Cognitive functioning in first-episode schizophrenia: MATRICS Consensus Cognitive Battery (MCCB) Profile of Impairment

Abstract

Background

Although many studies have assessed cognitive functioning in first-episode schizophrenia (FESz), the pattern and severity of impairment across cognitive domains remain unclear. Moreover, few studies have directly compared the pattern of cognitive performance between FESz and chronic schizophrenia (CSz). In this study we examined the cognitive impairment profile in FESz using a standardized neurocognitive battery (MATRICS Consensus Cognitive Battery; MCCB).

Methods

MCCB data were compared from 105 FESz patients, 176 CSz patients and 300 non-psychiatric (NP) participants. Mixed model analysis evaluated group differences in MCCB profiles and relative strengths and weaknesses in the MCCB profiles of patients. Clinical implications of MCCB performance were also examined; we compared the proportion of participants from each group who exhibited clinically-significant global cognitive impairment based on the MCCB Overall Composite score.

Results

FESz and CSz showed impaired performance across all MCCB domains relative to NP. With the exception of relative preservation of working memory and social cognition in FESz, the MCCB domain scores were similar in FESz and CSz. The distribution of impairment on the Overall Composite score did not significantly differ between FESz and CSz; compared to NP, both patient groups were overrepresented in moderate and severe impairment categories.

Conclusion

The pattern, magnitude, and distribution of severity of impairment in FESz were similar to that observed in CSz. However, early in the illness, there may be relative sparing of working memory and social cognition.

Introduction

Marked cognitive impairment is a core, enduring feature of schizophrenia, prompting some researchers to posit that the illness can be conceptualized as primarily a disorder of cognition (Rund, 1998, Green and Nuechterlein, 1999). Among patients with chronic schizophrenia (CSz), cognitive impairment is diffuse and pervasive, with deficits typically 1 to 2 standard deviations below non-psychiatric control samples across various cognitive domains (Heinrichs and Zakzanis, 1998, Gold, 2004, Green, 2006). Cognitive impairment appears to be a relatively stable feature of schizophrenia (Hoff et al., 2005, Barder et al., 2013b), even across acute vs. remitted states (Nuechterlein et al., 1992), with limited response to antipsychotic treatment (Rund, 1998, Keefe et al., 2007). Cognitive impairment is present in first-episode schizophrenia (FESz; Gold et al., 1999, Hoff et al., 1999, Mohamed et al., 1999, Addington and Addington, 2002, Barder et al., 2013a), and likely precedes the onset of illness in an attenuated form (Cornblatt et al., 1999, Hawkins et al., 2004, Lencz et al., 2006, Simon et al., 2007, Seidman et al., 2010, Lewandowski et al., 2011). However, the findings have been mixed regarding the magnitude of cognitive impairment in FESz compared to CSz, with some studies reporting negligible differences between FESz and CSz (e.g., Hoff et al., 1992), and others reporting less impairment in FESz relative to CSz, at least on some cognitive tests (e.g., Saykin et al., 1994, Albus et al., 1996, Addington and Addington, 2002, Townsend and Norman, 2004, Braw et al., 2008). Hence, the pattern and severity of impairment in FESz across cognitive domains remain unclear.

A meta-analysis of 43 studies of cognition in FESz (Mesholam-Gately et al., 2009) provided strong evidence for notable impairment across all cognitive domains, with medium to large mean effect sizes on par with those reported in CSz (Heinrichs and Zakzanis, 1998). Impairment was most pronounced in immediate verbal memory and processing speed, in addition to the global impairment observed in schizophrenia. Indeed, a selective deficit of verbal memory and processing speed has been reported elsewhere in the schizophrenia literature (e.g., Dickinson et al., 2008, Gonzalez-Blanch et al., 2010). The meta-analysis noted significant heterogeneity of effect sizes within cognitive domains and between studies, likely due to methodological variations across studies, such as differences in the operational definition of FESz, properties of the patient samples (e.g., diagnoses, demographic variables), medications, and possible cohort effects. Inconsistency in cognitive test batteries across studies provides an additional source of variability in that tests may differ substantially in reliability, which impacts estimates of effect size (Baugh, 2002). Moreover, given that the various tests were not normed on the same sample, it is difficult to directly compare differences in effect sizes across tests/domains (Russell et al., 2005).

The Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) Consensus Cognitive Battery (MCCB; Nuechterlein and Green, 2006) is a compilation of independently owned and published tests that initially had normative data from non-overlapping samples. The MCCB has the advantage of being co-normed (i.e., normative data for each test collected from the same sample) on the non-psychiatric (NP) community sample from the MATRICS Psychometric and Standardization Study (MATRICS PASS; Kern et al., 2008). The MCCB includes seven cognitive domains: speed of processing, attention/vigilance, working memory, verbal learning, visual learning, reasoning and problem solving, and social cognition. A brief description of the 10 tests comprising the MCCB can be found in Table 1. Five MCCB domains are assessed with one test each, thus the T-score and percentile for those domains are based on those individual tests. The speed of processing and working memory domains are assessed with multiple tests, and these domain scores are based on a composite of the included tests. In addition to the seven domain scores, the MCCB also provides an Overall Composite score, an index of cognitive functioning across domains. The Overall Composite score is derived through equal weighting of the seven MCCB domain scores (Nuechterlein and Green, 2006).

The profile of MCCB impairment for a CSz sample was previously reported (Kern et al., 2011). Compared to the NP sample, CSz patients were impaired across all MCCB domains, with greater relative impairment in speed of processing and working memory, and less relative impairment in reasoning and problem solving compared to their average performance across the remaining MCCB domains.

Adoption of the MCCB as a neuropsychological test battery in schizophrenia research is on the rise. At the time of this writing, ClinicalTrials.gov, an online registry of clinical trials across the globe maintained by the National Institutes of Health and National Library of Medicine, lists over 50 studies that employ the MCCB. The aim of this paper was to examine the overall magnitude and profile of cognitive impairment in FESz using the MCCB. Specifically, we compared MCCB domain scores for a FESz sample to data from CSz and NP participants in MATRICS PASS (Nuechterlein et al., 2008, Kern et al., 2011). Four hypotheses were tested: 1) that FESz patients would show significant impairment averaged across MCCB domains compared to NP; 2) that FESz patients would show similar magnitude of impairment as CSz patients; 3) consistent with the Mesholam-Gately meta-analysis (Mesholam-Gately et al., 2009), that FESz patients would exhibit particular weakness in speed of processing and verbal learning relative to performance in the remaining MCCB domains; and 4) finally, that the proportion of participants from each patient group who exhibit clinically-significant global cognitive impairment will not differ.