The short answer
In the delivery robot vs medication courier question, the robot wins on repeated, point-to-point runs of sealed supplies between fixed departments, such as pharmacy to a ward. It does not fit urgent doses, controlled substances that need a witnessed handoff, or routes where elevators and crowded halls make timing unreliable. In most hospitals the answer is a robot working beside the courier and nursing staff, not a robot that removes them.
Key takeaways
- A robot fits repeated, fixed-route runs of sealed supplies, while a courier still covers urgent and controlled medication handoffs.
- Nurses walk several kilometers per shift, so trips a robot takes over can return time to patient care.
- Shared elevators and crowded halls slowed robot deliveries in the studies reviewed here.
- Turnover is high enough that a missing courier or runner shift is hard to cover, but a robot never replaces the whole team.
- Lease payments are quoted only after a free site assessment of your routes, elevators and staffing.
Delivery robot vs medication courier: when does a robot make sense at a hospital?
The American Hospital Association counts 6,120 hospitals in the United States in its Fast Facts on U.S. Hospitals, 2024 edition, using 2022 survey data. That is a large and varied group, from small rural sites to academic medical centers, so the decision turns on the building and its routes more than on the hospital's size or name.
A delivery robot makes sense when the same few routes repeat on every shift, such as pharmacy to a ward, supply room to a unit, or sterile processing to a department. The loads are sealed, the trips are point-to-point, and nobody needs to ride along. That is the pattern a robot handles best.
It does not make sense when the work is mostly urgent, when items need a witnessed handoff, when the route depends on crowded shared elevators, or when trips are rare enough that existing staff already cover them. Count the trips before you look at any machine.
What does a medication courier do during a hospital shift?
In most hospitals the medication courier is a pharmacy technician, a pharmacy runner, or a support staff member who carries sealed doses, orders and supplies from pharmacy to nursing units and brings empty or pending items back. Some units fill the role with nurses or unit clerks when the courier is short. The Bureau of Labor Statistics counts about 490,400 pharmacy technicians employed in 2024, and hospitals employ 17% of them, according to its Occupational Outlook Handbook.
The work arrives in bursts. Morning medication rounds, discharge orders and evening restocking create peaks, and the trips follow the same corridors each time. Couriers also handle exceptions: a wrong item, a missing dose, or a stat order that cannot wait for the next scheduled run.
That mix is why the job is hard to plan. A scheduled run can be automated. The exception still needs a person who knows the unit, the chart and the pharmacy.
Which parts of the medication run can a delivery robot take over?
A delivery robot can take scheduled runs: restocking carts, moving non-urgent sealed doses in locked compartments, returning empty totes, and carrying linens, meal trays and supplies between fixed points. It works best on routes it can drive many times a shift without a person holding a door at every stop, which means a clear path and an elevator it can call and ride.
It cannot take the parts of the job that need judgment or a signature. Controlled substances usually need a witnessed handoff and a logged count. A robot cannot confirm identity at a nursing station or check a patient at the bedside. Stat orders need someone who can walk straight to the unit and change the plan. IV compounding stays with pharmacy staff.
The honest split is that the robot takes the repeated trips and the courier keeps the exceptions. Write that split into the route plan before you choose a route.
- Scheduled restock run
Delivery Robot
Runs the same route on a fixed schedule and logs each trip
Medication Courier
Walks the route, but gets pulled into exceptions partway
- Stat or urgent order
Delivery Robot
Not suited, because it cannot change the plan mid-route
Medication Courier
Walks straight to the unit and reroutes
- Controlled substance handoff
Delivery Robot
Cannot give a witnessed or signed bedside handoff
Medication Courier
Does the count, the signature and the check
- Crowded shared elevators
Delivery Robot
Slower and less reliable when elevators are full, per published studies
Medication Courier
Waits, then adjusts the order of stops
- Night shift
Delivery Robot
Runs without a person on the floor when the route is clear and someone is on call
Medication Courier
Needs someone on shift, and sick call-outs cut coverage
- Consistency across trips
Delivery Robot
Same route and handling every time
Medication Courier
Varies with workload, but brings judgment the robot lacks
How much nursing time goes to walking?
Walking is the clearest place to look for time a robot can return. In a time-and-motion study of 36 hospitals that tracked staff with RFID tags (Hendrich et al.), median walking distance was 4.8 km per 10-hour day shift and 3.5 km per night shift. The study also recorded time by task and location, which is the level of detail a site review should copy.
An older ICU study (Nahk and Frolova, 2009) reported a mean of 10.7 km per 12-hour shift, rising to 12.3 km when nurses cared for two patients and falling to 8.9 km for one. That sample was small and dated, so use it to show that intensive units walk the most, not as a benchmark. A 2025 study from Japan (Nakashima et al.) found weekends changed walking distance only on long day shifts.
A simulation of a 30-bed medical-surgical unit estimated walking could fall by up to 67.9% with decentralized support spaces. That is a simulated estimate, not a measured result, and it suggests that moving supplies closer to the bedside can matter as much as adding a robot.
How high are nurse turnover and vacancy rates?
Trade press coverage of the 2026 national RN retention report, covering calendar 2025, puts national RN turnover at 17.6%, up from 16.4%, across 527 hospitals in 40 states. Behavioral health RN turnover was 22.5%, emergency 20.7%, telemetry 19.5% and pediatrics 13.4%. The spread between units is wide, so a staffing gap rarely hits every floor the same way.
The 2025 edition of that retention report, covering calendar 2024, found RN turnover of 16.4% and RN vacancy of 9.6%, with 41.8% of hospitals reporting a vacancy rate of 10% or more. The 2026 vacancy rate was 8.6%, but the report changed its calculation method, so the year-over-year comparison is not clean. Read it as direction, not as a precise change.
On injury, the Bureau of Labor Statistics snapshot for 2020 counts a rate of 1,023.8 nonfatal injury and illness cases with days away from work per 10,000 full-time equivalent nursing assistants, against 390.6 for registered nurses. Private hospitals reported 6.0 cases per 100 full-time workers in 2015, according to a Bureau of Labor Statistics Monthly Labor Review article from 2017. These figures are old, and the data does not break cases down by cause, so do not count on a robot to change injury rates. Measure how many lifts and carries your own staff make on each route first.
What do the medication delivery studies show?
The strongest evidence is a six-month, single-center controlled study at a tertiary hospital in China, published in a peer-reviewed journal. Robot delivery cut delivery time by 32% to 36% compared with manual delivery, with p below 0.001, and nurse and patient satisfaction were higher in the robot group. It is one hospital, so treat it as a strong signal rather than a forecast for your building.
A feasibility study at a tertiary hospital with shared elevators found autonomous medication delivery worked, but performance dropped as elevator congestion rose, and frequent robot users reported more time-related pressure. That finding matters more than the headline result. A route that depends on elevators shared with visitors is the first thing to test.
A medication delivery pilot at a Veterans Affairs site shows what process change alone can do. It was lean and Six Sigma work, not autonomous. Average delivery time to nursing units fell from 34 minutes 43 seconds to 27 minutes 39 seconds. Process fixes come first, and a robot on top of a clean process is easier to evaluate, because you can separate the two effects.
How does the team work with a robot?
The team never goes to zero. The robot takes the repeated runs, while the same staff still load, check, sign and handle exceptions. In practice that means fewer trips for the courier or nurse, not an empty pharmacy window or an unstaffed floor. Schedule a person on call for the robot's route, the exception queue and the elevator fallback.
Acceptance depends on fit with existing habits. In the China study, nurse satisfaction was higher in the robot group. A vendor-reported case study at a Texas hospital found most nurses said the robot gave them more time with discharged patients, while some struggled in crowded halls during morning rush. Train the people who load the robot, the people who meet it on each floor, and the people who handle its exceptions, not only the pharmacy team.
We are the integrator on this. We pick the robot for the site, lease it through financing, deploy and integrate it, train the staff and service it nationwide. Our first deliverable is the route map and the staffing plan, not the machine.
What can a delivery robot not do in a hospital?
It cannot take a stat order, a witnessed controlled substance handoff or a bedside check. It cannot clear a crowded hall or an elevator that is full at the wrong time. When these hit, the robot waits, reroutes or stops, and a person has to step in. Any plan that removes that person will fail on the first bad morning.
Downtime is a real planning issue. A robot stopped in a hall takes its route out of service until someone clears it or a technician repairs it. The staffing plan needs a fallback path and a service response before go-live. Our service includes on-site dispatch and remote triage, but a fallback still has to exist on the floor.
Results vary by building. The evidence here comes from one peer-reviewed controlled study, one feasibility study, a lean process pilot and vendor-reported cases. Treat the numbers as a range and test your own routes before you commit.
How do you get started with a delivery robot at a hospital?
Start with a route count. List each fixed run that repeats on a shift, its start and end point, the elevator it uses, and who does it now. Routes with many repeats and few exceptions are the candidates. Our free site assessment maps the route, the elevator access and the staffing plan before any lease is discussed.
Lease the robot through financing, with the lease quoted after the site assessment, once the route and staffing are known. Start with one unit on one or two routes, measure trips per shift and delivery time, and expand only when the numbers hold. We train the staff who load, check and handle exceptions, and we service the robot nationwide, with spare unit coverage for downtime.
Ask any integrator, including us, for the route data and the fallback plan in writing. A good first pilot is one unit, one route and one measure: trips per shift.
Frequently asked questions
Want a delivery robot at your hospital?
Tell us about your site. We will recommend the robot, quote the lease, and keep it serviced.
Sources
- American Hospital Association, Fast Facts on U.S. Hospitals 2024
- NSI national health care retention report, 2025 edition
- Becker's Hospital Review, 2026 coverage of nurse turnover
- Bureau of Labor Statistics, Occupational Outlook Handbook, Pharmacy Technicians
- Bureau of Labor Statistics, Monthly Labor Review, hospital workers injuries (2017)
- Bureau of Labor Statistics, nursing assistants injury snapshot (2020)
- Peer-reviewed controlled study, tertiary hospital in China, robot medication delivery
- Medication delivery lean and Six Sigma pilot, Veterans Affairs site (repository record)