A study of a pilot has found the effectiveness of case management is being limited. Martin Roland and Ruth Boaden report

A study of a pilot has found the effectiveness of case management is being limited. Martin Roland and Ruth Boaden report

Case management has been widely promoted as a way of cutting emergency admissions for frail elderly people - and expectations are high for the 3,000 community matrons being appointed as case managers.

However, a National Primary Care Research and Development Centre study has found that a pilot form of case management, Evercare, did not deliver the benefits hoped for.

According to our evaluation, the overall pattern of hospital admissions was unaffected even though care was improved and nurses and patients gave individual examples of admissions that had been avoided. So how can we make case management more effective?

The job of a case manager, often a nurse, is to support people with frail health and high risk of hospital admission, improving their quality of life and reducing costs. The support involves planning and co-ordinating services, and in some cases providing clinical care.

We found a number of factors that limited the effectiveness of Evercare, which ran from April 2003.

A key problem is that the NHS has real difficulty selecting patients who need case managers, because of poor information systems.

Ways of using information to identify patients at high risk of admission are not good enough. Details of patients are held on practice computers, but these cannot easily be collected or analysed across a primary care trust. The latest risk adjustment models from the King's Fund are an improvement, but no model based on routine data is going to be adequate.

Room needs to be left for professional judgement. In addition, few admissions can be predicted far in advance. If case managers could intervene acutely with patients at imminent risk of admission, this would increase the chances of an overall impact on admissions.

Once case management is up and running, existing information systems are again little help in monitoring what is going on. Solutions proposed by the national IT programme are not always sufficient, and some sites are developing their own systems. Case managers often have difficulty in getting information about case-managed patients added to GP systems.

More than an add-on

We found that most case management was seen as an add-on to primary care. This limits its effectiveness - case managers will be most effective where they work closely with district nurses, GPs and practice nurses. Good relationships with GPs are vital to provide and co-ordinate care. These relationships work best where case managers are attached to particular practices rather than geographical areas.

Many case-managed patients have chronic obstructive pulmonary disease and other conditions for which specialist nurses are available. Case managers need to work closely with these staff.

We also found that patients with multiple complex problems did not fit care pathways for single diseases. These need to be interpreted flexibly.

Effective co-operation among out-of-hours services is vital, but we found it was often poor. Community matrons are geared up to keep patients out of hospital. A GP flown in from Germany to cover a weekend is not.

However, we found examples where details of patients at risk of deterioration were communicated effectively to out-of-hours services, so good communication is possible, and patient-held notes help.

Many patients and case managers said they would like case management to be provided out of hours - because of the quick and sympathetic response it gives to patients. However, this would be expensive and would not give the continuity of care a single case manager provides. In the absence of better evidence that case managers can cut admissions, it is probably not justifiable to extend the service out of hours. What is needed is a more appropriate response from existing out-of-hours services.

Co-ordinating services across health and social care is the most challenging part of case management. We found relationships with hospitals were the most difficult. There was little enthusiasm to reduce admissions, partly because accident and emergency targets give little time to respond with admission alternatives. It was often easier to admit the elderly person.

Hospital staff could work much more effectively with staff in the community to keep patients at home. They could get patients home more quickly, too, if there was more effective joint working. In fact, there is plenty of potential for case managers to work with hospitals to reduce length of stay - it is not all about avoiding admissions.

Sometimes small things - equipment or same-day urgent social care - are all that is that is needed to avoid admission. Yet we found that processes for accessing services and appliances in the community that might keep patients out of hospital were often difficult and complex.

In theory, case managers should be able to access alternatives to admission. In practice this is difficult. It can be that there are simply not enough beds or vacant places in community services - but often the problem is that the process of accessing alternatives is too complex to be of any use when a patient gets ill.

Approaches which give immediate access to a wider multi-disciplinary team look promising, for example the 'virtual wards' being used by Croydon, which recently won several categories in last year's HSJ Awards.

The notion that a case manager has to be an experienced nurse is not borne out in practice. Most Evercare nurses found that their work depended less on clinical skills, and more on psycho-social support and accessing services. In the Eldercare project in Cornwall, case management was sometimes
co-ordinated by an allied health professional, and in the Castlefield model, a nurse/social worker team seemed to be effective in getting patients back home from hospital. This highlights the need for case management to include access to a multidisciplinary team.

The most important message of the study is that case management should be seen as a way of improving care, not just avoiding admissions.

Unrealistic expectations

Expectations of Evercare were always far too high in terms of avoiding admissions. This is not least because the US version of Evercare - which reduced admissions by 50 per cent - included intensive nursing for old people in nursing homes. This has never been part of case management in England.

Evercare's own US evaluation concluded that the intervention did not actually stop old people getting sick, it just provided a cheaper way of looking after them.

One final point on measuring outcomes - it is not enough to show that older people who had several emergency admissions last year have fewer this year. That happens on its own. Although this is a convenient way for PCTs to 'prove' that case management cuts admissions, one has to look at the overall numbers of admissions to know if one has been successful.

Martin Roland is director of the National Primary Care Research and Development Centre and Ruth Boaden is professor of services operations management. Both are based at the Manchester University and contributed to the NPCRDC's evaluation of Evercare. Read the national evaluation of Evercare at www.npcrdc.ac.uk/es42