Publication: Potential of Assistive Robots in Clinical Nursing: An Observational Study of Nurses' Transportation Tasks in Rural Clinics of Bavaria, Germany
Authors: D. Sommer, J. Kasbauer, D. Jakob, S. Schmidt, F. Wahl
Published in: Nursing Reports 14 (1), pp. 267–286 (2024)
In brief
Transportation tasks in nursing are common, often overlooked – and they come straight out of time at the bedside. That is where this observational study begins. Instead of adopting robot vendors' claims about what is supposedly automatable, we measured reality: which goods are transported in a hospital, how often, for how long and in what way?
Over seven observation days in two rural clinics, we recorded 1,830 transports using a standardised app-based form. The result is one of the few detailed data foundations on nursing transportation tasks – and thus the basis for developing and deploying assistive robots to actual need.
Why transportation tasks are a blind spot in nursing
When people picture nursing work, they think of care, observation, attention – the direct work with another human being. What rarely comes to mind is the logistical scaffolding that makes this work possible in the first place: consumables have to come from the store room, medication has to travel from the ward to the pharmacy and back, samples to the lab, meals to the bedside. Each of these movements looks trivial on its own. Taken together, however, they tie up precisely the skilled staff who were trained to care for patients – not to cover distances.
This blind spot has a simple reason: transportation tasks do not appear as a line item in any job description. They are spread across the day, each one lasts only a moment, and none of them seems important enough to document. That is exactly why a solid data foundation has been missing until now. Discussions about deploying service robotics in hospitals have largely rested on assumptions rather than measured need. Our study closes that gap by making the invisible countable.
What is transported
Transports fall clearly into a few categories:
- Non-medical supplies: 27.05 % (n = 495)
- Medical supplies: 17.32 % (n = 317)
- Pharmacotherapy: 14.10 % (n = 258)
- Other, such as meals and drinks: 12.68 % (n = 232)
Notably, most transports actually take under a minute – so these are many short, repetitive trips rather than a few large ones. And 77.15 % of all transports are done by hand. It is precisely this small-scale, manual logistics that adds up over a shift to substantial walking distances and physical strain.
This structure is more revealing than it first appears. The fact that the bulk of transports concentrate in a handful of categories means automation need not founder on countless special cases but can target a few clearly defined task types. And the fact that the individual trips are so short shifts the focus: the problem is not the length of any single transport but their sheer frequency. A task that takes under a minute each time yet recurs dozens of times per shift is exactly the pattern in which repetitive interruptions create the real burden – not least because every interruption tears apart focused work at the bedside.
Where robots pay off
The economically most interesting finding concerns meal delivery: meals are the costliest transport item at roughly EUR 9,596.16 per year in the observed clinics. A low-cost transport robot would pay for itself within a year through meal transport alone.
That is the real point of the study: robots do not have to do everything to be worthwhile. They have to take on the right, frequently recurring tasks. The data shows which ones those are.
It matters what such a calculation can and cannot do. It does not replace a full economic assessment that also accounts for acquisition, maintenance, infrastructure and organisational change. But it provides a clear starting point: anyone wanting to know where automation makes sense first should begin where recurring tasks meet high bound costs. Meal delivery meets both criteria – it is predictable, regular and scheduled in fixed time windows. That makes it not only the most expensive but also one of the most readily automatable transport items.
What the data means for device design
Measured need tells you not only whether a robot makes sense but also how it must be built. Concrete requirements can be derived from the distribution of transport categories: pharmacotherapy calls for secured, verifiably locked containers; meal delivery for solutions that preserve temperature and hygiene; non-medical supplies for robust, easily loaded transport containers. A device that tries to meet all of these requirements equally becomes expensive and complicated. A device tailored to the most frequent and most costly tasks can stay lean, reliable and affordable.
This is exactly where evidence-based development pays off: it prevents the build-up of features that are barely used in practice and concentrates the investment on what genuinely relieves the daily work in the hospital.
Why this study matters to us
Good robotics does not start with the robot but with the requirements analysis. This study provides the solid evidence base our platform builds on: it tells us which transports hospOS should prioritise, which requirements to place on the devices (such as lockable or cooled containers) and where the economic lever is greatest. Instead of "robots for robots' sake", at Athegus we take the reverse path: first the measured need, then the automation.
This path is harder than reaching for the next product promise, but it is the only one that holds. A hospital investing in automation must be able to trust that the decision rests on solid numbers and not on marketing assumptions. That trust is built precisely by measuring first and building second. The data presented here is one building block of that stance – and an example of how we understand technology: as a means to give skilled staff their time back, not as an end in itself.