Managed care: Disease managementBMJ 1997; 315 doi: https://doi.org/10.1136/bmj.315.7099.50 (Published 05 July 1997) Cite this as: BMJ 1997;315:50
- Correspondence to: Professor Hunter
The disease management approach to patient care seeks to coordinate resources across the healthcare delivery system. The growing interest in evidence based medicine and outcomes, and a commitment to integrated care across the primary, secondary, and community care sectors, all contribute to making disease management an attractive idea. A combination of patient education, provider use of practice guidelines, appropriate consultation, and supplies of drugs and ancillary services all come together in the disease management process. But its effectiveness is largely untested, so evaluation is essential.
Disease management is often regarded as one of the ways of achieving managed care, but it can also be viewed as a stand alone mechanism aimed at improving the cost effectiveness of care. Clinical pathways and integrated care packages are other terms used to describe a disease management approach. Examples of how a good disease management programme would work for a patient with diabetes mellitus is shown in the case history given in boxes throughout the paper.
What is disease management?
Disease management views patients as entities experiencing the clinical course of a disease, rather than viewing their care as a series of discrete episodes or as fragmentary encounters with different parts of the healthcare system. It has three parts:
A knowledge base that quantifies the economic structure of a disease and includes guidelines covering the care to be provided, by whom, and in what setting for each part of the process;
A care delivery system without traditional boundaries between medical specialties and institutions; and
A continuous improvement process which develops and refines the knowledge base, guidelines and delivery system.1
It is a structured systems response to a set of problems which are evident to some degree in all health- care systems. These include a fragmented and uncoordinated set of arrangements for delivering care, a strong bias towards acute treatment, a neglect of preventive care, and inappropriate treatment. In an attempt to overcome some of these problems, disease management is outcomes led. This is its major strength and its major weakness. The weakness lies in our incomplete knowledge base. For many conditions there is no consensus about outcomes, or whose outcomes should prevail—those of professionals or patients.
For which diseases is a managed approach suitable?
Disease management is most suitable for the diseases about which most is known, for which it is easy to develop disease protocols that are evidence based, and in which it is possible to measure outcomes. The most typical disease favoured for this approach is diabetes, followed by heart disease and cancer. Stroke, asthma, mental health (including depression), prostate disease, and dermatological diseases are also often candidates. Angina, AIDS, cystic fibrosis, hypertension, renal dialysis, substance misuse, and peptic ulcers are less commonly managed. Senior NHS managers surveyed in 1995 said that the principal reasons for choosing diseases are a high local incidence of disease; the need for integrated guidelines and systems in primary and secondary shared care; the high cost of treatment; a requirement to improve guidelines; the lack of certainty in best practice; and the need to improve patient outcomes.2
The spectrum of disease management extends from health promotion and disease prevention, through diagnosis, treatment, and rehabilitation, to long term care. No published research has explicitly evaluated disease management programmes.3 The claims for disease management made in most papers have not been tested empirically, and research has been confined to hospital based interventions. This tends to undermine the population based approach which is central to the disease management philosophy.
Setting up a disease management programme
To an extent, disease management is little more than a marketing or packaging device whereby familiar and often long standing concepts are combined into a single philosophy or approach and offered as a complete package. The success of a disease management programme depends on several factors: committed managers, an organisation prepared and willing to take this route, a structured process of change management, a structured approach to analysis, a well developed performance management system—with the patient at the centre of the process.
The purpose of the programme must be clear at the outset, and the organisational structure within which the programme will operate must be established. The skills and resources required must be identified, as well as the diseases to be managed. Links and alliances must be in place at the outset and everyone involved in the process must understand what is going on. It is also vital to review the evidence for each disease in order to ascertain what is known about the disease in each sector of care. Each disease should be broken down into its constituent elements and protocols produced for each stage. How outcomes are to be measured for each stage of the disease and across all the stages must be determined. Finally, the whole process should be piloted and independently evaluated.
As with any major change, top managers have to be seen to be taking a lead. But without the ownership and commitment of staff, any new developments will quickly fail or become distorted. Good communication is critical to the success of disease management.
Preparing for the change process requires analysis and an objective assessment at the outset of what the problem is and how it might be tackled. Putting the patient at the centre of the process allows bottlenecks and problems of coordination to be identified. Finding solutions to some of these obstacles and blockages may entail radical change—for example, it may be necessary to address poor patient preparation, inappropriate staffing, and work planning routines, none of which are new to the NHS or to its managers, but unless these problems are confronted and resolved, disease management will not succeed.
Features of successful disease management
Patient centred and outcome focused
Objective and evidence based, to ensure credibility and acceptance
Flexible and pragmatic, to account for normal variations and uncertainties in medical practice at patient level
Dynamic, to enable evolution and durability
Based on what happens in the real world and not on what ought to be happening
Designed to minimise difficulties and maximise benefits
Skills and tools needed for disease management
Disease management programmes depend on a diverse set of skills and tools.6
Knowledge base—Up to date information about disease in terms of epidemiology screening, prevention, pathology, and treatment options is clearly essential. Many stakeholders will already be knowledgeable but there may need to be even more specialisation around particular conditions for greater cost effectiveness.
Outcomes research—In many ways outcomes research is the raison d'être of disease management. It means measuring quality, service or satisfaction, and cost outcomes for any course of treatment, and disseminating this information to develop clinical guidelines and protocols. But this poses a dilemma. Few systems based outcomes data are available, and even when they are, changing clinicians' behaviour will demand management skills of the highest order. Some managers are attracted to the idea of partnerships or joint ventures with pharmaceutical companies that believe they possess these skills.7
Framework for assessing merits of joint ventures in disease management
Patient issues—patients' interests should be paramount
Ethical issues—there should be no conflict with ethical requirements of practitioners to provide whatever treatment they consider clinically necessary for an individual patient
Protection and use of patient information
Legal issues—NHS parties should satisfy themselves as to the legality of any proposed venture
Transparency and accountability—services specified in a joint venture should be published and NHS parties held accountable for them
Finance issues—joint ventures should represent value for money to the NHS
Evaluation—joint venture schemes should include arrangements for monitoring and evaluation
Information systems—Disease management needs all stakeholders to have access to integrated information so they can all understand treatment options, long term costs, and outcomes. Collecting and sharing data must be a priority—currently, information is often unreliable and inaccessible.
Tools for influencing behaviour—Effective, successful integrated health care systems require stakeholders to change their behaviour. For instance, providers will need to respond to clinical practice guidelines—and provide good reasons if they are not prepared to follow them—and patients will need to take compliance more seriously. Some pharmaceutical companies believe they can contribute by influencing the behaviour of patients and providers through educational programmes and other devices.
Continuous quality improvement—Measuring performance against accepted benchmarks will allow the system to be continually refined through regular evaluation.
Ability to share and manage financial risk—With stakeholders collaborating in new ways to care for patients, there will be a different distribution of risk, and incentive structures will need to ensure that all stakeholders are working for the same ends. Obstacles include the lack of useful information on the health status of populations in local areas and the difficulty of assigning diseases which have a number of sequelae and comorbidities into specific patient pools (eg diabetes, asthma) and of calculating the expected risk and financial liability.
A diagnosis of diabetes mellitus
Jane, aged 12 years, presents to her general practitioner with polyuria, polydipsia, and weight loss. The doctor makes a rapid diagnosis of diabetes mellitus and refers her to the diabetes centre in line with locally agreed guidelines. The next day Jane attends the diabetic centre with her mother. The consultant diabetologist recommends an insulin regimen and introduces Jane to the diabetic specialist nurse, who teaches her how to inject herself using a pen device and how to monitor her blood glucose. Jane is introduced to the diabetes dietician and watches a video on diabetes supplied by a pharmaceutical company, which is accompanied by a helpful information pack on compliance with treatment. The diabetic team discuss the symptoms and management of diabetes with Jane, and targets for her self care are developed. She is given her patient held record as well as patient information packs, including patient guidelines appropriate to her age.
Over the next two weeks the diabetes nurse visits Jane at home to monitor Jane's progress and contacts the practice nurse, who helps administer the monthly diabetic clinic at the practice's surgery. Jane's school is contacted to reassure the teachers that Jane can take part in all normal activities and to provide them with information about diabetes. The primary and secondary care teams are in close contact.
There are two further critical elements of a disease management approach. It demands a long term perspective, but results are focused on short term gains and improvement—and although there should be a focus on values rather than cost, the danger is that in practice this will be reversed. Cost data are more reliable than data on quality, and most information systems are designed around financial rather than clinical outcomes. However, the price for focusing on costs will be to alienate the essential support from clinicians, who are likely to be motivated by quality improvement and service development rather than cost control.
Pros and cons of disease management
The three main stakeholder groups in a disease management approach are the health system, clinicians, and patients. In the survey mentioned earlier, respondents listed what they thought were the benefits and disadvantages for each of these groups (see box). Over and above the perceived disadvantages, there seem to be three principal barriers to its introduction: a lack of clarity about how disease management will tie in with NHS structures, in particular the separate budgets in primary and secondary care and social care; possible professional resistance to change; and an absence of clinical information systems.
Unless the tie-in with NHS structures is sorted out, there is a danger of establishing initiatives that counter the values and strategic aims of the NHS. The present structural and budgetary divisions are likely to lead to a lack of incentives for integrated and seamless care and for scrutinising the quality of clinical care provided. Moreover, unless rigorous evaluation is built in from the outset there is the risk of failure to learn from the experiences of others.
The litmus test for a successful disease management programme will be the extent to which the patient's interests are given primacy and then met. But there remains an issue over whether the focus is, or should be, on individual patients or whole populations of patients. Another concern is the doctor-patient relationship. Many fear that disease management might constrain the freedom of choice.
A pregnancy test is positive
Jane adjusts well; she seems well motivated and is seen at the surgery once a month by the practice nurse and at three monthly intervals at the diabetic centre. Once a year she has a full assessment, including funduscopy, and also sees a chiropodist. Everything seems to be going well until, four years after diagnosis, the primary care information systems pick up that Jane (who is now 16) has missed an appointment. A further appointment is sent but also not attended. The practice nurse visits Jane at home and after discussion enlists the help of the clinical psychologist at the diabetic centre, who establishes a rapport with Jane to the benefit of her diabetic control.
Jane leaves home and moves in with her boyfriend. At her next clinic visit she is found to have microalbuminuria (confirmed on three occasions). Despite her normal blood pressure she is about to be given an angiotensin converting enzyme inhibitor, according to the protocol developed in conjunction with the renal team; then it transpires that she had stopped taking oral contraceptives three months previously due to a “pill scare.” A pregnancy test is positive. She had received preconception advice six months earlier, as set out in the agreed guidelines, but at that time she had not been sexually active.
Professional resistance to change may be overcome only if doctors themselves drive, or feel as if they are driving, disease management.8 Clinicians need to be convinced that the disease management approach will heighten their professionalism because it entails adopting best practice.
Within the NHS, disease management comes up against the classic demarcation between primary and secondary care. But diseases span all these services and are no respecters of boundaries. Given the focus on a primary care-led NHS, it is worth asking the question as to whether primary care is capable of taking on the extra load of disease management.
An integrated healthcare system will require long term alliances to be established between primary and secondary care which will, in turn, require improved information, information sharing, and collaborative working. None of this will occur quickly and all the changes will have implications for training and development. It will also be a key challenge for managers to avoid unacceptable variations across the country and to try and ensure reasonably consistent and equitable progress.
Mother and baby do well
Jane is happy to be pregnant. She attends the diabetic antenatal clinic where she sees the multidisciplinary team, which includes a diabetologist, obstetrician, midwife, diabetes specialist nurse, and dietician. Her diabetes control is good due largely to the home visits of the diabetic specialist nurse and the practice nurse, who Jane trusts. At 24 weeks a routine check discovers retinal changes and Jane is seen by the ophthalmology team. Her blood pressure becomes difficult to control and her baby fails to thrive. At 37 weeks the baby is delivered by caesarean section. Mother and baby do well and are discharged into a flat found for her through the intervention of the social worker.
Jane's diabetes continues to be monitored in agreement with shared care protocols. Over the long term, care is well coordinated by the primary care team. Jane is well informed and feels involved, in control, and an equal partner in her own care. All those involved in Jane's care feel happy that they are delivering a patient centred, cost effective service and that Jane is a success story. This is reflected in her biomedical, psychological, and social outcomes. The overall local population outcomes and the health team's performance indicators are also satisfactory. It is, however, acknowledged that while Jane is a success story the disease management approach cannot be guaranteed effective for every individual.
The government's approach
The NHS Executive's 1994 guidance on disease management was negative about the prospect of the NHS doing deals with the private sector, especially the pharmaceutical industry. 9 But the revised guidance which was to have been issued some months ago but got delayed by the election, will take a slightly more liberal approach.10 The discussion paper which preceded the final guidance focused on the area which has generated most political concern, namely, the desire for joint disease management ventures between the NHS and private sector companies. According to this document, the government “neither encourages nor discourages such ventures.”
In its guidance, the government is concerned to respond to the interest which has been generated by overtures to the NHS from the private sector and sets out some safeguards to help the NHS to assess individual proposals for joint management before deciding whether to enter into them. A possible framework is provided to help NHS purchasers assess locally the merits and risks of individual schemes and help them to decide whether to enter into disease management agreements.
Disease management will improve the delivery of care if its limitations are honestly acknowledged and it has the full support of clinicians and others to drive quality and improve outcomes. It represents good practice and common sense in the provision of effective care—but it will only be as successful as the robustness of its evidence base and the calibre of its managers. It is not a panacea. But the inherent reasonableness and common sense of the approach, and the growing frustration with a health system that is becoming more fragmented and compartmentalised, points to an enhanced role for disease oriented approaches in future.
We thank Dr S Gilbey of St James's University Teaching Hospital, Leeds, for help with the case history.