Despite mental disorder being so common, detection by general care teams is poor. The diagnosis of delirium is missed in 32–67% of cases¹⁵, and cognitive impairment in over half¹⁶. A meta-analysis of older medical admissions reported a median detection rate for depression of 10%¹⁷. In England, a national report found that only 41% of people with dementia had any form of mental assessment recorded¹⁸.

General care teams have difficulty understanding the concept of syndromes, and their response to mental disturbance is, predominantly, a reaction to symptoms. Consequently, crying, not eating, and thoughts of death are synonymous with depression, and any cognitive errors with dementia. Jackson and Baldwin¹⁹ reported how nurses on medical wards recognize symptoms of depression rather than the syndrome. A small qualitative study in the USA found that 75% of general nurses could not distinguish dementia from delirium, despite 75% saying that they had received formal education on the subject or had attended a conference about confusion in the elderly²⁰.

The limitations of general care teams discussed above highlights the need for education and training, and suggest that whatever training is currently employed seems to have little effect. Teaching and training is core business for liaison teams and there is evidence that the introduction of liaison services into general hospitals is associated with changes in referral that might be taken to indicate a more informed change of behaviour²⁸. And so, it is likely that their presence, in itself and by clinical contacts, has an educative effect.

However, the best way to deliver more formal training to general care teams remains uncertain. Principles of adult learning which may be usefully incorporated into educational programmes include timely and relevant information based on participant identification of learning needs and integration of new information into existing knowledge base(s) with time for consolidation and appropriate non-judgemental feedback to the individual²⁹.

Case-based discussions that bring immediate relevance, supervision that reinforces knowledge and skills, and audit to evaluate change will be important to effective education. Care pathways can be helpful. This might suggest that an approach which focuses on the needs of specific departments and specialities involving complete clinical teams could be more effective than random teaching sessions for hospital staff and particular professional groups. The circumstances and subject matter should dictate the approach.

Whatever the approach, training time must be built into job plans and the resourcing of liaison teams. The time needed is always underestimated, and when clinical staff are under pressure, training will be relegated to low priority. In one audit of a nurse-led liaison service, 7.5% of time was spent directly on formal education and 33% giving guidance, advice and making recommendations to ward staff³⁰.

Delivery to large hospitals, including night staff, with staff who move through in training, is difficult and a perpetual process. Incorporating mental health into established teaching programmes will give it credence and importance. The presence of liaison teams in general hospitals offers the prospect of mental health being absorbed into the culture of the hospital and not seen as a separate activity just provided by mental health services. The importance of improving the mental health knowledge and skills of general care teams cannot be overestimated because the vast majority of older people admitted to general hospitals with mental health needs will not be seen by specialist liaison teams. The management of non-referred cases will be the ultimate measure of how successfully liaison services have changed the general hospital’s response to mental health.

Delirium

Delirium is the most common non-specific presentation of physical illness among older people and the most common new condition in the hospitalized elderly. It will complicate 5–10% of routine surgery. Delirium present at the time of admission is referred to as prevalent delirium, and that beginning in hospital as incident delirium. The distinction is important when discussing prevention, as 30–40% of incident delirium may be hospital acquired and preventable. Rates of delirium will vary between hospital departments, being highest with the most ill patients (Intensive Care), higher with emergency than elective surgery and higher with more complex conditions.

It is a cause of death, prolonged admission, loss of independent function, high cost and is associated with an increased risk of hospital-acquired complications including incontinence, falls, pressure sores and infection. Estimates from the USA suggest that it affects more than 2.3 million older people each year, is responsible for more than 17.5 million inpatient days and costs over $4 billion (1994 dollars) each year³¹.

Furthermore, it is now clear that not all delirium resolves, and cognitive impairment persists 12 months after a medical admission when 15% still meet the diagnostic criteria for delirium³². Patients developing delirium during a medical admission are twice as likely to have died, and six-times more likely to have acquired a diagnosis of dementia three years later¹². Reversibility seems crucial for outcome. A prospective study of older medical patients found that the 6 and 12 month outcome for delirious people who survive and recover was no different to no-delirium groups in relation to cognition, function or institutional status³³.

Onc established, the treatment is that of the underlying condition, with supportive care and management of behavioural and psychotic symptoms³⁴. There is a theoretical basis to developing pharmacological prophylaxis and treatment with cholinesterase inhibitors^35,36 and manipulation of the immune system³⁷, but small controlled trials of the former have failed to demonstrate efficacy.

There is compelling evidence for interventions which prevent incident delirium that is closely linked to processes of care. Clinical prediction rules have been described for medical, non-cardiac thoracic surgery and elective orthopaedic populations^38–40. These enable people at risk to be identified on admission to hospital and pre-ventative care strategies employed to reduce delirium risk. Because delirium mostly develops within the first seven days of admission or surgery, this approach can concentrate on that period⁴¹.

The most developed prevention approach arose from studies of older medical patients, identifying the strongest independent predisposing and precipitating factors that predict incident delirium^42,43. These large studies found the relative risk was 9.5-times greater for people with a high-risk predisposing profile, and 8.9-times greater for those with a high-risk precipitant profile, than those with low risk.

This formed the basis of a controlled trial of a preventative intervention involving 852 older medical inpatients, employing a multi-component intervention strategy (The Elder Life Program). The subjects were identified on admission as being intermediate or high risk, and in the intervention arm the incidence of delirium was 34% lower than usual care. The intervention group experienced significantly lower total days with delirium, lower total number of episodes, improved cognitive function in those already cognitively impaired and lower use of sleep medications⁴⁴.

This intervention was shown to be cost effective for intermediate risk (72% of subjects) but not high risk. Cost of the intervention was offset by savings accrued from treatment intensity, including nursing and diagnostic procedures³¹.