Negative symptoms have greater impact on functioning than positive symptoms in schizophrenia: Analysis of CATIE data

Abstract

Increased attention has been given to treatment of negative symptoms and its potential impact on functional outcomes, however previous inferences have been confounded by the fact that measures of functional outcomes often use items similar to those of negative symptoms. We attempted to discern the relative effects of negative symptoms on functioning, as compared to other symptoms, using data from the National Institute of Mental Health CATIE trial of chronic schizophrenia (n = 1447) by examining correlations of Positive and Negative Syndrome Scale factors, Calgary Depression Rating Scale and select items from Heinrich's and Lehman's Quality of Life Scales measuring aspects of functioning that did not overlap with negative symptoms. Baseline functioning and change in functioning were more strongly related to PANSS negative factor than any of the other symptoms — though the amount of variance explained by symptom changes in general was small. The data suggests that improvement in negative symptoms may have a distinctive and independent effect on functional outcome relative to other symptoms. This should be further tested in studies where negative symptoms improve without concomitant improvement of other symptoms.

Introduction

Recently, increased attention has been given to treatment of negative symptoms and to the potential impact that this could have on functional outcomes. Concomitantly, there has been a heightened recognition of functioning as a target for pharmacotherapy. Some work has suggested that negative symptoms are inversely correlated with functional outcomes (McGurk et al., 2000, Herbener and Harrow, 2004, Milev et al., 2005, Rocca et al., 2009). This is encouraging about the possibility that treatments targeting negative symptoms could lead to additional improvement on psychosocial functioning. It has, however, been difficult to interpret prior findings as measures of psychosocial functioning in general, and those specifically used in those studies, included items measuring social functioning that overlap with negative symptoms. Herbener and Harrow (2004) used the Strauss and Carpenter Outcome Scale (Strauss and Carpenter, 1972) which includes four questions, two of which measure the quantity and quality of social relationships. Another used the Global Assessment of Functioning scale that evaluates functioning across three domains one of which is social functioning (Rocca et al., 2009) and Milev et al. (2005) used a global psychosocial functioning assessment that included interpersonal relations as a key area. Thus these scales capture two major negative symptoms, passive and active social withdrawal.

Another concern about studies examining negative symptom improvement is pseudospecificity (Leber, 2002). Specifically, the concern is that improvement in negative symptoms is claimed to drive the change in functioning without ruling out the possibility that positive or other symptoms may have actually improved to the same extent, or even more and that changes in functioning may be attributed to those changes. These concerns have hampered attempts to examine the effectiveness of antipsychotic drugs on negative symptoms (Kirkpatrick et al., 2006). For this reason, previous analysis of symptoms and functioning in CATIE (Swartz et al., 2003) either did not include negative symptoms, due to a concern that measures of functioning were capturing negative symptoms, or excluded the following key negative factor symptoms: “difficulty in abstracting,” “emotional withdrawal,” and “passive/apathetic social withdrawal” (Mohamed et al., 2008). The current study attempts to discern the relative effects of negative symptoms on functioning, as compared to other symptoms, while addressing the methodological limitations of previous research. Specifically we chose measures of functioning that do not include social functioning and controlled for pseudospecificity by examining other symptoms concomitantly.