What mechanical engineering taught me about clinical workflow
The first thing mechanical engineering teaches is that the user should never see the complexity.
A well-designed hinge moves smoothly. The tolerances, the material selection, the geometry that prevents failure under load. None of it is visible to the person opening the door. The experience is simplicity. The engineering is not.
Jony Ive described this as designing so that the complexity is resolved before the product reaches the user. The object feels inevitable. The work that made it inevitable is hidden.
Clinical workflow design follows the same principle, though it is rarely described this way.
A doctor opens a patient record. Prescribing history, dispensing data, clinical notes, and care pathway context are visible in one view. The integration across systems, the data normalisation, the logic that determines which information surfaces and which stays in the background. None of that is visible to the clinician. The experience is clarity. The architecture underneath is anything but simple.
When clinical systems expose their complexity to the user, the cost is not frustration. It is clinical risk. A prescribing interface that requires the doctor to navigate between systems to check dispensing status introduces a gap. Not a UX gap. A safety gap. The complexity that should have been resolved at the infrastructure layer has leaked through to the point of care.
Mechanical engineering has a term for this: tolerance stack-up. Every component in an assembly has a tolerance. Individually, each is acceptable. But tolerances compound. A system that passes inspection at every individual interface can fail at the system level, because the accumulated variation exceeds what the assembly can absorb.
Clinical platforms have the same problem. The booking system works. The prescribing system works. The dispensing integration works. But the tolerances between them compound. A patient falls through a gap that no individual system created but the combination makes possible.
The discipline that prevents this is the same whether the material is aluminium or software: design for the system, not the component. Resolve the complexity before it reaches the user. Test the assembly, not just the parts.
Mechanical engineers learn this through physical failure. A part that passes individual inspection but fails in the assembly teaches the lesson permanently. In clinical systems, the equivalent lesson is quieter. The failure is a patient who falls between systems. A prescription that was written but never dispensed. A follow-up that was scheduled but never triggered.
The reduction of complexity. The coherence across the whole system. The restraint in what you expose to the user. These are not software principles borrowed from hardware. They are engineering principles that apply everywhere the user is more important than the system.
Healthcare is one of those places.