Effects of Psychosocial Interventions for Behavioral and Psychological Symptoms in Dementia on the Prescription of Psychotropic Drugs: A Systematic Review and Meta-Analyses.

BACKGROUND

Dementia is often accompanied by neuropsychiatric symptoms. Psychotropic drugs for the treatment of neuropsychiatric symptoms are frequently used to manage these problems, but they are of limited effectiveness and can have serious side effects. Psychosocial interventions are advocated as first line treatment and may help to reduce psychotropic drug use.

AIMS

To assess the effect of multidisciplinary psychosocial interventions in nursing homes on the psychotropic drug prescription rate.

METHODS

Literature obtained from searches in 9 electronic databases was systematically reviewed. In addition, the pooled effects of specific psychosocial interventions in homogenous subgroups were analyzed (meta-analysis).

RESULTS

Eleven randomized controlled studies that investigated multiple psychotropic drugs interventions (psychotropic drugs in 3, antipsychotics in 9, and antidepressants in 5 studies) as well as different types of psychosocial interventions were included. We separately analyzed interventions directed at the care staff level (educational programs in 3, in-reach services or consultation in 1, cultural or process change in 6 studies) and the individual resident level in 1 study. In 7 out of 9 studies reporting on antipsychotic drug use, the physician was actively involved. Nine studies in which antipsychotic drug use was specified reported a significant decrease in prescription rate as a result of psychosocial interventions [relative risk (RR) 0.71, 95% confidence interval (CI) 0.59-0.88], whereas meta-analysis of 5 studies investigating antidepressant drug use failed to show a significant effect (RR 0.82, 95% CI 0.64-1.02). Pooled effect sizes of 6 studies investigating cultural change, showed a significant decrease in antipsychotic drug use (RR 0.65, 95% CI 0.57-0.73). Effect sizes of 2 studies on educational programs on antipsychotic use were nonsignificant (RR 1.50, 95% CI 0.49-4.64). Sensitivity analysis of 7 studies reporting on antipsychotic drug use involving prescribing physicians showed a more robust decrease (RR 0.66, 95% CI 0.54-0.80).

CONCLUSIONS

The results of this study show that psychosocial interventions may lead to a substantial reduction of antipsychotic drug prescription, especially in studies that reported on cultural change and that involved prescribing physicians. Conspicuously, a profound lack of information was observed in many studies as to what exactly constituted the care-as-usual treatment in the control group.