Patient Flow in Action

A talk delivered at the Whole of Health Program Masterclass

New South Wales Parliament House, Sydney

Before I begin, I would like to acknowledge the traditional custodians of the lands on which we meet – The Gadigal people of the Eora Nation – and to acknowledge that Indigenous Australians have poorer health outcomes and experiences than non-Indigenous Australians. Patient flow problems contribute to health inequities. I ask everyone present today to commit to doing their very best to close the gap.

Thank you for the invitation to speak today. I think this means I have been officially outed as a patient flow nerd!

I’m an emergency physician. I’m currently working in medical administration roles across Northern Sydney Local Health District, following a close encounter with cancer in 2025. I previously worked as Director of Emergency Medicine at Hornsby Ku-ring-gai Hospital. I served as President of the Australasian College for Emergency Medicine (ACEM) from 2021 to 2023. During this time, I visited more than 80 hospitals across Australia and New Zealand, and most of the conversations I had centred on patient flow. When you are President of ACEM, you become the official spokesperson on access block. Ministers, staffers, health departments, hospital executives, and clinical leaders, as well as journalists, all want to talk with you about patient flow problems, and to discuss potential solutions.

Conversations about access block have now moved well beyond emergency departments (EDs). I find this really encouraging. I’m sure that everyone in this room knows that while patient flow problems manifest in EDs as overcrowding and ambulance ramping, EDs themselves are not usually the root cause of the problem.

Today, I have been asked to give you some practical advice based on what I have seen and heard in my various roles over the years. I would like to thank everyone who has shared their expertise and ideas with me. I must admit that when it comes to solving patient flow, there are no shortcuts or simple solutions out there. If there were, I think we would have implemented them by now. But don’t lose heart, there are still some meaningful things that we can do.

Design care for patients

Let’s start with a simple, but important principle. Don’t build a new unit, or introduce a new service, with the primary goal of improving flow. Instead, design models of care that serve real patients, with real problems and needs. Otherwise, you just add complexity – more little boxes in series, or in parallel, and more opportunities for congested traffic or wrong doors – in an already overloaded patient flow circuit.

The only exception I make to this rule is for transit units, which I think can serve a genuine need for departing patients, for example the person who is waiting for a lift home from their daughter, or for their discharge medications to be delivered. I’m sceptical about the use of transit units for arrivals – I’m not sure that the needs of new patients from ED, who might not be completely worked up, are well served by transit lounges. I think new patients need medical and nursing expertise on appropriate inpatient wards.

Of course, there are some models of care that meet clinical needs and also improve flow. Geriatric Emergency Medicine (GEM) Short Stay Units are a great example. They give frail, elderly people improved access to senior medical decision-making and allied health advice, while minimising time spent in hospital, which we know, for this group, is associated with adverse events including delirium and falls. If you’re interested in improving outcomes for elderly patients who present to ED, have a look at the GEMSSU models in operation at Hornsby and Blacktown Hospitals.

One size does not fit all

Every health service is different. They exist in different socioeconomic contexts, with different community supports around them, and have different workforces with different skills. They have different politics and serve different patients with different care needs.

It is really tempting to see a model that seems to work somewhere else and just transplant it into your context. It is superficially appealing to try to make everything, everywhere, the same. But it just doesn’t work. Instead, borrow the principles. Adapt them to local circumstances, then communicate about your model very clearly, and to all stakeholders, as publicly as possible. Monitor your model’s performance and tweak it as you go.

In Australia, we love to look to the United Kingdom’s National Health Servance for guidance. I’d advise caution when trying to implement a model of care from the UK here. My colleague, Dr Peter Roberts, and I recently convened a conference about patient flow. We invited Dr Louella Vaughan, an acute physician working in London, to speak about patient flow initiatives in England. To cut a long story short, many of the models don’t work. A video of her talk is available via the GOODEM website, and I encourage you to watch.

Untangle the tangles and bridge the gaps

Clinicians will tell you where the friction points and discontinuities in their service are. Trust their feedback and work with them to improve and integrate care. Please don’t ask clinicians to try a new model to prove that it will fail, instead of incorporating their suggested changes. That’s like whipping a horse until it runs across a broken bridge. When you do this, clinicians will disengage, and the model is doomed from the outset. There are too many examples of services that don’t align with patient demand, such as medical assessment units that don’t accept ED admissions at times of peak presentations, or clinics with referral processes that are too difficult to navigate.

Professor Clair Sullivan, from Princess Alexandra Hospital in Brisbane, has done a lot of work on improving care transitions. She also spoke at the GOODEM patient flow conference.

I dream of untangling the ED waiting room, using a concierge and computer-aided decision support, a bit like the systems that have been implemented at Service NSW. Using nurse triage, and some patient or carer entered information about past medical history and activities of daily living into an iPad or computer kiosk, we could stream patients according to clinical urgency (their triage score), complexity (need for early senior doctor input), and whether they need a bed, a chair, or a cubicle for ‘see and treat’. The Patient Experience Officer could guide each patient, in order, to the next available appropriate clinical space, freeing up all the ED clinician time that goes into finding (and tidying up) a treatment area. It sounds like a fantasy – but have any of you every checked into a private hospital for surgery? See – it can be done better.

Bring people together

And not just for meetings or to talk about problems. This is the secret sauce of smaller hospitals, which often work better than large ones. People know each other, belong to a community, and work together to drive positive change.

Culture is like a garden. You need to water it, weed it, and make sure it gets plenty of sunshine.

Good processes are based on relationships. Foster them, across professions, teams and specialties, but with clear boundaries and roles so people don’t burn out.

Don’t let digitised processes block real life interactions. I wrote a story about this once, after an intern described to me how it felt to be tasked by electronic messages. When I was a junior doctor, I knew all the nurses, because the only way to find out which jobs needed to be done was to check the book at the front desk of each ward. Please don’t just message someone. Walk the wards or pick up the phone and talk.

‘Hi Clare. It’s the bed manager. The 7B NUM asked me to call you to ask you to call the Gastro Reg to ask her to call her consultant to ask if the patient in bed 6 is ready to go home’. I’m confident that some version of this phone call plays out dozens of times each day in your hospital. In my current medical admin role, it seems that some patient flow problems might evaporate if nurses and doctors spoke with each other more, and the nurse unit managers felt empowered to speak directly with the consultants.

Plan the unplanned care and smooth the planned care

It’s time to bust a myth. Unplanned hospital care is surprisingly predictable. That’s right – we know what comes through emergency departments each day, how many appendicectomies we do each week, and how many femurs get fractured. We should plan for it better – with emergency theatre lists, after-hours staffing, and proactively allocate the resources we need to provide that care.

Most bottlenecks in patient flow are caused by peaks and troughs in planned care. The graph that Justin presented earlier, with ambulance offload times across each week looking a bit like atrial flutter, demonstrates precisely that. If you’re not already aware of it, look up the work of Dr Eugene Litvak and the Institute for Health Optimisation in the USA.

Most hospitals run big surgical cases, such as spinal lists, early in the week. From the surgeon’s perspective, this probably makes sense. Those patients are most risk of post-operative complications during the remaining days of the working week, not over the weekend. But – this practice has big knock-on effects on hospital-wide capacity. Think of the associated demand for ICU beds and follow-up diagnostic tests.

We can smooth this out. Spread out elective admissions and make the working day on wards more predictable, with regular consultant ward rounds, assigned slots for family meetings, and prioritised allied health reviews. I remember a NSW Health program from around 2005 called ‘Barbara’s Journey’ that attempted to do exactly this. I couldn’t find any trace of it left on the NSW Health website.

Timing of outpatient clinics matters too. Think about fracture clinic at your hospital. Sixty to eighty patients, all arriving at once, each needing an x-ray, most needing a splint, some needing physiotherapy, some needing surgery or an admission – all lumped into two or three afternoons per week. As an aside, I remember a clinical redesign project at my hospital that aimed to improve processes in our fracture clinic. It was proposed that fracture clinic would be renamed the ‘Orthopaedic Follow-Up Clinic (O.F.U.C.). Funny! That was exactly how patients felt when they saw the queue.

Redesign outpatient clinics

Outpatient clinics are ripe for reform. There are too many clinics designed around specific diagnoses, not symptoms or syndromes, resulting in lots of wrong doors, with patients bouncing around the system. Clinics need good administration, with senior clinical input into triage processes, and authorisation for forward referral, when needed. Clinics need realistic appointment times or they back-up – see the work of Dr Mark Mackay from South Australia, who spoke at the patient flow conference. They need senior doctors present – so decisions are not just deferred to the next clinic by more junior team members – and clear criteria for referral back to primary care. They need to be designed and timed so that associated services, such as diagnostic tests, allied health reviews, theatre lists and inpatient beds, can be accessed when required.

Outpatient clinics can be very useful as step down from ED or inpatient care. To work in this way, they need broad inclusion criteria, simple referral processes, and senior clinical input. The Paediatric Acute Review Clinic at Hornsby Hospital is a fantastic service that allows a child who is on the border of needing admission to go home with close follow up by paediatric nurses and doctors. Feedback from parents is very positive. I wish we had an equivalent service for adults.

Sort the governance

This is important but not obvious. Most hospitals have a divisional structure – usually medicine, surgery, and women’s and children’s health. These divisions were imagined before EDs and intensive care units existed, and don’t reflect current conditions. It makes more sense to group services with similar operating models together. EDs have high volume and activity, and during my hospital visits, I noticed that health services with the best patient flow practices tended to have EDs represented directly, not buried within the division of medicine or surgery. Clinicians should be embedded in the executive, with enough non-patient-facing time remunerated to allow them to lead well. Business and project managers, with appropriate qualifications and expertise, should be embedded in clinical teams, to support clinical leaders with service planning, change management and clinical redesign.

It is important to align KPIs across the entire health service. For example, tight KPIs on diagnostics in ED, but not on the wards, drive ED delays. If you can get a CT scan in ED within a few hours, but it takes two days on the ward, then inpatient teams will not accept admissions unless patients are completely worked up.

The Tragedy of the Commons

 I would like to thank Dr Kendall Bein, an emergency physician from Royal Prince Alfred Hospital, for devising this elegant analogy for access block.

The Tragedy of the Commons is an economic concept, describing the situation where individuals, acting in their own short-term interest, deplete a shared resource, even though it is no-one’s best interest to do so in the longer-term. Professor Elinor Ostrom won a Nobel Prize for describing a solution. People sharing a resource need to create clear boundaries and rules, monitor them, be accountable to them, and sanction individuals who don’t cooperate.

This makes sense in the hospital context. Surgeons understand patient flow – because they need beds to operate. Paediatricians also manage flow well, because they can only admit patients to beds on designated children’s wards. Maybe we need to get back to home wards? Or, at least, very deliberately allocate a bed base to each inpatient medical team?

Access to senior decision-makers

Lack of timely access to senior decision-makers, especially consultant doctors, is a major problem that creates duplication and delays. Access is hampered by clinical hierarchies and competing priorities. The on-call gastroenterologist might be busy doing a scope. The orthopaedic registrar is in clinic. The mental health clinical nurse consultant has to wait to talk to the psychiatry registrar who has to review the patient himself before talking to the consultant psychiatrist.

There are potential economies of scale to be achieved from running an on-call roster for senior clinical input at regional, or even statewide, level – with specialists available to give advice by phone or video call, not busy in their rooms or doing other things. We are seeing this sort of model emerge in services like vCare in regional NSW, the WA Country Health Service, or the Patient Access Coordination Hubs across Queensland.

Some medical teams have moved to team-based care, for example the renal unit at Royal North Shore Hospital, or neurology at Hornsby – and intensive care at every major hospital in Australia. The on-call consultant rounds on all inpatients, regardless of who was on when the patient was admitted. They are rostered to do this in a predictable pattern and move other commitments to be available for hospital care. This model also makes life much simpler for registrars, residents and interns, who do not have to work around multiple consultant rounds each day.

Patient flow initiatives need to ask the right questions of the right people. There is no point asking an intern when a patient will be discharged, they don’t have enough experience to know yet. You need senior medical engagement to make a difference – and that might mean running or timing meetings about patient flow differently.

A word of caution. You need to be mindful of providing teaching and learning opportunities for clinicians in training. Most health professions still follow an apprenticeship model. While you need senior clinicians to make decisions, you need to make sure that junior clinicians are not cut out of the loop entirely. They need to learn how and why clinical decisions are made.

Money matters

Too often, clinical redesign projects fail because money is not considered at all, or because the financial analysis lacks sophistication. The way we fund healthcare in Australia is complicated. Activity-based vs block vs historic funding. Medicare and the pharmaceutical benefits scheme. Private health insurance. Out-of-pocket costs for patients and carers. Commonwealth vs state vs local government responsibilities.

Clinical leaders need to be across health finance and economics. In particular, they need to understand opportunity cost and moral hazard.

At clinical unit level, we tend to save pennies to spend pounds. We don’t employ enough staff to cover planned and unplanned leave, so costs blow out on premium labour. We don’t have enough working equipment. Walk into any ED during the afternoon shift and mentally calculate how much clinicians get paid to queue up to use a computer.

When it comes to patient flow, how doctors are paid really matters. Some doctors earn a salary. Some are paid by the session. Others are paid fee-for-service – they only receive payment when they see a patient admitted under their own card in a hospital bed. Most patient flow initiatives fundamentally misunderstand how doctors work and how they earn an income. I think this is a topic for a whole talk another day – one I am happy to give at a future Whole of Health Program workshop.

Less is more

Have you read the paper ‘Getting Rid of Stupid Stuff?’ It is my favourite ever article from the New England Journal of Medicine.

There is a lot of work as imagined happening in hospital board rooms. Work as imagined is very different from work as done. If you haven’t already, explore the writings of Dr Steven Shorrock on his Humanistic Systems website. So much energy goes into working around processes that don’t work. Dr Nick Taylor, an emergency physician from Canberra, gave an excellent talk about the normalisation of deviance in healthcare at a conference a few years ago. When the prescribed way of doing something doesn’t work, clinical practice drifts from guidelines and evidence, impacting patient and clinician safety.

The first step of a clinical redesign project is service mapping. The second step should be simplification – removing every unnecessary step.

Please trust and respect clinicians who provide challenging feedback. Don’t just assume they are being difficult or are resistant to change. Find out what they know and use that information to design better ways of getting things done.

Be mindful of the work pressures clinicians are under. Beware allocating administrative tasks to clinicians. ‘It will only take a few minutes’ becomes hours when multiplied across many patients per day.

Go gentle with unwritten rules that have arisen from adverse outcomes. Every hospital has odd things that spook people. Clinicians have long memories and deep emotional ties to past patients. Trust is important. If something doesn’t make sense to you, please ask why it is done that way, and be careful with how you manage change.

Nagging doesn’t work

I want you all to reflect on a time when nagging made you change your mind or do something you didn’t want to do. See – it really doesn’t work. Instead, ask how you can help.

Negative feedback doesn’t work well either. Shame is a negative motivator.

I remember a time when a hospital executive called me to complain about a small drop in one rating on the patient experience survey. That was the year that Hornsby ED was on the front page of the Sydney Morning Herald for topping the state.

Find the positives – always. Create a shared vision of better ways of working, with patient care at the heart.

Leadership should bear witness and offer genuine support.

Don’t nag clinicians to shave off minutes in a single patient’s journey. Do the deep structural work required to free up hours or days for everyone – patients, carers, and clinicians.

Digital done well

I know you know this, but the data dashboard does not tell the full story. The map is not the territory.

The ED navigator does not need you to look at the portal and call her to tell her there is an ambulance that needs offloading. She knows. She’s there. She can see it. Get down to ED and have a look around. Listen. Help. Don’t just document that you made a call.

These days, it is easy to create digital tools. We have a data display or a checklist for almost every aspect of patient care. But information is no longer enough. To be useful, digital tools must predict problems and facilitate prevention. We need to think of the clinician user experience – every alert creates a risk of diversion or distraction from a more critical task. We also need to think of the patient and carer experience – do digital tools reflect what matters to them?

The big picture

At the start, I said there is no simple fix for patient flow. Here are some things we need to do.

We need enough capacity in hospitals. For now. Not for some perfect, predicted future state.

We need to fix primary and community care – especially aged, disability and mental health care. That is likely to require some investment in residential and semi-residential or drop-in models. I would love a local health district to partner with a primary health network to build a wrap-around health ecosystem using shared commonwealth, state and local funding. We have some new hospitals coming online in New South Wales that might present this opportunity.

We need to invest in innovation. Not just band aids and slashed budgets. We need to genuinely reimagine care systems with expertise, resources and care.

Patient flow is a wicked problem. Hospitals don’t work like freeways or factory production lines. Assistant Professor Kieran Le Plastrier, an academic General Practitioner, spoke about this at the patient flow conference - watch his talk here. To fix patient flow, we need to learn about complexity science and look to soft systems methodology for meaningful solutions.

Patient flow is not easy, but it’s important. If we work together, we can make a difference.

Thank you for the opportunity to speak with you today.

Dr Clare Skinner is a specialist emergency physician, communicator and health reform strategist. She is former President of the Australasian College for Emergency Medicine and works as a clinician, educator and health executive. Clare has special interests in clinical leadership, governance and workplace culture. She is deeply committed to improving population health using an inclusive, person-centred and evidence-informed approach. Clare was selected in the Top 50 Public Sector Women NSW in 2018.