Understanding and fighting alert fatigue
The danger of sub-optimal alert systems
In 2013, a 16-year-old boy at one of the US’s top hospitals was given a 3800% overdose of his medication.
The hospital’s built-in alert system noticed the overdose order and sent alerts to a doctor and a pharmacist. And yet, a short time later, the overdose was administered and the seizures, full-body numbness, and struggle for the boy’s life began.
How could this happen—especially when the safety system caught the problem before the medication arrived at the boy’s bedside?
The answer is alert fatigue.
Both the doctor and the pharmacist ignored the system’s alert because that same system generates alerts for about 50% of the hundreds of prescriptions they deal with each day. They’d learned that most of those alerts were false alarms, and, as a coping mechanism, they’d started giving them a cursory glance at best.
And so a boy who should have taken a single pill took 38. And while he ultimately survived, the consequences to his health were significant.
Stories like these are common—and too often fatal—in hospitals and the aviation industry. In fact, a 2013 survey found that 19 out of 20 hospitals rank alert fatigue as their number one safety concern.
And while the risks are different, alert fatigue is also common for IT and DevOps teams as they monitor the always-on technology that drives our businesses.