Skip Navigation
Department of Health and Human Services
Slide Tray
0 slides

Return to Slide Library


Add Presentation to Slide Tray Presentation:

Off-Label Use of Atypical Antipsychotics: An Update

Slide: 19 of 47

Results: Atypical Antipsychotics for Dementia

Combined results are presented from meta-analyses of all atypical antipsychotics that have been studied as treatment for dementia in placebo comparisons. The effect size is the standardized mean difference between the means of treated and control groups after the treatment period. The published study results came from a variety of assessment instruments and, as a procedure of the meta-analysis, they were standardized to a common scale before determining the mean difference across studies. The results are presented together with the 95-percent confidence interval, which is the statistically valid range of results that includes the true population mean in 95 of 100 experiments. For mean differences, the result is statistically significant at p < 0.05 when the 95-percent confidence interval does not cross 0.0.

As a class, atypical antipsychotics improved global scores by 0.17 points (statistically valid range from 0.08 to 0.25). Psychosis scores improved by 0.12 points (statistically valid range from 0.04 to 0.19), and agitation scores improved by 0.20 points (valid range from 0.12 to 0.27). The strength of evidence for these findings is high. Among atypical antipsychotics examined individually, the difference in means was an improvement of 0.12 points with olanzapine alone (valid range from 0.00 to 0.25). No statistically significant difference was found for psychosis scores. Agitation scores were reduced by 0.19 points (valid range from 0.07 to 0.31). Risperidone improved global scores by 0.19 points (range from 0.00 to 0.38), psychosis scores by 0.2 points (range from 0.05 to 0.36), and agitation scores by 0.22 points (valid range from 0.09 to 0.35). The strength of evidence for olanzapine is moderate and for risperidone is high.