Medical Devices: Who Needs to Read Device Instructions?

By William A. Hyman, PhD

Essentially all medical devices used in hospitals come with a user instruction manual, commonly called Instructions for Use (IFU) or Directions for Use. Typically an IFU includes basic operational “how to” information as well as pages of warnings, cautions, and other general or device-specific information beyond simply how to operate the device. Operational trouble shooting, contingency instructions, and maintenance information may also be included. All of this is consistent with the Food and Drug Administration’s (FDA) requirements for “prescription” devices, which state that the device’s labeling (including the IFU) must supply information about indications, effects, routes, methods, and frequency and duration of administration, as well as any relevant hazards, contraindications, side effects, and precautions under which the device can safely be used. As used here, “prescription” includes those devices for which safe use requires use by or direction of a practitioner licensed by law.

It is not unusual (in my experience) for users to state that they have not read the IFU, they cannot recall ever having seen the IFU, and they do not know where the IFU is kept. In the case of some rental equipment, the IFU might not have made it to the end user because the rental company retained it. Discussion about who did or did not read the IFU may occur subsequent to an adverse event that is directly related to information in the IFU that, if known and followed, would have prevented or mitigated the event. If the event becomes the subject of a claim or litigation, both the plaintiff and the manufacturer will be happy to point out that the user had not read the IFU. The manufacturer would also find some protection against any claim of inadequate warnings since it could be argued that if the IFU was not read then it cannot matter what it did or did not say. Absolving one potential defendant tends to shift blame and cost to the remaining defendants.

I asked the National Patient Safety Foundation’s email discussion list how often staff members read IFUs, and most responses confirmed my belief that IFUs are not routinely read. Some responded by explaining that IFUs are often too long and too complicated, that in-service or peer training obviated the need to read the manual, and that a short, possibly locally produced “cheat sheet” replaced having to read the complete IFU. Some device manufacturers provide quick instructions, sometimes directly on device labels, but these cannot be considered to replace the full IFU. Beyond the scope of any written material, the adequacy of in-services in general is open to substantial discussion, especially when they are squeezed into shift schedules and are expected (often promised) to be brief.

In any of the circumstances described above, an important question is what was covered or not in the in-service, training, or locally written instructions. The latter also raises the question of whether creating new instructions for a medical device rises to the level of a modification, or if the process meets the same standards as instructions created by the manufacturer. Perhaps in some circumstances, one could argue that the locally produced instructions are better because they are understandable and succinct. This argument, however, would be challenged if something relevant to an incident was in the original IFU but not in the local instructions.

If an IFU is indeed too complicated to be understood by the staff, it is important to convey that information to the manufacturer. Such feedback should be stated as a “complaint” because “complaint” has specific FDA regulatory meaning that includes tracking and, if necessary, responding to the reported information. Perhaps to avoid having negative user feedback in their complaint files, some companies have been known to dissuade people from saying they have a complaint in favor of just “expressing a concern.” It is therefore important to insist that you are complaining. Poor instructions may also be the basis for a voluntary report to the FDA. In turn, the FDA could increase its efforts to review IFUs for understandability and usability. The quality of the instructions is also something to consider when purchasing medical devices. In this regard, a post-incident assertion by the users that the IFU was unclear can lead directly to the observation that the device had been selected despite the poor IFU or without knowledge of it.

A few responders in the email discussion noted that if they were having a problem using a device, they might look in the IFU for troubleshooting information, perhaps secondary to asking a colleague or a vendor representative. Also cited was an analogy to personal behavior, e.g., “I don’t read the manual that comes with my car; why would I read the manual that comes with a medical device?” However true that may be, I find it hard to rationalize that this justifies not reading the user manual for a medical device; staff members have a professional responsibility to be knowledgeable and to use devices safely and effectively.

The Disconnect

There is an obvious disconnect between manufacturers investing heavily in generating IFUs, the FDA insisting on them, and at least some users routinely ignoring them and justifying doing so. If this is true, then the potentially important role of the information contained in the IFU becomes a fiction. Alternatively, perhaps the value of the IFU is a fiction perpetrated by manufacturers and the FDA, while also serving to protect the manufacturer when use errors are made.

This disconnect could be mitigated in part by better human factors design, resulting in safe use that is more intuitive and less reliant on the instructions, and also perhaps simpler instructions. However, even where this is the case, there will often still be IFUs that contain important information.

A Proposal

It appears to me that users of medical devices must be expected to read the IFU as part of their training since the user, at least sometimes, needs to know not just how and when to turn the knobs and push the buttons, but to also understand the correct application of the device and the associated risks and risk mitigations. If this is the case, then staff members must be provided and credited with sufficient time to read the IFU. That is not the same as just making the IFU available to them, as it should not be expected that such reading will be done, on their own time. Given expectation and opportunity to read the IFU, there should also be a means to record that staff members have actually done so. Self-reporting may be considered, but, of course, creates opportunity for less-than-forthright responses. In addition, a group discussion might be beneficial, with everyone encouraged to ask questions. It is not desirable for “No, I never read it” to be replaced with “Yes, I read it but didn’t understand it.”

I can hear the howling now about the time and resources this would require, adding additional cost to today’s economically constrained environment. I do not see, however, how the alternative —not reading the IFU—is acceptable. That may become evident the next time a staff member honestly testifies after an adverse event that he never read or even saw the IFU, no one ever told him he was supposed to read it, and he doesn’t even know where to find a copy.

William Hyman is professor emeritus of biomedical engineering at Texas A&M University. He now lives in New York where he is adjunct professor of biomedical engineering at The Cooper Union. Hyman may be contacted at