Hospitals put on notice: Alarm management is a top priority for 2014
Too many false alarms cause alarm fatigue, jeopardize patient safety
As promised, The Joint Commission (TJC) has unveiled a new National Patient Safety Goal (NPSG) on alarm management. This follows a Sentinel Event Alert that the agency issued in April 2013, citing evidence that providers are becoming overwhelmed by the alarms that emanate from ventilators, infusion pumps, heart monitors, and other medical devices.
The NPSG on alarm safety will be implemented in two phases. The accrediting agency explains that on January 1, 2014, hospitals will have to demonstrate that they have established alarm safety as an organizational priority. Further, they must begin to identify the most important alarms to manage based on input from medical staff, the potential for patient harm according to internal data, patient risk if the alarm malfunctions or is not attended to, and evidence concerning whether specific alarms are unnecessary and may contribute to alarm fatigue.
By the second phase of the NPSG, which begins on January 1, 2016, hospitals must show they have developed and implemented specific alarm management-related policies and procedures that cover such issues as clinically appropriate settings for alarm signals, the conditions under which alarm signals can be disabled or the parameters changed, which individuals have the authority to set or change alarm parameters, and policies related to alarm monitoring and maintenance.
Also required by 2016: Hospitals must demonstrate that they have educated staff and independent licensed practitioners about the proper operation of any alarm systems that they are responsible for. And that is just for starters. The agency notes that hospitals should stay tuned because the requirements for 2016 could be strengthened based on hospital experience with the phase one requirements as well as any new evidence about alarm safety that could emerge. (Also see "The Joint Commission is tapped to delve into IT-related adverse health events,")
The crux of the issue: False alarms
Raj Ratwani, PhD, the scientific director at the National Center for Human Factors Engineering in Healthcare in Washington, DC, suggests that the new NPSG is a good approach, although he would like to see more guidance on how hospitals can safely and effectively reduce the number of false alarms that take place. "That is the crux of the issue here," he says. "There are simply too many false alarms, and not enough true positives, and any time that happens, from a cognitive perspective it is very, very challenging."
Ratwani notes that other industries struggle with this same issue as well, and that there are no simple solutions. "One thing that is going to be very important is to really go through all the devices and equipment, see what is absolutely critical, and start weighing the patient safety risks for each of those alarms," he explains. "It is also critical to find good ways of getting the operator, the nurse, and the clinician to quickly accept that information."
Essentially, what hospitals have to do is set up devices and environments where clinicians can always see the information that they need to see without all of the other noise, says Ratwani. "It will require a very deliberate process, and hospitals are going to have to go device by device."
Investigators at the Regenstrief Center for Healthcare Engineering (RCHE) at Purdue University in West Lafayette, IN, already have personal experience with all the steps involved with managing the alarms that are placed on infusion pumps. In fact, the center has started to develop a system that downloads alarm information from infusion pumps into a common database to facilitate analysis.
"One of the interesting things we found in working with infusion pumps is that the chips embedded in the pumps actually have information about the conditions which generate the alarms, so you can get very high quality data from the technology and eliminate the need for human reporting, which in busy areas like the ED, can really become a detriment to getting good quality data," explains Steven Witz, PhD, MPH, director of RCHE.
Establish data-driven solutions
With this data, investigators have been able to create a better understanding of the situations around which alarms have been occurring. "The second thing that came out of our experience with infusion pumps that I think could be generalized to other devices is that often times the technology has relatively good guidelines for use, so in the case of infusion pumps, there is good information about dosing," says Witz. However, clinicians also have to enter upper and lower parameters around which the alarms will be triggered, and in many cases the judgment on these settings is very subjective, he says.
By accumulating all of this information and sharing it with clinicians so that they can see all the variation that exists across different sites, practitioners have been able to move toward practices that minimize the unnecessary alarms, explains Witz. "The work has taken us a couple of years to pull together, and we are still trying to sort through the information, but trying to make things as data-driven as possible seems to have really helped," he says. "It still introduces tremendous subjectivity in managing the alarms, but at least there is a logic structure to try to do a better job."
One trend that concerns Witz is that as vendors become aware of the value of information from their devices, some of them are trying to sell it back to the clinician. "What they are doing is building capacity to report on the data collected, but only for a fee, and I think that is the wrong way to handle it," he says. "This actually may deter hospitals and clinicians from trying to get into [the management of alarm information] because it just adds another expense."
Consider greater coordination
Longer term, there needs to be greater integration of information across all devices, observes Ratwani, noting that the current approval process for equipment is very siloed. "You look at one device and you say it is acceptable, but you are not looking at it in consideration of all the patient devices in the clinical environment. And until there is that greater integration, it is going to remain a complicated problem."
For instance, a clinician may be receiving information from a blood pressure monitor, a heart rate monitor, and a glucose monitor, but often times those devices are not talking to each other, explains Ratwani. "If you begin to coordinate those, you can have an overall management process," he says. "You can build a computational algorithmic method of making that process more coordinated and easier for the human user to actually accept that information and know what to do with it."
Health care can look to aviation as a model for how devices can be managed in a coordinated way, adds Ratwani. "The cockpit of an airplane is a completely regulated environment where every device is closely examined by the FAA [Federal Aviation Administration] to make sure that it is coordinated so that the flight management system, the GPS [global positioning system], and every other function all come across as a coherent whole to the human operator, the pilot," he explains. "The same is not true for the physician or the nurse. There is no regulating body and there is nothing that is looking at coordination across multiple devices that are in a patient environment."
- Raj Ratwani, PhD, Scientific Director, National Center for Human Factors Engineering in Healthcare, Washington, DC. E-mail: firstname.lastname@example.org.
- Steve Witz, PhD, MPH, Director, Regenstrief Center for Healthcare Engineering, Purdue University, West Lafayette, IN. E-mail: email@example.com.