F133. A LONGITUDINAL COMPARISON OF TWO NEUROCOGNITIVE ASSESSMENT PACKAGES IN PATIENTS WITH SCHIZOPHRENIA AND HEALTHY VOLUNTEERS: TIME EFFECTS ON NEUROPSYCHOLOGICAL PERFORMANCE AND ITS RELATION TO FUNCTIONAL OUTCOME

Abstract

Background: Cognitive impairment is a major unmet need in the treatment of schizophrenia. Over the last decade, the MATRICS Consensus Cognitive Battery (MCCB) and the cognitive-neuroscience-based Cambridge Neuropsychological Test Automated Battery (CANTAB) have been used to assess the effects of novel treatments for cognitive impairment in schizophrenia. Although both batteries purport to assess cognitive function in psychosis (Lees et al., 2017a, Keefe et al., 2013), no previous study has attempted to examine their validity longitudinally and the potential overlap between the two batteries over time. The aim of the current study was to assess the relationship between the MCCB and the CANTAB in the longitudinal assessment of cognitive impairment in schizophrenia. Also, it was targeted to examine the predictive effect of each battery on the psychosocial functioning of subjects. Method(s): A sample of 39 stable schizophrenia outpatients (mean age= 35.79, SD=10.06; males=27, 69.2 %) and 18 controls (mean age=35.94, SD=8.14; males= 12, 66.7 %). completed the MCCB and the CANTAB battery at baseline, and at 2, 4 and 8-weeks follow-up. UCSD Performance-Based Skills Assessment- brief version (UPSA-B), including total score, financial skills and communication skills indices, and Work and Social Assessment Scale (WSAS) scale were administered at baseline to examine psychosocial functioning. Mixed-model repeated measures approach was used for the timepoints mean comparison of each battery independently and the time*battery interaction (including the subjects ID as random variable). Backwards-stepwise linear regressions were performed to determine the best set of cognitive tests/domains derived from the MCCB and/or the CANTAB that would explain the maximum variance for functional outcomes. All the analysis controlled for age including it as covariable. Result(s): The present study detects practice effects from baseline to follow-up assessments in the CANTAB cognitive battery in visual memory (F3,146= 4.37, p=0.006), reasoning and problem-solving (F3,147= 4.58, p=0.004) and social cognition (F3,239= 2.96, p= 0.033) in patients, and social cognition in controls (F3,66= 10.92, p<0.001). In contrast it finds stability in the MCCB battery over time. In particular, a significant time*battery interaction was found for those cognitive domains emphasizing the main time effect on CANTAB. Furthermore, MCCB verbal learning predicted global psychosocial functioning in patients and MCCB verbal memory and CANTAB processing speed were the best predictors of financial skills. Moreover, performance in CANTAB processing speed, verbal memory, working memory and reasoning, and problem-solving best predicted UPSA communication skills in patients. Finally, MCCB verbal learning predicted SWAS working skills. Discussion(s): These findings suggest there are specific differences across cognitive tests to assess cognitive impairments in schizophrenia and that measures derived from the CANTAB appear to be more prone to practice effects in these patients. However, these results should be interpreted cautiously. There are some variables such as the number of included subjects, the placebo effect and patient's clinical symptomatology that could be taken into consideration. As previously (Kumar et al., 2016), a strong association was displayed between both batteries and the psychosocial functioning of the patients. In particular, verbal learning as measured with MCCB and a composite of various domains as assessed by CANTAB were the most related to working, financial and communication skills. These results underline the importance of cognitive performance in the daily function of patients with schizophrenia.