July / August 2008
Human Factors Design and FDA Regulation
Companies that produce medical devices such as breast implants and implantable defibrillators now have federal protection against lawsuits arising from deaths or injuries associated with their products as long as the product has been approved for use by the FDA. On February 20, 2008, the United States Supreme Court produced a landmark decision that makes it nearly impossible to hold medical device manufactures accountable even if the medical device malfunctions and causes an injury. The device at issue was a balloon catheter that burst during an angioplasty. The mishap caused serious injury to the patient, Charles R. Riegel. He and his wife, Donna, sued the manufacturer, Medtronic, which had received approval to market the device in 1994, 2 years before the incident. Two lower federal courts in New York dismissed the suit on the ground that the FDA’s “pre-market approval” precluded the imposition of liability under state law.
The Supreme Court last looked at a medical device case in 1996, when it ruled that devices approved by the United States Food and Drug Administration (FDA) under a less rigorous, more expedited process were not shielded from state liability. At that time, the federal government itself argued against pre-emption. A question in this case, Riegel v. Medtronic Inc., No. 06-179, is whether the court will give the government’s position the usual deference it accords an agency’s interpretation of its basic statute. The 8-to-1 Supreme Court decision was a victory for the Bush administration, which for years has sought broad authority to pre-empt tougher state regulation. In 2004, the administration reversed longstanding federal policy and began arguing that “pre-market approval” of a new medical device by the FDA overrides most claims for damages under state law. Because federal law makes no provision for damage suits against device makers, injured patients have turned to state law and have won substantial awards. Left unreported is transparent connection to safety science and the role of human factors engineering in producing safer medical devices. Several questions remain about the public concerns with the safety of medical devices and instrumentation.
1. Why should Americans be personally concerned about their individual safety with the recent Supreme Court decision? There has been a wide range of adverse safety events caused by medical devices, involving nearly all hospitals in the United States. The Institute of Medicine (IOM) in 1999 reported that at least 44,000 Americans die from medical errors. The IOM report estimated that 10% of those deaths were caused by medical device malfunction.
2. What amount of new device safety is both legally required and valued by consumers? Modern healthcare is complex and produces increasingly interrelated human and technological errors, which go unreported. Yet both Medtronic and the FDA have received hundreds of user concerns and mal-event reports long before the 1996 legal challenge brought by Mrs. Riegel. Such legislative delays are a predictable (but reparable) occurrence in the present environment of “shoot the messenger, not the message.” The interests of investors, innovators, and industrial manufacturers as stakeholders increasingly impact healthcare quality. The ability of budget constrained hospitals, physicians, and nurses to efficiently balance the totality of the “standard of care” required in a nation governed by the rule of law proves this fact. If patients value relentless, rapid, and rigorous information transparency, then Representative Waxman and Senator Kennedy (with a Congressional majority in both houses) should enact legal immunities for healthcare professionals related to safety database reporting of near misses and adverse events. Such reporting has altered aviation safety practices. Until reporting of device safety data in the public interest is assured by the rule of law, the risks of litigation, loss of job security, and retribution as payback for negative impacts on hospital and industrial profits and losses, the obstruction to safety in healthcare will continue.
3. What testing is required to avoid more biomedical device injuries and deaths? The science of human factors engineering (HFE) is needed to identify design flaws in medical devices. Because of potential to predict adverse events related to design, HFE methods (e.g., heuristic evaluation, ergonomics, and usability testing) have become standard device development processes in all domains, and are supported by biomedical device national standards organizations and the FDA. All devices for medical usage should go through 3 basic steps:
|Step I.||Safety exploration and thorough review of existing information of the existing product and the overall product category;|
|Step II.||Failure modes and effects analysis;|
|Step III.||Usability design testing.|
The focus is on how human users of the device actually behave in using the product. It is an objective measure of the “real world snapshot” to analyze how the human/product interaction error rate may be minimized. The popular iPod music device was designed in this fashion. These methods are starting to be used in medical device design and are making healthcare safer (Table 1). There is a compelling need to standardize user interfaces terminology, programming, and capabilities to address cost, inventory management, and safety.
|Suicide-resistant door-jamb for
inpatient psych room
|New devices with
|Automated defibrillator||3, 4|
|IV tubing auto-clamp when pump door opens||5|
|Map and simplify the
|Remove unnecessary steps for LP preparation||6|
|Standardize on equipment or process||Standard defibrillator on all code carts||7|
of standard CVC
|Redundancy||Abnormal X-ray F/U to physician & separate tech.||8|
|Computer alerts for
|Checklist/cognitive aid||Ensure all anesthesia equipment is
|Eliminate look and
|Losec and Lasix not stored near each other||12|
|Readback of critical lab value||13|
|Medication name and dose highlighted on IV bag||14|
|Education||Knowing cognitive biases decreases misdiagnoses||18|
|Double checks||One person calculates dosage, another person reviews their calculation||15|
|Warnings||Adding audio alarms
or caution labels
* Modified from Gosbee, John, Hierarchy of Treatments, Evidence based patient safety, University of Michigan, 2008.
One important but often neglected point is that proprietary hardware and software are in direct conflict with any standardization. The competitive nature of the business, consolidation of device companies, and the long useful life of our technologies all contribute to this issue. This actually makes it more difficult for medical personnel because they are required to learn a new system for every new device.
4. What is the role of industry? Companies are not only responsive to their customers but they want to market medical devices that meet customer’s needs for functionality, safety, and effectiveness. Fortunately, industry and other stakeholders have begun to absorb this message. The ANSI/AAMI HE-74 is a guide to less error-prone design of devices, including infusion devices. Designing efficient, yet safer, devices will require that industry fully embrace HFE methods.
5. Why do nurses, physicians and patients have little input into medical device selection? Physicians and patients have little sway or influence over equipment purchasing decisions. Device manufacturers continually add new capabilities according to marketplace dictates. In the multibillion dollar medical devices market, as long as a minimum standard of adequacy does not exist, patients and caregivers can neither feel safe nor benefit from “the invisible hand of market morality” that might put some equilibrium between industrial monopoly and purchaser “monopsony.”
Monopsony means a purchaser monopoly (e.g., buying group) that exists when one large purchaser, rather than the seller, controls a large proportion of the market and drives the prices down. The enhancements manufacturers address through software are much easier to address than those related to hardware and fundamental user interface/human factors differences.
6. What is the consumer’s role? (“Homework”) The FDA is but one element of the problem and an all too frequent whipping post for special interests and lobbyists who tie the hands of Representative Waxman and his fellow representatives and senators. Americans seeking better healthcare safety should demand political and legal facilitation of greater transparency in approving, reviewing, and evaluating medical devices and equipment. FDA regulations, industry standards, and HFE methods will never be sufficient. All of them depend on you, the healthcare providers and your patients to report all adverse biomedical device events and near misses to independently maintained data registries.
There is still much debate about the implications of the Medtronics mishap. “In enacting legislation on medical devices, Congress never intended that FDA approval would give blanket immunity to manufacturers from liability for injuries caused by faulty devices,” Senator Kennedy publicly stated. He added, “Congress obviously needs to correct the court’s decision.” In the meantime, the practical impact of the ruling is clear: If you’re injured by a medical device and/or its user-unfriendly design that’s been pre-approved — whether it’s a defibrillator, a heart pump, or a drug-coated stent — there’s not much you can do.
The legally controlling decision will come this fall, when the court takes up another case, Wyeth v. Levine. That case, which involves a woman whose arm was amputated after a drug gave her gangrene, involves a pharmaceutical company, not a medical device maker. There is no specific federal pre-emption in statutes governing drug companies, so the justices may have more room to decide for themselves whether the company can be held liable for causing harm. Barring new legislation, the standard is set: When it comes to medical devices, the FDA is now, officially, the last word.
Michael Rie is an academically tenured associate professor of anesthesiology, specializing in critical care medicine at the University of Kentucky College of Medicine, Lexington, Kentucky. A graduate of Dartmouth College, he received his MD degree from Harvard Medical School. Rie also has formal training in ethics from the Kennedy Institute of Ethics at Georgetown University. He is a board-certified internist, board-certified anesthesiologist and critical care medicine specialist. He received his internal medicine resident training at Albert Einstein College of Medicine in New York City and anesthesiology and critical care medicine training at Massachusetts General Hospital in Boston. From 1973 to 1991 he was on the anesthesia faculty of Massachusetts General Hospital and Harvard Medical School concentrating in critical care medicine in the respiratory surgical intensive care unit. He moved to the University of Kentucky College of Medicine in 1991 and has been there in the departments of anesthesiology and surgery, specializing in surgical and trauma critical care. He may be contacted at firstname.lastname@example.org.
Paul Barach is a board-certified anesthesiologist, with fellowship training in cardiac anesthesia, critical care medicine and human factors, at the Massachusetts General Hospital and Harvard Medical School, where he trained and practiced. He founded the Center for Patient Safety and Simulation at the University of Miami Jackson Memorial Hospital, where he was associate professor and associate dean for patient safety. He chaired the Jackson Memorial Hospital Patient Safety Committee and was medical director for quality for 3 years. He is associate professor at the University of South Florida where he is senior advisor to the dean of the Medical College on patient safety. He helped write the legislation and co-founded the state supported Florida Patient Safety Corporation. Barach has been a visiting professor in 47 universities including 20 outside the United States. He has worked extensively with the World Health Organization, and the UK National Patient Safety Agency (NPSA) in advancing patient safety and quality improvement initiatives. He is currently a visiting professor at the Utrecht University Medical Center in the Netherlands, focusing on developing innovative research using human factors and team training, and may be contacted at P.email@example.com.