The history of the GP referral.
“… by 1900 The London Hospital [outpatients department] was seeing, on average, over 4200 out-patients a week or (excluding Sundays) over 700 a day. These patients were seen in the space of 2–3 hours each morning with the average consultation lasting about 1 minute or less, during which they were seen and prescribed a bottle of medicine by junior hospital staff.”
“Anyone with an eye for figures reading the Clerk’s report for St. Bartholomew’s Hospital for the year, say 1877, must, if previously unused to such facts, feel his mind somewhat unhinged by the statement that 157 947 patients were relieved at that institution in the course of the year. He might cut out the 5000 and odd in-patients, and the 780 women attended at their houses, but he would still be left with a total of 151 836 persons who were actually registered as visiting the hospital presumably for the first time, and having their complaints investigated and treated …”
Bridges, [who later became poet Laureate and worked as a casualty doctor] added that follow-up attendance were not included in the totals and a sizeable number escaped registration, so that the true total was nearer 200 000 than 150 000.
“The surgery [that is, out-patient department] is a large hall, 90 feet by 30 feet with rooms opening off it at the four corners … One half of the hall, with the rooms off it, is devoted to the women, the other half to men … The work is done in the mornings and patients are admitted at 9 o’clock [when] the doors are thrown open … [and] forcibly held ajar by the porters … This goes on until, at 10 o’clock precisely, the doors are forcibly shut. If anyone should go into the hall at about 20 past 9, he would see some 100 persons standing in an orderly manner … the women engaged in conversation, the men waiting in silence. If he goes out and comes in again at 11, he will frequently find the room nearly or quite empty.”
Bridges was deeply shocked by having to deal every day with:
“… some 200 paupers, who are many of them seriously ill, some mortally, many but slightly, but nearly all with considerable bodily inconvenience or pain which, unless disease be a joke, and this the whole constitution of our hospital forbids us to suppose, entitles them to his patient attention and
investigation, and demands”
Senior hospital staff were ambivalent towards the out-patient problem. They could, indeed should have seen that outpatient chaos was incompatible with good medical care, but they firmly believed that the larger the numbers of patients they claimed to have ‘treated’, the more subscriptions they would attract. But those who were loudest in their condemnation of this scandalous state of affairs were the GPs. In areas such as the East End of London, until the 1840s GPs earned (or scraped) a living by charging the working classes sixpence or a shilling for a visit to the surgery and sixpence for the inevitable bottle of medicine. Out-patients however was free, and as the number of patients increased GPs were forced to reduce their fees. Many went bankrupt. By the end of the 19th century it was clear that something had to be done because, as one correspondent put it in 1894: ‘The abuse of the hospitals’ out-patients departments is an evil so gigantic that the tendency is to regard it as being unavoidable … ’ He recommended that only cases which were ‘certified by a medical man as requiring special consideration’ should be admitted to out-patients.
The origin of the GP referral and the gatekeeping role was born out of a crisis bought on by exponential demand on hospital outpatient departments.
“… at The (Royal) London Hospital the annual number of out-patient attendance was (in round figures) about 1000 in the decade 1800–1809, 17 000 in 1850–1859, and 220 000 in 1900–1909”
The solution was that ‘the physician and surgeon kept the hospitals and the GPs kept the patients’ meaning that the only way to access a hospital outpatients was with a letter of referral from a GP.
Unlike the university educated physician or surgeon, the origin of the GP in England was the apothecary or pharmacist, and the professionals had long looked down on them.
In the mid-nineteenth century, sensing competition, the powerful Royal Colleges, notably the Surgeons, did all they could to suppress the foundation of the Association [of General Practitioners], and contemporary accounts of the acrimonious debates which ensued make today’s jibes about arrogant hospital consultants and golf-playing GPs look distinctly insipid. General Practice, Past, Present and Future
After the National Insurance act of 1911 GPs earned enough to provide basic care for their poorer patients and avoid the need for specialisation, but nothing was done by the British Medical Association to tackle the variation in quality of General Practice and few GPs showed any interest in providing care in hospitals.
The reason for this historical preamble is to show how the divisions between GPs and specialists became entrenched, because how we address GP referrals depends almost entirely on the quality of this relationship.
A century later we face another crisis.
The reasons for the extraordinary rise in medical demand in the decades up to 1911 are still not understood. Neither, for that matter are the reasons for the rising demand in recent decades which has occurred at the same time as the population, by most measures has become healthier. I offer some suggestions in another blog post, ‘The Cost of Chronic Disease and the lack of NHS reform’
This crisis is not one of excessively busy outpatient departments bought on by an exponential increase in GP referrals, nor of an exponential rise in patient demand on GPs, but is a crisis of funding, manufactured by our coalition government who have flat-lined NHS spending. So-called ‘efficiency savings’ amount to £20bn over the next 4 years and according to a Tory peer who wrote to me about the health bill, ‘they will destroy the NHS’.
It is proposed that significant savings can come from reducing GP referrals. The bulk of the NHS budget is being given to GP commissioning consortia (GPCC) to spend ‘commissioning’ services for their patients. This means that whenever one of my patients is referred to hospital, either to an outpatient clinic or an A&E department in an emergency, or even if they go to A&E without a referral, the cost comes out of the GPCC budget. Any follow-up appointments, investigations, procedures, operations, inpatient stays, prescriptions and so on all contribute to the costs. Therefore the GP has little control over most of the costs associated with a referral.
Why do GPs refer patients?*
When a GP refers a patient it is usually for one of three reasons:
- investigation and/or diagnosis,
- advice and/or reassurance for the patient and/or GP.
There are enormous variations in the rates at which GPs refer their patients. Up to 40% of the variation can be explained by patient characteristics, for example some GPs see more elderly patients, or work in more deprived areas. Up to 10% is explained by doctor characteristics including ability to tolerate uncertainty or cope with patient pressure.
Not referring patients
The first patient I ever made a decision not to refer had a headache. As an A&E doctor I could have referred him to a specialist, but I thought he was fit to go home. A few hours later his family bought him back to hospital because the headache was worse and he was confused and drowsy. On arrival at the A&E department he had a fit and despite emergency surgery for a bleeding blood vessel in his brain, he died not long after. It was a devastating start to my medical career and has influenced my practice ever since.
When Sally bent down to pick up her 3-year-old daughter, she suddenly felt a pain in the middle of her back. She was a healthy woman in her late 30s with no other symptoms. Three months later we sat down as a practice to discuss how it was that the breast cancer which had spread to her spine had been missed when she came in to see me complaining of back pain. I still feel awful when I think about her. As the responsible clinician I prepared for the critical incident meeting by researching all the recent guidelines for the assessment of patients presenting with back pain to make sure that neither I, nor any of my colleagues would make the same mistake again. I discovered in my research that the chance that a woman in her late 30s with breast cancer would present with back pain due to a spread of cancer to her spine is about one in a million. Back pain is one of the commonest complaints patients bring to GPs and accounts for between 5 and 10% of consultations. A GP is expected to be an expert at managing the vast majority of cases that do not need a referral and at recognising the one in a million that does.
The consequences to the patient, of a doctor failing to spot a serious disease will vary according to the length of delay. In some cases it will be fatal and in others it will make very little difference to the prognosis. In some cases, it will seriously damage the relationship between the doctor and the patient, but frequently it does not. The effect that it has on a doctors’ psychology and behaviour is not often examined.
I have made headaches one of my special interests and last year I spent time teaching other local GPs about how to manage patients with headaches. Some GPs were arranging a lot more brain scans and referring a lot more patients than others. Many of these GPs simply struggled more than others to manage the niggling doubt that their patients’ headaches might represent something really serious. Only one had ever seen a patient with a brain tumour, but all felt really pressured by patients who they thought were more anxious and harder to reassure than ever. They thought this was due to several contemporary trends including:
- the loss of medical authority and dwindling respect for professional opinion
- the conversion of patients into ‘consumers with rights’ to be referred if that is what they choose. A phenomenon called ‘healthism’
- The tabloid obsession with exposing ‘GPs who missed something rare but serious’ which always concludes with the simple message to their readers, ‘if you’re worried about something insist your GP refers you to a specialist’.
- The fear-mongering medical industry which uses patients’ anxiety to sell them unnecessary scans, backed up by alarmist media.
- faith in technology which suggests that for every symptom there must be a physical cause and a technological solution.
- Easier access to investigations which has led to doctors investigating more frequently and then raises patient expectations of investigations.
- Many of our patients are from countries where doctors have a financial interest in scanning or prescribing and consequently expect to have a scan or a prescription for every complaint. Our patients are suspicious that we scan less in order to save money.
- The NHS cuts and growing awareness of GPs as NHS budget holders in a time of austerity.
A detailed review of patient pressure for referral for headaches in 2006 showed wide variation in GPs ability to tolerate uncertainty:
This was particularly, but not exclusively, emphasised by GPs who described less clinical confidence:
‘ … I might refer people just for my own anxiety and fears of missing something, even if they’re not anxious. So, really, I suppose it depends on lots of factors.’ (GP 1803)
‘If we could refer everybody for an MRI scan tomorrow, we’d all have a lot more peace of mind. But we cannot, because there aren’t the facilities available. And so we have to live with uncertainty and we can, all of us, can only live with a certain amount — there comes a point where you just get uneasy and think “I need to be more sure about this”.’ (GP 2511)
However, problems of tolerating uncertainty were not fully explained by actual clinical competence or experience, as this GP observed:
‘We used to have a partner who is retired now who had pretty well double everyone else’s [referral rate], in spite of being very experienced, good at diagnosis and so on, but he couldn’t stand uncertainty.’ (GP3911)
Pressure to refer
Peggy has had terrible headaches for the last 2 years, a consequence of severe osteoarthritis in her neck. It keeps her awake for 2 or 3 nights a week and has not improved with physiotherapy, osteopathy or a range of medications.
Peggy is very anxious, she comes to see me almost every month and telephones for advice and reassurance almost every week. Together we have been through worries about breast and ovarian cancer, cancer of the spine and more recently concerns about her husband’s health. I was surprised that it took so long for her to admit that she was worried about a brain tumour. I had asked her before if she was worried about cancer, a stroke or some other serious underlying cause for her headaches. Even though we had agreed that the arthritis in her neck was to blame and she reassured me that my reassurance had reassured her. When eventually she phoned to say that she had arranged a brain scan privately I couldn’t help feeling disappointed, not with her, but with myself for failing to allay her fears. She apologized and said it wasn’t my fault, she said she trusted me and was sure that I was right, she hoped, sincerely that I would forgive her. She promised to come back and see me after the scan.
Peggy understands that one of the main roles of a general practitioner is to manage their patients’ anxiety. She, like millions of others like her, knows the pain and distress of coping with her own anxiety and she is embarrassed and upset by the effect it has on others at the same as she respects those who are able to cope with it, and haven’t pushed her away or abandoned her. She has bought in cards and small gifts of appreciation over the years, always expressing how sorry she is to keep bothering me with her worries and constantly afraid that one day I’ll say ‘enough’ because she has ‘cried wolf’ one time too many. Among my consistent reassurances is the assurance that she is not wasting my time. There is a therapeutic limit, a point at which seeking reassurance becomes an obsession, an end in itself, and we have bought this up and remind ourselves of it from time to time to ensure the relationship is therapeutic.
Like many of my most anxious patients, Peggy is socially isolated. She has very few close friends left. Not wanting to bother them who have enough worries of their own she keeps herself to herself. She self-medicates with alcohol and cigarettes, ‘anxiolytics’ in the pharmacological lexicon, drugs that relieve anxiety. The combined effects of poverty, financial insecurity, low levels of literacy and numeracy, poor housing and street violence add up to a high proportion of my patients suffering chronic, severe anxiety. My strongest impression of working as a GP for a couple of weeks in better off Salisbury was that the patients there were so much easier to reassure than in Hackney.
Peggy and her husband are better off than most of my patients and they have private health insurance which will pay for Peggy’s MRI brain scan. I could order an MRI scan myself, paid for by the NHS. I chose not to order one in part because I had no reason to suspect it to show any abnormalities, but also because investigations rarely provide more than a very short-term relief from anxiety and then raise the expectation that new symptoms need a similar level of investigation. Cost is not an issue at present. For the last few years I have been able to arrange MRI scans at no additional cost to the practice or PCT because the provider, InHealth was paid a block contract and we are still are a way off ordering all the scans that were paid for. It has changed the way local GPs order scans; many are ordered for knees, spines, and brains that previously would have first required a consultant referral and an expert opinion. Only rarely do they change clinical management, but they are raising our patient’s expectations. It will not be long before contracts like this finish, and every scan will incur a cost to the CCG.
Our local neurologist/ headache specialist explained that when a GP referred a patient with a headache they used to arrange a brain scan as a matter of course because a. they could, b. it was what the patient expected and c. it was what (they assumed) the GP wanted. Now he says he rarely arranges a scan because a. the GP has already done one or b. the GP has spent time and effort explaining to the patient why a scan is unnecessary. Just as verbal reassurance seeking can become obsessive, so can radiological reassurance seeking. Scans can cause physical and psychological harm. This ranges from exposure to radiation to ‘false-positive’ results that lead to more unnecessary investigations and increasingly risky procedures. Anxious patients are too numerous and too easily exploited to prevent the market growing in commercial organizations offering MRI and other scans.
Over-investigation in the NHS has never before been a commercial enterprise, but in recent years, because of the internal market, hospitals are paid for each procedure and ‘scoping for money’ has become a real risk. There are other more benign reasons for over investigation. Humans are social creatures and emotions are contagious, we infect others with a little of our joy, frustration, serenity and angst. Anxious patients make doctors anxious. When we are faced with one patient after another, day after day, infected with the anxiety associated with deprivation and social exclusion it can weigh heavily and influence our clinical management and so anxious doctors working in deprived areas tend to have above average referral rates.
All doctors, especially those who deal with the uncertain business of making a diagnosis out of their patient’s undifferentiated symptoms worry about missing a serious illness. It is a rarely challenged rule of medicine that faced with an anxious patient with physical symptoms, we first have to rule out a physical cause’ before dealing with the psychological distress. How far we go in ruling out a physical cause depends on our knowledge and confidence and on what it takes to reassure our patient.
Dr Mike Chester is a cardiologist with a special interest in ‘refractory angina’. Angina is a condition due to poor circulation to the heart that causes chest pain. The symptoms are almost indistinguishable from a heart attack. Refractory angina refers to the most severe cases with chest pain coming on unpredictably including at rest. Dr Chester discovered that by using psychological interventions he could significantly reduce his patients symptoms with a reduction in medical and surgical interventions and a rapid and sustained reduction in unscheduled admissions and heart attack rates. The key to this was ‘patient centred care’, letting patients take a lead in deciding what treatment they wanted. It is estimated that up to there are up to 50 000 unnecessary medical interventions for angina in the UK annually and one aim of his work is to try to reduce these excessive, high-risk interventions.
Why not refer?
The risks of over-investigation are particularly grave for some patients. Up to 50% of patients’ symptoms defy medical explanation and a significant number of patients repeatedly present with these so-called, ‘medically unexplained symptoms’. Research conducted by the Tavistock in conjunction with GP practices including our own has found that there are patients who repeatedly present to heir GP and to hospital with physical complaints for which no medical cause can be found. Although many do have chronic anxiety and/or depression, many do not. Identifying the most severe patients is often retrospective. In the days before electronic records they would have had the thickest sets of medical records, testament to the efforts of the medical establishment to find a cause of their symptoms. Accounting for approximately 1% of our practice population they outnumber patients with rare cancers many times over. What both groups have in common is that the diagnosis is too often delayed.
Carol eventually found a private consultant who offered a drastic solution to her severe abdominal pain and constipation. Over a period of about 20 years she had seen gastroenterologists and surgeons, dieticians, nutritionists, homeopaths and other alternative therapists. She had been scanned dozens of times and had endoscopies to inspect the inside of her bowels and operations to examine the rest of her abdominal and pelvic organs. Her gall-bladder and appendix had been removed, gut hormones and secretions measured and fragments of her bowel repeatedly studied. Eventually the surgeon suggested that he removed half of her large bowel. Even after this extreme intervention her symptoms remained. Finally she was referred to a psychologist who traced her symptoms back to an incident in her adolescence when she woke up after a very drunken party in which she believed she had been raped, but could not prove it and had never told anyone about it. After a year of therapy her symptoms had significantly improved and she was able to enjoy sex with her husband for the first time.
This demonstrates the risk of doing everything possible to rule out a physical cause without considering the psychological causes of ill-health. Medicine still suffers from a Cartesian divide between the mind and the body. Patients problems are divided into either physical or physical at a very early stage and those with physical ailments are psychologically neglected and vice-versa. Not only are the psychological needs of patients with biological illnesses neglected, but patients with serious mental illnesses like schizophrenia die prematurely from treatable medical conditions. Doctors are increasingly specializing in single organs or a smaller, more technical range of procedures with the result that they are becoming poorer at looking after the whole person. This is what the psychoanalyst Michael Balint referred to as ‘the collusion of anonymity’ in which, “the patient is passed from one specialist to another with nobody taking responsibility for the whole person”
As I have detailed in a recent post “A world without health professionals”, the most significant part of a GPs job is to manage the interface between suffering -what the patient presents with- and illness or disease. Only rarely is it necessary to refer patients to a hospital specialist for medical management. Our job, far more than diagnosing disease, is diagnosing ‘non-disease’, in other words, helping patients to cope with illness and suffering without imposing a medical diagnosis. Because a GP sees patient-defined suffering, the incidence of serious illness is much lower than in patients who present to hospital specialists, ‘gatekeeping is an effective way to increase the prevalence of serious disease in the population of patients whom hospital doctors see. Since a high prevalence of disease in a population ensures that the positive predictive value of signs and symptoms is increased, having effective gatekeepers turns out to make the diagnostic task of hospital doctors easier.’ The problem arises when patients with medically unexplained symptoms present to hosptial doctors, unlike GPs their emphasis is on exploring the biological possibilities, and it is here, rather than at the point of GP to specialist referral that the costs of excessive medicine arise.
A successful referral?
What then of a more straight-forward referral, where the GP has made the diagnosis and has referred the patient for treatment?
I visited Gordon at home last week. He has been housebound since his knee replacement a year ago. When I initially referred him to the orthopedic surgeons 2 years ago he was just about able to hobble into my surgery on crutches, in severe pain from osteoarthritis. The operation was delayed for a year because the anaesthetist wanted a cardiologist to review his heart failure before the operation. The cardiologist wanted to repeat a barrage of investigations which involved trips to different hospitals all over london to get the scans they wanted. Gordon was advised that the operation itself was safe, but the anaesthetic carried high risks because of his heart failure. He was torn between pain and disability and the possible fatal consequences of an anaesthetic and we spent several consultations over the year discussing his fears. We never considered the possibility that his knee might not improve; that he would be more mobile and in less pain was taken for granted. According to Hospital Episode Statistics released this year, only about 50% of patients report an improvement after a knee replacement. I know Gordon wishes he had never had the operation. As a GP I need to know the likelihood of my patients’ improving after surgery so that I can help them make an informed decision. But telling Gordon that there was a 50:50 chance of improvement may have been much less significant that the risk of dying under the anaesthetic. This raises questions about the limitation of the government’s obsession with patient choice of hospital provider. The orthopedic surgeon is only one part of a complicated network of different specialists working together in a different way for every unique, individual case.
Can GP referrals be reduced?
GPs are naturally cautious and every review of GP referrals has shown that they refer far more patients than could be justified by evidence-based guidelines alone. My review of headache referrals suggested that up to 40% of referrals were unnecessary and this is in line with other research. What I could not tell from reading the referral letters was how many of these apparently unwarranted referrals were to reassure the GP and/or the patient and how many were for management that the GP could or should have been able to do themselves. A decision to refer a patient is only one of a range of choices that GPs and patients make together (described in a previous blog, What’s the point of patient choice? ) After all these other decisions, the decision to refer may not have been made lightly, but after considerable negotiation and overladen with great anxiety.
Guidelines are rightly described as ‘guidance for the wise and rigid adherence for the intellectually moibund’. Clinical judgment cannot be simply over-ridden by guidelines as a reason for referral. One reason GPs defend using clinical judgement to sidestep evidence based guidelines is that patients such as young people with cancer present with atypical symptoms that do not fit the guidelines.
At the moment GPs throughout the NHS are reporting ‘enormous pressures’ to reduce referrals. Methods range from the draconian; being limited to four referrals a week, to the bureaucratic; referral management centres run by international businesses like United Health or Humana, whose profits depend on reducing the numbers of referrals. Earlier this year, the UK division of US health insurance giant UnitedHealth sold its GP practices to The Practice, focusing instead on providing commissioning support (referral management) to GP consortia. UnitedHealth also own a program called Script Switch, which automatically suggests cheaper alternatives when a GP types in a prescription. It can also be used to restrict the range of prescribable drugs. UnitedHealth have identified that commissioning, rather than general practice, is where the profits are. They have a history of restricting care in the US where were found guilty of fraud after they underpaid millions of people in New Jersey, Florida and California after determining insurance reimbursements for out-of-network care.
Update 11.11.2011: GP commissioneers impose locum referrals ban Pulse
Update 17.11.2011: Serious incidents at referral gateways Pulse.
A review of the literature on GP referrals concluded:
Targeting high or low referrers through clinical guidelines may not be the issue. Rather, activity should concentrate on increasing the number of appropriate referrals, regardless of the referral rate. Pressure on GPs to review their referral behaviour through the use of guidelines may reduce their willingness to tolerate uncertainty and manage problems in primary care, resulting in an increase in referrals to secondary care. The use of referral rates to stimulate dialogue and joint working between primary and secondary care may be more appropriate.
Many practices are telling patients that they cannot refer them until they have had a meeting with the other GPs in the practice to decide whether or not it is necessary. But there are too few examples of GPs and local hospital specialists working together.
My experience of headache referrals suggests that if the quality of referrals is improved, then the quantity will reduce. This needs peer support and appraisal from GPs who understand the pressures they are under. It also needs close collaboration with the local hospital specialists, so that both may benefit from the eachothers very different expertise, the GP with their knowledge of the patient and the specialist with their knowledge of the disease.
The pressure to refer patients is likely to increase due to factors beyond the control of GPs. These include the effects of social deprivation, health care marketing, ‘healthism‘ and the changing relationship between public and professionals.
Although there is evidence that many GP referrals could be avoided, there are patients who might benefit from referral who are not presently referred. The emphasis should therefore be on improving the quality rather than reducing the quantity of GP referrals.
There is a lot that can be done to improve the quality of GP referrals but the longstanding rift between GPs and specialists discussed at the beginning of this post remains a significant part of the problem. Payment by results and the purchaser-provider split adds to the problem by incentivising hospitals to over investigate and over treat, whilst pressuring GPs to under-refer, leading to mutual suspicion.
The ideological political ambition to open up the NHS to any willing provider of care, force patients to choose between these providers and have them compete with each other is anathema to collaborative, integrated care between GPs and their local hospital.
The enormous pressure that GPs are under, to save money rather than improve care, risks undermining the relationship of trust between doctors and patients. It risks referral decisions being overturned without consideration of the complex reasons behind the referral, in particular the psychological motivations of both GPs and patients.
And the perogative to reduce referrals in order to cut costs will put patients’ lives at risk.
The principle of referral: The gatekeeping role of general practice http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2233970/
General Practice: Past, present and future http://www.blackwellpublishing.com/content/BPL_Images/Content_store/Sample_chapter/9781405170703/Hastie%209781405170703_4_001.pdf
*Variation in GP referral rates: what can we learn from the literature? http://fampra.oxfordjournals.org/content/17/6/462.full
Patient pressure for referral for headache: a qualitative study of GPs’referral behaviour http://www.ncbi.nlm.nih.gov/pubmed/17244421
For an explanation of the issues, risks and benefits of screening see: http://www.healthknowledge.org.uk/interactive-learning/screening/chapter2
Why patients get unnecessary referrals. KevinMD blog
often leads to specialist referral and the use of
investigations, but little information is available on
the outcome of hospital referrals for haemoptysis and
even fewer data are available to guide cost effective
decision making in primary care”. Alarm symptoms in early diagnosis of cancer in primary care:
cohort study using General Practice Research Database
Breast and prostate screening NY Times http://nytimes.com/2011/10/11/opinion/cancer-screenings-are-a-gamble.html
Why hasn’t integrated care developed more widely in the USA and not at all in the UK? http://jhppl.dukejournals.org/content/36/1/141.abstract
Limits to demand for healthcare http://www.bmj.com/content/322/7288/734.full