In preparing medical-device incident reports, attention to detail, staff cooperation, and follow-ups are key.

Eager to better manage their medico-legal risks, many hospitals routinely require a formal investigation whenever malfunctioning medical equipment and devices are involved in adverse incidents.

This strikes some as needless extra work, but it isn’t, experts assure.

“Adverse events provide a unique opportunity to examine entire systems and make adjustments that, ideally, will prevent a repeat of the incident in the future,” says Marvin Shepherd, PE (safety), FACCE, with Devteq Consulting in Walnut Creek, Calif.

Izabella Gieras, president of the American College of Clinical Engineering, agrees and adds that such investigations not only help shield against liability but also contribute to improved quality of care.

Triage System
What qualifies as an incident requiring investigation could be anything that deviates from the intended plan of operation—at least it is at each of the two William Beaumont Hospital campuses in Michigan, where Beaumont Services Co LLC provides clinical and biomedical engineering support.

“We initiate a probe whenever a medical device malfunctions or simply doesn’t perform according to manufacturer specifications and results in—or had the potential to result in—patient injury or illness,” says Gieras, manager of Beaumont Services’ clinical engineering department in the city of Royal Oak.

The majority of incidents typically investigated by Gieras and her department involve failures of catheters, syringes, and other everyday products. But problems also come up with big-ticket items, and that’s where things get complicated.

“We have a triage system in place to help us properly allocate investigative resources in response,” she says.

The department’s seven-tiered triage methodology categorizes incidents from mild to severe (the latter being those that result in injury or the initiation of special action), Gieras says.

Aided or not by triage, adverse-event investigators need to guard against failure to explore all possible angles. Shepherd says it’s tempting to prematurely conclude a probe report with “no problem found” after a cursory examination finds the equipment or device in good working order.

“The mistake here is to not recognize that the device is actually just one piece of a five-component system that delivers a clinical benefit,” Shepherd says. “The other components are the operator, the facility, the patient, and the environment—any one or a combination of those might be the cause of the adverse event. They, too, have to be considered during the investigation.”

Of course, even when all angles are taken into account, an event postmortem will only turn out well if good information is readily available during the fact-finding process.

“Investigations can be hampered when information needed isn’t available up front,” Gieras says. “For example, the longer it takes to collect input from the people involved in the incident, the cloudier their memory of it becomes.”

Working Relationships
As a rule of thumb, investigations that proceed smoothly and end propitiously require the cooperation of staff outside the biomedical department. That includes nursing, other clinical specialists, and even housekeeping—anybody who might have had knowledge of the circumstances involved in or leading up to the adverse event.

However, if a staffer suspects that your investigation has a hidden agenda to assign blame and bring about disciplinary action, he may hold back and offer little information of value. “Where criminality is not an issue, investigations have to be perceived by everybody as nonthreatening activities,” Shepherd says. “Disciplinary action arising from an investigation only has to occur once and, from that time forward, no future investigation will ever have voluntary cooperation.”

Shepherd contends that it is wise to establish good relations with nurses and other personnel before starting an investigation. Part of that exercise should involve you explaining in a friendly, sympathetic manner how things will proceed and what they need to do to help you.

In particular, be sure to meet with the hospital’s risk manager. His interest in the investigation is to ensure that you conduct it in accordance with the hospital’s policies, procedures, and protocols. However, Shepherd says, recognize that some risk managers are averse to allowing the biomedical or clinical engineering department to take the lead in an incident probe.

Not infrequently, an incident investigation will need to be conducted by a neutral third party. For example, if your equipment maintenance is contracted to an outside service company, that provider may want to avoid conflict-of-interest charges by foregoing a probe of its own and instead bringing in an independent investigator. (An independent investigation is advisable, too, for avoiding appearances of anything untoward in situations where serious injury or death occurred and equipment maintenance is an in-house responsibility, Shepherd hints.)

Follow-Up a Must
Bryanne M. Patail, BS, MLS, FACCE, biomedical engineer with the Department of Veterans Affairs, National Center for Patient Safety, in Ann Arbor, Mich, believes that the work of the investigator does not end when the inquiry closes.

“The outcome of all investigative work should be an action plan to provide both short- and long-term remedies that prevent a repeat of the incident that triggered the investigation in the first place,” he says.

However, be sure not to miss the mark by allowing finished reports to languish on the desks of recipients.

“You’ve got to follow through on your recommendations and work to see to it that the right resources will be brought into play for purposes of implementation,” Patail says.

Patail says you can promote adoption of your recommendations if you conduct some or all of the investigation with the help of a multidisciplinary team, one that includes members from fields that routinely make use of the equipment or device at the heart of the inquiry. “It’s easier to obtain user buy-in to the recommendations that way,” he says.

Among the recipients of your investigatory report should be the equipment or device manufacturer.

“Some problems can only be remedied by a redesign of the product,” Patail says. “Sharing your findings with the manufacturer helps them know what needs to be changed and how.”

However, don’t expect manufacturers to respond with changes to the equipment you’re using at the moment. The changes they’ll make—assuming they accept your findings—will more likely show up in next-generation models. Says Patail, “They’re probably not going to expend resources to retool current and older models, the exception being if the problem is serious enough that lives are endangered and the Food and Drug Administration (FDA) is considering ordering the product off the market.”

A good investigatory practice entails consulting various publicly accessible databases operated by the FDA, Joint Commission on the Accreditation of Healthcare Organizations, and other entities that keep tabs on adverse events nationwide. You can use these databases to see whether the equipment or device under your scrutiny has a checkered past and produced similar woes at other institutions. Patail says this can also help you become proactive in your thinking about adverse events.

Perhaps the most proactive step you can take is to share your database-developed insights among your hospital’s purchasing decision-makers so that they can be steered toward equipment and devices with reputations for the safest, most reliable operation.

“Prevent poorly engineered products from ever reaching your institution, and you’ll be conducting far fewer incident investigations and—by extension—writing up far fewer investigatory reports,” Patail says. “That, in and of itself, is a worthy goal because if you’re having to write fewer reports, it’s evidence that you’ve done your job well and are contributing in a very positive way to improved patient safety and lowered medico-legal risk.” 24x7

Rich Smith is a contributing writer for 24x7.