Moderator variables of executive functioning in schizophrenia: meta-analytic findings

The literature regarding executive functioning in schizophrenia was examined to determine effect size differences between schizophrenic, normal control, and psychiatric groups. Effect sizes (n = 176) from 71 studies were cumulated across measures including the Wisconsin Card Sorting Test (WCST), Halstead Category Test (HCT), verbal/design fluency, Trail Making B (TMB), and the Stroop Color-Word Test. Schizophrenic groups were impaired relative to controls (delta = -1.45). Schizophrenic groups also performed 0.40 standard deviations lower than psychiatric groups. Within the schizophrenic versus normal control comparisons, the effect size based on complex measures, such as the WCST/HCT (delta = -1.42), was similar to that based on less complex measures, such as TMB and the Stroop (delta = -1.58). Chlorpromazine equivalents and illness duration were not significantly correlated with effect sizes (p > 0.05). Effect sizes and number of hospitalizations (p < 0.05) were related. Effect sizes and Brief Psychiatric Rating Scale (BPRS) were not significantly correlated. Negative and positive symptoms were significantly correlated (p < 0.05). The results as they apply to research and clinical practice are discussed.