-
When this checklist applies
-
Step 1: Confirm where the device was actually made
-
Step 2: Verify the specs that actually matter on a syringe pump
-
Step 3: Check the whole data chain on diagnostic equipment
-
Step 4: For laparoscopes, check reprocessing compatibility before optics
-
Step 5: Train staff before the equipment arrives — including how to read vital signs on the new UI
-
Step 6: Get service and parts coverage in writing
-
Step 7: Lock the specs into the purchase contract
-
Common mistakes I see
When this checklist applies
If you're the person who has to sign off on infusion pumps, patient monitors, surgical scopes, or any other piece of critical care gear before it rolls onto a floor — this is for you.
I'm a quality and brand compliance manager at a medical equipment distributor. I review every device batch before it reaches a hospital customer — roughly 200 line items a year. In 2024 I rejected about 14% of first deliveries, most of them over spec labels that didn't match the actual device on the pallet.
Here's the checklist I run, in order, every time. Seven steps. The fifth one is the one most sourcing teams skip.
Step 1: Confirm where the device was actually made
Labels lie. Not usually maliciously — but secondary suppliers, repackagers, and contract manufacturing blur the paper trail fast.
The first thing I do is match the "manufacturer of record" on the device label to the establishment registration in the FDA's public database. If the carton says the device was made at Smiths Medical's Plymouth, MN site, that address should map to a registered establishment — and the product code should map to a matching 510(k) clearance. If either lookup fails, stop. Ask for the traceability file before you accept delivery.
This takes about ten minutes online. It has saved me two full shipments in four years.
Step 2: Verify the specs that actually matter on a syringe pump
The brochure for a Smiths Medical syringe pump (or any competitor's, honestly) lists a dozen specs. Four of them matter for clinical safety:
- Occlusion alarm threshold — adjustable, and documented in mmHg. Not "adjustable" in marketing copy. Actual numbers in the test report.
- Flow accuracy at low rates — ±2% or better at the rates your unit actually uses (neonatal and cardiac units need different verification than general med-surg).
- Drug library size and update cadence — how many entries, and how fast can your pharmacy team push a new protocol.
- Battery runtime under load — not idle. Loaded. Running.
Ask for the test report, not the datasheet. The datasheet is marketing. The test report is what your biomedical team will demand anyway.
Here's where my experience diverges from the conventional wisdom. Everything I read said premium syringe pumps are always worth the delta. In practice, on a 40-pump ICU order we ran last year, our mid-tier option produced fewer occlusion false alarms than the flagship — because the alarm threshold was tunable to our line pressures and the flagship's default settings were tuned for a different use case. Same specs on paper. Different behavior in our unit.
Match the spec to your actual use case, not the marketing tier.
Step 3: Check the whole data chain on diagnostic equipment
A Holter monitor isn't a device. It's a data pipeline: capsule, download dock, analysis software, and the report your cardiologist actually reads. Most sourcing teams check the recorder and stop there.
Run a sample. Start a recording, let it run an hour with a known signal source, download it, and confirm the artifact rejection doesn't eat real ectopic beats. I've seen monitors with beautiful hardware and analysis software that flagged every baseline wander as a PVC. That device is worthless no matter how good the capsule is.
I'm not a cardiologist, so I can't tell you what your patient population needs. What I can tell you is how to verify the data path end to end before you buy 30 of them.
Step 4: For laparoscopes, check reprocessing compatibility before optics
Optical clarity is easy — put it on a benchtop next to a reference scope and compare. That's the test everyone does.
The spec that actually bites you is sterilization compatibility. If your facility runs autoclave cycles and the scope is ETO-only, you've bought a shelf ornament. Confirm the reprocessing method against your SPD's actual capability, not the vendor's "compatible with common methods" line.
I've never fully understood why some vendors still bury autoclave compatibility in a footnote. Best guess: legacy of the transition from fully reusable to semi-disposable optical designs. Either way, footnote or not, verify it in writing.
Step 5: Train staff before the equipment arrives — including how to read vital signs on the new UI
This is the step that gets skipped. Every new patient monitor is a new user interface. How to read vital signs on the old unit — color coding, alarm hierarchy, trend window position — does not transfer to the new one.
I've watched experienced ICU nurses take 30 seconds longer to spot a falling SpO2 because the new monitor moved the trend window from the right side of the screen to the bottom. Thirty seconds. In a critical care unit, that's a long time.
Build a 90-minute hands-on session in the actual unit, with the actual device, before go-live. Not a webinar. Not a quick-start card. Hands on. Have your clinical educator walk through how to read vital signs on the new interface, then have staff read them back under time pressure. If they hesitate, you haven't finished training.
Step 6: Get service and parts coverage in writing
Response-time SLA. Loaner unit availability. Parts lead time. The "global service network" slide is fine, but I want a commitment letter with names and phone numbers attached to it. Especially for infusion pumps. A down pump takes a bed offline, and a bed offline is a capacity problem the CFO will ask you about.
I once had two hours to decide between two pump vendors because the PO window closed at 5 p.m. Normally I'd have run a full demo cycle. There was no time. Went with the incumbent based on service history alone. It worked out — but I don't recommend flying that way if you can avoid it.
Step 7: Lock the specs into the purchase contract
Every rejection I've processed in four years traced back to a spec that wasn't written into the contract. Occlusion threshold, flow accuracy, reprocessing method, alarm volume, durometer on any soft-goods component. Put it in the PO.
If the vendor pushes back, that's data too.
Example: in 2023 we rejected a batch of 120 IV sets where the tubing durometer measured roughly 15% harder than spec — normal tolerance was ±5. The vendor claimed it was "within industry standard." We sent the batch back; they redid it at their cost. But it cost us a six-week launch delay on a new line, and the project sponsor remembers. Since then, every contract includes durometer requirements. Every one.
Common mistakes I see
- Buying on unit price. Total cost of ownership includes setup, consumables, training hours, and downtime. The lowest quote is almost never the lowest total.
- Skipping the demo. Always demo with your own staff, on your own floor, with your own consumables.
- Ignoring consumable compatibility. A pump may "take standard syringes," but the drug library was validated on a specific brand, and the tubing was qualified at a specific durometer.
- Accepting vendor claims without substantiation. Per FTC advertising guidelines, product performance claims must be truthful and substantiated with evidence. If a vendor says "zero failure rate" or "best-in-class occlusion detection," ask for the test report. No report, no claim.
- Not asking for the previous generation's service records. The best predictor of the next model's field reliability is the last one's. Ask for the failure data.
The fundamentals of this checklist haven't changed much in a decade — verify the manufacturer, verify the spec, verify the service, verify the training. But the execution has transformed. Public FDA databases, downloadable test reports, and remote demo capability mean you can do in an hour what used to take a site visit. Use that. The checklist gets shorter. The scrutiny doesn't.