How the rebirth of general medicine will save lives
June is 83, lives alone and still drives herself to the shops. When she doesn’t answer the telephone one morning, her daughter drops by to check on her and finds her lying naked in bed, in a soak of urine, staring at the ceiling. June turns her head at her daughter’s call but does not seem to recognise her, nor does she respond to questions. Her daughter panics and calls an ambulance; she runs to the bathroom, throws June’s pills into a plastic bag and gets her into a nightgown. Then she sits on the edge of the bed and cradles her mother’s head in her lap. The ambulance officers arrive within half an hour; they note that, apart from a low-grade temperature, all the patient’s vital signs are normal and she can spontaneously move all four limbs. They bring her to the hospital where I work as a doctor.
Excepting elective admissions for surgery and the like, beds in a public hospital are guarded by the Emergency Department; if you are sick and need admission you have to come through the ED. Most people have some idea of the drill: the triage nurse ranks the severity of your problem, and that ranking determines how quickly a doctor will see you. If you turn up in the morning and the nurse or ambulance suspects you are having a heart attack, you will be deemed a Category 1 patient and will be seen to immediately. If it is a large, acute heart attack you will likely be on the table having your angiogram and stents within the hour. If all goes well in the angiography suite, by late afternoon you’ll be in a large white room, hooked up to monitors, watching the news, swallowing all your new pills and eating dinner. If you are June, or someone equivalent to her on the triage scale – an older person who has fallen down and can’t walk, someone who’s had a faint, has a fever, is dizzy or delirious or looks starved to death – you are considered a Category 4 or 5. This means that by dinner time you may still be lying on the trolley, perhaps in a corridor of the emergency department if things are really frantic, waiting to be seen by a doctor.
Informally, these patients are known as ‘crumbles’: they are not crashing towards their death like the guy with the heart attack; they are merely engaged in a slow, crumbling demise. When the elderly and not-crashing patients are finally seen, it will be by the most junior doctor in the department. This doctor may take a long time to work out what is going on. Meanwhile, the inexorable crumbling continues.
In theory, the triage system seems sound and unavoidable: if someone is dying in front of you – exsanguinating from an amputated limb, unable to breathe because they have a hole in a lung – you cannot ask them to lie down and wait their turn. If they are not helped immediately they will certainly die. If you don’t immediately help the crumble – June, for example – she will not die, or at least she will not die in front of you. However, according to a report published by the Australasian College for Emergency Medicine, the mortality associated with excess waiting time in emergency departments in Australia exceeds 1500 deaths a year – more than the national road toll. The crumbles are engaged in slow-motion crashes that we cannot see until the final metres, by which time it is too late.
In 2011, the government commissioned an investigation into emergency department waiting times. The result was the formulation of National Emergency Access Targets. The stated aim was to “improve patient safety and quality of care by removing obstacles to patient flow that contribute to emergency department overcrowding”. By 2015, 90% of all patients must move through emergency departments within four hours. I do not know if I can convey how radical this proposal is. It is like telling someone who jogs an easy 5 kilometres in 45 minutes every weekend that in a few years they will be expected to do the same distance in 15 minutes. Changes have been trialled in Western Australia since 2009 and there has been progress. Emergency waiting time targets are being met and patient mortality is down. But these improvements have come at a significant cost. An army of administrative staff spend their days on phones and on the floor policing and pushing patients through the funnel of the ED. Doctors and nurses are exhausted and relationships between medical and administrative staff are strained. Are we sacrificing good training in our desire for efficiency? Meeting the targets within the current triage system is proving extremely difficult. What is needed is a radical new way of running emergency departments so that everyone is seen quickly, so that somehow everyone takes precedence.
There is no waiting room in front of the ED at the Royal London Hospital. There is no triage nurse. If you arrive sick, no matter your degree of morbidity or your age, you will be seen within ten minutes by the most senior doctor. She will look at you, talk to you, briefly examine you and, though she may not come up with a definitive diagnosis, she will at the very least be able to decide where you will best be managed and thus where you should go: to the medics, to the surgeons, to the ED proper for further stabilisation and investigation, or home.
It makes sense. I have a friend who is an art curator. After completing her basic degree, she hung out at art galleries, volunteered at the Heide museum, wrote a couple of theses and landed a job in a major public gallery. Her visual apparatus from eyes to occipital cortex is different from mine: it has been trained. She sees more and better. Where I see a pretty object on a stick, she sees a complex creation with a place in history, heavy with implication in the present and future. She assesses most of this in the time it takes me to appreciate that the object is suspended from a wire that reminds me of the line my dad used in the ’70s to catch flathead. She is the senior emergency department physician and I am the intern doctor. The patient is the work of art.
The idea of adopting this streamlined model in Australia has met with some resistance: it turns the seemingly natural order of things on its head. Senior doctors have done their time; they don’t want to be on the ground running with the pack. They want to supervise, be offered pre-digested summaries, teach, hand out gems here and there, some advice, point out their registrar’s blind spots. They are the bosses.
At the Royal London, if you are the equivalent of a Category 4 or 5 general patient, you will not languish in the ED corridor for five hours before the intern takes your history, your blood, comes up with an interim diagnosis, discusses it with a consultant and then rings a ward medical registrar to come and admit you. At the Royal London the emergency consultant will deliver you directly into the hands of the general medicine team for immediate assessment and treatment in their acute ward.
Of course, there are a number of barriers to the flow of patients from the ED to the ward, such as the numbers of beds available and staff levels. But a major barrier is also the traditional model of what an emergency department does: sort, package and sell patients to teams. These are ugly words to describe sick people with, but they are the words that are used and they represent the practice as it stands. Ward registrars (the most senior of the trainees) will not traditionally accept (buy) any patient who has not been sorted (thoroughly investigated) and packaged (diagnosed or at least had a single organ system nominated as the main problem). This all takes time. After four hours the patient may well remain unseen, undiagnosed or diagnostically undifferentiated – and if you can’t nominate an organ system in which the pathology lies it will be difficult for you to sell the patient.
When I first started out as a medical registrar there would be daily arguments in the emergency department or over the telephone about which teams should accept the care of ‘non-differentiated’ patients with ‘general’ or multi-system decline. To be clear, registrars were not fighting to take the patients; they were fighting to avoid taking them. One of the most common arguments I heard was from the sub-specialties that perform interventional procedures such as angiograms (cardiology) or gastroscopies (gastroenterology): ‘We won’t take the patient as we don’t need to do anything for them.’ Doing something for a patient was reduced to a procedural intervention, as if everything else the patient needed (assessment, monitoring, medicine, care) could be offered by anyone, and therefore not by them.
One of the last times I engaged in this behaviour myself was over a middle-aged patient with a rare neurological disorder – degenerative, untreatable – who came to the ED with worsening confusion and seizures that had probably been precipitated by a urinary tract infection. The neurology registrar and I faced off outside the patient’s cubicle. (Neither of us had seen the patient yet, but we knew his story and, when it came down to it, we both knew how to treat him.) I argued that neurology should take the patient as they had cared for him over the years, knew about his underlying disease and could best manage his seizures. The neurology registrar argued that the patient’s problem had a non-neurological cause (an infection) and so, even though it had resulted in a worsening of his neurological condition, someone else should manage the patient. “That’s crazy,” I said. “We’re not taking him,” the neuro reg said. The nurse pulled back the curtain and there was the patient, an emaciated man-boy in neat navy-blue pyjamas, his elderly parents sitting anxiously at either side of his bed. They’d heard everything.
There are many reasons why intelligent, hardworking and generally humane doctors might argue fiercely in order to avoid taking patients: we may have a huge patient load already; we may be working with a less-than-physicianly consultant (boss) who would disparage us for accepting patients with problems outside her organ of interest; we may feel the patient will be better managed by someone else, as we may have no idea what to do. Also, for patients to be moved to a ward bed and be seen by a treating team quickly we need two things: a team to accept the patient and a bed for them to go to.
When a patient is in the ED, we registrars feel secure in the knowledge that someone is looking after them. The patient cubicles all open to a central area crowded with doctors and nurses. The ratio of staff to patients is high. Once we bring a patient to our ward they become our responsibility. This can be a large burden if our list is already full of patients who are still chaotically unwell. The undifferentiated patient is usually complicated: they take more time to sort out; they are generally older and may be frail, which means more things will go wrong and there is more chance of causing inadvertent harm with any treatment we give. And if we accept them early we must have the staff and the time to work out what is wrong, organise investigations and formulate a treatment plan. One solution to this has been the development of acute medical assessment units. Though these are set up to accept the patients early from the ED, they are still a work in progress at most hospitals in Australia. How many doctors and nurses do you need to look after a group of un-worked-up patients? We do not want acute medical assessment units to become chaotic, crowded holding bays with the atmosphere of a developing-world clinic: people twisted up in sheets and hanging skewed from beds, calling for a non-existent nurse.
The general physician in Australia had all but died out by the 1980s, everywhere except in the rural and remote hospitals that had neither the workforce nor the need for representatives of multiple sub-specialties. In the ’80s and ’90s there were virtually no general medicine physician trainees in Australia. The cities were in love with super sub-specialisation. Chapters were formed, training pathways developed. You didn’t just become a cardiologist, you became an electrophysiology cardiologist, or an interventional cardiologist, or an echo-cardiologist. You specialised in one kind of lung disease, or at the very least you specialised in a single organ. This was necessary in the face of a vast expansion in knowledge. The physician’s bible, Harrison’s Principles of Internal Medicine, has 4012 pages. It is far more manageable to have to know only 300 of them in detail.
In theory, all physicians receive a solid early training in general medicine: the holistic management of a patient, the focused juggling of the problems. We all sit the same exams after five years of working as a junior doctor, and then we all do three or four more years of specialty training. It is those final years that have become less focused on general medicine. And physicians differ in the extent to which they leave their generalist training behind them. I was on a ward round once, presenting a patient’s medical history to the consultant respiratory specialist, when he interrupted me with a huge theatrical yawn and asked when I was going to get to the bit about the lungs. He had no interest in hearing about the other things contributing to the patient’s decline: her heart, joints, bones and sugars. As far as he was concerned, I (the registrar) could fix them up myself or I could ignore them.
I found it difficult to choose a sub-specialty. I chose neurology at first. I thought I might sub-sub-specialise within the discipline in multiple sclerosis, or acute stroke or psychosomatic disorders. What could be cleverer than specialising in brains? I’d get to carry around gleaming equipment – ophthalmoscope, tuning forks, tendon hammers – in a shiny briefcase. I’d need a Chanel-red hat pin on hand at all times, to check a patient’s visual fields. I could be the next Oliver Sacks.
For a year I participated in a general neurology clinic. It took me that long to admit it to myself: I loved my patients, but I was bored. No patient came because they could taste the colour green; no one mistook their wife for a hat. Patients turned up with two things: dizziness or headache. We’d rule out dangerous stuff like venous sinus thrombosis and cerebellar strokes and then give them reassurance or a pill.
I tried nuclear medicine. It sounded very high-tech, and my mum loved that. I got to sit in a comfy wheelie chair in a quiet office with ambient lighting and endless cups of tea in my own cup and saucer, while I dictated reports about fuzzy scans of people with cancer or clots or broken bones. I looked for the black in the scan – that was the cancer. If someone had cancer metastases all through their bones, liver and lungs, we’d call it a Dalmatian scan and know they’d be dead in a few weeks. Lots of the people I scanned were slowly dying, but I didn’t know them. I was in another room, and the scanner didn’t pick up facial expressions. I would never have to be involved with a patient’s actual death – unless someone had a cardiac arrest in my scanner, and if that happened to a passenger on a bus, you wouldn’t expect the bus driver to fix her, would you? So there I sat, alone in a dark room with a bunch of fuzzy ugly scans, sipping another cup of tea.
Next up I thought I’d try endocrinology, where I’d specialise in diabetes and out-of-control hormones. I like diabetes, especially the kind you get if you’re fat; I can relate to people who are struggling with the consequences of having done stuff they shouldn’t have. And I like to intervene in a disease process before the consequences become irreversible. But I just couldn’t get excited about the thyroid gland in the way all the endocrine bosses were, keen for treatment breakthroughs, keen to discuss whether we should palpate it, inject it, ablate it, scan it, irradiate it, or just watch it.
There are many advantages to sub-specialisation for the doctor: when you say you’re an oncologist, everyone knows you administer medicine that dissolves bone marrow, fat and hair; people will open the door and let you go through first. And there are advantages for certain kinds of patients: if you are going to get an organ transplant, for example, it’s probably a very good idea to be treated by someone with expert knowledge of what happens to you when your blood is pumped by a heart harvested from another body and sewn into the cavity of your chest. But what if you are getting old and have a bit of this and a bit of that? What if your kidneys pack it in at the same time as your heart and so you can’t get an angiogram and there’s more going on than either your GP or the nephrologist with a special interest in the autoimmune glomerulonephritides is comfortable with?
That general medicine was the only specialty for me became clear when I was treating an 84-year-old patient named Maria. I was working as a registrar on the respiratory unit and had been asked by another sub-specialty unit to take over her care as, in their opinion, her main problem was a chest infection. Before I met the patient I flicked through her notes. Her medical problems included emphysema requiring her to use continuous oxygen at home, congestive cardiac failure, multiple small strokes that had left her with a weak arm and chronic dizziness, atrial fibrillation, hypothyroidism, chronic daily headache and hypertension, and she had recently sustained a subdural brain haemorrhage in a fall. She was, at the time of my review, sporadically attending five separate sub-specialty clinics for the management of these problems. She lived with her daughter, who took sole care of her.
Maria was lying in the hospital bed. I introduced myself and asked her why she had come to hospital the day before. “I have a chest infection,” she said. “Yes,” I said, “but what did you feel that made you come in to hospital yesterday?” She told me again that she had a chest infection, and that her GP said he couldn’t help her, that she needed medicine in her vein. “But can you describe to me what it was you felt, your symptoms?” “I felt a chest infection, a chest infection,” she repeated, like I was stupid not to see the obvious. “When did you last feel well?” “Why are you asking me all of these questions?” she asked grumpily. I said, equally grumpily, “You have a lot of health problems and I am trying to keep an open mind about what is wrong, and if you want me to help you then you have to answer my questions and there are going to be a lot of them.” We faced off for a moment. “Two months ago,” she said. “And what has changed since then in the way you feel?” I asked. She closed her eyes and sighed, then said she felt very tired and weak, she couldn’t walk around the house easily anymore, she’d fallen over a few times, she’d had a terrible cough a few months back but the sputum was now clear, her headaches were bad and she felt her heart palpitating in her chest sometimes. She opened her eyes and looked at me. “If my daughter was here she could tell you better.” I picked up her hand and told her I would examine her, look at her blood tests and then call her daughter.
Even if Maria did have a chest infection, it was obvious that it was not her main problem: she needed to lose a bit of fluid; her heart rate needed slowing; her thyroid hormone levels needed checking; she needed to stop being prescribed so much prednisolone, which was contributing to her main problem of muscle weakness, which itself was probably a result of the de-conditioning that came with the immobility she had experienced during a chest infection a few months earlier. All of these medical problems needed sorting out, but above and beyond any medical management we could throw at her, what Maria needed was a course of physical rehabilitation if she was to return home with her daughter – which is what they both very much wanted. “I told Mum,” her daughter said to me on the telephone, crying, “if you can’t get out of bed, I can’t take care of you any more.”
As a representative of the respiratory unit my job was to take Maria under our bed card and prescribe antibiotics for her chest infection, if she had one. If she did not, in my opinion, have a chest infection, then it was my job to reject her, to leave her care in the hands of some other sub-specialty. But I saw that there was no sub-specialty that Maria fitted into neatly. No one would want her under their bed card.
Stories abound about patients who suffer the consequences of being treated ‘sub-specially’ by a sub-specialty. It happens on the surgical wards, too. I know of an elderly woman who recently fell and ripped a huge flap of skin off her elbow. She also had a sore hip. An X-ray of the hip was arranged in ED. The patient was admitted to the plastic surgeons who operated on her elbow. She recovered on the ward and was discharged home, but the pain in her hip worsened so she came back. The hip X-ray was reviewed in the ED – a week after it was taken. The plastic surgeons had taken exemplary care with her elbow. Too bad she had a snapped femur at the same time. They could not see beyond their own suture margins.
The problem is clear enough: in massive hospitals demarcated into care silos there has been a loss of a holistic approach to the patient. This means that the medical care of the elderly and the crumbling has to be artificially fragmented into the care of separate organ systems. From a best-practice perspective, a health resource perspective and from the perspective of Maria’s daughter, who was having to bring her to multiple appointments, what was chiefly needed was a good general physician to look after her, both as an inpatient and as an outpatient.
Some hospitals without a general medicine unit roster on a daily ‘physician of last resort’. This consultant doctor and her team must take all the patients rejected by the other sub-specialist teams; for that day they cannot say no. In other hospitals the registrars just argue and argue until one team gives in and accepts the ‘undifferentiated patient’. Either way it takes a long time for the patient to be admitted to the last resort. This situation is untenable. It is also inhumane and dangerous. Someone needs to want to look after these patients – the crumbles, the mysteries. A single team, headed by a doctor with expertise in treating a patient holistically, should direct their care. Yet these doctors – the general physicians – had become almost extinct.
Fortunately, it’s dawning on authorities that the public needs hospitals and doctors to serve an ageing community among whom chronic diseases are on the rise; that hospitals need large general medical units with the staff, facilities and funding to scoop the chaotically unwell and the crumbling patients out of ED, to assess and treat them promptly and to go on caring for them till they are well. In Victoria, every major tertiary hospital has a general medicine department run by a mix of dual-trained physicians, general physicians and sub-specialists who either have a genuine interest in general medicine or who can’t get a job in their chosen field.
Although there is still a dire shortage of committed generalists, there are now almost 300 registrars currently training to be general physicians. This shift in the delivery of specialist and hospital health care in Australia has brought with it a number of challenges. Having managed to attract this new generation of doctors to the practice of general medicine, we find we are unable to train them. To train as a general physician you are required by the Royal Australian College of Physicians to work two six-month sub-specialty terms. Finding departments that will employ general medicine trainees is proving extremely difficult, even when the trainee brings funding for most of her own salary. It seems sub-specialty departments on the whole do not wish to foster these strays; after 50 years of rapidly increasing medical knowledge, and the concomitant division of that knowledge into separate areas of practice based on organ systems, they want to produce doctors in their own image. Some sub-specialty associations are even putting in barriers to discourage their trainees from training both as a sub-specialist and a general physician, a combination that is especially valuable in rural areas where there may not be a full-time need for a cardiologist.
June arrives in my hospital’s emergency department. Fifteen minutes later the emergency consultant rings me: “We have a patient for you. Eighty-three-year-old woman from home alone, found incontinent and confused. Stable, with a low-grade temperature. We’ve taken blood, urine, cultures and done a chest X-ray. May we send her up?”
We have 15 patients on our list, we’ve just received two new ones and they want us to accept June too. But yes, he may send her up. This is the rebirth of acute and general medicine as a specialty in Australia. We will see June in our acute assessment unit, start her on fluids and antibiotics and check her test results as they come through. She most likely has a urinary tract infection with associated delirium. Early treatment will increase her chances of getting better and getting home. We split our team and I’m relying on my junior registrar to let me know if anything dire or unexpected shows up in any of the patients I don’t have time to see. There’s no way my intern’s going home on time today; she’s juggling pathology forms and X-ray requests and faxes from other hospitals and scribbling like mad in the chart, translating my questions and the patient’s answers into a smooth narrative that explains why they’re here. Meanwhile her pager is going off and the nurses are harassing her for discharge summaries and scripts for other patients who are ready to leave. I’m thinking that another doctor or two and a few more nurses wouldn’t go astray here. The last resort may have had a makeover, but we’re still a bit thin on the ground.