Lack of an inverse relationship between duration of untreated psychosis and cognitive function in first episode schizophrenia

https://doi.org/10.1016/j.schres.2008.11.003Get rights and content

Abstract

This study assessed the relationship between duration of untreated psychosis (DUP) and cognitive measures in order to assess if longer DUP was associated with worse performance. One hundred two patients with first episode schizophrenia or schizoaffective disorder were assessed on cognitive measures of speed of processing, episodic memory, executive function, and visual spatial processing at baseline (when patients were drug naive and after 16 weeks of olanzapine or risperidone treatment), so that a change score could be derived. DUP was defined by the emergence of psychiatric symptoms and the emergence of psychotic symptoms. Data were analyzed correlationally, parametrically (after the group was divided into long and short DUP by median split), and by regression. We found that DUP for psychotic symptoms in this group of patients was long, with a median of 46 weeks. Neither correlational, parametric analyses in which DUP served as a class variable, nor multiple regression indicated that longer DUP was associated with worse cognition at baseline or smaller magnitude of improvement in cognition. Our results suggest that while early intervention may be critical for symptom amelioration by shortening DUP, early intervention for treatment of psychiatric symptoms may have little or no impact on cognitive function. Furthermore, assuming that cognition is a core symptom of schizophrenia, the notion that ongoing psychosis is somehow toxic for a variety of information processing domains appears questionable.

Introduction

Duration of untreated psychosis (DUP) is a potentially important construct for understanding schizophrenia. Interest in DUP increased when Wyatt, 1991, Norman and Malla, 2001 proposed that it is neurotoxic by reducing neuronal connectivity. If there is a strong inverse relationship between DUP and illness course, outcome, symptom improvement, or cognitive level, this would support the idea that psychosis has a cumulative impact on plasticity-related phenomena. On the other hand, the absence of such a relationship would suggest that psychosis, while obviously reflecting aberrant neural circuitry, is not deleterious to such circuitry. The implications of this account are important because they suggest that early intervention is not only humane, but has neuroprotective effects.

Much research since has addressed DUP and symptom severity and short term outcome. The field was comprehensively reviewed by McGlashan (1999) and more recently subjected to a meta-analysis of 26 studies that found a significant, albeit modest, relationship between DUP and treatment response for such variables as symptom severity, positive symptoms, overall function, and remission, especially at six months post-initial treatment (Marshall et al., 2005). Strikingly, the relationship between baseline symptoms and DUP was found to be quite small. Our own previous work in a demographically similar sample to the one studied herein (Robinson et al., 1999, Robinson et al., 2004) suggests that DUP had a small but measurable association with functional recovery dimensions five years after the first episode of psychosis.

Several large scale studies have examined the relationship between DUP and cognition at study entry. In one of the initial studies, Hoff et al. (2000) found no relationship between these classes of measures. Ho et al. (2003) examined the relationships between DUP and cognition in a large group of first episode (FE) patients. Of nine cognitive domains, verbal memory was the only one to demonstrate a significant relationship with DUP. Similarly, in a Scandinavian sample, Rund et al., 2004, Rund et al., 2007 also observed negligible associations in a large sample of FE patients with schizophrenia or schizophrenic spectrum disorders using multiple cognitive indices. An important study utilizing carefully measured DUP indices and a comprehensive selection of cognitive tests that included current IQ, premorbid IQ, fluency, problem solving, updating, and attention, reported a single, albeit counterintuitive, correlation between DUP and intellectual deterioration (Norman et al., 2001). In an interesting approach to the issue Addington et al. (2004) examined DUP and cognition after two years of treatment, but found no relationship. In a large sample of psychotic patients recruited in Brazil, Ayres et al. (2007) found no relationship between DUP and a variety of cognitive measures. Positive studies have been less frequent (Amminger et al., 2002, Joyce et al., 2002, Lappin et al., 2007); however, the inconsistency across studies, with regard to both methods and results, have made these data difficult to interpret.

Specifically, interpretation of these studies has been hampered by several methodological issues. First, a key implication of DUP suggests a reduction in plasticity-related phenomena, presumably including those related to learning and memory. Moreover, DUP has been associated with treatment response, i.e., a change in, but not the baseline severity of symptoms. Thus, assessment not only of baseline cognitive scores might be necessary, but also assessment of changes in cognition over time and with treatment. The literature has also examined different definitions of DUP. These have generally centered around the emergence of any psychiatric symptoms (e.g., anergia, dysthymia, isolation, oddities in speech, belief, or perception), or frank and ratable positive symptoms. Furthermore, the length of DUP varied across studies, with some studies (Rund et al., 2004, Norman et al., 2001) reporting a relatively short median DUP, perhaps minimizing the likelihood of observing relationships among key clinical variables. In this study we were able to address these issues by utilizing an FE sample with a very high proportion of drug naive patients assessed at baseline, who had relatively long median DUPs, and for whom we were able to generate cognitive change scores over a 4 month period. In addition, through careful clinical tracking and structured assessment, we were able to assess two distinct measures of DUP (one based on emergence of any psychiatric symptom, the other based on emergence of psychotic symptoms).

Section snippets

Subjects

One hundred and two patients in their first episode of schizophrenia or psychosis participated in the study. By study's end 74 were diagnosed with schizophrenia, 10 with schizoaffective disorder, and 18 with schizophreniform disorder. Of these patients 80% were antipsychotic medication naive while assessed at baseline. Patients were then randomized to treatment with the SGA olanzapine or to the SGA risperidone. For the purposes of this study the treatment groups were collapsed, because in

Results

The results of the zero order correlational analyses examining the relationship between DUP and cognition are listed in Table 2. We used Spearman's rho for rank order to minimize the impact of outliers. Two zero order correlations between DUP (psychosis) and cognition at baseline were significant: a measure of verbal story recall (WMS-R Logical Memory) and a test of visual-perceptual skill (Judgment of Line Orientation — JLO). Longer DUP was associated with better performance on both of these

Discussion

We did not discern a relationship between longer DUP and worse cognition, irrespective of whether DUP was treated as a continuous variable or as a categorical variable or whether DUP was measured from the emergence of psychotic symptoms or any psychiatric symptom. Thus, our findings are not consistent with the hypothesis that psychosis per se is “toxic.” These results came in the context of neurocognitive measures that involved both baseline and change scores and DUPs that were sometimes quite

Role of the funding source

Funding for this study was provided by grants MH60004, NIDA K23 DA015541, MH41960 (The Zucker Hillside Center for Intervention Research in Schizophrenia), and RR018535 (Feinstein Institute for Medical Research General Clinical Research Center) from the National Institutes of Health. the NIMH had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

Contributors

Drs. Goldberg, Goldman, Kane, Lencz, Malhotra, and Robinson were involved in the design of the study. Drs. Kane, Malhotra, and Robinson were involved in the execution and implementation of the study. Drs. Burdick, Goldman, Lencz, Patel, Sevy, and Robinson and Ms. McCormack participated in the collection of data. Dr. Goldberg, Ms. Napolitano, Dr. Lencz, and Dr. Robinson were involved in data analysis. Drs. Goldberg, Burdick, Malhotra, and Robinson were involved in the writing of the study. All

Conflict of interest

Dr. Goldberg has consulted for Merck, Organon, Pfizer, and Wyeth. Dr. Kane is a consultant for Abbott, BMS, Pfizer, Janssen, and Lilly and lectures for BMS and Janssen. Dr. Robinson has received honoraria from Astra Zeneca Canada and grant support from Janssen and BMS. Dr. Goldman is now an employee of Pfizer, but at the time this study was designed and implemented, he was an employee at Zucker Hillside Hospital. All other authors have declared that they have no conflicts of interest.

Acknowledgements

We wish to thank Margaret G. Woerner, Ph.D., Nina R. Schooler, Ph.D., Handan Gunduz-Bruce, M.D., Jose M Soto-Perello, M.D., Alan Mendelowitz, M.D., Ali Khadivi, Ph.D., and Rachel Miller, LCSW for their important contributions to the study from which our data were obtained.

References (20)

There are more references available in the full text version of this article.

Cited by (0)

View full text