​The Challenge

For large, multihospital systems, hospital bed purchasing is a continuous process: Beds for newly constructed facilities need to be obtained, and beds at existing or newly acquired facilities occasionally need to be replaced. To handle this challenge, one large health system, Banner Health (Phoenix, AZ), developed a robust bed assessment process—an initiative that made it a finalist for the 2014 Health Devices Achievement Award.

The need to develop a bed assessment process grew from the organization's desire to standardize the bed models used for three applications—medical/surgical, critical care, and labor and delivery—in conjunction with its implementation of a proactive technology refresh program, which supplied the funding to update the beds. Before committing to purchase specific models for enterprise-wide use, the organization had to understand its needs and compare available alternatives. Poor bed selection decisions, after all, could lead to staff injuries (e.g., musculoskeletal injuries that result from patient-handling activities), patient injuries (e.g., falls, pressure sores), or patient dissatisfaction.

In keeping with the organization's philosophy of providing optimal equipment to caregivers so that they in turn can provide optimal care to patients, Banner Health sought to develop a bed assessment process that drew heavily on end-user feedback. The organization wanted beds that were easy to use and had intuitive features. Thus, it needed to develop an assessment process that would identify bed designs that "made it easy to do the right things, and difficult to do the wrong things," as explained by Bridget O'Hare, human factors clinical informatics senior consultant and a key member of Banner Health's bed assessment team.

The Solution

In formulating a bed assessment process, the Banner Health team paid close attention to factors surrounding the use of beds by frontline staff, both to clarify the organization's needs and to ascertain whether (and how well) various alternatives would meet those needs.

To understand what features might be needed for specific types of beds, the team observed caregivers in their work environment and collected information about the frequency and severity of bed-related injuries sustained by the organization's staff, sorting the results by job title and facility department. These activities helped the team identify what tasks were being performed and which of those posed a risk of employee injury (e.g., forceful exertions, sustained motions and prolonged exertions, repetitive motions, awkward postures). This analysis helped the team determine, for example, that the beds needed to have sufficient clearance underneath to fit floor-based patient lifts, and that they needed to be equipped with a power drive so that staff would not have to exert pushing/pulling forces beyond the acceptable limit when moving an occupied bed. 

To obtain feedback from frontline staff about the alternative bed models under consideration, Banner Health designed and conducted a human factors study that brought key usability factors to light. For this study, more than 60 bed users performed simulated scenarios with beds from various vendors. The study was conducted at three testing sites with multiple sets of users. Banner Health operates numerous facilities across a wide geographic region; thus, "it was important to get feedback from a representative sample of users," noted Perry Kirwan, senior director, clinical technology assessment and planning.

The study was conducted as follows:

• The team developed a protocol for different groups of users, including nurses, transporters, respiratory therapists, and the staff responsible for cleaning and sanitizing the beds between patients.

• The protocol for each group consisted of bed-related tasks representative of those that users would be expected to complete during the course of their shift.

• Staff participating in the study were asked to work through the protocol in a simulation center using the different beds from various vendors. To represent realistic, worst-case conditions, staff were asked to operate the beds without prior training or coaching. "In a perfect world, everyone would have sufficient training," observed J.P. Purswell, a consultant who worked with the team. "But we know that in the real world that is not always the case." This requirement helped reveal bed features that were intuitive and those that might lead to confusion when staff have to operate the bed under the stress of patient care.

• While performing each task, users were encouraged to think aloud to communicate how easy or difficult it was to complete the required task (e.g., to release the brake, to push the bed, to maneuver corners at a rapid pace). User interactions with a bed were observed by staff conducting the studies, as well as by representatives from the bed manufacturer, who were allowed to observe the proceedings but not coach the users. Most interactions were also filmed for later analysis.

• At the conclusion of testing, users were asked to identify features on each bed that they found intuitive or helpful, as well as features or control operations that they had difficulty locating or operating.

The study allowed the bed assessment team—along with bed manufacturer representatives—to witness how staff interacted with the bed under realistic conditions. For example, one scenario for evaluating labor and delivery beds involved a simulated case in which a patient was admitted for a normal vaginal delivery, but ultimately needed to be transported to the OR for an emergency C-section. Nurses were asked to push the bed down to the OR, going around corners and down a long corridor to mimic a real-life scenario. "The nurses were really tuned in," noted O'Hare. "The simulations were stressful and real for them, because they could relate to the situation."

The information from the usability study was coupled with fixed criteria for the three different types of beds (medical/surgical, critical care, and labor and delivery) to objectively evaluate bed purchasing selections.

Lessons Learned

In discussing the value of Banner Health's human factors usability study, O'Hare remarked that the testing helped the organization assess usability at "a whole different level." Rather than "the salesman driving the process," telling facilities what the manufacturer wants them to know, the test scenarios revealed the needs and concerns of the end users. Such studies could be adapted for other medical device purchasing decisions and replicated at virtually any facility.

In addition, Banner Health noted the potential for influencing device design. The organization shared the information it learned about difficult-to-use features and functions with the bed suppliers as a way to help spur the development of improved bed designs. In one instance—which O'Hare referred to as her "Aha! moment," revealing the power of this process—almost every test subject misunderstood the icons on a bed's control panel to represent the exact opposite of the intended function. The bed manufacturer has taken this feedback and expressed plans to modify the design.

Congratulations and thanks to the people who submitted Banner Health's application for the Health Devices Achievement Award: Bridget O'Hare, Perry Kirwan, Barbara Campbell, Merl Miller, and J.P. Purswell.