Missing Patient Drills: How to Run an Elopement Response Exercise That Finds Real Gaps

By the MGM Solutions team – life-safety RTLS specialists since 2002.

Every hospital has a missing patient procedure. Far fewer have tested it in a way that would
survive a real event. The usual drill is announced in advance, run on day shift, and ends when someone
“finds” a volunteer standing in a predictable stairwell. Everyone signs the attendance sheet, the
exercise is logged for the surveyor, and nothing about the actual response gets better.

A real elopement does not work like that. The patient who leaves is usually the one nobody was
watching closely at that moment. They leave at shift change, during a code on the same unit, or while
the only sitter is on break. The first minutes decide the outcome, and those minutes are exactly what
a comfortable drill never measures.

This guide covers how to design a missing patient drill that finds problems, what to time, and how
real-time location technology changes the shape of the search. If you are building the prevention
program itself, start with our guide
to preventing patient elopement in hospitals. This article is about the response when prevention
fails.

Why Elopement Response Deserves Its Own Exercise

The Joint Commission treats an unauthorized departure that leads to death or serious harm as a
sentinel event, and CMS surveyors look closely at how a facility supervises at-risk patients. Those are
the formal reasons. The practical reason is simpler: an eloped patient with dementia, a psychiatric
hold, or a medical device still attached can come to harm within a short walk of the building. Parking
structures, retention ponds, loading docks, and busy roads are all close to most hospital entrances.

A missing patient search also behaves differently from other emergency codes. A fire or a medical
emergency has a known location. A missing patient code starts with the one fact nobody has: where the
person is. Every minute spent searching the wrong area widens the area that has to be searched next.

The Four Gaps Drills Usually Miss

Time to discovery. Most procedures begin “when a patient is discovered missing.”
That step is the slowest and least controlled part of the chain. If discovery depends on the next
scheduled check, the patient may have a fifteen or thirty minute head start before the code is ever
called.

Who searches where. The page goes out and staff spread out without assignments.
Three people check the same lobby. Nobody checks the exterior stairwell or the garden courtyard.
A search with no zone assignments is a search with blind spots.

The description. Responders at exits are told to watch for a missing patient, but
the clothing, height, and photo sit with the charge nurse on the unit. A door posted by someone who
does not know what to look for is not really posted.

The handoff outside. Once the search reaches the property line, who calls local
police, who notifies the family, and who decides when to escalate? Many plans go vague at exactly the
point where an outdoor search becomes the only search that matters.

Designing a Drill That Measures the Right Things

A useful exercise is built around timestamps, not attendance. Before the drill, decide which
intervals you will capture and assign an observer to record each one.

1. Departure to awareness: from the moment the volunteer leaves the unit to the
   moment any staff member realizes it.
2. Awareness to code called: how long it takes to confirm the patient is not simply
   in the bathroom or at radiology, then announce.
3. Code called to exits covered: when each assigned exit actually has a person at it
   who knows the description.
4. Code called to located: total search time, and which search zone produced the
   find.
5. Located to safe return: how the patient was approached and escorted back.

Run at least some drills unannounced, on nights and weekends, and during shift change. Vary the
scenario: a patient who walks out the main entrance, one who uses a stairwell, one who reaches the
parking area. The point is to find the condition where your response breaks, then fix that
condition.

Where RTLS Changes the Clock

A procedure can only be as fast as the information feeding it. With door alarms alone, the
facility learns that a door opened, not who went through it or where they went next. With periodic
checks, the facility learns a patient is gone only when someone looks.

Continuous location tracking changes two of the five intervals above. Departure to awareness
shrinks because the system alerts when an at-risk patient crosses a boundary, rather than waiting for
the next check. Code called to located shrinks because responders start from a last known location and
direction instead of a whole building. That is the difference our
article on RTLS
alert latency describes: seconds at the start of a search are worth minutes at the end of it.

The outdoor portion matters most and is where many systems stop. A patient who clears the exit
door has not been found; they have just become harder to find. SecurTRAK
PatienTRAK is designed to
keep tracking past the door, through courtyards and parking areas, on the same infrastructure used
indoors. We explain why that matters in
Campus-Wide
Elopement Prevention: Why Indoor-Only Systems Leave Patients at Risk.

Technology also gives the drill something it rarely has: an objective record. When the location
system logs the volunteer’s path and the alert timestamps, the after-action review no longer depends
on people reconstructing a stressful ten minutes from memory.

Test the Technology Inside the Drill

If your facility uses an electronic elopement system, the drill should test it, not route around
it. Include at least one scenario where the volunteer wears a tag and one where the tag is removed or
its battery is low. A system that silently loses a tag is a system that fails without telling anyone,
which is why tag
battery and tamper supervision belongs on the drill checklist.

Also confirm that alerts reach the right people on nights and weekends, that escort or override
functions work when staff accompany a patient out, and that alerts can be acknowledged and closed
cleanly so the next one is not lost in the noise.

Running the After-Action Review

The review is where a drill pays for itself. Keep it short, hold it within a day, and focus on the
timeline rather than individual blame. For each interval, ask what slowed it and what one change would
shorten it next time. Typical fixes are concrete: a laminated exit assignment card at each nurses’
station, a photo captured at admission for every at-risk patient, a standing script for calling local
police, or a change to which alerts go to which badge.

Write down the changes, assign an owner to each, and retest the weakest interval in the next drill.
A drill program that measures the same intervals quarter after quarter will show whether the response
is actually getting faster. That record is also far more persuasive to a surveyor than an attendance
sheet.

Frequently Asked Questions

Related reading:
Code Pink: Building
an Infant Abduction Response Plan  • 
Full-Floor RTLS vs.
Door-Only Monitoring  • 
Compare Campus-Wide RTLS
Systems