing FDA Law, CDRH Says | Drug-Abuse Tests Need 510(k) Clearance | Research Misconduct a Top FDA Concern | Mandatory Device Bar Coding a Contentious Issue |

Product recalls have traditionally been a touchy subject for medical device makers—especially in the context of FDA regulation. The industry is typically slower than drug companies to file recall reports with the agency, which is no doubt a result of a very litigious marketing environment.

So it came as something of a surprise to hear Steven I. Gutman, MD, tell an industry audience in June that recalls can actually be a good thing. Gutman, director of the Office of In Vitro Diagnostic Device Evaluation and Safety, CDRH, is widely regarded as one of the agency's brightest officials.

“Recalls are not bad,” he told the Orange County Regulatory Affairs (OCRA) Discussion Group annual meeting in Irvine, CA. “I actually would be much more worried about a company that never had a recall than I would be about a company that has a Class 1, 2, or 3 recall. I would say a response to a Class 1 or 2 recall means there's a QA system that actually works. If you have no recalls, I would worry that you're either clueless or that you're actually fraudulent. We're all humans, and all systems will eventually break down.”

Before risk-based enforcement, FDA officials frequently saw a company's product recall rate as a trigger for a for-cause inspection, or an indication that its quality system must be out of control. In the medical device arena, companies were often slow to report recalls out of fear of the attention it would provoke. So, a reporting lag time of several years was not unusual.
“Even a Class 1 recall isn't bad,” Gutman emphasized. “What's bad is not taking it seriously and not responding. We will work with you on press releases. If you have been responsive and pulled out all the stops, we want your press release to reflect the fact that you've been like Tylenol. That's an important message to get out.”

Gutman's thinking didn't sit well with some old FDA hands. Former FDA Denver District director John Scharmann quickly challenged Gutman. He said this was the first time in 37 years with the agency that he had seen a new interpretation of the meaning of a recall.

“His position that recalls are an indication that a QA system is working is difficult to understand,” Scharmann said. “I was always under the impression that a QA system was in place to keep violative products off the market, and that when this did not happen, a firm's QA system was not working.

“If FDA is, as Gutman contends, worried about firms that do not have recalls, it may explain the current action by FDA against Utah Medical Products Inc. (Midvale, UT),” said Scharmann. The firm has never had a recall, but it is now defending itself against an attempt by FDA to enjoin its operation. “This must be happening because, as Gutman contends, all systems fail. So, one could consider the Utah Medical action preventive medicine.”

FDA Tries Third-Party Inspections Again

Third-party inspections can be a win for both industry and FDA, CDRH director Daniel G. Schultz told the OCRA annual meeting in June.

There has not yet been a single volunteer for these paid-for inspections by FDA-accredited persons in lieu of an FDA inspection. However, the CDRH director said he believes “it will happen.” Such inspections will “hopefully allow us to allocate our limited resources where they really need to go.” He also hopes they will provide efficiency to the industry by having one inspection done for multiple regulatory bodies in this country and worldwide.

Schultz said that so far, nine FDA-accredited persons have also been accredited by ISO. In their case, two inspections could be done simultaneously. He noted that FDA's Canadian counterpart has adopted ISO for device inspections. The two North American agencies are in discussions aimed at selling this approach to device companies with facilities in both countries.

“I hope we can get [your] companies to engage with us to allow these third-party inspections to take place in your facilities,” Schultz told the OCRA audience. He wants firms “to allow these people to get experience in doing inspections, to get used to having our people go out with them and make sure the inspections are in fact up to the levels that we think are necessary.”

Third-party inspections were authorized but not required by the Medical Device User Fee Modernization Act of 2002. FDA sees them as a way of freeing up increasingly scarce inspectional resources. The agency may then be able to focus its inspections on cases that represent the highest risk to public health.

FDA Guidance on Postmarket Studies

CDRH is developing a guidance on condition-of-approval studies required for many new medical devices. The center expects to issue the guidance in draft form by the end of the summer. Center director Daniel G. Schultz gave this estimate while addressing the OCRA Discussion Group's annual meeting.

He cited a plethora of recent headlines criticizing both FDA and industry over product safety. He also cited acting commissioner Lester Crawford's public vow to make the agency more transparent on what it knows about product risks.

CDRH's response to this has been to change its policy on condition-of-approval studies by moving responsibility for them from the premarketing group to the postmarketing group.

In addition, he said, there is a new tracking system for these postmarketing study commitments. The system was put in place so that the center can be sure that when there are commitments made, it will know what the commitments are. It will also know when the reports are due and that CDRH staff can communicate quickly and efficiently with sponsors to make sure that the reports come in on time.

ASR Claims Circumventing FDA Law, CDRH Says

Current marketing of analyte-specific reagents (ASRs) represents “a dazzling Arabian bazaar of activity,” said director of the CDRH Office of In Vitro Diagnostic Device Evaluation and Safety Steven I. Gutman to the OCRA Discussion Group's annual meeting. “Indeed,” he continued, “it's fair to say that some companies have gotten a little carried away and are marketing wolves in sheep's clothing. They're essentially taking ASRs and converting them into pseudo kits and, in my view at least, circumventing FDA law. It is my view that's particularly egregious in some examples of microarrays and inappropriate use of research-use-only products. FDA's response in this area has been slow, but I assure you that we are not unaware that this problem exists.” Gutman added that FDA's response will be certain and steady.

Gutman said FDA intends “to clarify the ASR rule, which is an imperfectly written rule.” AdvaMed has now taken a leadership role and is working on frequently asked questions about ASRs. It plans to submit them for use as guidance, hopefully in the fall, Gutman said. “We'll take the guidance, probably massage it and make it our own, and attempt to use that as at least a bridge to create some clarity. And we are considering issuing our own draft guidance, or reopening the rule.”

He indicated that FDA will not accept ASRs as an across-the-board Class I exempt product. The Food, Drug, and Cosmetic Act at Sec. 510(l) and the regulatory IVD classifications show that “when you push an exempt product too far in terms of its risk profile, you take away its exemption. That's not a new regulation or a new law—they've both been around for a long time. They've just not been enforced with much vigor. So we need to decide how and whether we're going to enforce those parts of our law. If we were to do that, it would be on a risk-based basis—it would be an effort to establish parity in this colorful marketing arena.”

Gutman emphasized that FDA has no enmity toward analytical-based lab tests. “In fact, the truth of the matter is that most of the lab tests that we clear are on an analytical base. And we have no animus toward analytical lab tests that are looking for a clinical use. We just have a difference of opinion on the right terminology. Instead of calling those IVATs (in vitro analytical tests), we would call them investigational devices.”

Drug-Abuse Tests Need 510(k) Clearance

FDA says tests used to detect drugs of abuse (e.g., cocaine, marijuana, opiates, PCP, or methamphetamine) should be cleared through the agency's 510(k) process. The agency says it has become aware that some companies are marketing tests that are not 510(k) cleared, and companies are making claims that such clearance is not required.

An FDA notice points out that regulations address premarket review of these screening tests, which involves assessing both performance data and labeling. The notice directs businesses and consumers who use the tests to a database search engine to confirm whether the tests they use are legitimate. The notice is available online at

Research Misconduct a Top FDA Concern
FDA's bioresearch monitoring (BIMO) inspections are focusing on “allegations of research misconduct,” said Tim Ulatowski. The CDRH Office of Compliance director made these comments to the OCRA Discussion Group annual meeting in Irvine, CA, in June. The allegations, which are often received in complaints to the agency, are a “ripe area for OAI [official action indicated] findings—where there is smoke, there is fire,” he said. Research misconduct allegation inspections have a 60–70% OAI rate, compared with a 20–30% OAI rate for all other BIMO inspections, he said. A newly created branch, the Allegations Inspection Branch, is leading the charge in the BIMO area.

CDRH's issuance of BIMO warning letters has increased significantly over the past two years, Ulatowski said. Forty-four BIMO-related warning letters were issued in 2004, a 42% increase over the 31 letters issued in 2003. He said 2005 is on track to surpass even the 2004 numbers. “To see the fundamental problems with regard to clinical studies is absolutely amazing,” he said. “Informed consent is missing time and time again. IRBs are out of the loop time and time again.”

Ulatowski provided breakdowns of BIMO warning letter citations and the percentages of letters containing each citation. In sponsor BIMO inspections, inadequate monitoring was cited in 40% of the letters, followed by failure to secure investigator compliance (21%). Inadequate device accountability followed at 16%, and 11% did not obtain FDA or IRB approval.

The top citation on investigator deficiency inspections was failure to follow investigational plan, cited in 54% of the letters. This was followed by inadequate subject protection (24%) and protocol deviations (16%). And for IRB-related inspections, the top citations were inadequate initial or continuing review (50%) and inadequate minutes (28%). An inadequate membership roster was cited by 21%, followed by inadequate SOPs (19%).

Mandatory Device Bar Coding a Contentious Issue

Scarcely a week after AdvaMed urged the opposite action, seven members of the House of Representatives and seven hospital organizations asked acting FDA commissioner Lester Crawford to require device bar coding. This, the congressmen wrote in May, would be a “commonsense next step in our shared goal of improving quality, cost-effectiveness, and supply chain efficiency.” The letter came from Pete Sessions (R–TX), Tim Murphy (R–PA), Bill Jenkins (R–TN), Michael Conaway (R–TX), Bob Inglis (R–SC), George Radanovich (R–CA), and Mike Doyle (D–PA).

The hospital organizations said bar coding medical devices “has vast potential for improved clinical product and service innovation. As FDA itself has recently noted, a compelling patient safety interest also lies in requiring bar codes for medical devices that could be subject to recalls. Comprehensive data on—and the ability to conduct rigorous comparisons of—emerging health practices, products, and services is essential for both clinical and economic decision making.”

In its letter to FDA, AdvaMed asked the agency to “continue to recognize that the unique diversity of medical devices is so significant that they should only be considered for inclusion in a voluntary bar code standard.” AdvaMed said its 1300 members are committed to voluntary use of industry-standard automatic identification for medical devices. However, such coding should be reserved for situations where it is clinically beneficial, technically feasible, and economically advantageous.

The association added it was concerned that recent agency requests to require bar coding on all medical devices demonstrate “a fundamental misunderstanding of the nature of medical devices.” In the absence of any documented benefit to patient safety, AdvaMed said, the requests fall short of justifying the high cost of such a requirement.

According to AdvaMed, mandatory bar coding works for drugs because drugs look alike, have similar packaging, and may have similar-sounding names. All these factors can lead to confusion and errors. FDA's 2004 Bar Code Rule makes sense for drugs for two reasons, AdvaMed said. First, there is a well-documented problem in the misadministration of drugs due to similarity in appearance and name. Second, drug packaging is relatively uniform, lending itself to the application of bar codes.

Conversely, it said, devices come in all sizes and are packaged individually and by the hundreds. They are made of a wide range of materials requiring various sterilization methods and differing storage requirements. In addition, they may be designed for a single use or multiple uses with reprocessing.

Including medical devices in the Bar Code Rule would fail, it said, because there is a dearth of literature showing bar codes on medical devices improve patient safety. “If included in the Bar Code Rule, many medical devices throughout the hospital, such as surgical implants with finely polished surfaces and instruments that are exposed to repeated sterilization cycles, would require extremely expensive printing techniques or laser etching,” AdvaMed's letter said. “When there is demonstrable clinical benefit to the patient, however, the expense is not an issue. Furthermore, applying bar codes to some devices, such as tiny vascular stents, would be technically impossible.”

AdvaMed said it recommends that refurbishers and reprocessors of single- or multiple-use devices be required to remove or permanently obliterate the original equipment manufacturer's bar code to prevent misidentification. It also recommends that providers and purchasers use bar codes when they appear on medical devices. Finally, it says that there should be voluntary use of Universal Product Numbers on medical devices, allowing observance of industry-approved UCCLEAN or HIBC standards.

Copyright ©2005 Medical Device & Diagnostic Industry

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.