Adverse Event Reporting for Medical Device Trials

A Webinar on CD by Dr. Nancy J Stark

First published 2 September 2008

Events and Effects
When we compare the good clinical practice requirements for medical devices and pharmaceuticals, adverse event reporting is among the top three causes of confusion.

In the device world not all adverse events are reportable, instead the investigator makes the decision as to whether a given adverse event is device related or not. If so it is called an adverse device effect (think of cause and effect) and only adverse device effects are recordable and reportable.

As with drugs, an adverse event is any unwanted medical occurrence in a subject. And an adverse device effect is any device related adverse event. ISO/DIS 14155--Clinical investigations of medical devices in humans, good clinical practices, 2008, defines a serious adverse device effect as one that, just like any serious adverse event, results in hospitalization, permanent impairment, death, or other sequelae.

Unanticipated Effects
But here distinctions really start to diverge. FDA does not recognize a separate category of adverse device effect called a 'serious adverse device effect'. Instead, the concept of serious is embedded within the definition of an unanticipated adverse effect; so we have the definition that an unanticipated adverse device effect is a 'serious adverse effect on health or safety...caused by...a device...if...not previously identified in nature, severity, or degree of incidence in the investigational plan....' [Sec. 812.3(s)].

Serious adverse device effects are no more or less reportable than plain-old-vanilla adverse device effects. But unanticipated (and hence serious) adverse device effects play a special roll in the world of medical devices.

Unanticipated adverse device effects are reportable to the sponsor and IRB within ten days of the investigator's learning of them, and to FDA and other IRBs within ten days of the sponsor's learning of them.

Other Event Categories
The protocol or investigational plan may call out special reporting requirements for adverse events (not related to the device), serious adverse events (serious, but not related to the device), or anticipated serious adverse device effects (serious, but we were expecting them in our population). But these are considered on a case-by-case basis by FDA depending on the nature of the device and the study.

ISO/DIS 14155, 2008, reinforces a concept that has been around for awhile but not fully embraced by regulators or manufacturers: the concept of an adverse event that happens to a non-subject such as a caregiver or bystander.

Given the physical-mechanical nature of devices we can imagine that injury from some devices will occur to persons who are not trial subjects.

Injury to non-subjects will likely be reportable under the 2008 ISO standard. And reporting such an event under the new information clause of 812.40--General responsibilities of sponsors is certainly appropriate, even if not commonly done.

Terminating a trial because an unanticipated adverse device effect presents an unreasonable risk to subjects is a forward-looking concept unique to FDA. [Sec 812.46(b)(2)] There is no corollary in ISO 14155.

Data Safety Monitoring Boards
While not the norm, data safety monitoring boards are finding their way into more and more medical device clinical trials. These boards may be charged with reviewing adverse events for seriousness or for device-relatedness.

If called a Data Monitoring Board the charter might be limited to or expanded to assessing device efficacy as well as safety.

Whatever the charter, DSMBs make recommendations to the sponsor that are usually, but not necessarily, implemented.

Safety Assessment and Adverse Event Report Forms
I recommend using a two-part system for capturing and evaluating adverse events.

The first part consists of a case report form that is used whenever the subject visits the investigative site. It asks, one-by-one, if a risk on the risk assessment has occurred. If the answer is 'no', no further action is required and the sponsor slowly builds up data regarding the (in)frequency with which the risk occurs.

If the answer is 'yes', then an adverse event evaluation form is completed. This form asks for an event description and information about the date of onset, date of resolution, seriousness, and device relatedness. It asks if the event was anticipated or not; if yes, it asks if the event has increased in nature, severity, or frequency from what was anticipated--a change in nature or increase in severity or frequency may cause the anticipated event to become unanticipated. Finally, it asks what medical actions were required for treatment.

Non-significant Risk Studies
Many device clinical trials are non-significant risk studies, meaning the IRB reviews the protocol but FDA does not. Since FDA has no record of these studies the ratio of non-significant risk to significant risk studies is poorly documented. Our private research established this number to be about 50/50.

Because FDA does not review non-significant risk devices there is no reviewer assigned to the study. So the question arises: are unanticipated, serious adverse device effects that occur in non-significant risk studies reportable to FDA? And if so, to what individual?

Yes, unanticipated, serious adverse device effects are reportable to FDA with ten days.

But you can set your fears aside. A quick phone call to CDRH's Office of Device Evaluation will guide you to the right party for reporting. They will ask for the protocol and a description of the event and consider the following issues: 1) should the study have been categorized as significant risk, and 2) should additional safeguards be included in the protocol.

 

Nancy J Stark, PhD
President

Clinical Device Group Inc