Meta-analysis of psychosocial interventions for caregivers of people with dementia.

Abstract

To review published reports of interventions for caregivers (CGs) of persons with dementia, excluding respite care, and provide recommendations to clinicians. Meta-analytical review. Electronic databases and key articles were searched for controlled trials, preferably randomized, published in English from 1985 to 2001 inclusive. Thirty studies were located and scored according to set criteria, and the interventions' research quality and clinical significance were judged. Home or noninstitutional environment. Informal CGs-persons providing unpaid care at home or in a noninstitutional setting. The primary measures were psychological morbidity and burden. Other varied outcome measures such as CG coping skills and social support were combined with measures of psychological distress and burden to form a main outcome measure. The quality of research increased over the 17 years. Results from 30 studies (34 interventions) indicated, at most-current follow-up, significant benefits in caregiver psychological distress (random effect size (ES) = 0.31; 95% confidence interval (CI) = 0.13-0.50), caregiver knowledge (ES = 0.51; CI = 0.05-0.98), any main caregiver outcome measure (ES = 0.32; CI = 0.15-0.48), and patient mood (ES = 0.68; CI = 0.30-1.06), but not caregiver burden (ES = 0.09; CI = -0.09-0.26). There was considerable variability in outcome, partly because of differences in methodology and intervention technique. Elements of successful interventions could be identified. Success was more likely if, in addition to CGs, patients were involved. Four of seven studies indicated delayed nursing home admission. Some CG interventions can reduce CG psychological morbidity and help people with dementia stay at home longer. Programs that involve the patients and their families and are more intensive and modified to CGs' needs may be more successful. Future research should try to improve clinicians' abilities to prescribe interventions.