Are schizophrenia and schizoaffective disorder neuropsychologically distinguishable?

Available online at www.sciencedirect.com

Schizophrenia Research 99 (2008) 149 – 154 www.elsevier.com/locate/schres

Are schizophrenia and schizoaffective disorder neuropsychologically distinguishable? R. Walter Heinrichs a,⁎, Narmeen Ammari a , Stephanie McDermid Vaz b , Ashley A. Miles a a

b

Department of Psychology, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3 Cleghorn Early Intervention in Psychosis Program, St Joseph's Healthcare Hamilton and Hamilton Health Sciences, 25 Charlton Ave East, Suite 703, Hamilton, ON, Canada L8N 1Y2 Received 27 June 2007; received in revised form 16 October 2007; accepted 18 October 2007 Available online 26 November 2007

Abstract This study sought to objectify the distinction between schizophrenia and schizoaffective disorder in terms of standard tasks measuring verbal and non-verbal cognitive ability, auditory working memory, verbal declarative memory and visual processing speed. Research participants included 103 outpatients with a diagnosis of schizophrenia, 48 with schizoaffective disorder, and 72 non-patients from the community. Schizophrenia patients were impaired on all cognitive measures relative to schizoaffective patients and non-psychiatric participants. Regression-based prediction models revealed that cognitive measures classified schizophrenia patients accurately (91%), but not patients with schizoaffective disorder (35%). In addition, there was no statistical evidence for the unique predictive validity of any specific cognitive task. Patients with schizophrenia were significantly more symptomatic and had greater community support requirements than those with schizoaffective disorder. However, group differences in cognitive performance are insufficient to separate these syndromes of psychotic illness. © 2007 Elsevier B.V. All rights reserved. Keywords: Schizoaffective disorder; Neuropsychology; Diagnosis

1. Introduction Psychiatric nosology defines and distinguishes schizoaffective disorder from schizophrenia on the basis of a mood disturbance, which may occur in the presence of schizophrenic symptoms. Yet whether this clinical distinction also reflects a biological difference, or possibly a ⁎ Corresponding author. Tel.: +1 416 736 2100x66205; fax: +1 416 736 58145. E-mail addresses: [email protected] (R.W. Heinrichs), [email protected] (N. Amari), [email protected] (S. McDermid Vaz), [email protected] (A.A. Miles). 0920-9964/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2007.10.007

continuum, has been the subject of considerable debate (Kempf et al., 2005). Thus one view holds that patients with schizoaffective disorder have schizophrenia along with “incidental” mood-related symptoms. Another perspective is that schizophrenia and bipolar illness lie at opposite ends of a disease dimension, with schizoaffective disorder located roughly in the middle. Biological evidence in the form of genetic and population-based studies of the two conditions is limited and provides no definitive support for one view over another. However, recent reviews and theories propose that schizoaffective disorder is an intermediate condition with genetic links to both schizophrenia and bipolar illness (Hamshere et al.,

150

R.W. Heinrichs et al. / Schizophrenia Research 99 (2008) 149–154

2005; Laursen et al., 2005; Potash, 2006) or part of a single psychotic disorder that subsumes schizophrenia and the mood disorders (Lake and Hurwitz, 2006). Part of the challenge in understanding the etiology and pathophysiology of schizophrenia-related disorders stems from the use of illness definitions that are grounded in symptoms and history. Hence there is a corresponding reliance on patient self-report and subjective data. It has been argued that greater definitional sophistication is required before advances in genetics and neurobiology can clarify syndromes like the psychoses (Gottesman and Gould, 2003; Kendell and Jablensky, 2003). A variety of biobehavioral and neurocognitive measures are available that may objectify and validate, or modify, prevailing illness definitions (Conklin and Iacono, 2003). Can these measures be used to improve the definitions of schizophrenia and schizoaffective disorder? There is evidence that neurocognitive test performance may be useful in organizing psychiatric syndromes like the psychoses into more biologically homogeneous variants and subgroups (Heinrichs, 2005). For example, schizophrenia patients with and without generalized cognitive impairment have distinct genetic profiles (Hallmayer et al., 2005). Similarly, cognitive performance may distinguish schizophrenia from schizoaffective disorder if the diagnostic categories map onto different neurobiological entities. In support of this idea, Stip et al. (2005) reported significant differences between schizophrenia and schizoaffective patients on computerized measures of visuomotor speed and non-verbal memory. Furthermore, Gruber et al. (2006) demonstrated preserved auditory working memory in schizoaffective disorder relative to schizophrenia patients. These results conflict with earlier findings of no group differences on an extensive set of neurocognitive measures (Evans et al., 1999). Not surprisingly, lack of differentiation is interpreted as support for the view that schizoaffective disorder is a subtype of schizophrenia and not a distinct entity. However, unlike the more recent studies, Evans et al. evaluated middleaged and elderly patients and included a substantial number not receiving anti-psychotic medication. These characteristics imply that patient samples generating negative findings may be unrepresentative of the schizophrenia population. It is unknown whether different cognitive tasks vary in their sensitivity to diagnostic distinctions within the schizophrenia spectrum or all index the same broadly based dysfunction. Specific aspects of cognitive performance, including word recognition and vocabulary, resist many neurobiological disease processes and may therefore reflect premorbid abilities (Hawkins, 1998; Johnstone et al., 1995; Johnstone and Wilhelm, 1996). Hence it is

noteworthy that Reichenberg et al. (2002) found schizophrenia and schizoaffective disorder patients to be statistically indistinguishable on intellectual and language measures of premorbid functioning. Nonetheless, their schizoaffective disorder sample was small, reducing statistical power and the ability to detect group differences. Accordingly, whether generalized or more selective impairments distinguish these syndromes of psychotic illness remains unclear. The primary purpose of the present investigation was to determine if a set of commonly used neurocognitive measures yields performance differences in patients with schizophrenia compared to those with schizoaffective disorder. It was expected that schizophrenia patients would be significantly impaired relative to patients with schizoaffective disorder and healthy people. However, unlike previous studies, we were concerned not only with detecting group differences, but also with the ability of cognitive performance to recapitulate and thereby to validate the two diagnostic categories. By itself, a statistically significant difference in cognition is weak validation because extensive overlap may still exist between groups. In addition, the relative sensitivity of putative premorbid as well as present-state and combined cognitive indicators was assessed. 2. Materials and methods 2.1. Participants The clinical sample comprised 151 male and female patients who met the following criteria: 1) diagnosis of schizophrenia (n = 103) or schizoaffective disorder (n = 48) by DSM-IV criteria; 2) outpatient status; 3) age 18–65 years; 4) no history of serious neurological or endocrine disorder, including head trauma, epilepsy, Cushing's disease or thyroid disorder; 5) no concurrent DSM-IV [10] diagnosis of substance abuse; 6) no history of developmental disability; 7) minimum English reading level of grade 6; 8) willingness and ability to sign informed consent; and 9) normal or corrected vision. One hundred and forty-six patients were receiving antipsychotic medication and 5 patients were temporarily medication-free at the time of the study. Demographic and clinical characteristics are presented in Table 1. Nonpsychiatric control participants (N = 72) were screened for medical and psychiatric illness and history of drug abuse. All participants were paid for their time. Patients were recruited at the Community Schizophrenia Service (CSS) of St. Joseph's Healthcare Hamilton, the Hamilton Program for Schizophrenia (HPS), the Toronto, Ontario branch of the Canadian

R.W. Heinrichs et al. / Schizophrenia Research 99 (2008) 149–154

151

Table 1 Demographic, clinical and functional characteristics of patient and non-psychiatric participants Schizophrenia

Schizoaffective

Non-psychiatric

42.60 (8.98) 63 69 86

38.73 (9.38) 79 54 75

40.64 (11.4) 94 74 74

45 39 14 27

54 35 8 23

– – – –

PANSS Positive scale, M (SD) Negative scale, M (SD) General scale, M (SD)

21.27 (6.29) 20.56 (6.10) 43.64 (9.10)

17.31 (4.74) 16.45 (5.52) 38.25 (8.02)

MSIF Global support, M (SD)

3.96 (1.39)

3.38 (1.44)

Age in years, M (SD) High school completion, % Gender, % male First language, % English Antipsychotic medication Clozapine, % Other atypical, % Typical, % Anti-Parkinsonian medication, %

Statistics 2.20 a 23.48 b 5.19 c 5.19 d 1.60 e

0.43 f

3.88 g 3.89 h 3.52 i

1.15 (0.36)

126.01 j

Note: Anti-Parkinsonian medication data were missing for 6 patients. a F(2, 220), p N .05. b χ2(N = 151), p b .001. c χ2(N = 151), p N .05. d χ2(N = 151), p N .05. e χ2(N = 151), p N .05. f χ2(N = 145), p N 05. g t(149), p b .001. h t(149), p b .001. i t(149), p .01. j F(2, 220), p b .001.

Mental Health Association (CMHA) and satellite clinics of Whitby Mental Health Center. Non-psychiatric participants were recruited through advertisements in local media. Written informed consent was obtained from all participants. The project was approved by the institutional review board at each research site and by York University. 2.2. Assessments Patients were administered the Structured Clinical Interview for DSM-IV-TR (First et al., 2002) to determine diagnosis. Clinical variables measured included symptom severity and functional status. Symptoms were assessed with the Positive and Negative Syndrome Scale (PANSS; Opler et al., 1999). The PANSS provides severity ratings of positive, negative and general psychiatric symptoms based on a structured interview format. Functional status was measured with the Multidimensional Scale of Independent Functioning (MSIF; Jaeger et al., 2003). The MSIF is in part a structured interview and self-report measure, with verification of information provided by history, proxy reports and informant interviews. The Global Support rating reflects the amount of assistance a

person receives across residential, work-related and educational settings. A rating of 1 indicates no significant support beyond that considered a community norm for healthy people. In contrast, a rating of 4 indicates moderate levels of support in all settings or comprehensive support in one setting. A rating of 7 indicates that a person receives complete services in all settings. Ratings are based on detailed “anchors” provided for each environment. The Global Support rating has been shown to have good inter-rater reliability (intraclass correlation = .75) and its component ratings vary significantly and in expected directions with differences in patients' residential, work and educational situations. Neurocognitive measures included the Vocabulary subscale of the Wechsler Adult Intelligence Scale (WAIS-III; Wechsler, 1997) to measure verbal intellectual ability. Non-verbal ability was measured with the Matrix Reasoning subscale, auditory working memory with the Letter–Number Sequencing subscale and visual information processing and performance speed with the Symbol Search subscale of the WAIS-III. In addition, patients were administered the California Verbal Learning Test (CVLTII; Delis et al., 2000). The cumulative recall total over five learning trials with a 16-item word list was used as an

152

R.W. Heinrichs et al. / Schizophrenia Research 99 (2008) 149–154

index of verbal learning and memory. Response generation and word fluency in response to stimulus letters were measured with the Controlled Oral Word Association Test (Benton and Hamsher, 1983). Visual word recognition was assessed with the Reading subtest of the Wide Range Achievement Test (WRAT-3; Wilkinson, 1993). Vocabulary and reading tasks are regarded as indicators of premorbid cognitive ability insofar as performance is known to be robust in a variety of diseases and neurological events, including schizophrenia (Hawkins, 1998; Heinrichs and Zakzanis, 1998). In contrast, performance on non-verbal conceptualization, working and declarative memory, processing speed and response generation tasks is believed to be highly vulnerable to the pathophysiology underlying psychotic illness (Nuechterlein et al., 2004). All of the tests described were administered and scored by a trained clinical research assistant. 3. Results The clinical samples comprised patients ranging from 21 to 65 years of age (see Table 1). The patients were predominately male (64%), with English as their first spoken language (83%). Patients with schizophrenia and schizoaffective disorder and healthy non-psychiatric participants did not differ in age, first language or sex distribution. However, the clinical and healthy groups differed in educational achievement, with non-psychiatric participants having significantly higher high school completion rates. With p values adjusted for multiple comparisons, the two clinical samples did not differ significantly in terms of the proportion of patients completing high school or receiving different types of medication. Schizophrenia patients experienced significantly higher positive and negative symptom levels

(PANSS), general distress (PANSS) and functional support requirements (MSIF) than patients with schizoaffective disorder. Overall, nonetheless, both clinical samples demonstrated symptom and functional levels considered average for treated schizophrenia patients (Jaeger et al., 2003; Opler et al., 1999). The findings with respect to cognitive performance are presented in Table 2. The overall F tests for group differences were significant for each measure. Moreover, pair-wise comparisons indicated that the schizoaffective group significantly outperformed the schizophrenia group on all tasks. Effect sizes (ES) were calculated by dividing each mean group difference by the pooled standard deviation (Cohen, 1988). This produced the following rank order: WAIS-III Vocabulary (ES = .70), WAIS-III Symbol Search (ES = .65), CVLT-II Total Recall (ES = .65), WAIS-III Letter– Number Sequencing (ES = .54), WAIS-III Matrix Reasoning (ES = .52), COWAT Total Words (ES = .47) and WRAT Reading (ES = .46). Logistic regression, which is suited to making predictions when dependent variables are dichotomous, was used to examine the ability of the cognitive measures to predict diagnosis and accurately classify patients into schizophrenia and schizoaffective categories. This procedure was carried out in 2 steps, with premorbid ability tasks (WRAT-3 Reading, WAIS-III Vocabulary) entered as the first block of predictors followed by present-state tasks (WAIS-III Letter–Number Sequencing, Matrix Reasoning, Symbol Search; CVLT-II total recall; COWAT total words) as the second block. The equation based on premorbid ability tasks alone correctly classified 90% of schizophrenia patients, but only 23% of the schizoaffective group, resulting in an overall accuracy of 69%. The regression weight for WAIS-III Vocabulary

Table 2 Cognitive test results for schizophrenia (S), schizoaffective (SA) and non-psychiatric (NP) participants Schizophrenia

Schizoaffective

Non-psychiatric

ANOVA

Pairwise comparisons

M

M

M

F (2, 220)

Bonferroni ( p b .05)

WAIS-III Vocabulary Matrix reasoning Letter–Number Symbol search

SD

SD

SD

8.99 8.60 7.93 6.71

3.44 3.18 2.99 2.76

11.31 10.23 9.56 8.52

3.02 3.35 3.19 3.12

11.25 11.67 10.86 9.46

3.24 2.93 3.23 3.34

13.18b 20.46b 19.10b 18.30b

S b SA, NP S b SA b NP S b SA, NP S b SA, NP

CVLT-II Total recall

38.10

11.37

45.46

11.08

50.88

11.28

27.84b

S b SA b NP

COWAT Total words WRAT reading

32.42 93.09

11.29 14.05

37.71 99.01

11.71 11.98

38.81 103.18

14.18 9.86

6.52a 14.49b

S b SA, NP S b SA, NP

Note: a p b .01; b p b .001.

R.W. Heinrichs et al. / Schizophrenia Research 99 (2008) 149–154

differed significantly from 0 (Wald statistic = 7.61; p b .01). Hence Vocabulary contributed uniquely to the validity of the equation. Inclusion of present-state tasks (step 2) in the regression equation yielded a schizophrenia classification of 91% and an improved schizoaffective classification of 35%, with an overall accuracy of 74%. In addition, the amount of diagnostic variance accounted for increased from 13% to 21% with inclusion of all variables. However, none of the individual regression weights, including WAIS-III Vocabulary, differed significantly from 0 on the basis of Wald statistics. This indicates that no individual task contributed uniquely to validity once all test variables were in the equation. 4. Discussion The findings of this study show that, on average, schizoaffective disorder is a less cognitively compromised form of psychotic illness than schizophrenia. All group differences on a set of standard neuropsychological tests yielded statistically significant moderate-tolarge effect sizes. However, there was no evidence to suggest any selective sensitivity of individual tasks, including those regarded as indicators of premorbid ability memory. Moreover, significant group differences in task performance did not translate into high rates of diagnostic accuracy. Approximately 2/3 of patients with a diagnosis of schizoaffective disorder were misclassified as schizophrenia patients based on the full set of cognitive predictors. This contrasts with a relatively low misclassification rate for schizophrenia patients. Accordingly, the results indicate more severe average cognitive impairment in the population of patients diagnosed with schizophrenia, but very extensive overlap with the schizoaffective patient population. These results contradict reports by Evans et al. (1999), Bornstein et al. (1990) and Beatty et al. (1993), which found no or limited and highly test-specific group differences on comprehensive neurocognitive measures. Our findings are partially consistent with recent applications of auditory working memory paradigms that demonstrate preserved performance in schizoaffective disorder, but not in schizophrenia (Gruber et al., 2006). On the other hand, our data clearly reflect a broadly based cognitive impairment in both forms of psychotic illness and we found no evidence that impairment occurs preferentially in working memory or any other specific ability. In the absence of objective and independent confirmation that schizophrenia and schizoaffective disorder represent separable entities, the symptomatic and functional disability differences in our patient samples may reflect a disease severity continuum (Lake and Hurwitz,

153

2006). However, differences in functional status may be partly an artifact of the explicit requirement for significant social and occupational dysfunction in the diagnostic criteria for schizophrenia (American Psychiatric Association, 1994). Furthermore, our study did not include a group with a diagnosis of bipolar disorder, and thus cannot address the possibility that schizoaffective disorder is part of a single psychotic disorder spectrum rather than a schizophrenia spectrum. Our study was also limited in that specific measures of basic motor speed and non-verbal learning were not included in the test battery. Therefore the effect on diagnostic accuracy of including such measures cannot be determined. Finally, it has been noted that the diagnosis of schizoaffective disorder is often unstable over time and most studies, including our own, are crosssectional in design (Averill et al., 2004). Hence, some patients in the schizoaffective group might receive a different diagnosis on follow-up. However, unless movement between diagnostic categories systematically varies with degree of cognitive impairment along with shifting symptoms, such instability does not invalidate crosssectional results. In conclusion, patients with schizophrenia and schizoaffective disorder demonstrate significant group differences in objective cognitive performance, symptom severity and functional status. However, cognitive differences are insufficient in magnitude to provide objective neuropsychological validation for 2 distinct and separable psychotic syndromes. Such results are most consistent with severity continuum theories of the relation between these syndromes. Role of funding source Funding for this study was provided by the Ontario Mental Health Foundation (OMHF); the OMHF 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 W. Heinrichs designed the study and wrote the protocol. S. McDermid Vaz and A. Miles collected the data. W. Heinrichs and N. Ammari conducted the analyses. All authors contributed to and have approved the final manuscript. Conflict of interest There are no conflicts of interest with respect to this manuscript. Acknowledgments We thank Andrew Miki, Dr. Joel Goldberg, Susan Strong, Dr. Suzanne Archie, Dr. Peter Prendergast, Marta Statucka, the Community Schizophrenia Service, Hamilton Program for Schizophrenia, Cleghorn Program, St. Joseph's Healthcare Hamilton, Whitby Mental Health Centre and the Challenging Directions program for their assistance in data collection and study management.

154

R.W. Heinrichs et al. / Schizophrenia Research 99 (2008) 149–154

References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, fourth ed. American Psychiatric Association, Washington, DC. Averill, P.M., Reas, D.L., Shack, A., Shah, N.N., Cowan, K., Krajewski, K., Kopecky, C., Guynn, R.W., 2004. Is schizoaffective disorder a stable diagnostic category: a retrospective examination. Psychiatr. Q. 75 (3), 215–227. Beatty, W.W., Jocic, Z., Monson, N., Staton, R.D., 1993. Memory and frontal lobe dysfunction in schizophrenia and schizoaffective disorder. J. Nerv. Ment. Dis. 181 (7), 448–453. Benton, A.L., Hamsher, K., 1983. Multilingual Aphasia Examination. AJA Associates, Iowa City, Iowa. Bornstein, R.A., Nasrallah, H.A., Olson, S.C., Coffman, J.A., Torello, M., Schwarzkopf, S.B., 1990. Neuropsychological deficit in schizophrenia subtypes: paranoid, nonparanoid, and schizoaffective subgroups. Psychiatry Res. 31 (1), 15–24. Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences, second ed. Academic Press, New York. Conklin, H.M., Iacono, W.G., 2003. At issue: assessment of schizophrenia: getting closer to the cause. Schizophr. Bull. 29 (3), 405–411. Delis, D.D., Kramer, J., Kaplan, E., Ober, B.A., 2000. California Verbal Learning Test, second ed. The Psychological Corporation, San Antonio, TX. Evans, J.D., Heaton, R.K., Paulsen, J.S., McAdams, L.A., Heaton, S.C., Jeste, D.V., 1999. Schizoaffective disorder: a form of schizophrenia or affective disorder? J. Clin. Psychiatry 60 (12), 874–882. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2002. Structured Clinical Interview for DSM-IV Axis I Disorder-Patient Edition (SCID-I/P). Biometrics Research Department, New York State Psychiatric Institute, New York, NY. Gottesman, I.I., Gould, T., 2003. The endophenotype concept in psychiatry: etymology and strategic intentions. Am. J. Psychiatry 160 (4), 1–10. Gruber, O., Gruber, E., Falkai, P., 2006. Articulatory rehearsal in verbal working memory: a possible neurocognitive endophenotype that differentiates between schizophrenia and schizoaffective disorder. Neurosci. Lett. 405 (1–2), 24–28. Hallmayer, J.F., Kalaydjieva, L., Badcock, J., Dragovic, M., Howell, S., Michie, P.T., Rock, D., Vile, D., Williams, R., Corder, E.H., Hollingsworth, K., Jablensky, A., 2005. Genetic evidence for a distinct subtype of schizophrenia characterized by pervasive cognitive deficit. Am. J. Hum. Genet. 77 (3), 468–476. Hamshere, M.L., Bennett, P., Willmans, N., Segurado, R., Cardono, A., Norton, N., Lambert, D., Willmans, H., Kirov, G., Corvin, A., Holmans, P., Jones, L., Jones, I., Gill, M., O'Donovan, M.C., Owen, M.J., Craddock, N., 2005. Genomewide linkage scan in schizoaffective disorder: significant evidence for linkage at 1q42 close to DISC1, and suggestive evidence at 22q11 and 19p13. Arch. Gen. Psychiatry 62 (10), 1081–1088. Hawkins, K.A., 1998. Indicators of brain dysfunction derived from graphic representations of the WAIS-III/WMS-III technical manual

clinical samples data: a preliminary approach to clinical utility. Clin. Neuropsychol. 12 (4), 535–551. Heinrichs, R.W., 2005. The primacy of cognition in schizophrenia. Am. Psychol. 60 (3), 229–242. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12 (3), 426–445. Jaeger, J., Berns, S.M., Czobor, P., 2003. The multidimensional scale of independent functioning: a new instrument for measuring functional disability in psychiatric populations. Schizophr. Bull. 29 (1), 153–168. Johnstone, B., Wilhelm, K.L., 1996. The longitudinal stability of the WRAT-R Reading subtest: is it an appropriate estimate of premorbid intelligence? J. Int. Neuropsychol. Soc. 2 (4), 282–285. Johnstone, B., Hexum, C.L., Ashkanazi, G., 1995. Extent of cognitive decline in traumatic brain injury based on estimates of premorbid intelligence. Brain Inj. 9 (4), 377–384. Kempf, L., Hussain, N., Potash, J.B., 2005. Mood disorder with psychotic features, schizoaffective disorder, and schizophrenia with mood features: trouble at the borders. Int. Rev. Psychiatry 17 (1), 9–19. Kendell, R., Jablensky, A., 2003. Distinguishing between the validity and utility of psychiatric diagnoses. Am. J. Psychiatry 160 (1), 4–12. Lake, C.R., Hurwitz, N., 2006. Schizoaffective disorders are psychotic mood disorders; there are no effective disorders. Psychiatry Res. 143 (2–3), 255–287. Laursen, T.M., Labouriau, R., Licht, R.W., Bertelsen, A., Munk-Olsen, T., Mortensen, P.B., 2005. Family history of psychiatric illness as a risk factor for schizoaffective disorder: a Danish register-based cohort study. Arch. Gen. Psychiatry 62 (8), 841–848. Nuechterlein, K.H., Barch, D.M., Gold, J.M., Goldberg, T.E., Green, M.F., Heaton, R.K., 2004. Identification of separable cognitive factors in schizophrenia. Schizophr. Res. 72 (1), 29–39. Opler, L.A., Kay, S.R., Lindenmayer, J.P., Fiszbein, A., 1999. Structured Clinical Interview: The Positive and Negative Syndrome Scale (SCI-PANSS). Multi-Health Systems Inc, North Tonawanda, NY. Potash, J.B., 2006. Carving chaos: genetics and the classification of mood and psychotic syndromes. Harv. Rev. Psychiatry 14 (2), 47–63. Reichenberg, A., Weiser, M., Rabinowitz, J., Caspi, A., Schmeidler, J., Mark, M., Kaplan, Z., Davidson, M., 2002. A population-based cohort study of premorbid intellectual, language, and behavioral functioning in patients with schizophrenia, schizoaffective disorder, and nonpsychotic bipolar disorder. Am. J. Psychiatry 159, 2027–2035. Stip, E., Sepehry, A.A., Prouteau, A., Briand, C., Nicole, L., Lalonde, P., Lesage, A., 2005. Cognitive discernible factors between schizophrenia and schizoaffective disorder. Brain Cogn. 59 (3), 292–295. Wechsler, D., 1997. Wechsler Adult Intelligence Scale, third ed. The Psychological Corporation, San Antonio, TX. Wilkinson, G.S., 1993. Wide Range Achievement Test WRAT3: Administration Manual. Wide Range, Inc., Wilmington, DL.