Article Text
PDF
Statistics from Altmetric.com
Learning objectives
Gain an overview of the evidence and guidelines relating to a multidisciplinary approach to heart failure care.
Understand what constitutes multidisciplinary care.
To consider how heart failure care might be organised, understanding that this will vary in different healthcare settings.
Introduction
Heart failure (HF) is a complex, relapsing, severe chronic disease. It causes multisystem dysfunction and results in a high burden of morbidity, mortality and healthcare costs. It is common, affecting ≥10% of the people aged over 70,1 and despite advances in cardiac care the incidence probably remains stable.2 In the UK, it is estimated to be the primary diagnosis in 5% of acute admissions and costs the National Health Service approximately £2 billion/year (~2% of the total budget).3 Furthermore, outcomes for those with both chronic stable disease and acute HF remain relatively poor, comparing unfavourably to many malignant conditions.4
A multidisciplinary team (MDT) approach is considered the gold standard model for the delivery of HF care. However, to shape effective multidisciplinary HF services, it is crucial to understand the relevant evidence base, the essential components of an MDT approach and how services might evolve.
Main text
Since HF results in multisystem dysfunction, it seems intuitive to adopt an MDT approach to care, which recognises this complexity and allows various aspects of the illness to be addressed by the most appropriate healthcare professionals. The aim is to provide individualised, holistic care, which is responsive to the changing needs of patients throughout the course of the illness and seamlessly transitions primary and secondary care. In other words, it allows patients to receive the right care from the right person at the right time.
MDT care is often part of disease management programmes—a term covering a broad range of structured treatment plans and interventions for patients with chronic disease. However, while these programmes are frequently delivered by an MDT, this is not always the case and therefore the terms are not interchangeable.
MDT care is strongly recommended by the National Institute for Health and Care Excellence (NICE),3 the European Society of Cardiology (ESC) (class 1A recommendation)1 and the American College of Cardiology Foundation/American Heart Association (ACCF/AHA) for high-risk patients (class 1B recommendation).5 However, little detail is provided in the guidelines as to what constitutes MDT care.
The evidence
Since the 1990s, over 10 000 patients with HF have been included in the studies of MDT care. These range from small cohort studies to randomised controlled trials and meta-analyses. Since there is no standard definition for MDT care, the studies are diverse with the timing, nature and intensity of the intervention varying considerably. It can also be difficult to understand whether some interventions truly represent MDT care. Pooling the evidence on an MDT approach is also made difficult by the heterogeneous nature of the studies. Attempts to do so inevitably result in oversimplification of the literature; interventions are categorised into a few broad categories which are inconsistent between meta-analyses. Categorising studies differently will also affect any conclusions about the efficacy of various types of MDT care.
Further complexity is added because what constitutes standard care in control groups is frequently ill-defined and is also variable. The UK National Heart Failure Audit6 demonstrates that patients admitted with HF receive extremely variable care within a relatively small geographical area under a single healthcare provider. The standard of usual care in the control group will influence the magnitude of benefit conferred by the intervention. Patients receiving lower quality usual care are more likely to benefit from MDT care and thus the benefit of the MDT will be exaggerated. The relevant evidence must be interpreted in the context of these limitations.
Study patients
Most studies have included modest numbers of patients, mainly 50–350 subjects,7–9 although one of the more recent single-centre randomised controlled trial included 1023 patients.10 In keeping with the broader HF literature, subjects tend to be younger than real world patients (mean/median age range of 56–80, mostly late 60s or 70s).
Studies have been conducted almost exclusively in developed countries. Many are single centre with the associated limitations. Patients with serious comorbidity, such as cognitive impairment, are also among the groups underrepresented.
In contrast to most drug and device therapy studies, left ventricular ejection fraction has usually not been an eligibility criterion and frequently not measured or reported. Consequently, a substantial number of HF with preserved ejection fraction (HFPEF) patients will have been included. However, separate analyses evaluating the magnitude of the benefit in these groups are lacking.
Timing of the intervention
Investigators have usually targeted patients at high risk of adverse events. These patients will potentially benefit the most and higher event rates allows for smaller sample sizes. Thus, the vast majority of published data relate to MDT interventions in patients who have been hospitalised with HF. Most of the MDT interventions are usually outpatient programmes, typically beginning at the time of patient discharge. This is reflected in the ACCF/AHA guidelines5 which recommend multidisciplinary HF disease-management programmes for patients at high risk of hospital readmission. ESC guidelines recommend MDT care for all patients with HF, although a footnote specifies that high-risk individuals should be targeted.1 Given the high mortality among patients hospitalised with HF, it is surprising that studies of MDT intervention early during a hospital admission are lacking.
Personnel delivering the interventions
Nurses, in particular specialist HF nurses, are central to the delivery of MDT care in the majority of studies.10–12 Physicians are usually involved, although the extent of their involvement is often not clear. Models in which other healthcare professionals, such as pharmacists, deliver care have also been described.13 14
Nature of the intervention
Interventions can be delivered by telephone alone, in hospital/community clinics or in the patient’s home. Studies often employed more than one approach and with varying frequency.
The exact nature of the intervention varies but includes some of the following:
Education, self-monitoring and management: Education about HF symptoms, signs and prognosis and lifestyle advice. Teaching patients to monitor parameters such as weight and blood pressure. The MDT can define changes in these parameters which should prompt the patient to respond, for example, a weight gain of more than 2 kg in a short period prompts an increase in loop diuretic.
Telemonitoring: Collection and transmission of physiological parameters to the MDT, for example, weight, blood pressure, heart rate, pulse oximetry, respiratory rate. Healthcare professionals respond to deviations.
Structured telephone support: Education, advice and support delivered via the telephone according to a predefined protocol. For example, delivered at set time points with clear goals to achieve each call and set actions in response to the findings.
Clinical review: Can include basic investigations including blood tests and ECGs in addition to history and examination.
Medication management: Encouraging patient understanding and ownership of drug therapy. Monitoring for clinical response and adverse effects. Non-medical prescribers are frequently used.
Cardiac rehabilitation: Exercise-based interventions, alone or as part of a comprehensive programme. Individualised, but often delivered as a group intervention.
Additional components of MDT care: There are other aspects of MDT care including psychological support and palliative care which are less well defined in studies but occur to some extent with all patient interactions. They are not usually listed as part of the protocol in trials or measured as outcomes but are undoubtedly valuable components of the model.
Effectiveness of interventions
In common with many drug and device studies, those evaluating the impact of an MDT approach to HF care have used hospitalisation (both HF related and all-cause) and mortality as end points. To a lesser extent, they have also assessed patient well-being.
Hospitalisation and mortality
A minority of studies have demonstrated that MDT care reduces hospitalisations11 12 15 and mortality.12 15 The majority of individual studies, however, have shown that MDT care has a neutral effect on these outcomes. This may reflect relatively small sample sizes. However, the Coordinating Study Evaluating Outcomes of Advising and Counselling in Heart Failure (COACH) study, which is probably the largest study of a genuine HF MDT intervention,10 was also neutral. This single-centre Dutch study of 1023 patients demonstrated that the addition of basic or intensive support by specialist nurses to routine follow-up with a cardiologist after a HF admission did not result in a reduction in the combined end point of HF hospitalisation or mortality. It is, however, noteworthy that standard care post-HF admission was high in both quality and frequency, involving at least four follow-up visits with a cardiologist during 18 months of follow-up. Control group follow-up was also intensified for many patients—presumably those who were decompensating. While the event rate was quite high (29% mortality in the control group), it was relatively low compared with many other studies and real world data from the UK National Heart Failure Audit.6 High quality standard care and lower event rates may therefore have negated any significant additional benefit from MDT care.
On closer review, the majority of neutral studies show the same signal, namely, a non-significant trend towards reduced hospitalisations and/or mortality with MDT care. Indeed, when the literature is pooled and analysed collectively in meta-analyses and systematic reviews, the CIs narrow and these trends become significant prognostic benefits7–9 16 (table 1). Two relevant systematic reviews on the impact of MDT care were published about a decade ago.7 8 Pooling such diverse studies has limitations as discussed above and has been challenged.17 In recognition of the heterogeneity of what constituted MDT care, both analyses attempted to categorise interventions into groups. However, interventions were categorised very differently between the reviews making comparisons difficult. McAlister et al 7 concluded that interventions involving specialist MDT follow-up (in clinic or at home) reduced mortality and both all-cause and HF hospitalisations. Furthermore, strategies involving enhanced patient self-care or telephone-based follow-up reduced hospitalisation (all-cause and HF hospitalisation and HF hospitalisation only, respectively) but not mortality. Holland et al 8 also concluded that MDT interventions reduce mortality and both all-cause and HF hospitalisation. However, in contrast, the mortality benefit was driven by those who received remote monitoring and telephone follow-up rather than those receiving face-to-face input in the hospital or in the community.
Meta-analyses on the impact of MDT care on hospitalisation and mortality in HF
The most contemporary meta-analysis is a Cochrane review9 on clinical service organisation for patients with HF. Studies with a minimum of 6 months of follow-up were included. Interventions consisting of purely telemonitoring/telephone support, exercise/rehabilitation, primarily education and those targeting generic chronic diseases were excluded. Interventions were classified as case management, clinic or multidisciplinary and could be performed in an inpatient, outpatient or community setting. Using this classification scheme, 17 studies (and one arm of an 18th study) were categorised as case management studies, six as clinic models and only two studies were described as employing MDT interventions. However, all 25 studies employed what would generally be recognised to be multidisciplinary interventions and studies defined as MDT interventions in other meta-analyses were categorised differently in this review. The review contains a comprehensive table summarising each of the studies included in the analysis.
In keeping with the previous meta-analyses, the overall finding of the Cochrane review was also a significant reduction in mortality and HF hospitalisation. The greatest benefit was seen in patients receiving interventions categorised as case management, while the data for clinic-based interventions was less strong, but also more limited. More intensive interventions (regardless of study category) conferred a mortality benefit, whereas the least intensive interventions did not.
Overall, the studies included in the review were judged to be at a low risk of bias, although there was insufficient information to determine this for many of the criteria examined. It seems unlikely that any one study skewed the findings given that study sizes were all comparable and modest in size. Furthermore, the larger COACH10 study described above, in addition to being neutral, was subdivided into basic and intensive intervention groups for the purpose of this analysis.
There is some overlap between the three meta-analyses described but perhaps not as much as might be expected. Fifty-three different studies were included overall with nine studies included in all three analyses and 13 in two of the analyses, whereas 31 studies were only included in one of the meta-analyses. Despite including different datasets and classifying interventions differently, the findings in favour of MDT care remained remarkably consistent. Both meta-analyses showed a similar relative risk reduction in the order of 20%–25%. The Cochrane review reports an OR, which will be of comparable magnitude but may be somewhat lower than relative risk. Although direct comparisons cannot be drawn, similar magnitudes of benefit have been demonstrated with other important medical and device interventions in HF18–22 (figure 1). While the MDT data comes from meta-analyses, drug and device studies have enrolled patients with HF and a reduced ejection fraction (HFREF) who are potentially at higher risk of adverse events, and have involved longer follow-up periods. Patient follow-up is short in MDT studies with almost all ≤1 year. Furthermore, it is conceivable that the benefits may be enhanced further if the MDT is involved earlier during a HF admission. One recent single centre, non-randomised, observational study suggested that this may be the case with the potential for large benefits in these patients.23
Relative risk reduction associated with various interventions for heart failure. Data from Studies of Left Ventricular Dysfunction (ACE inhibitors)18; Cardiac Insufficiency Bisoprolol Study (betablockers)19; Randomised Aldactone Evaluation Study (spironolactone)20; Cardiac Resynchronisation in Heart Failure Study (cardiac resynchronisation therapy (CRT))21; Sudden Cardiac Death in Heart Failure Trial (implantable cardioverter defibrillator (ICD))22 and meta-analysis data (multidisciplinary team (MDT)).7
Another Cochrane review24 concluded that exercise-based cardiac rehabilitation also reduced HF and all-cause hospitalisation but not mortality. Further studies, with mixed results, have also been published, some with very large numbers of patients.25 26 However, many of these studies are not MDT interventions for patients with HF. Some are not HF specific, while other interventional disease management programmes do not always involve MDT care.
Other outcomes
The MDT approach is also likely to have an impact on patient well-being and the cost-effectiveness of care delivery. These, more complex outcomes, have been evaluated less rigorously. It is also more difficult to pool results and thus harder to draw firm conclusions. Many studies have suggested a favourable impact on quality of life, although this has not always been statistically significant. Cardiac rehabilitation specifically, which is delivered by MDT members, improves quality of life.24 Equally, many studies have reported interventions to be cost saving, but formal cost-effectiveness analyses are lacking.7 27 28
What aspects of MDT care confer benefit?
Unfortunately, it is simply not possible to accurately distil from the data which exact component(s) of the MDT provide the benefit. However, the majority of interventions, and therefore the weight of evidence in favour of an MDT approach, used a model delivered by nurses, with physician support. It could be speculated that a major contribution to the benefit will be better prescription rates, and adherence to, evidence-based pharmacotherapy. Holland et al 8 argue that as the benefit of the MDT approach has been consistent over years when prescribing rates of evidence-based therapy were increasing that this suggests that other aspects of MDT care are important. While this is an attractive hypothesis, good data on prescribing rates of these drugs is absent and so this remains speculative.
One aspect of HF care, with a limited evidence base,29 is the prescription of diuretic therapy. Treating congestion is one of the key aims of HF care, but having the confidence to give high-dose diuretic therapy, particularly in the setting of renal impairment, requires specialist experience. Masters et al 23 reported a striking mortality benefit with specialist MDT care in the inpatient setting. Patients managed by the MDT received much higher doses of loop diuretics, while other drug prescription rates were more comparable between the control and intervention groups.
Translating the evidence into clinical practice
Since the evidence must be applied to a range of diverse healthcare environments, there will not be a single optimal model. Rather a programme including some of the interventions described above should be tailored to the local healthcare system and individualised for each patient.
The MDT should be led by a HF specialist, usually a Cardiologist. HF specialist nurses are a vital part of the clinical team and are responsible for bridging the gap between primary and secondary care. They need to be easily accessible to the patient and other MDT members. Care needs to be coordinated carefully with the primary care doctor. This integrated approach, transitioning artificial healthcare boundaries is fundamental for success. We consider these healthcare providers to be key members of all HF teams. Numerous other healthcare professionals can be valuable members of a HF MDT. A team tailored to their operating environment will include many of those listed in box (which is not exhaustive).
Heart failure multidisciplinary team members. Bold indicates key multidisciplinary team members.
Heart failure physician specialist.
Heart failure nurse specialist.
General practitioner/primary care doctor.
Pharmacist.
Physiotherapist.
Palliative care.
Psychologist.
Occupational therapist.
Administrators.
Teams with access to the broadest range of skills are best placed to respond to the varying requirements of the patient. However, the exact roles of individuals within the MDT will vary and tasks traditionally performed by certain professionals are frequently performed successfully by other team members. For example, many HF nurses are independent prescribers, while pharmacists can perform clinical assessment. Furthermore, while HF physicians and nurses will be involved throughout the patient’s illness, input from others will predominantly be required at certain times, for example, those involved with rehabilitation in the earlier stages and those with palliation at end of life.
To be most effective, HF care should be delivered by the MDT as early as possible in a patient’s journey. For outpatients, this requires primary care to have easy access to screening tools including natriuretic peptides and rapid access to a HF specialist as needed. HF specialists need to be able to lead management but also support and facilitate the MDT to take a central role. Inpatients with HF are at highest risk and require most intensive management. Hospitalisations are also an opportunity for intense and repeated interventions including education, support and drug and device optimisation. The MDT should be involved as early as possible in this process with input from doctors, nurses, pharmacists and therapists all likely to be valuable. At discharge, transition of care to community-based team members needs to be carefully planned and communicated to all relevant team members and the patient and their family/carers. During the entire course of the disease, patients require easy access to the MDT both for routine and unscheduled care. The team needs to be responsive to the evolving needs of the patient over the course of their illness. Several professionals will be involved with each patient but it is also necessary to provide continuity of care so that the prerequisite trust and therapeutic relationships can be built.
The future of the MDT approach to HF
Service delivery
There is the potential for substantial gains through improving existing HF services. HF has historically had a relatively low profile despite the associated poor prognosis and the very significant impact that appropriate care delivered by experts has on morbidity and mortality. There is currently almost universal underprovision of HF services. Education of the public, the mainstream media, policymakers and other healthcare professionals is urgently needed to influence change. The short-term goal should be the establishment and strengthening of specialised MDTs in all healthcare institutions managing patients with HF. There is considerable variation in the services that are currently available with patients receiving inequitable access to care. In the UK, for example, we are starting to see guidelines30 and policies to encourage this service development. While these standards do not represent complete HF care and may skew priorities, they are relatively easy to measure and are seen as demonstrating that good clinical care is being applied more broadly.
In keeping with many healthcare systems, HF MDTs tend to be reactive. Both patients with de novo diagnoses and those with known HF tend to have contact with the MDT after decompensation, whether in the community or in hospital. Given the poor prognosis associated with decompensation, and the associated suffering for the patient, another challenge for the MDT is to develop a more proactive approach to try and prevent acute HF episodes. For example, a common scenario is when patients undergo surgery. The HF MDT should be involved early to provide appropriate advice about fluid balance and medication management perioperatively and postoperatively. Unfortunately, it is all too common for the HF MDT to be contacted too late when the patient has already decompensated.
Since decompensation frequently results in prolonged and costly admissions, and with the anticipated rise in the prevalence of HF, there is a need to develop alternative models of care. This includes managing ambulatory patients with acute HF in an outpatient setting. The HF MDT will be a key to providing the intensive support that such patients will need to be managed safely. Limited early data suggest that intravenous diuretic therapy can be safely and effectively delivered by an MDT to outpatients.31 Changing systems of care for patients with acute HF will undoubtedly represent a significant organisational challenge to many.
Remote monitoring
This includes telemonitoring and haemodynamic monitoring via either an implantable device (cardiac resynchronisation therapy and implantable cardioverter defibrillators) or a dedicated implantable monitor. The largest and most contemporary studies have failed to demonstrate that telemonitoring alone can reduce hospitalisation or mortality.32 33 MDT interventions using telemonitoring alone cannot therefore be recommended as effective. However, many HF MDT studies used telemonitoring and telephone support as part of their approach. The increasing burden of HF means that telemonitoring is likely to be increasingly used as part of a broader MDT management strategy.
Remote monitoring via pacing devices has also been disappointing and studies have failed to demonstrate an improved prognosis.34 35 Conversely, limited data from a dedicated implantable pulmonary artery device (CardioMEMS) in selected patients is more encouraging. Device use was associated with reduced HF hospitalisation in patients who were both very symptomatic (New York Heart Association III) and at high risk of adverse outcome.36 More refined use of remote monitoring coupled with the advent of smartphones, allowing for data collection by patients, may yet improve outcomes. However, it also has the potential to generate a huge data burden which could easily overwhelm existing systems. Appropriate MDT involvement in the response to the data generated has been lacking in many of the neutral remote monitoring studies to date. HF MDT members will be a key to any potential successful application of these technologies and will need to make time to interpret and respond to these data.
Conclusions
HF is a complex condition and an MDT is best placed to manage the multiple needs of patients and to optimise outcomes. There is an extensive published literature evaluating the role of the nurse-centred MDT. The exact nature of who should make up an MDT and what they should do cannot be defined too precisely. Moreover, it is not desirable to exactly replicate a single model rather the principles of an MDT approach should be broadly applied to fit local needs. More research to better understand which MDT interventions are most effective and evaluating the role of inpatient HF MDTs is urgently required to shape developing services. However, a successful MDT approach to HF care results in a sizeable prognostic benefit with reduced rates of hospitalisation and death for patients with HF, regardless of left ventricular systolic function.
Key messages
Multidisciplinary team (MDT) care is the gold standard approach for managing heart failure (HF) which causes multisystem morbidity.
The MDT should be accessible throughout the illness and responsive to changing patient needs.
Pooled analyses of MDT care demonstrate significant reductions in mortality and/or all-cause and HF hospitalisation in patients at high risk of adverse outcomes.
There is significant heterogeneity between the nature, duration and intensity of MDT interventions.
The exact nature of the MDT intervention and the quality of routine care will have a substantial effect on the impact of an MDT approach.
It is not possible to tell which component(s) of the MDT provide the benefit. However, the weight of the evidence supports a HF-nurse centred model.
Studies demonstrating the prognostic benefit of MDT care included unselected patients with HF including those with both HF with reduced ejection fraction and HF with preserved left ventricular ejection fraction.
Most MDT studies recruited patients at high risk of adverse events, at the point of, or shortly after discharge. There is a paucity of data investigating the impact of an MDT approach during a HF hospitalisation.
All major guidelines strongly support an MDT approach for high-risk patients.
An investment in care systems is needed to provide the necessary HF MDTs.
MDT care needs to transition to a more proactive approach.
CME credits for Education in Heart
Education in Heart articles are accredited for CME by various providers. To answer the accompanying multiple choice questions (MCQs) and obtain your credits, click on the ‘Take the Test’ link on the online version of the article. The MCQs are hosted on BMJ Learning. All users must complete a one-time registration on BMJ Learning and subsequently log in on every visit using their username and password to access modules and their CME record. Accreditation is only valid for 2 years from the date of publication. Printable CME certificates are available to users that achieve the minimum pass mark.
References
Footnotes
Contributors GM is responsible for drafting and revising the article. JM and PC critically appraised the manuscript and were responsible for significant editing of important content.
Competing interests None declared.
Provenance and peer review Commissioned; externally peer reviewed.
Author note References which are listed with a * are key references for this paper.