An Alarming Matter
Medical-device alarm notification outside of critical care units is a scandalso much so that the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) created a national patient-safety goal to deal with the problem (National Patient Safety Goal 6B).
In the intensive care unit or high-dependency unit, where the nurse-to-patient ratio ranges from 1:1 to 1:3, alarms arent a big problem; the caregiver is right in the area and can easily hear and respond to alarms. In units with higher nurse-to-patient ratioswhere private rooms are the trendnurses must respond to nurse calls, phones, overhead pages, and medical devices.
Many tools and techniques have been employed over the years to improve alarm notification. Central stations have been a fixture for years at nursing stations and in centralized monitor tech rooms. Many nurse call systems take alarm outputs from medical devices and relay them to the appropriate personnel. Pagers were hot several years ago to deliver alarms directly to caregivers, but they have fallen from favor (for good reason). And many units sport signal lights outside patient rooms and message panels in hallways to ensure that caregivers get the right information quickly. In the last few years, some device vendors have been showing alarm notification on personal data assistants (PDAs).
The approaches above have a variety of problems. The absence of a closed-loop alarm-delivery method is a key limitation of pagers, message panels, and other indicators. Alarm fatigue is a common problem on units, resulting from too many false-positive alarms and the exposure of each caregiver on the unit to every patients alarms. Medical devices wired into nurse call systems and central stations can result in alarm fatigue. Many hospitals employ monitor techs to screen out false-positive alarms from patient-monitoring systems, but this is expensive. A single-vendor solution is not possible because no one vendor makes patient monitors, IV pumps, vents, and the other alarming medical devices used outside of critical care units. Getting vendors to create solutions that coexist and are validated as such presents numerous problems, one of which leads us right to the US Food and Drug Administration (FDA).
Alarm notification (when it really counts) is all about life-threatening alarms (LTAs). The notification and management of LTAs is covered by the indications for use in a medical devices 510k filing with the FDA. The methods or capabilities described in the 510k filing represent the primary means of alarm notification. Any method of alarm notification that is not covered by a devices 510k filing is a secondary means of alarm notification. Nursing units must be organized, staffed, and managed based on the devices primary alarm notification. Secondary alarm notification doesnt count, and any hospital that runs its units based on secondary-alarm-notification capabilities is at risk.
All of the foregoing is not to say that vendors arent interested in bringing alarm-notification systems to market to sell upgrades or displace competitors. Nurse call systemsand perhaps even health care information technology (IT) vendorswould love to become the data hub for bedside device communications.
So whats the solution? Any successful solution will have to work with most of the legacy devicesincluding nurse call systems, phones, and medical devicesfound in hospitals today. Very few hospitals can afford the wholesale replacement of equipment necessitated by single-vendor solutions or limited interfaces.
The typical use model for alarm notification necessitates a nurse-carried device that supports audible annunciation, text, and graphics to screen false-positive alarms. Nurse-to-patient assignments must be supported so caregivers only receive alarms from the patients for whom they are responsible. And a system must include escalation in the event a caregiver is otherwise occupied and cannot respond quickly to an alarm.
Given the plethora of vendors, devices, and systems at the point of care, an alarm system must provide an enterprise solution that eliminates the unnecessary duplication of multiple-point solutions. Nurse-to-patient assignments must be entered once; caregivers must get all of their alarms through one device, rather than through a tool belt of devices.
Alarm notification is just one part of point-of-care automation, and it represents a real opportunity for biomedical engineering to collaborate with IT to ensure that the needs of caregivers are met and patient safety is maximized. How the market develops is anyones guess, but making progress in improved alarm notification is becoming a real possibility.
Tim Gee is principal, Medical Connectivity Consulting, Beaverton, Ore.