Cognitive impairment in schizophrenia is a strong predictor of the functional outcome and no effective treatments are available. MATRICS Consensus Cognitive Battery (MCCB) is approved by the FDA as outcome measure for trials of cognitive-enhancing drugs in schizophrenia. CogState Schizophrenia Battery (CSB) provides a briefer cognition assessment with minimal practice effects and a strong correlation between the CSB and MCCB composite scores. We tested the sensitivity of CSB as a cognitive outcome measure in a clinical trial in schizophrenia, where a cognitive-enhancing drug and cognitive training were combined.

49 participants with schizophrenia were enrolled in a double-blind, placebo-controlled study. Participants were randomised to modafinil (200mg/day) or placebo and underwent a cognitive training program for 10 weekdays. CSB was administered twice at baseline to minimise practice effects, at the last day of the intervention and two weeks after the completion of the intervention.

There was a significant time effect at the end of the intervention on the CSB composite score (p=0.042). There was no significant treatment effect on CSB composite score at the end of the intervention (p=0.686) or at follow up (p=0.120).

Multiple administrations of CSB were well tolerated by participants. The significant time effects on the composite score may suggest the operation of practice effects. Several factors could have contributed to the lack of treatment effects on CSB, such as the burden of multiple neuropsychological testing in a relatively brief study, the duration of modafinil treatment and also the intensive nature of cognitive training.

Cognitive deficits in schizophrenia have major functional impacts. Modafinil is a cognitive enhancer whose effect in healthy volunteers is well-described, but whose effects on the cognitive deficits of schizophrenia appear to be inconsistent. Two possible reasons for this are that cognitive test batteries vary in their sensitivity, or that the phase of illness may be important, with patients early in their illness responding better.

A double-blind, randomised, placebo-controlled single-dose crossover study of modafinil 200 mg examined this with two cognitive batteries [MATRICS Consensus Cognitive Battery (MCCB) and Cambridge Neuropsychological Test Automated Battery (CANTAB)] in 46 participants with under 3 years’ duration of DSM-IV schizophrenia, on stable antipsychotic medication. In parallel, the same design was used in 28 age-, sex-, and education-matched healthy volunteers. Uncorrected p values were calculated using mixed effects models.

In patients, modafinil significantly improved CANTAB Paired Associate Learning, non-significantly improved efficiency and significantly slowed performance of the CANTAB Stockings of Cambridge spatial planning task. There was no significant effect on any MCCB domain. In healthy volunteers, modafinil significantly increased CANTAB Rapid Visual Processing, Intra-Extra Dimensional Set Shifting and verbal recall accuracy, and MCCB social cognition performance. The only significant differences between groups were in MCCB visual learning.

As in earlier chronic schizophrenia studies, modafinil failed to produce changes in cognition in early psychosis as measured by MCCB. CANTAB proved more sensitive to the effects of modafinil in participants with early schizophrenia and in healthy volunteers. This confirms the importance of selecting the appropriate test battery in treatment studies of cognition in schizophrenia.