Alert Fatigue Simulator | mysimulator.uk

Alert-system sensitivity

Few, well-targeted alerts.

Shift

Speed 1.5×

Clinician fatigue

Fatigue 0%Attentiveness 100%

This shift

Charts done

0 / 0

Alerts fired

Critical present

Critical fired

Caught

Missed (fatigue)

Overall critical catch rate

–

Low-value alert (noise)

Critical alert

Critical, caught

Compare settings

Setting	Alerts	Catch %

Run a full shift on each setting to compare.

Why more alerts can mean more misses

Each alert costs a moment of attention regardless of whether it turns out to matter. A "high sensitivity" configuration flags far more low-value warnings — interactions that are fine in context, allergies the clinician already knows about, generic dosing thresholds — so the same shift produces a much longer stream of alerts to click through.

fatigue += cost_per_alert  (every alert, regardless of value)
attentiveness = 1 − fatigue  (clamped)
P(catch critical alert) ≈ P(system fires it) × attentiveness

Low sensitivity fires few alerts, so fatigue barely rises — the clinician stays attentive, and the alerts that do fire are read carefully.
High sensitivity technically fires an alert for a larger share of genuine critical issues — but the flood of low-value noise it also generates drives fatigue up fast, so by the time a critical alert appears it's often skimmed and dismissed like all the others.
"Critical fired" (system caught it) and "Caught" (clinician actually acted on it) are different numbers — the gap between them is fatigue-driven misses, not detection failure.

This is the core patient-safety tradeoff in clinical decision support: comprehensiveness of automated checking and clinician attentiveness pull in opposite directions once alert volume crosses a threshold.

Low-value Critical Caught

Press "Start shift" to begin

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.