Tube or Robot for Hospital Specimens: A Route Decision Guide

Which system fits an older hospital?

For most older hospitals, the right answer depends on the route more than the brand. A pneumatic tube suits fixed, repeated runs between departments that already have stations at both ends. A delivery robot suits routes that change, cross floors the tubes never reached, or need a controlled handoff at the far end.

BayCare Health System's June 2026 expansion at its Winter Haven hospitals shows the second case in practice, and it shows the two systems can share one building. The sections below cover routing, specimen quality, chain of custody, elevators and staff time, the four tests a facility leader should run before deciding.

What did BayCare add in June 2026?

As of June 2026, Robotics Tomorrow reported that BayCare added two Relay robots, Stork and Miles, at Winter Haven Hospital and Winter Haven Women's Hospital. Both carry specimens between departments.

Stork moves specimens from the first floor clinic to the second floor lab at Women's Hospital. Miles serves main hospital departments that lack pneumatic tubes, carrying pharmacy orders and supplies as well as specimens. Assigning a robot to the spots the tubes miss is a useful model for any facility weighing where a robot should sit.

Relay's blog projects more than 500 deliveries over a month and about 150 nursing hours saved in that period. Those are vendor projections, not measured results. Relay's case study also reports that Stork completed 18 deliveries in her first two days, another vendor reported figure. Read both as launch stage claims until BayCare publishes its own counts.

How do routing and tube layouts differ?

A tube network is fixed by design. Its routes follow the stations and carrier paths that were installed, so a new destination usually means new construction. Older buildings can leave a tube serving the inpatient wards but not the outpatient wing down the corridor.

A robot route lives in software and a floor map. A new destination can be added by remapping rather than rewiring, though the robot still needs clear corridors and doors it can open or pass. Routing flexibility is real, but it is only as good as the floor plan the robot can actually travel.

Lab validation guidance published in Clinical Chemistry and Laboratory Medicine holds that each tube system must be evaluated individually. Two hospitals with matching floor plans can still differ in door hardware, elevator control and staffing, so the same caution applies to a robot route.

Does specimen quality hold up on each route?

Published tube studies are reassuring on quality. In one hematology study, white cell counts showed no meaningful difference between tube transported and hand delivered samples. A study of bloodstained cerebrospinal fluid found tube transport had no clinically relevant effect on hemolysis compared with manual transport.

A 2006 emergency department study found tube delivery shortened turnaround against human couriers. Hemoglobin turnaround came in at 33 minutes (range 4 to 230) and potassium at 64 minutes (range 34 to 208), with no loss of sample quality. Hemolysis affected 7 of 121 tube samples (5.79%) and 20 of 200 courier samples (10%), and the study reported no significant difference.

We found no published study testing robot transport against tubes or couriers for sample quality. That gap is why a paired test on your floors matters. Validation guidance calls for paired testing of tube routes, comparing tube or hand delivery for each analyte and route, and a robot route deserves the same design. A paired test should record the following.

• One analyte and one route per pair of runs
• Time from draw to lab receipt for each leg
• Hemolysis or other quality flags counted by group
• Every rejected or redrawn sample, with its cause

Who signs for a specimen when no one walks it?

Chain of custody runs from collection to final disposition. Each handler signs and dates the record, and the chain should show no gaps. Legal or forensic specimens also get tamper evident seals. A phlebotomy study guide describes these steps, but it is teaching material, not a standard, so compliance staff should confirm the requirements that apply to your lab.

We found no source comparing handoff signatures, custody forms or tamper evidence for tube against robot transport. Public evidence on this point is thin, and any claim that one method is more defensible than the other runs ahead of the data.

A robot can timestamp each leg, but only if the facility designs the handoff, such as a scan or a signature at each end. A tube run needs the same signature at both stations, and the procedure should say so in writing.

How do elevators and staff time change the math?

Elevators are where robot routes often stall. Robots need elevator integration, and one reported approach has the robot call the elevator, wait for an empty car and direct it to its floor by radio. A University of Virginia simulation, reported by ScienceDaily in 2000, noted that robots add load to elevator systems partly because their programming keeps them from sharing a car with a person.

In an older hospital with a few shared cars, that rule can cost more time than the robot's travel speed. Tubes avoid the elevator question when the stations sit on the same floor, which is one reason they suit stacked departments.

Staff time is the other line item. The same simulation recorded staff time spent loading and unloading specimens and estimated savings of up to $600,000 over five years against human couriers. That is an older model, not a benchmark for today's hospitals. Neither method removes the loading step, so measure it on the route you are testing.

Where should a facility leader start?

Start with the routes, not the vendor. Map the specimen flows that repeat most, mark where tubes stop, and time the elevator waits on each leg. A short, fixed run usually belongs to a tube. A route that changes, reaches an uncovered wing or depends on a busy elevator is the robot's ground.

Service Robot Co. is an OEM neutral integrator. We compare robots across manufacturers, then lease, deploy, integrate, train and service each unit through a nationwide US engineer network. Our offer is a three year lease with one fixed lease quote after a site assessment, so the quote reflects the floor the robot will actually cover.

Four questions belong on the site assessment checklist for any route.

• Which specimen legs repeat most each shift, and which have no tube station?
• How long does each elevator leg wait for an empty car at peak hours?
• Which doors can the robot open or pass without staff help?
• Where does each handoff get signed, and who keeps the record?