Background: Antipsychotic drugs are often used to treat behavioral and psychological symptoms (BPSD) in adults aged 65 years and older with dementia, although there is uncertainty about the effectiveness of long-term use for this indication and there are concerns that they may cause harm.

Objectives: To evaluate whether discontinuation of long-term antipsychotic drugs for BPSD is successful in adults aged 65 years and older with dementia. This article is based on a Cochrane review updated in 2018.

Design: A Cochrane systematic review and meta-analysis.

Setting and participants: Eight databases were searched in January 2018 to identify 10 randomized controlled trials with 632 older adults. MEASURES: We used standard methodological procedures according to the Cochrane Handbook for Systematic Reviews of Interventions. We assessed the number of patients completing the study. We considered sustained withdrawal of antipsychotics until the end of the study period as successful outcome.

Results: Based on assessment of 7 studies (n = 446), discontinuation may make little or no difference to whether or not participants complete the study (low-quality evidence). In 2 trials, including participants with psychosis, agitation, or aggression who had responded to antipsychotic treatment, discontinuation of antipsychotics was associated with a higher risk of leaving the study prematurely because of symptomatic relapse or a shorter time to symptomatic relapse. We found low-quality evidence from 7 trials (n = 519) that discontinuation may make little or no difference to overall BPSD, measured using various scales. There was some evidence from subgroup analyses in 2 trials that discontinuation may be associated with a worsening of BPSD in participants with more severe BPSD at baseline.

Conclusions: Our meta-analysis revealed that there is low-quality evidence that long-term antipsychotic drugs for BPSD may be successfully discontinued in most adults aged 65 and older.