Digitise the referral, retire the fax, connect the dots between services, share the info and send it to the MHR. Use what we already have, better. That's the low-hanging fruit, and we can pick it now.
Every now and again, another story surfaces in the media about a referral lost or left sitting in a pile. It happens so often that Consultmed has built a whole page to collect them: Axe the Fax.
But the past few weeks have been exceptional. Three cases, in two states, have landed in quick succession.
I should declare my interest up front: I’m not neutral. I’m a paediatrician and the founder of Consultmed. I’m pro-digital and anti-fax, and I’ve developed our referral platform with colleagues who are GPs, nurses and allied health professionals, because everyone who sends, receives or acts on a referral has felt this problem firsthand.
Read what follows with that in mind – that there are a group of Australian clinicians which has spent the better part of five years designing and developing a solution to help address this problem.
I also don’t know the specifics of these cases beyond what has been reported. But at face value, the common threads are hard to ignore.
What’s been reported
Albury Wodonga Private Hospital. Kenneth Toll, 62, died in 2019, three days after a double knee replacement. As reported by Health Services Daily, last week NSW Coroner Sally McLaughlin found he developed tachycardia that needed timely cardiology review, but there was no evidence a referral was ever made. She described repeated failures to record important information and a complete lack of documentation of his postoperative care.
An infant in Victoria. A baby of eight weeks was killed by his father after repeated hospital presentations. At the first, a nosebleed at the Royal Children’s Hospital in 2021, inflicted injury was considered but the injury was put down to chafing. State Coroner Judge Liberty Sanger found multiple missed opportunities, cited barriers to information sharing, and called for steps towards a statewide EMR.
Mildura Base Public Hospital. As far as I can tell, this one hasn’t gone to a coroner; it was reported to Safer Care Victoria as a sentinel event, as reported by HSD. Around 150 GP referrals, many for suspected cancer, sat untriaged in cardboard boxes, some for more than a year. One patient has reportedly since died and another is terminal, though no one yet knows whether earlier triage would have changed that. Hospital insiders say there was no central way of tracking referrals at all.
The common thread
In each case the question is the same: did anyone know?
A referral that should have been made wasn’t, and nothing flagged its absence. What one hospital visit learned about a patient that didn’t reach the next. Referrals arrived, and for more than a year nobody knew they were there.
These don’t read to me as failures of clinical skill. They read as failures of visibility. A referral is a clinical handover, and a handover nobody can see isn’t really a handover.
As a paediatrician, I know how hard it is to spot inflicted injury at a single presentation. The pattern usually only shows across several, which is exactly why the information has to follow the child.
To be clear, I am not saying a referral system would have prevented this infant’s tragic death. I don’t know enough about the case to say that.
But I am absolutely saying that if the numerous clinicians who saw this baby could have seen, easily, any and all referrals already in place to other specialties and to child safeguarding services, it would have raised their suspicion further.
That is what transparency and data sharing gives you.
Why a referral?
It’s fair to ask why a referral specifically rather than records in general. The answer is that a good referral is a beautiful summary document. In a page, it sets out the presenting complaint, the relevant history, current medications and recent presentations, written by one clinician for another, with a clear question attached.
In practice, that often makes it more useful than a discharge summary. Depending on who wrote it, and increasingly that includes AI scribes, a discharge summary can be long-winded, heavy on detail and sometimes light on what matters.
A referral is short by design. It says why this patient needs someone else’s attention, which is exactly what the next clinician needs to know.
Start local, then join up
I understand why the Victorian coroner suggested a statewide EMR, and sure, it would help. But implementing statewide EMRs is slow and expensive – take the Single Digital Patient Record in NSW Health as a case in practice.
Mildura committed to digitising its records in 2023 and its inpatient notes are still on paper; upgrading the Loddon Mallee region alone has been costed at around $70 million.
Connected referrals are a far simpler and cheaper answer, and they don’t have to start at state level. They can start at home. For any health service leaders reading, here is a quick playbook.
- Start inside your own walls. Digitise referrals into your health service and connect every clinic and service, not just the willing few or ‘pilots’. And ban the fax, for goodness’ sake. Nobody wants it, and the few who still send faxes are enabled by the health services that keep accepting them.
- Connect with your neighbours. Link your service with the ones next to you, so a referral can be shared or forwarded rather than lost.
- Join up the region. If you are up for it, connect across your region: public and private, GPs and community providers, with the PHNs involved.
Do that, and every referral has a sender, a receiver and a status, with a copy in the patient’s My Health Record.
This is already happening, organically, and it shows what is possible when leadership sees the problem and acts.
In southeast Melbourne, Consultmed is joining up Alfred Health (as part of the newly formed Bayside Health), Monash Health and the Gippsland Health Alliance. Every non-admitted service across these three regions, which together serve more than two million Victorians, will be connected, so referrals can be received, shared and forwarded, and waitlists managed, across all of them.
GPs and other primary care referrers stay in the loop, with updates landing back in their practice software. South Eastern Melbourne PHN is engaged and has funded the system for its palliative care referrals, and Gippsland PHN is engaged too.
In NSW, the Sydney Children’s Hospitals Network and South Western Sydney LHD are connected, Murrumbidgee and Southern NSW LHDs are now joining, and so is the statewide NSW Refugee Health Service. Paediatric and child health referrals are now largely connected across the state.
In each case, someone in leadership saw the problem and chose to fix it.
I want to be clear: this is not about one system to rule them all. We have worked hard to make Consultmed interoperable, to meet Australian Digital Health Agency standards, to connect with the major practice management systems and to integrate with whichever secure messaging solutions services have already invested in. Nobody should have to rip out what already works.
But secure messaging on its own is not referral management.
Too often the “secure message” lands in a hospital email inbox, gets printed and ends up in a box for filing. That doesn’t connect anything. What matters is what happens after the referral arrives: who owns it, where it is up to, and whether anyone can see it, including the patient. It’s their referral.
None of this needs new technology, and it can grow organically. But it happens faster, and more evenly, with state help: setting the standard, backing it with funding and expecting services to meet it.
Related
When does it become negligence?
Coroners don’t decide civil liability, and I’m not suggesting anyone here has been found negligent. But the question is worth asking of health services, not just individual clinicians.
Under section 48 of Victoria’s Wrongs Act 1958 and section 5B of the NSW Civil Liability Act 2002, a failure to take precautions is negligent only if the risk was foreseeable, was not insignificant, and a reasonable person in the same position would have taken those precautions. In deciding that, courts weigh the probability and likely seriousness of the harm against the burden of the precaution and the social utility of the activity.
On referrals, that test is getting harder to pass. The risk isn’t just foreseeable; coroners and safety reviews keep describing it. The harm can be serious. And the precaution is available now and, next to most of what health services spend money on, modest.
There are of course defences, and I don’t want to overstate this. And a patient still has to prove the failure caused the harm.
Even so, I struggle to see a court accepting that running referral intake out of cardboard boxes is widely accepted competent practice.
So, my answer, falling to the trusted pub test, is this: it becomes negligence when a health service knows the risk, knows the fix, can reasonably afford it, and still hasn’t acted. After this month, I think a lot of services are close to that line.
Start with the simple things
No health service should be able to receive a referral and not know it has it. No patient should wait 17 months for an appointment that should have happened within 30 days because a box was forgotten.
In an age when every conversation in health seems to be about AI, it’s worth remembering that some of the biggest gains are far simpler.
Most referrals still depend on manual handling: printed, faxed, scanned, sorted into piles and pigeonholes, keyed in again and chased by phone.
Every one of those steps is a chance for a referral to go missing.
So, take the manual handling out. Digitise the referral, retire the fax, connect the dots between services, share the information and send it to My Health Record. Use what we already have, better. That’s the low-hanging fruit, and we can pick it now.
Then, by all means, let AI do its magic. It will work far better on referrals that arrive digitally, complete and connected than on a fax sitting in a cardboard box.
Vikram Palit is a paediatric respiratory physician founder and CEO of Consultmed, and associate professor of paediatrics and child health at Western Sydney University.
This article was first published on Professor Palit’s LinkedIn feed. Read the original here.