In discussions and studies about access to health care — from prevention to inpatient management — the lower end of the socioeconomic spectrum rarely rates a mention. This sizeable subgroup, especially in cities but also in rural settings, is over-represented among people attending emergency departments.1

Frequent attenders to emergency departments are a heterogeneous group — a mix of patients with chronic medical, mental health, alcohol and drug problems, as well as other psychosocial issues.2,3 Individual patients often have a combination of some or all of the above problems. Interestingly, the frequency of attendance may change with time, although patients with psychiatric and substance misuse issues tend to be persistent heavy users.4 Despite the fact that they may be as sick as other, non-frequent attenders, frequent attenders are often perceived as time-consuming “illegitimate” users of emergency department resources.1 From the patient’s viewpoint, despite having good access to primary health care, the emergency department is seen as the most appropriate place to seek help because of a perceived need for urgent care.5 Rather than as a substitute, emergency departments are often used in conjunction with primary care services, with the emergency department providing appropriate higher level care.6,7

Some of the debate about the resourcing of very expensive 24-hour services, such as stressed emergency departments, focuses on “diverting” patients away or “streaming” them somewhere else (eg, general practice) after appropriate triage. Multidisciplinary teams (including allied health workers) that target high-risk patient subgroups — the elderly, patients with respiratory illnesses, patients with mental health problems — have been introduced in many hospitals to address the swamping of emergency services.

But does all this actually make a difference? In this issue of the Journal, Phillips et al (page 602)8 evaluated the effect of multidisciplinary team case management on frequent attenders’ utilisation of an inner city emergency department in Melbourne, compared with their utilisation before case management (ie, the study population acted as their own historical controls). Emergency department utilisation and scores for housing stability, alcohol and drug use, and primary and community care engagement were determined. Perhaps surprisingly and paradoxically, despite intensive case management, Phillips et al found an increase in emergency department utilisation and overnight stays. A potential interpretation of this increased utilisation is that it may well reflect an increase in engagement of these patients, with the desired increased attention and improvement in their health. Furthermore, the study found that linkage with primary care and community services improved. Interestingly, in 2002, Moss et al, at another Melbourne inner city emergency department, found that a similar strategy produced a significant fall in hospital admissions.9

In an ideal world, there would be no debate about allocation of care teams to sick patients or the effectiveness of preventing health deterioration. Nor would certain groups of patients only be really visible once they have deteriorated — and then only mainly visible to certain parts of the health care system (eg, emergency departments, and community and mental health services). In health care, as elsewhere, planners can largely underestimate the need for, and the effect of, a new service where none previously existed. This would be easy to imagine when dealing with a marginalised, and at times psychosocially challenged, group of people.

Quoting Malone:

Research, measuring appropriate outcomes including accurate assessments of cost, has to be increased to allow resources to be rationally allocated.