The Bigger Opportunity for AI in Healthcare Is the Whole Patient Journey

I spend a lot of my time thinking about the operational side of general practice, and lately, about where AI fits into it.

My view is simple. The future of healthcare in Australia will not be defined by how quickly AI can write a clinical note, generate a referral or answer a doctor's question.

Those things are useful. But they are also the smallest version of what AI could become.

The much bigger opportunity is to use AI to understand and support the whole patient journey.

Patients experience one life

Healthcare is still largely organised around individual interactions. A patient sees a GP. Then perhaps a specialist. Then a physiotherapist, psychologist, pharmacist or another allied health professional.

Each clinician is focused, understandably, on the problem in front of them. That focus is part of what makes good clinical care possible.

But it also creates a gap.

Patients do not experience healthcare as a series of isolated consultations. They experience one life.

The sleep problem, the blood pressure reading, the stress at work and the caring load at home all belong to the same person. In practice, they often get handled in separate rooms, by separate people, at separate times.

An occupational therapist's lens

I originally trained as an occupational therapist, and that training has stayed with me.

OTs are taught to look beyond the immediate diagnosis. What is happening at home? At work? What matters to this person? What are they trying to get back to doing? What barriers are stopping them? Who else is involved in their care?

Those questions aren't a replacement for clinical judgement. They sit alongside it. They help explain why a plan works for one person and quietly fails for another.

I think that same lens should shape the next generation of healthcare AI.

What connecting the dots could look like

Imagine an AI system sitting across a patient's primary care journey. Not replacing clinicians, but connecting the dots between them.

It knows that a hypothetical patient has repeatedly discussed poor sleep, rising blood pressure, weight gain and work stress. It recognises they have not completed the pathology requested three months ago. It notices their preventative health screening is overdue.

It also knows they have seen a psychologist previously, have stopped attending physiotherapy, and are caring for an elderly parent at home.

None of those details is dramatic on its own. Together, they tell a story that no single consultation is likely to capture.

Instead of waiting for the next ten-minute appointment to deal with whichever issue is most urgent that day, AI could help identify what needs attention earlier.

It could help the practice coordinate care. It could prompt the right conversation, engage the patient between appointments, and bring the relevant information to the right clinician at the right time.

That is very different from asking AI to answer: "What should this doctor do in this consultation?"

The better question is: "What does this person need across their entire healthcare journey?"

Put plainly, the shift is from AI helping a clinician manage one appointment → to AI helping a practice support one person's whole healthcare experience.

Why general practice should lead

Primary care is uniquely positioned to lead this shift. It already sits at the centre of a patient's healthcare ecosystem.

General practice knows the patient over years, not episodes. It sees the referrals go out and, ideally, the results come back. It is often the one place where the whole picture can come together.

AI could make that longitudinal relationship dramatically more powerful. Not by changing what good general practice is, but by giving it better support to do what it already tries to do.

As someone who owns and operates two general practices, this is the layer that interests me most. The clinical encounter matters enormously. But so much of what determines outcomes happens in the spaces between appointments.

Of course, this future requires strong safeguards. Patient consent, privacy, data security, clinical governance and transparency cannot be optional extras. They need to be built in from the start.

Trust will be fundamental. Patients need to know how their information is used. Clinicians need confidence that the system supports their judgement rather than overriding it.

But if we solve those challenges, the opportunity is enormous.

The healthcare AI systems that create the greatest impact may not be the ones that make an individual clinician 20% faster. They may be the ones that recognise the patient who is quietly falling through the gaps.

The future of AI in healthcare should not just be clinician-centred. It should be person-centred, journey-centred and proactive.

And perhaps healthcare could learn something from occupational therapy along the way. Sometimes the most important question is not simply, "What is wrong with this patient?"

It is: "What is happening in this person's life, and how can we help them live it better?"