Comparing the effect of the subcategories of atypical antipsychotic medications on cognition in schizophrenia using a meta-analytic approach.

OBJECTIVES: The aim of this study was to compare the two most commonly prescribed classes of atypical antipsychotic medications (i.e., -pines and -dones) with regard to their effects on cognition in patients with schizophrenia.

DATA SOURCES: Ovid Technologies web-based software was used to search the Medline and PsycINFO computerized databases to identify articles that met the inclusion criteria.

REVIEW METHODS: The search was limited to papers published after 1990, written in English, employing human subjects, using atypical antipsychotics, using a within-subjects design or control group of patients with schizophrenia for comparisons, using participants aged from 18-65, and employing standardized neuropsychological measures.

RESULTS: A total of 996 eligible studies were identified, and of these 19 were finally analyzed. Nine domains of cognitive functioning were assessed. The two groups of agents produced equivalent overall effects (-dones = .254 versus -pines = .202). The -pines were found to improve the domains of attention/working memory, executive functioning, fluency, nonverbal memory, processing speed, and verbal memory, each with a significant, small effect size. The -dones were found to improve attention/working memory, executive functioning, motor function, nonverbal memory, processing speed, and verbal memory, each with a significant, small effect size. Failsafe N was robust for all of the domains for the -pines, but only for the verbal memory domain for the -dones, suggesting that the significant findings for the other domains with the -dones are more tenuous.

CONCLUSION: The results indicate that the agents were largely equivalent and that there was no clear evidence that the pattern of cognitive effects differed as a result of the agent applied. The effects themselves, while statistically significant, were small, indicating that some or all of the differences may be attributable to practice effects on the instruments employed.