Multidisciplinary team interventions for delirium in patients with chronic cognitive impairment

Delirium is common in hospitalized elderly people. In the frail elderly, delirium may occur in 60% of those hospitalized. In the cognitively impaired, 45% have been shown to develop delirium and these patients have longer lengths of stay and a higher rate of complications which, amongst other things, together contribute to an increase in cost of care. The combination of being elderly and chronically cognitively impaired leads to a high risk of delirium with the associated increased risk of prolonged hospital stay, complications, and poor outcomes. The management of delirium has commonly been multifaceted - the primary emphasis has always been on the diagnosis and therapy of the precipitating factors, but as this may not be immediately resolved, symptomatic and supportive care may become of major importance.

The objective of this review is to assess the available evidence for the effectiveness, if any, of multidisciplinary team interventions in the coordinated care of patients with delirium superimposed on an underlying chronic cognitive impairment compared with the usual care of older cognitively impaired patients.

The Cochrane Controlled Trials Register (Cochrane Library, up to and including Issue 1, 1998) was searched using the terms 'delirium, controlled trial, cognitive'. MEDLINE, EMBASE and Psychlit (Ovid via Winspirs up to Feb 1998) were also searched with the same terms. Other sources including personal communications, ongoing trials, conference proceedings, handsearching and reference lists of published papers and books were all searched for relevant randomized controlled trials. The total yield from searching was 157 from which 8 (eight) were retained for consideration in the review.

From the initial search yields, all randomised controlled trials involving the management of elderly patients with delirium were identified. A single reviewer (AMB) discarded irrelevant publications based on the title of the publication and its abstract. In the event that the article could possibly be relevant, it was retrieved for further assessment. All references were compiled in a list with a commentary on type of article, eg review, prospective study etc and this was independently considered by the second reviewer (RR) who agreed to review all randomised controlled studies reported on patients with delirium. Selection for possible inclusion in this review was then made on the basis of the participants reported as having chronic cognitive impairment, who then developed incident delirium and were randomly assigned to either coordinated multidisciplinary care or usual care. The outcomes of interest were length of stay in hospital, morbidity (including complications), patient distress & impact on care environment, mortality, discharge arrangements and follow-up including assessment of cognitive function at 6 months. Studies in which patients with chronic cognitive impairment or dementia, managed for incident delirium, according to ICD 9 criteria (see note) were considered eligible for inclusion in the review. Studies of risk factors and non-randomized studies were excluded. Note: this classification has been widely utilised throughout the English speaking medical literature over the past 20 years: ICD 10 is still being incorporated into clinical coding systems and has not been utilised in studies published in 1996. (ABSTRACT TRUNCATED)