Paul McGurgan: Using Data to Solve the Alarm Problem

Whether you work in a 75-bed community healthcare organization or a 1,000-bed teaching hospital, you’re probably putting considerable effort into determining how your facility will address the requirements of The Joint Commission’s National Patient Safety Goal on alarm systems safety while trying to understand how it will affect patient outcomes and satisfaction as well as your staff’s performance and job enjoyment.

Predictably, most hospitals are approaching the alarm safety problem using classic quality improvement techniques. Whether called “quality management,” “process improvement,” or something else, they all depend on having a measurable process. Unfortunately, the inherent first high hurdle in alarm management is the capability to cast a large enough net to gather all the data needed to effectively define the existing processes and take measurement of the problem. Alarm messages, waveforms, vitals, and alarm limit settings are all necessary data components in making those determinations.

“If you don’t measure it, you can’t improve it,” and “In God we trust. All others bring data,” are quotes often attributed to W. Edwards Deming, the original authority in quality management. Indeed, having the alarm and physiologic data is an essential step in the alarm management process. To make an evidence-based determination of what, if any, changes should be made in alarm settings, escalation processes, or clinical work flow, a simple count of the alarms your monitors are producing is not enough. And merely filtering which alarms are forwarded to a pager or smartphone does little to reduce the number of alarms that occur on the unit.

Analysis of hospital data shows that millions of alarms occur every month, likely far more than even the most burdened provider would estimate. It has been reported that 85-99% of cardiopulmonary monitor alarms are either false or clinically non-actionable (Talley, et. al., Horizons, spring 2011). Additionally, system alarms, which indicate the device requires attention rather than the patient, can in some cases comprise 70% of the alarms generated. We have seen many examples where a single atrial fibrillation event is sustained for four or five days. While this accounts for only one alarm event, its contribution to audio pollution and work-flow interruption goes far beyond that one tick mark. But, without all the data, how can you know your hospital’s alarm predicament?

The questions we are hearing from hospitals include:

  • How many alarms do we have?
  • What different kinds of alarms do we have?
  • Do we have different alarm situations in different care areas?
  • Can we get the waveforms and limits associated with the alarms?
  • What are we going to do to address the problems?
  • How are we going to measure the impact on any changes we make?

The Alarm Management Workshop at the AAMI 2015 Conference & Expo in Denver promises to provide a lot of insight and perspective into those questions—both from industry and hospital experience. I look forward to seeing you there.

Paul McGurgan is the national program director for alarm fatigue initiatives at Excel Medical.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s