Multicomponent Geriatric Intervention for Elderly Inpatients With Delirium: A Randomized, Controlled Trial

Journal of Gerontology: MEDICAL SCIENCES 2006, Vol. 61A, No. 2, 176–181 Copyright 2006 by The Gerontological Society of America Multicomponent Geriatric Intervention for Elderly Inpatients With Delirium: A Randomized, Controlled Trial Kaisu H. Pitkälä,1 Jouko V. Laurila,1,2 Timo E. Strandberg,1 and Reijo S. Tilvis1 Helsinki University Hospital, Clinics of Internal Medicine and Geriatrics, Finland. 2 Helsinki City Hospitals, Finland. Background. Delirium is a common syndrome with poor prognosis affecting elderly inpatients. Treatment is mainly based on common sense with wide variations in practice. We investigated whether intensified, multicomponent geriatric treatment could improve the prognosis of delirious patients. Methods. We performed a randomized, controlled trial of 174 patients with delirium in six general medicine units from an acute hospital in Helsinki, Finland. The intervention group received individually tailored geriatric treatment. The primary endpoint was the sum of those deceased individuals and the patients permanently institutionalized. Secondary endpoints included the number of days in hospitals and other institutions, delirium intensity, and cognition. Results. The mean age of patients was 83 years, and 31% had previous dementia. The intervention group (N¼87) received significantly more acetylcholinesterase inhibitors (58.6% vs 9.2%), atypical antipsychotics (69.8% vs 30.2%), specialist consultations (49.4% vs 28.7%), hip protectors (88.5% vs 3.4%), physiotherapy (87.4% vs 47.1%), and fewer conventional neuroleptics (8.0% vs 23.0%) than did the control group (N ¼ 87). During the 1-year follow-up, 60.9% of the intervention group and 64.4% of controls were either deceased or permanently institutionalized ( p ¼ .638). The intervention group spent a mean of 126 days in institutions, and the control group 140 days ( p ¼ .688). Delirium was, however, alleviated more rapidly during hospitalization, and cognition improved significantly at 6 months in the intervention group. Conclusions. Faster alleviation of delirium and improved cognition justify good, comprehensive geriatric care for these patients although treatment produced no significant improvements in hard endpoints of prognosis. D ELIRIUM is a serious neuropsychiatric syndrome complicating somatic illnesses. It affects approximately 25% of older medical inpatients (1). Prospective studies have related delirium to increased mortality (2–4), need for permanent institutional care (4–7), longer hospital stays (4–7), poor functional outcome (7,8), and dementia (2,8). Especially, delirium has been shown to have an independent effect on mortality (2–4) and admissions to permanent institutional care (4). Increased need for hospital and institutional days produces substantial costs for health and social care (9). Prevention of delirium by comprehensive geriatric intervention has been effective (9,10). Prevention has, however, left 10%–30% of elderly inpatients suffering from delirium (9,10) with little known about their best possible treatment. Only a few randomized studies have focused on treatment of full-blown delirium. Cole and colleagues showed that geriatric consultation for delirious inpatients may be effective in more rapidly improving their cognition (11), but showed no effect on mortality, institutionalization, or length of hospital stay (11,12). Treatment of delirium is thus mainly based on common sense, with large variations in practice (13). A fairly large body of evidence indicates that comprehensive geriatric assessment and intervention are effective (14). Since the studies by Cole and colleagues, more evidence has accumulated on pathophysiologic mechanisms of delirium. Central cholinergic deficiency with simultaneous relative dopamine excess may play a central role in the development of delirium (15,16). Thus, acetylcholinesterase 176 inhibitors (ChEIs) and small-dose dopamine antagonists might be effective in treating symptoms of delirium. The aim of this randomized controlled study was to investigate whether a comprehensive geriatric assessment and individually tailored treatment are effective in reducing mortality and permanent institutional care among patients with delirium. We also wanted to determine whether this treatment is beneficial in reducing the number of days spent in institutions, alleviating delirium, or improving cognition or physical functioning of these patients. METHODS Setting and Participants Potential participants were consecutive patients (.69 years) admitted to the general medicine service at one Helsinki City hospital from September 20, 2001 through November 24, 2002. This hospital, with 156 acute beds and six units for general medicine, serves a population (.100,000 inhabitants) of the western area of Helsinki. Systematic methods on screening or preventing delirium are not used in this hospital. Exclusion criteria included life expectancy of less than 6 months (e.g., metastatic cancer, severe stroke), inability to obtain informed consent within 2 working days, admission from permanent institutional care to the hospital, or refusal. Because all participants were delirious, informed consent was obtained from each patient’s closest proxy (100%). The 1 GERIATRIC INTERVENTION FOR ELDERLY INPATIENTS WITH DELIRIUM 177 study was approved by Helsinki University Hospital and the Helsinki City ethics committees. Screening and Confirming Delirium Patients were screened within 2 working days of their admission by two study nurses who had undergone detailed training and followed standard procedures. The initial screening included the Confusion Assessment Method (17), Mini-Mental State Examination (MMSE) (18), Digit span (19), a proxy interview, and a review of medical records (Figure 1). Those patients with a positive result on the Confusion Assessment Method test were assessed by the study physician, who confirmed the inclusion criteria and the diagnosis of delirium by the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (20). Design and Randomization Patients were randomly allocated by means of computergenerated random numbers to undergo the intervention or to receive the usual care. When delirium was confirmed, the study assistant called by telephone to a randomization staff member who had not seen the patients or their clinical records. She assigned the next number from the computer and the group assignment to the patient. Assessments All participants underwent interviews and assessments at baseline, at 3 months, and at 6 months. These included assessment of cognitive functioning by the MMSE (18), of activities of daily living by the Barthel index (21) and the instrumental activities of daily living scale (22), of depression by the Geriatric Depression Scale (23), and of nutrition by the Mini-Nutritional Assessment (24). Severity of delirium was measured daily by the Memorial Delirium Assessment Scale (MDAS) (scored 0–30, 30 being worst; 25 (...truncated)