Writing from the pathway.
Perioperative operations, clinical AI architecture, and the information problems behind preventable failures in surgical care.
All articles
-
Inside the PAINITE pathway: four arms, one journey
PAINITE (Perioperative Assessment & Intelligent Navigation, Integration & Triage Efficiency) is a pathway proposed at Royal North Shore Hospital. It adds four arms to the existing pre-op pathway, a voicebot, a snifferbot, a rule-based perioperative harness and day-prior calls, and is staged to start with an ophthalmology voicebot pilot and process evaluation.
-
The platform didn't know what the hospital knew
An RNSH audit of 1,532 patients who missed their digital pre-op questionnaire found 21 invited while admitted, 9 after death and 23 after surgery, while 4.9% had been transferred and 5.6% had declined or been removed. Every fact was in the medical record. None stopped the invitation, because the questionnaire platform was not listening to the record's events.
-
Four days: urgent patients get the least time
Among patients at Royal North Shore Hospital who did not complete their digital pre-operative questionnaire in 2025, the median gap between invitation and surgery was 4 days for Category A, 49 for Category B and 149 for Category C. The most urgent patients, with the most to gain from early assessment, had the least time. Invitation timing is a system setting, and it can be changed.
-
What is perioperative AI?
Perioperative AI is the use of artificial intelligence to support the care of surgical patients before, during and after an operation. Most of it today sits before surgery, where the work is gathering a patient's history, identifying risk and getting them ready. It isn't a robot anaesthetist; it's closer to infrastructure that gets the right information to the right clinician in time.
-
Who a digital-only pre-op pathway leaves behind
In an RNSH audit of 1,532 patients who missed their digital pre-op questionnaire, 64% were over 65, 14.5% preferred a language other than English, and ophthalmology was the largest group at 385 patients. Incompletion rose with a specialty's average age. A web form asks a lot of these patients; a voice channel, fewer questions and a human fallback ask far less.
-
Clinical safety rules should not be model output
Safety-critical clinical screening should be implemented as deterministic, versioned rules rather than model output, because clinical governance requires reproducibility rather than accuracy alone. A rule that fires whenever its conditions are met can be tested, audited and re-run months later to produce the same answer — a property a probabilistic system cannot offer.
-
The medication history is the weakest link
Pre-operative medication histories are usually assembled from patient recall and a GP summary that may be months old. The drugs patients are least able to name — anticoagulants and SGLT2 inhibitors described as 'a blood thinner' or 'something for sugar' — are precisely the ones whose management must be planned days ahead, which is why they cause cancellations.
-
What data sovereignty actually requires
Data sovereignty means patient information stays within Australian jurisdiction across its whole lifecycle — storage, processing, backup and any third-party service it passes through. Claiming an Australian region for the primary database is the easy part. The paths that matter are backups, subprocessors, support access, and, for AI systems, what is sent to a model provider.
-
Why clinics fill with the wrong patients
Pre-admission clinics fill with low-risk patients because slots are allocated in the order bookings arrive rather than by screened risk. The result is a clinic of largely well patients while the frail patient with poor functional capacity gets a short phone call. The constraint is allocation, not capability, so adding clinic capacity does not fix it.
-
Cancellations are an information problem
Most day-of-surgery cancellations are caused by information that existed weeks before the operation but reached the anaesthetist too late to act on. Because the limiting factor is timing rather than clinical thoroughness, cancellation rates respond to moving assessment to the point of booking far more than to assessing more carefully at the clinic.
See it working.
Book a 30-minute demo with a real pre-admission workflow.