Splitting in-patient and out-patient responsibility does not improve patient care

Summary

Over the past 15 years there has been a move away from consultants having responsibility for the care of patients both in the community and when in hospital towards a functional split in responsibility. In this article Tom Burns and Martin Baggaley debate the merits or otherwise of the split, identifying leadership, expertise and continuity of care as key issues; both recognise that this move is not evidence based.

For

Separating in-patient and out-patient responsibility in UK adult psychiatric care is now routine. What is striking about this reversal of long-standing practice is that it occurred with almost no detailed discussion. ¹ Splitting community mental health team functions in the 2000 NHS Plan ² was preceded by wide consultation following the National Service Framework. ³ Everybody is now aware of the ‘functional’ split and everyone has their opinions about it, but nobody seems to know who is behind it or quite why it happened. ¹

An entirely pragmatic decision

The split is certainly not evidence based, and there is not any theoretical basis for it that I can find. Nobody has shown in any scientific manner that patients (or staff) fare better with either system (although a recently initiated European study, COFI, comparing the two, may shed some light on this). In lieu of theory we find a vague belief that any increased specialisation in medicine is, in itself, a ‘good thing’. It is seen as a marker of improved knowledge and higher skills. It is not, however, a substitute or guarantee of such improvement, and the case for it must ultimately be pragmatic. Such a pragmatic decision must carefully balance the potential advantages of concentrating resources and refining practice against the consequent fragmentation of care. Fragmentation inevitably reduces access and brings a loss of both continuity and the whole-person perspective. This is something we know that patients (and most clinicians) value highly.

Three reasons for the split are commonly heard. First, articulated by a former College president, that UK in-patient wards had become unworkable and chaotic. ⁴ Second, usually from service managers, that it will somehow improve throughput and reduce bed occupancy. And third, from psychiatrists, that it will make our workload more manageable and reduce burnout.

Ward environment

It is hard to argue against the observation that our acute in-patient wards have become unacceptably fraught environments. However, there are important reasons for this that need to be confronted, not sidestepped. There are too few beds and no reconfiguration will magic that away. ⁵ Same-gender wards and the proliferation of different teams (early intervention services (EIS), assertive outreach teams (AOTs) etc.) mean that each consultant will have very few in-patients on a given ward at any time. Consequently our classical, rather theatrical, ward round may need to be rethought. How carefully have the alternatives been considered? The private sector has managed this situation for decades. I have seen much lighter-touch, individualised in-patient multidisciplinary reviews in other countries.

A more contentious issue, which few psychiatrists speak openly about, is the changing relationship between medical and nursing authority. As nursing has become a graduate profession with a more managerial structure, the ward has come to be seen as exclusively their domain. In some ways, of course, it always has been. However, senior medical authority with one foot outside the institution has previously exerted a powerful influence. The lessons of what happens when it is absent are all too clear from psychiatry's history. ⁶ An in-patient consultant is one response to this changed relationship but there are others that could possibly have worked just as well but with less disruption. Several options deserve exploring. These include identifying a coordinating lead from among the admitting consultants, or a dedicated senior staff-grade doctor. These alternatives may have been hastily dismissed precisely because they stir up concerns about authority.

Improved bed management

The belief that the split would lead to a reduction in bed-usage strikes me as utterly perverse. Even a cursory glance at variations in European practice shows that it is the countries that do not split this function (the UK until recently, Italy and now some parts of Scandinavia) who manage their beds most effectively. ⁷ Several trusts have already reported finding that there has been no impact since the split. This should come as no surprise. Admitting and discharging psychiatric patients is always a judgement about relative need, not simply the presence or absence of illness. It involves balancing the relative clinical needs of current in-patients with each other and also with those patients urgently needing to come in. ¹ It stands to reason that a consultant who is familiar with patients in both groups, with the strengths and weaknesses of the community mental health team, and with the immediate ward pressures will manage these transitions more justly and effectively. This is a much more difficult task for exclusively in-patient staff, never mind a hapless bed manager.

A more attractive job

Does it make the work easier? Maybe, but anecdotal reports suggest not always. ⁸ Some psychiatrists will certainly relish this more contained, acute style of work and I have noted previously the threat of two conflicting cultures evolving within our profession. ¹ However, there are also stories of dissatisfaction and even some boredom with an exclusively in-patient role and frustration with an exclusively out-patient one. The clinical skills needed for both in-patient and out-patient work are hardly beyond most of us. Although there may be satisfaction in a more contained, predictable and specialised role, professionals generally perform better when stretched. ⁹ A major service configuration introduced for the convenience to staff rather than for patient benefit is also surely something we should feel a little uncomfortable with.

Continuity of care – much ado about nothing?

Continuity of care, like beauty, may be hard to define but we all recognise it when we see it. ¹⁰ Patients (and that means all of us) want to be understood as individuals with our own unique personalities and histories. We do not want to be continually repeating our symptoms or being treated as ‘a case’. Henry Ford's production line is a poor model for modern medicine. ⁹ This is true of all medicine, but nowhere more so than in psychiatry. Building trusting relationships takes time and personal investment but it pays off. It pays off in improved patient satisfaction but also in more effective containment of anxiety, making for smoother transitions into and out of hospital.

A fundamental issue is raised here about what sort of profession we are. The all-too-easy surrender of continuity of care, devalues the central importance of the therapeutic relationship in our work. Building such relationships has long been the hallmark of a good psychiatrist and the cornerstone of psychiatric practice. If we jettison commitment to it, our profession could face a bleak future with stagnating recruitment. Endless ‘refinements’ in service structures may not be so much a manifestation of psychiatry's progress as a distraction from the current lack of progress. ¹¹

Time to rethink the functional split

So does anyone benefit from splitting in-patient and out-patient care? The numerous practical problems are obvious. These include the burden of extra information transfer, increased risks during transitions ¹² and its structural incompatibility with current community treatment order (CTO) legislation. ¹ If, and when, psychiatry becomes so technologically sophisticated that the necessary knowledge and skills are beyond one consultant then, of course, such a split will make sense and is to be welcomed. However, I can see no evidence that we are there yet. In the process we seem to have abandoned one of the UK psychiatry's most admired and formative qualities, its seamless continuity of care, for nothing substantial in return. It is time we rethought the functional split.

Tom Burns

Against

The majority of acute in-patient wards in England and Wales have moved to having dedicated consultant psychiatrists who spend all or the majority of their time on a ward (the so-called ‘functional’ split). Previously the model was typically sector based, with consultants responsible for the care of patients both in the community and when on the ward. This change has occurred over the past 15 years or so. Like many changes in health systems it is not based on any empirical research and lacks an evidence base. However, there is equally little of an evidence base to suggest the previous model was superior. There is also a tendency to look back on previous arrangements with ‘rose tinted’ glasses and forget the problems associated with them.

There has been a change in the character of acute wards that has contributed to the need for a functional split. There has been a gradual reduction in the numbers of in-patient beds (17% in the past 3 years) and this has been associated with an increase in the morbidity of the patients on the ward and the proportion of patients on the wards under a section of the Mental Health Act. ^13,14 Dedicated in-patient consultants are superior in my opinion to the traditional model in terms of ward leadership, throughput, patient safety, expertise and job satisfaction. Furthermore the criticism of this leading to a lack of continuity is overstated.

Ward leadership

Wards consume a significant proportion of the money allocated to mental health (over 50% of the budget for the 3% of patients under the care of secondary mental healthcare who are admitted at any one time). In-patient beds therefore are a precious, expensive and shrinking resource. They are also, especially in inner-city areas, often disturbed, stressful places both for patients and staff. In my trust, the move to a functional split was largely driven by the difficulties of providing effective leadership to the wards. In my experience the most highly performing wards are those with an excellent ward manager working in conjunction with a consultant psychiatrist with good leadership skills. John Dixon ward at Guys' was one of the first and Dratcu et al wrote of the benefits to the ward of have dedicated medical staff. ¹⁵ In a traditional model, with fewer beds and the effect of single-gender wards, it is likely to mean that each ward has to relate to a number of consultants, each with few patients. This creates difficulty for the nursing staff, psychologists and occupational therapists, as there is a requirement to support multiple multidisciplinary team meetings. Also there is the problem of which consultant (if any) is in charge of the ward as a whole. Sometimes there needs to be a discussion to discharge or transfer one of more patients to reduce conflict or prevent exploitation on the ward between patients, which is more difficult with multiple consultants. There are also likely to be differences in appetite for risk, prescribing practices and other idiosyncrasies in multiple different consultants that complicate life for the rest of the team. Finally, there is the question of junior doctors and who is their trainer and supervisor. If there are one or two core trainees they have to work with multiple consultants on one ward or with one consultant over many wards.

Throughput and patient safety through increased consultant presence on the ward

There also are the issues of throughput and patient safety. In the old traditional model typically a consultant would be on the ward only 2 days a week. This had the effect that if the consultant worked say Monday and Thursday on the ward, and a patient were admitted at 17.00 h on Thursday they might not receive a senior medical review until lunchtime on Monday. Of course there are ways to mitigate this but it will always be a potential issue. The consultant is a key figure in establishing the treatment plan and making decisions with regard to risk and discharge planning. The functional split allows a daily presence of a consultant. It might be possible to arrange this even with a traditional model but this is logistically difficult. Models such as the triage ward ¹⁶ were introduced with daily senior medical reviews that have demonstrated some reduction in length of stay (although not necessarily over the whole system). Likewise models of in-patient care in other parts of the UK have demonstrated more effective use of beds with daily multidisciplinary meetings with consultant presence.

With greater numbers of patients admitted under the Mental Health Act there is also an increasing requirement to attend Mental Health Act Review tribunals and assessments in the community that can be difficult to manage in split jobs. There is also the staff perception that the consultant is ‘never there’ and always in the other place (for example, ward or community).

Expertise

Many acute in-patient wards are more similar to psychiatric intensive care units in previous decades in terms of levels of violence and disturbance. These levels of violence in very disturbed patients require specialist expertise to manage and are in my opinion best delivered by a few specialised in-patient consultants who are expert in, for example, rapid tranquillisation. Such patients also require regular manipulation of their medication regime that requires daily not twice weekly attention. There are other particular skills, such as initiation of clozapine or detection of organic illness that would be potentially done better by consultants who do such things regularly.

Continuity of care

One frequent criticism of the functional model is a lack of continuity of care with in some cases different consultants looking after a patient in the community, sometimes with interim home treatment or crisis teams, and then the ward (and sometimes via a psychiatric intensive care unit too). However, in my experience for many patients the effective day-to-day care in the community is not provided by a consultant psychiatrist but by a care coordinator. The latter may then ask for a review of medication when required by a doctor on the team but the consultant is not in regular contact and therefore having a different in-patient consultant may not result in much practical difference to patients.

Increasing consultant satisfaction

There is evidence that contented staff provide higher quality care. Many in-patient consultants prefer the functional split rather than always being pulled between ward and community and feeling that they are never at either part of the job long enough to satisfy their colleagues and patients. The functional split can therefore prevent burnout, increase job satisfaction and hence patient care. There is also often a critical mass of consultants and trainees found in a hospital unit where the in-patient wards are that again is helpful in maintaining morale and improving consultant satisfaction.

Conclusion

The ‘functional split’ has evolved as a pragmatic solution to the changing nature of acute in-patient wards. There is little empirical evidence to support this development, but this is true of many service reorganisations. It is easy to look back wistfully at past practice, forgetting the associated problems. Beds are going to only become fewer in number and the patients admitted to them more disturbed. Admission will increasingly be an exception rather than the rule and therefore should be managed by consultants dedicated to this task with the time to do so and trained to have the specific expertise required.

Martin Baggaley

For: rebuttal

It is clear that neither Dr Baggaley nor I ‘know’ which of these two models works best. We have both considered the various issues and come to differing conclusions. Reassuringly we have based our conclusions on the same broad considerations – leadership and the ward environment, bed management and throughput, expertise, consultant satisfaction and lastly continuity of care. Neither of us, however, mentioned the inevitable and confusing differences of opinion on their management that patients will be exposed to across the divide. Nor the hours of fruitless wrangling about admission and discharge, or who should be responsible for the patient at any given moment. Both deserve consideration.

Bed management and consultant satisfaction

On bed management and consultant satisfaction Dr Baggaley is confident but I am not. There is really very little to go on and what there is, is contradictory. I hear stories of a significant turnover of in-patient consultants, so perhaps not so satisfied. Nowhere do I hear reports of revolutionary improvements in throughput leaving beds standing empty.

Undoubtedly a committed in-patient consultant offering strong leadership can do wonders for the ward environment (and probably, at least temporarily, for throughput). But how much of this is down to the new system and how much to the individual's personality? Psychiatric services research has a poor history in distinguishing between these two. ¹⁷ The result has been periodic upheavals chasing impressive outcomes reported by enthusiastic and charismatic pioneers, which then fail to stay the course when implemented by us ordinary mortals. Wise individuals generally wait and see whether the benefits of an innovation are sustainable and generalisable before recommending it as a blanket policy.

Outstanding individuals can make almost anything work well. As Dr Baggaley points out the best (and usually happiest) wards are those led by a strong consultant and an excellent ward manager. Long may it be so. But simply replicating what they do misses the opportunity of a careful exploration of the many different ways of running a ward. There may be many equally successful wards with a strong consultant and excellent ward manager who are managing things quite, quite differently.

Expertise and continuity of care

The issues of expertise and of continuity of care are what really divide Dr Baggaley's position and mine. What are we, as consultant psychiatrists, really supposed to be experts in? Of course we need skills in initiating clozapine or rapid tranquillisation but most of us can do this, we do not need our job re-engineered to ensure it. I would argue that our most important expertise is in engaging with, understanding, and helping very troubled, very ill, individuals. It is no mean skill to be able to negotiate with an angry manic patient or to persist with a negativistic depressed one. To do so requires an understanding of them that goes well beyond totting up their symptoms, and it takes time, sensitivity and commitment. That is where continuity of care comes in, being familiar with the patient across different phases of their illness, knowing what they have been through, and what matters most to them.

That there is much more to being a good psychiatrist than being a good mental health technician has been regularly pointed out from Jaspers over a century ago through to the radical psychiatry movement of today. We cannot treat our patients without relating to them. Building that therapeutic relationship is our profession's hallmark skill and some continuity of care is essential for achieving it.

Dr Baggaley says that ‘ … a lack of continuity is overstated’. I wonder. Cornwall Partnership NHS Foundation Trust is one of the few that have persisted with integrated teams, because they believe that continuity really matters. For a couple of the first 5 years of the Care Quality Commission National Inpatient Survey in mental health they found themselves top out of the over 60 trusts in England on some of the four questions on satisfaction with psychiatrists. They were also recognised as being very efficient with their beds. Was this patient satisfaction in part because they offered continuity? To find out they examined the results of the survey, identifying those trusts with either exclusively functional or integrated services. In the 26 who replied they found that, on all four questions, patients on wards served by integrated teams reported marginally higher satisfaction. ¹⁸ Remember these are in-patients being asked; those who should be favoured by the split. Of course, the survey is tiny and the authors avoid making any great claims, but it should give some pause for thought. There is certainly no evidence in the other direction.

I have little doubt that in 30 years our services will look very different from how they look today. There is no reason at all why integrated teams must survive. However, they have served us, and our patients, well (consistently performing as well as various expensive alternatives). ¹⁹ When changes do come let them be based on tried and tested improvements that are generalisable, convincing and sustainable. Such changes will be warmly welcomed. Unfortunately, none of these conditions can really be claimed to apply to our current functional split.

Tom Burns

Against: rebuttal

I would agree that the issues of expertise and of continuity of care are what really divide our positions. The reality, of course, is that there is not a perfect solution that satisfies everyone for all aspects of care. If we moved to a very USA model, with a very short length of stay and tightly protocol-driven admissions (also known as ‘managed healthcare’), it might be possible to do away with the functional split and have the ward effectively run by nurses, a middle grade to prescribe and monitor medication and the major psychiatric input provided by the patient's community consultant. However, while we have wards, often with a long length of stay, in which the throughput is driven largely by the consultant, my sense is that on balance, the system is more efficient with a dedicated in-patient consultant despite the adverse impact on continuity of care. Whatever we do, the focus must be on improving community teams and out of hospital care. My view is that our psychiatric resources should be concentrating on that and therefore to some extent whether or not we have a ‘functional’ split or not, is a distraction. The real gains are in keeping patients well such that they do not need an in-patient admission in the first place.

Martin Baggaley