The Smiths Medical Cleo 90 infusion set is the least expensive infusion set our hospital uses — and its quote was not the lowest we received in 2024. A competing supplier undercut it by $0.42 per set. At our annual volume of roughly 84,000 administrations, that priced out to $35,280 in apparent savings. It wasn't real. That set produced a 1.4-point higher unplanned replacement rate in our six-month pilot: about 1,176 extra nursing interventions a year at a documented cost of $35 each. That is $41,160 in costs that never show up on a purchase order. Net result: the lowest quote was $5,880 a year more expensive. Period.
I manage clinical supply contracts at a 520-bed regional medical center in New England. Since 2017, my team and I have tracked $7.4 million in annual supply spend line by line. This article is based on that 2024 pilot, four vendor quotes, and eight years of watching low bids age badly.
Here is the part I don't say often enough: I used to be the person who bought the low bid. When I first moved into procurement, I assumed unit price was the entire game. It took two failed vendor switches to change my mind. The first was infusion sets; the second involved SpO2 sensors, which I will get to in a minute.
The Four-Line Cost of an Infusion Set
If you only look at the quote, a set is a set. A clinician will tell you otherwise. A cost controller needs a framework that captures what clinicians already know. Here is the one I built after the 2024 test:
- Purchase price: the number on the purchase order.
- Replacement rate: sets that fail, get changed early, or trigger an occlusion alarm that causes a wasted visit.
- Clinical labour: calls to the on-call nurse, troubleshooting time, and the documentation that follows.
- Consequential cost: delayed therapy, line access issues, and patient confidence in home infusion.
Most vendor comparisons stop at line one. The entire difference between the competitor and the Smiths Medical Cleo 90 lived in the other three lines. In a CADD ambulatory pump system, the administration set must match the pump's occlusion and flow logic. When it doesn't, the pump generates alarms that waste everyone's time. Between May and November 2024, we recorded 42,000 administrations: occlusion-related line restarts ran at 2.6 per 100 with the low-quote set and 1.2 per 100 with the Cleo 90. After annualizing that 1.4-point gap, I got 1,176 extra events at $35 per event. That number is conservative. It does not count patient anxiety or the two cases that required an extra hospital visit.
Smiths Medical, Southington CT: Why a Ship-To Address Made My Scorecard
During the same review, my Value Analysis Committee asked all four suppliers where their product support actually sat. The Smiths Medical Cleo 90 packaging points to a Smiths Medical Southington CT operation. I am not claiming that geography makes a tube clinically superior. Geography matters to me for a dumber reason: accountability.
When a product is tied to a US-based operation with a named establishment and a support team that answers the phone, the contract has teeth. In our 2023 delivery audit, 23 of 24 Smiths Medical purchase orders arrived on time; the low-quote vendor managed 15 of 24, and the late ones averaged 11 days behind. In home infusion, an 11-day delay means patients wait, clinicians reschedule, and our cost tracking system fills with exception notes. That is the quiet cost that never makes the marketing brochure.
What Is SpO2? And Where My Cheapest-Sensor Plan Backfired
SpO2 stands for peripheral capillary oxygen saturation. It is the non-invasive estimate of oxygen saturation that a pulse oximeter calculates by shining light through a fingertip, toe, or earlobe. Clinicians treat it as a vital sign. Procurement treats it as a recurring sensor purchase. That is where I made my second mistake.
SpO2 is an estimate, not a direct measurement. It can drift from the true arterial value when perfusion is poor, and it has real limits on moving or sweaty patients. A vendor that does not tell you those limits is hiding the risk.
In 2019, I switched our hospital to a cheaper compatible SpO2 sensor and saved roughly 32% on sensor spend for two months. Then the complaints started. The sensors lost signal when patients moved, and the adhesive did not survive a night of sweat. Nurses began using tape to hold the sensors in place. Tape degrades the optical signal, so the monitor alarmed more often, which prompted more nursing calls. We burned through the savings in eight weeks and reversed the change. Total documented cost: over $13,000, not counting the trust we lost with the ICU nursing staff. There may be good third-party sensors on the market; after that experience, the burden of proof sits with them, not with my budget.
Where This Procurement Logic Stops: Pacemakers and Mass Spectrometers
I want to be clear that the cost-control framework I used for the Cleo 90 does not apply to every medical purchase. It is the right tool for high-volume disposables with a measurable failure rate. It is the wrong tool for at least two categories I manage:
Pacemakers. When our cardiology team selects a pacemaker, clinical indication and patient anatomy drive the decision. My role is not to pick the least expensive implant. My role is to negotiate the contract around the physician-led choice. Applying the low-bid logic to a permanent implant would be both dangerous and disrespectful to the clinical process.
Mass spectrometers. In capital lab equipment, the sticker price of a mass spectrometer is almost beside the point. The dominant cost is the service agreement, uptime, and the downtime penalty if the analyser fails mid-run. When we bought ours, we chose on the maintenance proposal, not the acquisition price.
When the Cleo 90 Recommendation Does Not Apply
Our Cleo 90 conclusion came from a CADD-pump, home-infusion, adult population. If your organization runs a different pump fleet, has low volume, or uses sets for gravity infusions, this data does not transfer automatically. The savings came from set-and-pump integration, not from magic plastic. If your volume is below 20,000 administrations a year, the $5,880 net difference is small enough that service and training might matter more.
None of this tells you that Smiths Medical products never fail. They do. What the 2024 data tells you is that price quotes measure only the first dollar. The expensive part of an infusion program lives in the failures, the delayed visits, and the alarms that send a nurse to a patient's home at 11 p.m. That is where the cheapest set in the hospital actually got its title.
