Disclaimer: This blog is independently written and published by me. The opinions expressed herein are my own personal opinions and do not represent my employer's view in any way.
Today I will be sharing some of the surprising things I learned about the regulation of Medical-Grade Devices in the US. Most people believe that when they are implanted with a medical device or other device that could negatively impact their health that those devices have undergone appropriate testing to demonstrate that they are safe and effective before they came on the market and doctors started using them. But for most moderate and high-risk devices, that may not be the case.
In the US, a ‘medical-grade’ health device is different than devices like personal fitness trackers because these devices must be cleared or approved by the Food and Drug Administration (FDA) and are Health Insurance Portability and Accountability Act (HIPAA) protected. Originally congress intended that almost all new devices go through a process called pre-market approval (PMA). A PMA Is similar to a new drug application, in that a manufacturer must test it first in humans, compile all this data and then present it to FDA scientists who will approve the device if in fact it is safe and effective. But the medical device industry pushed back on the cost of changing their products every year arguing that they need more flexibility to innovate. To address the feedback, congress established the 510(k) process.
Today’s process stems from the Medical Device Amendments added to the Food, Drug and Cosmetic Act in 1976, establishing the three regulatory classes for medical devices we have today. Because there were devices already on the market prior to 1976, all devices were considered either pre-amendment medical devices, or entirely new concept medical devices.
FDA Clearance vs. Approved
Since 1976, all manufacturers who seek to market medical devices in the United States must go through one of two pathways via the FDA —the section 510(k) process, where a device is “cleared” for distribution, or the Premarket Approval (PMA) process, where a device is “approved” by the FDA.
Today, the majority of low to moderate risk medical devices (class one and class two) go through the FDA 510(k) Clearance process which is required to market a device in the US. FDA-Clearance through a 510(k) does not require clinical validation studies. They get around this requirement by stating that the device is “substantially equivalent” to a pre-existing device that was already proven safe and reliable. Only quality measures (complying to CFR regulations for medical devices) are required but the manufacturer is only subject to inspection after FDA-clearance. That means, it’s on the honor system (!).
If you’re wondering if you read that right, I’m going to restate this information one more time. For the 510(k) pathway, all the manufacturer needs to demonstrate is that their device is "substantially equivalent" to another device that’s already on the market. Let’s discuss this phrase “FDA Clearance” for a moment. For Class II and Class I, the FDA doesn’t give “approval,” it just gives clearance. The FDA has simply allowed the manufacturer to market their product in the US. Ninety-Eight percent of devices today are regulated under this 510(k) framework giving “Clearance” with only 2% going through pre-market approval (PMA) because it is so much easier and less costly to get to market.
DE NOVO PATHWAY
There is one more pathway for low to moderate risk medical devices, called “De Novo” where there is no substantially equivalent device. De Novos are designed truly for new and novel, low and moderate risk devices. An advantage to the manufacturer is that they don’t have to prove that they are substantially equivalent and that they have more of a blank slate for marketing in terms of how they label their product. You do have to be able to substantiate your marketing claims. The De Novo is not a common submission type. The total number of De Novos that have come through the FDA since it was created a dozen years ago is under 200 while there have been 40,000+ 510(k)s.
Apple has stated that it has received a “de novo” classification for the EKG feature. That means that, although it’s still in Class II in terms of risk and hasn’t gone through as much testing as an “approved” device, it’s unlike anything else on the market. It is the first direct-to-consumer EKG wearable. (A year prior the FDA approved the AliveCor KardiaBand, a watch accessory that essentially does the same thing, but that wasn’t direct-to-consumer). The two FDA-approved Apple apps monitor for atrial fibrillation, a common heart rhythm irregularity that can lead to stroke. They can also detect unusually slow or unusually fast heart rates and record the data for a medical professional.
One of the disadvantages of a successful De Novo, is that it will likely lead to a predicate device, making it easier for competitors to enter the market. Essentially they would bring their device into the market as a 510(k) using the De Novo as a predicate. The labeling of the De Novo will determine how difficult it will be for a competitor to successfully use this strategy.
LITTLE TO NO TESTING IN HUMANS
Human testing of devices is often not a requirement of FDA 510(k) Clearance. I’ve personally talked to several device manufacturers that have told me that didn’t do any human testing or that there human testing was essentially equivalent to 3 human test subjects doing jumping jacks for 3 minutes. Often these companies don’t have the funding to do large scale testing or the access to patients. Getting through FDA clearance alone can take about 18 months and cost about $200,000 and if you’re a startup, this might eat up all your funding. And if the product has a new software update with a new version, it will needs to go through the clearance process again, which entails more cost and time. Hospitals won’t consider testing devices that aren’t FDA cleared. So the natural first step for devices is to become FDA cleared and then seek out relationships to do both human and ROI testing. Finding human test subjects especially in the US is difficult and that’s why many of these companies do their human testing offshore. Some hospitals have a venture arm and may invest in device companies that they believe in and work with them to do the testing. Sometimes government funding can be used to support testing.
Over time, what happens is that one medical device is approved on the bases of being substantially equivalent to a previous medical device that was approved because it was substantially equivalent to an earlier medical device and continues on for generations of devices. The original device might be seven or twenty layers up the chain (often referred to as a “daisy chain”). If a predicate device is recalled from the market because they have safety issues - say its potentially dangerous - a device manufacturer can still use it as a predicate device and get a new device cleared using the “substantially equivalent” clause. The FDA’s stance is not to judge what the prior device is.
Even the most stringent pathway to get a device FDA approved (PMA) but even that is less rigorous than the process for drugs. Most drugs have to have two clinical trials and large numbers of patients. With devices its only one study with small participant pools. Approval process for the riskiest process is not that strict. There are plenty of untested devices on the market. Despite very little data, devices are still approved even with open-ended questions remaining unanswered.
LOOKING AT A REAL Example
The Patient Status Engine (PSE) is a FDA cleared Class II medical device effective April 2018. It uses wireless body-worn sensors to automatically collect and analyze six vital signs in real-time including heart rate, respiration rate, temperature, oxygen saturation, blood pressure and coma (or pain) score. It then streams this data via a patient gateway to a central server and delivers it to the nurses’ station or remotely to clinicians.
510(k) Number: K172329
Device Name: Patient Status Engine
Indications for Use (Describe): […]
Type of use: Prescription Use
Legally Marketed Predicate Device: K152139 - Vital Connect Platform/VitalPatch/HealthPatch MD by Vital Connect Inc.
Device Description: [.]
Clinical Performance Data: There was no human clinical testing required to support the medical device as the indications for use is equivalent to the predicate device. These types of devices, including the predicate devices, have been on the market for many years with proven safety and efficacy for the use of the device. The non-clinical testing detailed in this submission supports the substantial equivalence of the device.
Statement of Substantial Equivalence: It has been shown through the documents provided in this 510(k) submission that the minor differences between the Patient Status Engine and the predicate devices Vital Connect Platform/Healthpatch MD/VitalPatch do not raise any new questions regarding its safety and performance. The Patient Status Engine has the same intended use and the same or similar indications for use and technological characteristics as the previously cleared predicate device, Vital Connect Platform/Healthpatch MD/VitalPatch. The Patient Status Engine as designed and manufactured, is determined to be substantially equivalent to the predicate device.
THE PSE DAISY CHAIN
I looked up PSE’s predicate devices to and followed the chain to its dozens of predicate devices before it became impossible to continue. It’s like trying to track down all 256 of your 6th-level great grandparents. I also made note of which ones stated they had submitting results of testing in humans.
510(k) Number: K172329 Isansys Patient Status Engine (2018) Predicate Device:
K152139 - Vital Connect Platform/ VitalPatch/ HealthPatch MD by Vital Connect Inc. (2015) Indications for Use: The Vital Connect Platform is a wireless remote monitoring system intended for use by healthcare professionals for continuous collection of physiological data in home and healthcare settings. Animal and Clinical Testing: Prior clinical testing was performed to support the clearance of the predicate device and is still relevant to the proposed device. The safety and effectiveness of the device associated with the product changes are demonstrated through performance testing. Thus, animal and clinical studies were not deemed necessary. Predicate Device: 1
K141167 Vital Connect Platform (2014) An IRB-approved "Long-term Field Study for Home Use" was conducted by Vital Connect with the VitalConnect Platform from September 2013 to March 2014, to assess the long-term use of the VitalConnect Platform in a home setting. The assessment includes the usability of the VitalConnect Platform in a home setting, including interaction with the module, patch, and relay. In addition, the assessment includes the long-term wearability of a series of patches on the same participants. Results indicate successful use of the VitalConnect Platform over a 50-day home use study on 76 participants. (A note here is that this study was performed by VitalConnect, and not a 3rd party, thus bias is a risk). Predicate Device: 1
K132447 VitalConnect Platform (2014) “Clinical study was performed on human subjects.” Here is the 2013 study I found: The sensor measures heart rate, heart rate variability (HRV), respiratory rate, posture, steps, and falls and was evaluated on a total of 25 adult participants who performed breathing exercises, activities of daily living (ADLs), various stretches, stationary cycling, walking/running, and simulated falls. Predicate Devices: 4
K110809 CareFusion, Wireless Monitoring System (2011) Predicate Devices: 2
Philips Intellivue Patient Monitor. K052961(2005) The names of the devices are the Philips MP2O, MP3O, MP4O, MP5O, MP6O, MP7O, and MP90 IntelliVue Patient Monitors with 802.11 Wireless LAN. The modified devices are substantially equivalent to previously cleared Philips devices marketed pursuant to K014159, K021778, K030038, K032858, K040304, K040183 and K040357, K041235, K41741, K042845, K050141, K050762, and K051106
Nihon Koden Multiple Patient Monitor. K071058 (2007) The ORG-9700A Multiple Patient Receiver is intended for use by medical professionals with Nihon Kohden telemetry transmitters and central stations to provide cardiac and vital signs monitoring for multiple patients within a medical facility. The device detects patient vital sign alarm conditions and includes an algorithm to detect cardiac arrhythmias. The intended use of the modified device has not changed as a result of the modifications. The device is available for use on all patient populations
K121197 Preventice, BodyGuardian (2012) Predicate devices: 1
AVIVO Mobile Patient Management System (k083287)
K113054 Hidalgo Ltd, Equivalent TMVital Signs Physiological Monitor EQo2
K083287 Corventis, Mobile Patient Management System. (2009) The AVIVO Mobile Patient Management System is intended to continuously record, store, and periodically transmit physiological data. The System is indicated for those patients who require monitoring for the detection of non-lethal cardiac arrhythmias. Predicate devices: 4
CardioNetTM ECG Monitor with Arrhythmia Detection, K072558; DSI, "Arrhythmia Detector and Alarm"
LifeShirtTM Real-Time, K043604; DQK, "Programmable Diagnostic Computer"
ZOETM Fluid Status Monitor, K042113; DSB, "Impedance Plethysmograph"
SenseWear Armband, 510(k) exempt, IKK, "Isokin-etic Testing and Evaluation System"
REPORTING DEVICE PROBLEMS
One might assume that the FDA has the data that they need to crack down on bad devices or to recall a product if needed. Indeed, there is a system for reporting complications but it’s a voluntary system of patient self-reporting. If a physician sees an adverse event, they do not have to report it. Thus much goes under-reported with only about 3% of all adverse events get reported to the FDA. The only bodies that are required to report adverse events are the device companies themselves and they are clearly not incentivized to do so. One study found that the worse the outcome was, the less likely a device manufacturer was to report it to FDA (yikes!). Adverse effects of a new implant are often not known until months or years after it’s on the market. By then, it may have affected hundreds of thousands of people.
SUMMARY
Like drugs, new device innovations are important to making strides in healthcare. But what I want you to take away with is that new devices on the market are often unproven, more so than drugs. Don’t assume that FDA clearance means that a device is safe, it simply means that it is equivalent to a predicate device and so on. Do your own research, be cautious, and use your own judgment. Ask your doctor if he or she understands what FDA clearance means and more importantly, what it doesn’t mean, before trusting their judgment on whether a device is safe for you or your family.