CPD PDSA guide
A worked PDSA cycle for closing the loop on unactioned results and correspondence
A worked PDSA cycle for closing the loop on unactioned results and correspondence
This PDSA supports the process of marking patient results and correspondence as 'notified'. Not notifying patients of their results poses a patient safety risk. Not marking results as notified when, in fact, wastes a doctors' time, as they have to close that safety loop to confirm the patient is aware of the result.
Test results and correspondence are often discussed with the patient but not marked as actioned in the record. The clinical decision was made, but the record does not reflect this. The next doctor to see that patient cannot tell whether the result was reviewed, whether it was discussed, or whether anything is still outstanding. Closing the loop, and making that loop visible to the whole team, is the goal of this PDSA.
GPs can meet a large share of their annual 50-hour CPD requirement without leaving the practice. Hours count across educational activities (EA), reviewing performance (RP) and measuring outcomes (MO). Nurses claim through AHPRA. Practice managers count it toward AAPM certification.
A documented PDSA is ready-made evidence of continuous quality improvement against the RACGP Standards for general practices. Presented to your local Primary Health Network, a completed cycle may also qualify as a Practice Incentives Program Quality Improvement (PIP QI) activity.
This PDSA stands on its own. It also underpins every recall, reminder and screening PDSA you run, because an unactioned result or correspondence is potentially a reminder that has not been started. If your practice is also working through cervical screening, lung cancer screening, Prolia monitoring or any other recall-driven topic, run this one first or alongside. A clean results workflow is a foundation that the others rely on.
Every practice accumulates results and correspondence that were seen but have yet to be marked as discussed with the patient. Some were actioned in a consultation and simply not marked off as notified. Some were seen by a duty doctor; the patient has since returned, and the loop was never closed. A smaller number were flagged urgent, and the notes do not make clear whether the matter was resolved. These are the most important to address.
In our practice, we found roughly 60 results flagged as urgent review since 2000 that had not been marked off, around 30 of them in the previous 12 months. Non-urgent reviews have run into the thousands since 2000, many involving doctors who have left the practice. Correspondence marked for discussion or no action added thousands more. None of this was necessarily a sign of poor care. In almost all cases, the result had been dealt with. The record just did not say so.
Failure to follow up test results is one of the more common sources of patient-safety incidents and medico-legal complaints in general practice.
The problem is measurable. Your clinical software can count unactioned results by urgency and by doctor. It is team-based. Clearing a backlog and keeping it clear requires GPs, nurses and receptionists to work to the same rules. And the change is small and repeatable: agree on a clearance protocol, record action results at each visit, and remeasure.
Telling independent doctors to mark off results doesn't clear a backlog of thousands, and it doesn't change practice in the long term. A PDSA sets a baseline, agrees a protocol, makes the problem visible, measures the reduction over a quarter, and embeds the habit. It also generates the documentation that counts for CPD, accreditation and PIP QI.
Education consisted of practice meetings and reviews of Automed's educational resources (or whichever vendor is used). Reviewing performance consists of logged activities, including time taken during consultations. Measuring outcomes involves downloading and analysing the reports.
These are the hours our practice recorded. Results handling generates far more reviewing-performance time than a typical PDSA because clearing a backlog of this size is part of the review.
| Category | Focus | Hours |
|---|---|---|
| Educational activities (EA) | Practice meetings and reviewing the Automed education resources | 4 |
| Reviewing performance (RP) | Logged clearance activities and time taken during consultations | 23 |
| Measuring outcomes (MO) | Downloading and analysing the reports | 3 |
| Total | 30 |
Claim the time you spend. Your figures will differ from ours and a smaller backlog will generate fewer reviewing-performance hours.
| Category | Focus | Hours |
|---|---|---|
| Educational activities (EA) | ||
| Reviewing performance (RP) | ||
| Measuring outcomes (MO) | ||
| Total |
The RACGP requires 50 CPD hours a year: a minimum of 12.5 hours of educational activities and a minimum of 25 hours across reviewing performance and measuring outcomes, with at least 5 in each. The RACGP classifies PDSAs under measuring outcomes. Submit as a GP-led activity, individually or as a group or practice. ACRRM members and members of other colleges self-report.
Two kinds of boxes appear throughout.
Shows how our practice ran this cycle, written from our experience. Use it as a reference, not a script. Your numbers and software will differ.
Blank space for your own figures, dates and notes. Fill these in as you go so the completed guide doubles as your CPD and accreditation evidence.
If our worked examples are relevant to your practice, you are welcome to paste into the box.
A brief description of the activity, and the key steps undertaken. Complete this for your own CPD submission.
What were the key steps undertaken?
Results and correspondence that are not marked as actioned create a clinical risk. The record does not show whether the loop is closed, so the next doctor cannot tell whether to raise the result or leave it. We test whether a shared clearance protocol, plus actioning results at each visit, reduces the backlog and keeps it down.
Audit the practice database for unactioned results. Categorise by urgency: urgent review, non-urgent review, and no action. Agree a clearance protocol at a team meeting. Track the time each doctor spends clearing their list. Re-measure monthly and record the reduction over one quarter. Note any patient-safety issue found along the way and act on it immediately. Data to collect: baseline count of unactioned results by category; time spent per doctor on clearance; a monthly count to track the reduction; any safety incident identified during the audit.
Describe what you will do and what you will measure. Consider: how you will run the search in your software, who runs it, what the baseline looks like by category, and how often you will re-measure.
Set specific, measurable targets. For example: clear all urgent reviews from the last 12 months within one month; reduce non-urgent reviews by an agreed percentage over the quarter.
We ran the cycle across the first half of 2026, starting with a planning meeting and holding regular learning-together meetings through to the closing report.
We ran a database search for patients with results not marked as actioned, and the actions arising were discussed at the team meeting on 19 February 2026. As a follow-up, a recall report for each doctor's patients was pulled out of Cubiko and provided to them individually. Each doctor was then asked to give the practice manager a time log of the hours they spent clearing their list. That is what produced the clearance figures below. If you do not run Cubiko, any clinical software report that lists outstanding recalls and results per doctor will do the same job.
| Activity | Date |
|---|---|
| Planning meeting | 5 January 2026 |
| Learning together meeting | 22 January 2026 |
| Learning together meeting | 19 February 2026 |
| Learning together meeting | 14 May 2026 |
| Learning together meeting | 16 June 2026 |
| Closing data report | 25 June 2026 |
Record your own meeting and extraction dates as you go.
Run the database search at baseline and again each month. Record the count in each category so you can see the trend, not just the start and end points.
| Measure | Baseline | Month 1 | Month 2 | Close |
|---|---|---|---|---|
| Urgent reviews unactioned | ||||
| Non-urgent reviews unactioned | ||||
| Correspondence: no action | ||||
| Clearance time logged (hours) | ||||
| Safety issues identified |
Our baseline confirmed the scale described earlier: roughly 60 urgent reviews outstanding since 2000 (about 30 in the previous year) and non-urgent reviews in the thousands. During the term of the PDSA, each GP reviewed results and correspondence. Each doctor logged the time they spent clearing their list from the recall report.
Time logged per doctor, from the recall report (doctors deidentified):
| Doctor | Time (hours) |
|---|---|
| Doctor 1 | 9 |
| Doctor 2 | 8 |
| Doctor 3 | 7 |
| Doctor 4 | 8 |
| Doctor 5 | 3 |
| Doctor 6 | 3 |
| Doctor 7 | 5 |
| Doctor 8 | 5 |
| Doctor 9 | 3 |
| Doctor 10 | 5 |
| Doctor 11 | 7 |
| Doctor 12 | 5 |
| Doctor 13 | 5 |
| Doctor 14 | 3 |
| Total (14 doctors) | 76 |
| Average | 5.4 |
In addition to the clearance work above, actioning results and correspondence inside consultations was estimated at one minute per consultation. Over the four months, that is an average of over 18 hours each. Collating both activities, that is over 23 hours in total per GP. The urgent backlog was cleared. The non-urgent and no-action backlog fell sharply once the print-and-mark shortcut was adopted (see below).
Making sure results and reports are actioned on each visit. The search function for unactioned results and reports is easy to use, but time-consuming. Marking off no actions other than in a consultation is probably just busy work and achieves little, but it does improve the database. Following up urgent and non-urgent reviews is important even outside a consultation. We agreed to address results and correspondence for patients of GPs who have left the practice during consultations. It is possible to tick off multiple investigations, but correspondence is a one-by-one process. We learned a new shortcut for marking off correspondence that has to be actioned one at a time.
Making sure results, reports and reminders are actioned on each visit.
What worked, what did not, what surprised you, and what still needs attention.
Un-notified results and correspondence previously got out of hand in a practice that prides itself on quality processes. It is best to always mark them off at every consultation.
What will you do differently as a result of this cycle?
At every visit, all doctors review the patient's results and correspondence and mark them off as discussed, review the action list, and make sure outstanding items are attended to or the patient is rebooked with their usual GP. The clearing happens while the patient is in attendance, face to face or by telehealth. For older correspondence marked as no action more than a year old, it may be appropriate to do this in bulk rather than one by one.
What will you standardise across the practice? Consider workflow rules, software shortcuts, who clears departed doctors' lists, and how new results are kept clear.
Much of the backlog can only be closed properly with the patient present. For the remaining backlog, we use a simple rule. Where a result is clearly urgent, or the notes do not show that an urgent matter was resolved, contact the patient. Where the result does not look important and the patient is not with you, mark it off and add a new action to discuss it at the next visit. That keeps the record honest without generating a phone call for every item. Where you have called or recalled a patient, flag it so the team can see.
These steps describe how our practice does this in Best Practice with Automed. The menus differ by software; the workflow does not. Screenshots are deliberately not included: the originals contained patient information. Where a step refers to a screen, the description tells you what to look for.
Clearing no-action items one by one is slow, and most of the backlog sits in correspondence. The quickest safe method: set a date range, choose print, select 'mark as given', then cancel the print job. The items are marked off without printing. Selecting multiple patients with shift-click is no longer available in current versions.
Where you have called or recalled a patient to action a result, note it by double-clicking the patient's name. A red phone or letter symbol appears to the left of the name so the team can see the patient has been contacted.
Investigations can be ticked off several at a time. Correspondence cannot: it has to be actioned one item at a time, which is why most of the backlog sits there. To move through it quickly, use the up and down arrow keys in the Correspondence In tab and press Alt + K, or use the mouse scroll function. It does not fix the underlying problem, but it speeds up clearing a backlog.
If your practice uses automated messaging (we use Automed), reviewing a result or correspondence triggers an SMS to the patient. Understanding this behaviour is part of the educational component of the PDSA. The detail is set out in the background section.
Education comes from practice meetings and reviewing your vendor's results and messaging training. Reviewing performance comes from the audit, the logged clearance time and the monthly counts. Measuring outcomes comes from the PDSA cycle itself. Submit via the RACGP CPD portal as a GP-led activity, individually or as a group.
Record time as you go. The RACGP CPD year runs 1 January to 31 December. If the project spans two years, allocate hours to the year the activity was completed.
List the doctors who took part. Each claims their own hours.
| Doctor | College and CPD number |
|---|---|
Go to your vendor's training. Whatever messaging and results system your practice runs, that vendor's material on results messaging, clinical reminders and recalls is what you want. It is the main source of the educational activity (EA) hours in this PDSA, so log the time your team spends on it and claim it. This is easily the most overlooked EA time in a results PDSA: people do the training and forget to record it.
The modules below are the ones our practice uses. They are specific to Automed and are included as a worked example of what to look for. If you run a different system, find the equivalent from your vendor rather than working from these.
The first module is the one that matters most for this cycle.
The remaining modules cover the wider messaging system and also count toward EA hours where your team reviews them.
Medius Global produces worked PDSA guides with real practice examples across clinical, preventive and administrative topics. Each includes data-collection templates, CPD claiming guidance and the reference material to run the cycle. See the full set at mediusglobal.com.au.
This section forms part of the educational-activity (EA) hours for the PDSA. It covers why unclosed results are a risk, and how automated patient messaging fits into a results workflow.
A result that is actioned but not documented looks identical, in the record, to a result that was never discussed. A doctor opening the file later cannot tell the two apart. They either re-raise a matter already resolved, which wastes time and worries the patient, or they assume it was handled and move on. Continuity of care depends on the record showing the decision, not just the doctor remembering it. Marking a result or correspondence as discussed with the patient closes the loop and makes the process visible to the next doctor.
This is a good time to be reminded of the automated processes. Automated messaging turns the act of reviewing a result into patient notification. Automed generates an SMS unless the patient has withdrawn consent. In our practice, it works as follows. Confirm the equivalent behaviour in your own system before relying on it.
These settings are not fixed. Automed preferences can be adjusted, but at the practice level, not by an individual user. An individual doctor cannot change the timing rules, the exclusions or what the patient sees. If a setting does not suit the way your practice works, raise it with your practice manager so it can be changed once, for everyone.
Encourage patients to opt in to messaging. It closes more loops automatically and reduces the manual follow-up load on nurses.
The single change that matters is marking results off during the consultation, even when the result belongs to another doctor. A backlog is cleared once. It stays clear only if every visit closes the loops that surface in it.
Under the Medical Board of Australia registration standard, medical practitioners complete 50 hours of CPD each year: a minimum of 12.5 hours of educational activities, and a minimum of 25 hours across reviewing performance and measuring outcomes, with at least 5 hours in each. Practitioners also record reflection on their practice. A group PDSA lets a practice meet a large part of this together.