Anyone who has been a patient or a visitor to a hospital knows they are noisy places.
But rarely do we acknowledge that the cacophony, in addition to being a source of irritation, can be downright dangerous.
The Economic Cycle Research Institute publishes an annual list of the top-10 technology-related safety hazards and, year-after-year, “alarm hazard” comes out on top.
The problem is not that alarms – which are used on all manner of medical equipment such as infusion pumps, feeding devices, ventilators and heart monitors – don’t work well. On the contrary, they work all too well. Or, more precisely, all too often.
There are a number of potential problems with alarms: The signal is not successfully communicated to staff; the alarm does not provide caregivers with enough information; or the caregiver does not know how to respond.
But there is one problem that overwhelms all the others: the frequency of alarms.
This results in alarm fatigue – the term given to the common practice of health professionals turning off alarms because they are deemed to be annoying or irrelevant.
Consider that a single patient in a critical-care unit, hooked up to a panoply of machines, can trigger up to 700 alarms a day, according to one study.
And research has shown, time and time again, that about 90 per cent of alarms are false alarms.
Is it any wonder that nurses and physicians – and often patients themselves – become inured to the noise?
Fire alarms work because they are a rarity – a clear signal of potential danger. When everything from electrodes falling off someone’s chest to a heart stopping sets off the same alarm bells, then nothing is an emergency.
When alarms are distracting or ignored, bad things can happen: Medication errors are made because health workers can’t concentrate; patients get up for help and fall; people suffer brain damage and death because their supply of oxygen is cut off, and so on.
Alarm fatigue is a serious patient-safety issue.
In the United States, between 2005 and 2010, there were 566 alarm-related deaths. In Canada, there were, in that same time period, 16 deaths related to failures of cardiac alarms.
In the grand scheme of things – adverse events cause hundreds of thousands of deaths in North America each year – those are tiny numbers. But everyone acknowledges that grossly understates the problem. Rarely do you have a situation where an alarm fails or is ignored outright and the patients dies in an A + B = C equation.
The real question is what can be done to address the pervasive and longstanding problem that is alarm fatigue?
It’s not so simple as to say: Nurses and doctors have to respond to every alarm. With the sheer volume and variety of technology and related alarms, that’s not humanly possible.
So the starting point has to be using alarms appropriately. Right now, they tend to be overused because equipment manufacturers and health workers want to cover their butts and avoid lawsuits by monitoring everyone. One study found that one in four patients getting cardiac monitoring actually needed it.
Manufacturers and hospitals also have a tendency of using standard end-points that trigger alarms. But patients have widely varying health status, so the alarm settings have to be personalized. This, in turn, requires training. But too little is invested in training, especially of the nurses who bear the brunt.
In the cockpit of a plane, there are numerous alarms, but the warning systems differentiate between true emergencies and issues that can wait. Alarms in hospital rooms need the same standards.
For that to happen, alarm fatigue has to become a priority – on the ward, in hospital administration and among accreditation bodies and regulators.
In the U.S., the Joint Commission, a group that accredits hospitals, has made reducing alarm hazards a priority, and it is bearing fruit.
Canada is way behind. There are few data collected on the extent of the problem – aside from anecdotes – and hence little incentive to correct it.
It’s not enough to be alarmed by the problem – we have to fix it.
Follow me on Twitter: @picardonhealthReport Typo/Error