Level 2: Mid-level evidence

In April 2008, the Dementia Antipsychotic Withdrawal Trial (DART-AD) showed that stopping antipsychotic treatment did not worsen cognitive, functional, or behavioral status in most elderly patients with Alzheimer’s disease (PLoS Med. 2008;5:e76; accessed April 8, 2009). Based on results of short-term trials that indicated an increased mortality risk, the FDA has previously warned against the use of both conventional and atypical antipsychotics for dementia-related psychosis (FDA MedWatch. 2008 Jun 16; accessed April 8, 2009). Now, a follow-up of DART-AD provides evidence of increased long-term mortality with anti-psychotics (Lancet Neurol. 2009;8:151-157). In a modified intention-to-treat analysis including 128 patients who started treatment, cumulative 12-month survival was 70% for the antipsychotic group vs. 77% for the control (P=.03, NNH 14). The difference in survival was greater at two to four years: 46% vs. 71% at 24 months (NNH 4), 30% vs. 59% at 36 months (NNH 3), and 26% vs. 53% at 42 months (NNH 3).