Home Device Studies E&I NSR Policy
E&I NSR Policy


The Food and Drug Administration (FDA) regulations permit a device to be shipped only if it has a marketing permit. Shipment of a device without a marketing permit is not allowed unless that device is exempt from the permit requirements by means of an Investigational Device Exemption (IDE). 

The FDA regulation about IDEs is located in 21 CFR 812. This regulation also specifies several exemptions from the IDE rules such as consumer preference testing and testing of a non-invasive diagnostic device where the usual procedure is also used as a back-up. 

These are exemptions from the IDE requirements. They should not be confused with exemptions from Part 56 (IRB review) or from Part 50 (informed consent requirement.) 

There are two types of IDEs. The first is obtained by the sponsor applying directly to the FDA. In the second is an “abbreviated IDE” which the FDA will consider to have been granted should an Institutional Review Board (IRB) give approval to the investigation under Part 56, and make a decision that it agrees with a sponsor's contention that the device is a "non-significant risk device." (812.2b) 

The regulation defines a "significant risk device" in 21 CFR 812.3(m). The FDA Information Sheet of October 1, 1995 defines it as "a study of a device" that meets one of the following criteria:

 1.  is intended as an implant

AND presents a potential for serious risk to the health, safety, or welfare of a subject;

2.  is purported or represented to be for a use in supporting or sustaining human life

AND presents a potential for serious risk to the health, safety, or welfare of a subject;

3.  is for a use of substantial importance in diagnosing, curing, mitigating or treating disease, or otherwise preventing impairment of human health

AND presents a potential for serious risk to the health, safety, or welfare of a subject;

4.  or otherwise presents a potential for serious risk to the health, safety, or welfare of a subject."

The difference between the two definitions is critical as the first evaluates the risks of the device and the latter interpretation evaluates the risks derived from all of the procedures in the study of the device.

The Information Sheets also suggest that the IRB should consider the NSR/SR question (Part 812.2) first and the protocol approval question (parts 56 and 50) second. The following are the guidelines of the E&I IRB for making decisions regarding the significance of the risk of devices.



In order to reach a consensus on the IRB review guidelines for making SR/NSR decisions, the IRB has agreed on several policies and presumptions. 

1. The IRB depends upon the sponsor to make an initial determination of the regulatory status of the device, to state that determination clearly and to provide justification for that position. The IRB does not have the sponsor's knowledge of the device's development or of the regulatory status with regard to the FDA. Therefore, it is a sponsor's duty to determine their marketing strategy and their decisions about the regulatory status of the device. 

2. The IRB will not routinely ask for outside expert consultation. Both Congress and the FDA understood the diverse membership and lack of specific expertise on IRBs when they assigned IRBs the responsibility for NSR review, therefore acknowledging that IRBs do not have the expertise to render a peer review of the same quality given by FDA panels. 

3. The IRB will consider the IDE question to be separate from the question of protocol approval and will render two separate decisions. The criteria for IRB approval and for NSR/SR decisions are quite different and use several different definitions of risk. Each risk definition requires a different analysis and has a different outcome. The risks considered under part 56.111(a) must be minimized and those in (b) include all of the risks in a study and are judged in relation to the benefits of the study. The risk in 50.25(a)(2) is simply to be explained. The risks in 812.(3)(m) are those of the device.

This policy could lead to approval of a protocol and a SR decision or to an NSR decision and disapproval of the protocol.

4. The IRB will consider the incremental risk posed to a subject by the device or the changed device over the risks of the device or procedure that would be posed to a patient by the standard or usual device or procedure.

Although it is reasonable to think that the FDA could not have wished an absolute criterion to be used since this would seem to be self-defeating, the October 1995 Information Sheets state:

"The amount of potential reduced or increased risk associated with the investigational pacemaker should only be considered (in relation to possible decreased or increased benefits) when assessing whether the study can be approved."

The intent of Congress was to allow a method of review that would speed the device approval process without adding undue risk to subjects. An absolute criterion evaluates the risks against a null background without comparison to existing (FDA cleared) devices or procedures or the harms that might befall the patient during ordinary care. It therefore yields a much higher risk analysis which means almost all devices are SR and must go to FDA.

5. The IRB will not consider the risks that might occur through negligent use of the device. Following release of a device, the FDA has no ability to control the skill of the device user. Therefore, analysis of risk in a trial must rely on the risks posed by the device as it would be used normally rather than as it might be misused. (For example, if surgery is involved, the evaluation will not include the skill of the surgeon beyond the skill usually required in practice.)

6. The risks to be considered in review of the study are all risks; those reviewed for the

NSR determination are those presented by the device used in the way it will be applied in the confines of that study. Using these criteria the IRB satisfies both the regulatory definition which asks for evaluation of the device and the FDA interpretation which asks for evaluation of the device study.



The questions involved in an IRB NSR decision are:

1) What are the possible harms to subjects in this study and which of those are potentially caused by the device as it is to be used in the context of the proposed study?

2) How frequently might those harms occur in the population described in the protocol?

3) How serious are those harms?


To reach their decision:

1.  The IRB will review the statements from the applicant concerning regulatory status and regulatory requests.

2.  The IRB will render two decisions: one concerning approval of the investigational protocol and one concerning whether the device as used in the context of the study poses a significant or non-significant risk. A decision approving a protocol is not a decision relative to NSR or SR.

3.  The IRB will consider all of the risks of the device as compared to similar devices already approved for use. That is, the IRB will consider the increment in risk over the prevailing risks the subject might experience.

4.  In general the IRB will consider the risks of the device itself rather than those risks plus those from all ancillary procedures. An exception will be made if a procedure is done solely due to the investigational nature of the device.

5.  The IRB will consider risks posed by possible device malfunctions; it will not generally consider risks posed by negligent use.

6.  As a general rule, a device causing systemic effects will be considered a significant risk device.

Last Updated ( Friday, 07 April 2017 09:38 )