When to Remove Medical Equipment From Service

Knowing when troubleshooting needs to stop and the device needs to come off the floor

A practical guide to recognizing safety concerns, intermittent failures, damaged equipment, questionable performance, and other situations where a medical device should not stay in clinical use.

Back to Biomed Basics

What This Page Explains

One of the most important decisions a biomed makes is not how to fix a device.

It is deciding whether that device should still be used.

Sometimes the answer is obvious.

A power cord is burned. A siderail is broken. A ventilator will not complete its pre-use check. A defibrillator fails to deliver a shock during testing.

Remove it from service.

Other situations are not so clear.

The nurse says the monitor shut off once, but you cannot duplicate it.

An infusion pump has a cracked housing but still passes every functional test.

A patient monitor works locally but will not send data to the EMR.

A bed has a broken plastic cover that does not appear to affect operation.

A battery only lasts half as long as it used to.

Those situations require judgment.

The goal of this page is not to create one universal rule for every device or every hospital. Your facility policy, manufacturer documentation, equipment risk, clinical use, and the specific failure always matter.

The goal is to explain how a biomed can think through the decision.

Because sometimes the most important question is not:

What failed?

It is:

Do I have enough confidence in this device to put it back on a patient?

Jump to a Section

The Simple Version

A medical device should generally stay out of clinical service when you cannot reasonably confirm that it is safe and able to perform the function someone may rely on it to perform.

That does not mean every failure makes the entire device dangerous.

It does mean you need to understand what failed and what that failure could affect.

Think about four questions:

  1. What is the reported or confirmed problem?
  2. Could that problem affect patient safety, treatment, monitoring, diagnosis, or a required safety feature?
  3. Can I verify that the device is functioning correctly now?
  4. Would I be comfortable explaining why I returned this device to service if the problem happened again?

That last question is useful.

If your explanation would basically be:

I couldn't make it happen again, so I figured it was probably fine.

You may not be finished troubleshooting.

Removing Equipment From Service Does Not Mean You Know What Failed

This is an important distinction.

You do not need a complete diagnosis before removing a device from service.

Removing equipment from service is a risk decision.

Repairing the equipment is a technical process.

Those are related, but they are not the same thing.

Imagine a patient monitor that randomly shuts off.

You do not yet know whether the problem is:

You may need hours of troubleshooting before you know the exact cause.

But you can decide much earlier that an unexplained shutdown makes the monitor inappropriate for clinical use until the problem is better understood.

Sometimes you know the device should not be used before you know why.

That is normal.

Start With the Risk, Not the Repair

Biomeds naturally want to troubleshoot.

We see a problem and immediately start thinking:

What board controls this?

Can I get into service mode?

Do I have another battery?

Is there an error log?

Can I reproduce it?

Those are good questions.

But before diving into the repair, ask what the failure actually means.

A broken printer on an ECG machine and a failed shock-delivery circuit on a defibrillator are both equipment failures.

They are not the same level of problem.

A network connection failure on a vital-signs monitor and an inaccurate NIBP reading are both faults.

They do not create the same risk.

A cracked decorative cover on a hospital bed and a cracked siderail are both physical damage.

Again, not the same thing.

The first step is understanding what function is affected and whether anyone could reasonably depend on that function during patient care.

Obvious Reasons to Remove Equipment From Service

Some conditions usually make the decision fairly easy.

Examples include:

The exact response still depends on the equipment and your procedures.

But if you are looking at the device thinking:

Someone could get hurt if this does exactly what it just did while connected to a patient,

that is a strong reason to keep it out of service until the problem is resolved.

Reported Problems You Cannot Duplicate

This is where things get harder.

A nurse reports:

This pump shut off during an infusion.

You bring it to the shop.

It powers on.

You run it for twenty minutes.

Nothing happens.

You run a basic functional test.

Everything passes.

What now?

The fact that you could not duplicate the failure does not prove that the failure did not occur.

It only proves that it did not occur during the conditions you tested.

That distinction matters.

You should consider:

A vague complaint like:

It was acting weird.

is different from:

It shut completely off twice while running on battery.

The more serious and specific the reported failure, the more confidence you should have before returning the equipment to service.

Intermittent Problems

Intermittent failures are some of the most frustrating problems in biomed.

They also deserve respect.

A device that is completely dead is often easier to troubleshoot than a device that works 99 times and fails once.

That one failure may happen:

Do not let a device convince you it is fixed just because it behaved while sitting perfectly still on your bench.

Try to reproduce the conditions surrounding the complaint.

Flex cables carefully.

Move connectors.

Run the device longer.

Operate it on battery.

Switch between power sources.

Use the accessories involved in the original complaint when appropriate.

Review logs if available.

Intermittent does not mean imaginary.

Physical Damage

Not all physical damage automatically makes equipment unsafe.

But all physical damage deserves evaluation.

A small crack in a cosmetic cover may have very little effect on operation.

A crack near:

can be a completely different situation.

Look beyond the crack itself.

Ask:

What is this part supposed to do?

Then ask:

What happens if it fails completely?

A broken piece of trim on a monitor cart is not automatically equivalent to a cracked structural weld.

The location and function of the damage matter more than how ugly it looks.

Also think about cleaning and infection-control implications.

A broken enclosure may create openings that are difficult to clean even if the electronics still work normally.

Electrical Safety Concerns

Electrical problems deserve a low threshold for taking equipment out of service.

Examples include:

Do not assume a device is safe because it powers on.

And do not assume a basic electrical safety test answers every electrical complaint.

If someone reports being shocked by a device, the important question is not simply whether you can obtain a passing reading five minutes later.

You need to understand what happened and whether the test you performed reasonably addresses the complaint.

A loose connection may only fail while the cord is moved.

Moisture may dry.

A damaged conductor may make intermittent contact.

Electrical safety testing is one tool.

Troubleshooting still matters.

Fluid Intrusion

Medical equipment lives around fluids.

IV fluids.

Cleaning solutions.

Water.

Blood.

Disinfectants.

Saline.

Condensation.

A small amount on the outside of a properly sealed enclosure may be nothing more than a cleaning issue.

Fluid inside the equipment is different.

If liquid may have reached internal electronics, power supplies, connectors, switches, batteries, or high-voltage areas, do not just wipe the outside dry and power it back on.

The device may need:

The response depends on the device and where the fluid went.

The important part is not assuming:

It dried, so it is fine.

Fluid can leave contamination, corrosion, conductive residue, and damage that is not immediately obvious.

Alarm and Safety-System Failures

If a device depends on alarms to protect the patient or alert staff, an alarm failure can be more serious than the main function appearing to work.

Imagine a ventilator that delivers breaths normally but has an unreliable audible alarm.

Or an infusion pump that runs correctly but does not generate the expected occlusion alarm.

Or a patient monitor with a speaker that cuts in and out.

The fact that the basic function works does not automatically make the device safe.

Medical equipment often relies on layers of protection.

Those layers may include:

When one of those protections fails, ask whether the device can still be used in the manner for which it was designed.

Do not treat safety systems like optional features.

Accuracy and Performance Problems

Some devices can function perfectly from an electronics standpoint and still be wrong.

A patient monitor may display NIBP readings.

An infusion pump may move fluid.

A thermometer may display a temperature.

A ventilator may generate pressure.

That does not prove the result is accurate.

If a device fails an accuracy, calibration, or performance verification, the affected function should not be trusted until the failure is understood and corrected.

This is especially important because inaccurate equipment may look completely normal.

A device that is ten percent wrong does not necessarily flash:

WARNING: I AM TEN PERCENT WRONG.

It may confidently display the wrong value.

That can be more dangerous than an obvious failure.

An obvious failure tells the user not to trust the device.

An inaccurate device may give them a number that looks believable.

Battery Problems

Battery complaints require some judgment.

A battery that has slightly less runtime than when it was new is not automatically unsafe.

A battery that shuts the device off unexpectedly may be.

Think about how the battery is used.

Is battery power:

For example, a stationary device that is almost always plugged in may create a different risk than a transport monitor that needs to operate reliably between departments.

Battery problems can include:

A swollen, damaged, leaking, or overheating battery is not just a runtime problem.

That is a battery safety problem.

Network and Integration Problems

This is an area where modern biomed gets interesting.

A device may work perfectly as a standalone medical device while failing part of the clinical workflow.

For example:

A patient monitor accurately displays vital signs at the bedside but is not sending those values into the EMR.

An ECG machine acquires and prints a correct ECG but cannot transmit it to the MUSE system.

An imaging device produces images locally but cannot send them to PACS.

Is the device out of service?

Maybe.

Maybe not.

This depends heavily on the clinical workflow and facility policy.

You need to separate two questions:

Does the medical device itself perform its clinical function?

and

Does the entire workflow the hospital depends on still function?

Sometimes manual workflows can temporarily replace a failed interface.

Sometimes they cannot.

Sometimes the interface is important enough that clinical leadership does not want the equipment used without it.

This is why not every network problem is simply:

Wi-Fi broken. Tag it out.

But it is also why you should not dismiss connectivity as:

That's just an IT problem.

If clinical information is expected somewhere and it is not getting there, somebody needs to understand the impact.

Missing or Damaged Accessories

Sometimes nothing is wrong with the main device.

The problem is the accessory.

Examples include:

If the failed accessory is replaceable and the device can be safely used without it, you may not need to remove the entire device from service.

But do not assume every accessory is optional.

Ask whether the missing or failed accessory affects:

Replacing the bad accessory may solve the problem immediately.

Removing a good $20,000 device from service because of a bad $100 cable does not make sense if the cable can simply be replaced.

At the same time, returning that device with the known bad cable still attached makes even less sense.

Can Part of a Device Be Used?

This comes up often with multifunction devices.

Suppose a patient monitor has:

The NIBP function fails verification, but everything else works.

Can the monitor still be used for ECG and SpO2?

There is no universal answer.

It depends on:

You need to think beyond your own knowledge of the failure.

You know the NIBP is bad.

Will the next nurse know?

Will the device clearly prevent someone from using it?

Could someone plug a cuff into it two shifts later and assume it works?

A workaround that only works because the biomed remembers the limitation is not a very good control.

Temporary Repairs

Temporary repairs require caution.

There is a big difference between a controlled, manufacturer-acceptable temporary solution and improvising something because the department needs the equipment back.

Medical equipment is not the place for:

The fact that something physically works does not mean it should be placed back into patient care.

A temporary repair should still meet the safety, performance, and service requirements that apply to the equipment.

Clinical urgency does not magically make an unsafe repair safe.

When to Stop Troubleshooting and Escalate

Good biomeds troubleshoot.

Good biomeds also know when to stop.

Escalation may make sense when:

Calling the vendor is not failure.

Neither is asking another technician.

Neither is escalating to your lead, manager, clinical department, IT team, facilities team, or risk department when the situation crosses into their area.

The mistake is pretending you have proven something you have not.

Tagging and Securing Equipment

If you decide a device should not be used, make that decision obvious.

The exact process depends on your facility.

Common approaches include:

The goal is simple:

Prevent somebody from accidentally using equipment you have already decided should not be used.

A tiny note taped to the side of a device in a busy hallway may not accomplish that.

Neither does verbally telling one nurse on day shift and assuming everyone else will know.

Follow your facility's process and make the equipment status clear.

Documenting the Decision

Good documentation explains not only what you did but why.

Weak documentation:

Checked unit. Could not duplicate. Returned to service.

That leaves a lot unanswered.

What was reported?

What did you check?

How long did you test it?

Did you recreate the reported conditions?

Why were you comfortable returning it?

Better:

Reported intermittent power loss while operating on battery. Inspected battery and contacts, reviewed device logs, and operated unit on battery under normal load for 90 minutes. Flexed battery connection and transitioned repeatedly between AC and battery power with no shutdowns observed. Battery capacity test passed. Unable to duplicate reported failure. Functional and safety checks completed with no additional faults found. Unit returned to service.

Now another technician can understand what happened.

For equipment being removed from service:

Reported intermittent shutdown during patient monitoring. Confirmed unit unexpectedly powered off twice while operating on battery despite indicated charge above 60%. Removed from service. Battery and internal power system require further troubleshooting before return to clinical use.

That tells the story.

You do not need to write a novel.

You do need enough information for the next person to understand the complaint, the findings, the decision, and what happens next.

Common Mistakes

Returning Equipment Because It Powers On

Powering on proves one thing:

It powered on.

It does not prove that every function works, that the measurements are accurate, that the alarms work, that the battery is good, or that the reported failure is resolved.

Treating “Unable to Duplicate” as a Repair

Unable to duplicate is a finding.

It is not a repair.

Sometimes a device can reasonably be returned after thorough testing even when the original failure cannot be reproduced.

But your decision should be based on the risk and the testing performed, not simply the absence of a failure during five minutes on the bench.

Replacing a Part Without Proving the Problem Is Fixed

You replace the battery.

The device powers on.

Done?

Not necessarily.

If the original complaint was random shutdown, reproduce the conditions and verify that the device stays powered.

A replaced part is not the same thing as a verified repair.

Ignoring the Original Complaint

The clinical user's description may not use technical language.

That does not make it useless.

If someone tells you:

Every time we roll it into CT it shuts off.

Do not test it plugged into the wall for five minutes and conclude that nothing is wrong.

The complaint contains clues.

Looking Only at the Device

Sometimes the device is fine.

The problem may be:

Removing a device from service may be appropriate while you investigate, but do not assume the box itself is always the cause.

Being Afraid to Take Equipment Away

Clinical departments need equipment.

Sometimes they need it badly.

That pressure is real.

But equipment availability and equipment safety are two different problems.

If a device should not be used, your job is not to make the problem disappear by putting it back on the floor.

Your job is to communicate clearly, document the issue, help locate alternatives when appropriate, and get the equipment repaired.

Real-World Examples

Example 1: Infusion Pump Randomly Shuts Down

Complaint:

Pump shut off during an infusion.

You cannot duplicate the failure immediately.

This deserves more than a quick power-on test.

Check the battery, battery contacts, AC transition, logs, runtime, connectors, and any known failure conditions.

Because an unexpected shutdown can interrupt therapy, you should have reasonable confidence that the cause has been addressed or the device has been adequately evaluated before returning it to service.

Example 2: Patient Monitor Will Not Send Data to the EMR

The monitor displays ECG, SpO2, NIBP, and temperature correctly.

The network connection works, but data is not reaching the patient's chart.

The bedside monitoring function may still be operational.

The integration workflow is not.

Whether the monitor itself must be removed from service depends on how your hospital uses that data, whether a safe alternate workflow exists, and facility requirements.

The important thing is identifying the failure correctly.

This is not necessarily:

Patient monitor broken.

It may be:

Bedside monitoring functional. Automatic EMR data transfer unavailable.

Those are different problems.

Example 3: Hospital Bed Has a Cracked Plastic Cover

The crack is on a cosmetic cover.

No sharp edges are exposed.

The frame is intact.

The siderails, brakes, casters, controls, and patient-support functions operate normally.

The crack may require repair without necessarily making the entire bed unsafe.

Now move that same crack to a load-bearing siderail.

Completely different decision.

Example 4: Ventilator Alarm Speaker Is Intermittent

The ventilator delivers the expected pressure and volume during testing.

But the audible alarm cuts out intermittently.

The fact that ventilation works does not erase the alarm failure.

The alarm system is part of safe operation.

This device should not simply be returned because:

It ventilates fine.

Example 5: ECG Machine Cannot Print

The ECG machine acquires a correct tracing and can transmit it electronically.

The internal printer does not work.

Whether that device needs to be removed from service depends on the clinical workflow and whether printing is required at your facility.

The printer failure should still be documented and repaired.

But not every failed feature creates the same risk.

Example 6: Transport Monitor Battery Lasts Ten Minutes

The monitor works perfectly on AC power.

The battery lasts ten minutes.

If that monitor is used to transport patients between units, the battery failure directly affects its intended use.

Plugging it into the wall and saying:

Everything else works.

does not solve the problem.

Example 7: Device Passes After a Reported Shock

A staff member reports feeling a shock from a piece of equipment.

You run an electrical safety test and it passes.

That is useful information.

But it does not automatically explain the complaint.

Inspect the cord, plug, inlet, chassis, accessories, and conditions surrounding the event.

Consider whether the problem could be intermittent or environmental.

A passing test should be part of the investigation, not an excuse to ignore the original report.

A Useful Question Before Return to Service

Before closing the work order, ask yourself:

What did I actually prove?

Did you prove:

Or did you only prove:

It turned on while I was looking at it.

Those are not the same thing.

Final Thoughts for Biomeds

Removing equipment from service is not about being overly cautious with every scratch, error message, or complaint.

It is about understanding risk.

Not every broken feature makes a medical device unsafe.

Not every device that powers on is safe.

Not every problem that disappears on the bench is resolved.

And not every problem needs a complete diagnosis before you decide the equipment should stay away from patients.

A good biomed learns to separate:

What failed?

from:

What does that failure mean?

and then from:

What do I need to prove before this goes back into service?

That judgment develops with experience.

You will get better at recognizing which failures are minor, which ones affect workflow, which ones affect clinical performance, and which ones require an immediate stop.

When you are unsure, slow down.

Understand the complaint.

Understand the equipment.

Test the function that actually matters.

Document what you found.

And do not claim more than your testing proved.

Sometimes the smartest repair decision you make is deciding that the device is not ready to go back yet.

— Jake

Important Note

This page is an educational overview for biomedical equipment technicians, clinical engineers, and healthcare technology staff. Equipment-removal and return-to-service decisions should follow your facility policy, manufacturer service documentation, applicable requirements, equipment-specific procedures, and the scope of work established by your organization.

Related Biomed Basics

Back to Biomed Basics