Clinical Documentation Guidelines in Australia: The Seven Checks a Reviewer Runs
By The Kitt Team
Most practices audit on completion: did every appointment get a note. That test measures almost nothing. This post takes clinical documentation obligations as read and moves straight to the audit.
What the guidelines actually require
There is no single clinical documentation guideline for all allied health in Australia. Requirements come from three directions, each tested by different people.
| Layer | Standard | Who tests it |
|---|---|---|
| Professional conduct | AHPRA shared code (registered professions) or ESSA standards (accredited exercise physiologists) | Regulators and professional associations |
| Privacy | Privacy Act 1988 (Cth) and Australian Privacy Principles | Applies to every health service provider; no small business exemption for health |
| Funder | Medicare, workers compensation, NDIS, private health insurers, DVA | Funder auditors and scheme reviewers |
The conduct standard requires accurate, factual and legible records covering history, findings, investigations and information given to the client, comprehensible to another practitioner, made at the time or as soon as practicable. The funder layer sits on top, not instead. Workcover progress reports covers WorkCover specifically.
Funders look first. The APA 2025 Workforce Census found 64 per cent of physiotherapists work in private practice, up from 57 per cent in 2023. That is a large share with no hospital team and funder audits as the most likely external test. Safe Work Australia’s National Return to Work Survey found 55.6 per cent of workers compensation claims are musculoskeletal, making the reviewer of allied health records something other than a hypothetical.
Having a note and being able to defend it are different tests
Completion is binary: did every appointment produce a note. Easy to measure, easy to pass, says nothing about whether those notes would survive review. Reconstruction asks whether someone reading the record cold can follow what you found, concluded, did and why. That requires reasoning written in the record, not carried in your head.
The specific failure is a complete, timely note that still fails reconstruction: the link between findings and treatment was never written down. The reasoning stayed in the room. A reviewer has to guess. The record has to say the care was good. According to the APA 2025 Workforce Census, 34 per cent of practice owners named regulatory and funding complexity as a business threat, up from 22 per cent.
The seven checks a reviewer actually runs
When a funder or regulator reads your records, the assessment runs along seven lines.
| The check | What fails it |
|---|---|
| Contemporaneous | Written at time of encounter or as soon as practicable. Fails: batch notes written a fortnight later, backdated entries. |
| Attributable and dated | Clear who wrote it and when. Fails: shared logins, unresolvable initials, undated entries. |
| Legible to a stranger | Another clinician can read and follow the note cold. Fails: private shorthand, undefined abbreviations, author-only notes. |
| Reasoning visible | Shows why this treatment for this presentation, not just what was done. Fails: action-only entries (“reviewed and treated” with no link from findings to plan). |
| Consent recorded | Record shows what client was told and that they agreed. Fails: consent assumed because client attended. |
| Amendments transparent | Corrections made as visible additions, not overwrites. Fails: silently edited entries (unexplained change is worse than the original error). |
| Retrievable and secure | Record can be produced when asked; access controlled. Fails: cannot be produced in usable form, access not controlled. |
None of these checks are exotic. They describe ordinary practice. The audit tests whether they held when notes were written late and thin. Sinsky et al. (Annals of Internal Medicine, 2016) found US physicians spend roughly two hours on EHR and desk work per hour of direct care, plus one to two hours after hours. That is US physicians, not Australian allied health. Documentation load gets compressed exactly when the day is hardest. The craft of writing notes well day to day is covered in progress notes.
Run it on your own last ten notes
Pull your last ten notes at random, not the ones you are proud of. Include one from a Friday afternoon and one from a day you ran late. The standard is set by your worst note, not your best.
Score each against the seven checks. Mark each present or absent. Read one note as a stranger would, and count how many times you add context from memory. Each is a gap the record does not cover.
Look at the pattern across all ten. One thin note is a bad day. The same check failing across seven of ten is a system problem, and system problems are the only kind you can fix.
The check that fails most often is reasoning visible. It is the only one no template fixes.
Closing the gap without adding admin time
The fix is almost never more discipline. The gap is between the encounter and the record: a note written at the time carries the reasoning, and a note written Sunday night can only approximate it.
That is the problem kitt addresses. kitt drafts the note, letter and treatment plan from the consultation itself so the record is made while the reasoning is current. kitt’s Client Contextual Memory surfaces prior findings, progression and goals before each session, so the record reads as a continuous narrative, not ten disconnected entries. Completed notes can be pushed to Cliniko or Nookal if you use either.
kitt drafts, you decide. Every note, letter and plan comes back for you to review, edit and approve; the clinical reasoning stays yours. kitt documentation supports clinicians’ AHPRA record-keeping obligations. On the Companion side, the plan goes to your client; kitt checks between visits, and that data comes back into the Clinician record so your next note starts from what actually happened. The Professional plan is A$69 per clinician per month with Companion included, there is a free tier, and a one-month free trial with no card.
Try kitt free for a month, no credit card required.
Frequently asked questions
What are the clinical documentation guidelines in Australia?
There is no single guideline. Requirements stack from your profession’s conduct standard, privacy law covering every health service provider, and the funder’s documentation rules. All three apply at once and may be tested independently.
How long do I have to keep clinical records?
Retention periods are set by state and territory legislation. The period generally runs from the date of the last entry. Records for a client who was a child when treated are kept longer. The clinical documentation reference guide covers the framework, but check your own state’s requirements. There is no single national figure.
What makes a clinical note fail an audit?
Most often it is not a missing note. It is a note that exists but does not show the reasoning connecting findings to treatment, or one written long enough after the encounter that clinical detail is gone. The record is present; it just does not say what happened or why.
Can I use AI to write my clinical notes?
It can draft the note from the consultation. You review, edit and approve it, and the clinical reasoning and responsibility for the record stay with you. The obligation does not transfer to the software.
The takeaway
- Requirements stack from three directions (professional conduct standard, privacy law, funder rules), tested by different people.
- Completion is the test you set yourself. Reconstruction is the test applied from outside, by someone who was not in the room.
- The seven checks above are what a reviewer runs. Try them on your own last ten notes this week.
- The check that fails most often is visible clinical reasoning. No template fixes it.
Your records say what happened when you are not there to explain it.