Every biomed shop faces the same recurring question: this device is down again, do we fix it or replace it? The wrong answer in either direction costs money — replacing serviceable equipment wastes capital, and repairing a dying device burns labor and creates clinical downtime.
Below is the framework most well-run HTM programs converge on, and the evidence you need to make the recommendation stick with the people holding the budget.
The four factors that decide it
Repair-vs-replace is rarely a single-number decision. Four factors carry most of the weight, and any one of them can override the others.
- Repair cost as a share of replacement cost — the classic threshold, but the weakest factor on its own
- Remaining useful life — age against the expected service life for that device class
- Parts and service availability — an end-of-support device with no parts pipeline is already replaced, whether or not anyone has said so
- Failure frequency and clinical downtime risk — a device that fails three times a year in a critical area costs far more than its repair invoices show
The cost threshold, used properly
A common rule of thumb: if a single repair exceeds roughly 50% of replacement cost, replacement usually wins. If cumulative repairs over the past 12 months exceed replacement cost, replacement definitely wins.
The trap is using the invoice number alone. Real repair cost includes tech labor, the rental or loaner during downtime, the clinical time lost, and the probability of a repeat failure. A $4,000 repair on a $12,000 device looks reasonable until you add the third occurrence and the rental.
Compare total cost of ownership over the remaining life, not the repair invoice against the purchase price.
Age and expected service life
Expected service life varies widely by device class, and your own fleet history is a better guide than any published table. Pull the service history for the same model across your inventory and look at when repair frequency starts climbing — that inflection point is your real end-of-life marker.
A device in the last 20% of its expected life rarely justifies a major repair. A device in the first half usually does, unless parts are gone.
Parts and support are often the actual answer
When a manufacturer declares end-of-support, the decision is largely made. Sourcing becomes harvested boards and gray-market parts, turnaround gets unpredictable, and the software or service keys may simply not be obtainable.
- Confirm the OEM support status before quoting any significant repair
- Check third-party ISO parts availability and typical lead time for that model
- Factor in whether service documentation and calibration software are still accessible to you
- Weigh cybersecurity: unsupported operating systems on networked devices are increasingly a replace trigger on their own
Downtime and clinical risk
A device in a low-acuity area that fails twice a year is an annoyance. The same failure rate on an OR or ICU device is a patient safety and throughput problem, and the replacement math changes accordingly.
Rank by where the device lives and what happens when it is unavailable. Facilities that skip this step end up with reliable equipment in storage and unreliable equipment in critical care.
Documenting the recommendation
Finance approves replacements that come with evidence. The recommendation that gets funded includes: complete service history for that asset, cumulative repair spend over 12-24 months, downtime hours, parts availability status, and the quoted repair against a quoted replacement.
This is where good service reports pay for themselves. If your technicians have been writing "repaired, returned to service" with no detail, you have no case to make. If every visit recorded the failure mode, parts used, labor time, and outcome, the case builds itself.
- Every service call on the asset, with dates and failure modes
- Cumulative parts and labor cost over the review window
- Total downtime hours and the clinical area affected
- OEM support status and parts lead time
- Repair quote vs replacement quote, with remaining useful life on each
A quick decision guide
When you need an answer in two minutes rather than two days:
- Repair under 30% of replacement, device under half its expected life, parts available — repair
- Repair over 50% of replacement, or device in its final years — replace
- Parts or OEM support unavailable — replace, regardless of the repair quote
- Third failure in 12 months in a critical area — replace, and escalate the fleet
- Anything in between — build the documented case above and let the numbers decide