Care Policy AskMe
Answers a carer's policy or procedure question at the point of care from your own approved library — quoting the section, naming the document and version, and saying so when the answer isn't there.
It searches only your approved policy set, quotes the section it found with document name and version, and refuses rather than guessing when nothing matches.
Where this works well
The slow, invisible problem this surfaces is the time floor staff lose hunting for an answer that already exists in writing. The retention rule, the medication SOP step, the restrictive-practices definition — it is all in a policy somewhere, but "somewhere" is a 200-page PDF on a shared drive, and the carer asking at 2am does not have the time or the access path to find the line. This pattern earns its keep when you have a defined, approved policy set and staff who keep asking the same procedural questions that have a documented answer.
It works best for the carer or enrolled nurse on the floor and for the clinical governance lead who currently fields those questions by hand. The grounding is the point: every answer names the document, the version and the section it quoted, so the person can read the source rather than trust a paraphrase. When the answer genuinely is not in the corpus — the demo's leave-entitlements question, which belongs to the Aged Care Award [MA000018] and not the clinical policy set — it says so and routes the person on, rather than improvising.
Where it works badly
It will confidently quote a superseded document if that is what you indexed. If your medication SOP was reissued last year but the old version is still in the corpus, the tool will cite the old one in good faith, with a version number that looks authoritative. The failure mode is not hallucination — it is faithful retrieval of the wrong source. This is most dangerous on exactly the things that changed recently: the strengthened Aged Care Quality Standards came into force on 1 November 2025, so any pre-November handbook in the corpus is now wrong about which Standard governs what.
It is also poor at questions that require judgment rather than lookup. "Is this a reportable incident under SIRS?" has a real answer that depends on the facts of the case — the tool can quote the eight reportable-incident types and the Priority 1 (24-hour) versus Priority 2 (30-day) timeframes, but it cannot decide whether a given event meets the threshold. The honest test: if the answer to your staff's question is a sentence already written in an approved document, this fits; if it requires weighing facts against a definition, it does not.
What it doesn't do — and shouldn't
It surfaces what your documents say. It does not decide what the policy should be, reconcile two documents that disagree, or interpret a grey area — and it should not. Deciding whether an incident is reportable, whether a restrictive practice was justified, or which of two conflicting procedures governs is clinical-governance judgment, and it carries regulatory consequence under the Aged Care Act and the SIRS. Those decisions stay with the clinical governance lead and the document owners.
The human-in-the-loop boundary here is deliberate. The tool's job is to put the right cited passage in front of a person quickly; the person's job is to apply it. We keep that line because a wrong call on a reportable incident or a medication procedure is a resident-safety and compliance matter, not a search-quality matter. The refusal state — "Not in your approved documents" — is part of that discipline: it would rather hand the question back than manufacture an answer.
What your data has to look like for this to work
You need a single, current, correctly-versioned set of approved documents — clinical governance policy, medication and care SOPs, the current strengthened-Standards material — with one authoritative copy of each, not three near-identical drafts across SharePoint folders. Each document needs a real version marker and date so a citation means something, and superseded versions need to be retired from the corpus rather than left to be retrieved. Scanned image-only PDFs need their text extracted before they can be indexed at all.
Most providers have only some of this in good shape. The policies exist, but they live in overlapping folders, the version history is in the filename if anywhere, and nobody is certain which copy is current. Fixing that — getting to one approved, current, machine-readable set with a re-index step tied to every policy update — is usually the real first job, and it is a matter of how documents are captured and governed, not of buying a tool. It is the work we help with, and it is typically bigger and more valuable than the AI layer that sits on top.
Could it give a carer the wrong answer and have them act on it?
The safeguard is that it only answers from passages it actually retrieved from your approved corpus, and it shows the document name, version and section it quoted so the carer can read the source for themselves. When nothing in the corpus matches, it returns 'Not in your approved documents' rather than inferring — the demo does exactly this with a leave-entitlements question, routing it to People & Culture instead of guessing. The carer is reading a citation, not taking an instruction; the person stays responsible for the action.
What if our policies are messy — old versions, duplicates, scanned PDFs?
Then the honest answer is that the tool will faithfully quote whichever version you indexed, including a superseded one. It cannot tell that a 2019 medication SOP has been replaced unless the current one is the document in the corpus. Getting the corpus to a single, current, correctly-versioned set is usually the real first job, and it is the part we help with before the AI layer earns its keep.
Does this replace our clinical governance lead or policy owner?
No. It surfaces what your documents already say; it does not decide what the policy should be, resolve a conflict between two documents, or interpret a grey area. Those judgments stay with your clinical governance lead and document owners. What it recaptures is the time they and floor staff lose hunting through PDFs for a line that is already written down.
How current does the policy library have to be?
As current as you keep it — the tool answers from whatever you last indexed, so a re-index has to follow every policy update or version bump. This matters acutely in aged care right now: the strengthened Aged Care Quality Standards changed on 1 November 2025 under the new Aged Care Act 2024, so any handbook predating that is out of date. We set the ingest so that a new approved version replaces the old one rather than sitting alongside it.
Where does our data go — does resident or staff information leave our control?
The corpus and the questions stay inside your tenancy; the demo is built on private RAG over your own SharePoint, not a public chatbot. Policy documents generally are not resident health records, but questions can contain identifying detail, so access is scoped to staff and the exchange is kept within your environment. We confirm the data-handling boundary with you before anything is indexed.
What happens when staff ask things the policies genuinely don't cover?
It refuses explicitly and logs the question as a gap for the document owner, and it can route the person to the right team — People & Culture for an award question, for instance. That refusal-and-log behaviour is deliberate: the unanswered questions become a list of where your policy set is thin, which is useful rather than hidden.
Estimated build: 3–4 weeks. Most of it is template work we've already done.
Fixed scope, fixed price, fixed dates.
The cost band reflects the engagement shape, not a per-feature line item. We work on fixed scope, fixed price, fixed dates — see the services catalogue for what falls inside each band.
Considering this for your org?
The honest place to start is a bite-sized first piece — one contained change, low risk. Tell us where it hurts; we’ll play it back, scope it, and show you what’s possible.