Cognitive Change in Schizophrenia and Other Psychoses in the Decade Following the First Episode

1 Objective: Schizophrenia is associated with a large cognitive impairment that is widely believed to 2 remain stable after illness onset. Yet, even to date, 10-year prospective studies of cognitive 3 functioning following the first episode with good methodology are rare. We examined whether 4 schizophrenia patients experience cognitive decline following the first episode, whether this decline 5 is generalized or confined to individual neuropsychological functions, and whether decline is specific 6 to schizophrenia. 7 Method: Participants were from a population-based, case-control study of patients with first- 8 episode psychosis that were followed prospectively up to 10 years post first admission. A 9 neuropsychological battery was administered at index presentation and at follow-up to patients with 10 a diagnosis of schizophrenia (n=65), or other psychoses (n=41), as well as to healthy comparison 11 subjects (n=103). 12 Results: The schizophrenia group exhibited declines in IQ and in measures of verbal knowledge, and 13 memory, but not processing speed or executive functions. Processing speed and executive function 14 impairments were already present at the first episode and remained stable thereafter. Magnitude of 15 declines ranged between 0.28 and 0.66 standard deviations. Decline in measures of memory was not 16 specific to schizophrenia and was also apparent in the group of patients with other psychoses. 17 Healthy individuals with low IQ, on the other hand, showed no evidence of decline, suggesting that a 18 decline is specific to psychosis. 19 Conclusions: Patients with schizophrenia and other psychoses experience cognitive decline after 20 illness onset, but the magnitude of decline varies across cognitive functions. Distinct mechanisms 21 consequent upon the illness and/or psychosocial factors may underlie impairments across different 22 cognitive functions.


Introduction
Cognitive impairment is a core feature of schizophrenia (1,2). Understanding the nature and course 25 of this impairment may have important implications for our understanding of the pathophysiology of 26 the disorder. 27 Research has shown that individuals diagnosed with schizophrenia experience cognitive decline from 28 the premorbid to post-onset period. There is clear evidence for moderate cognitive deficits in 29 children and adolescents who later develop schizophrenia, with meta-analyses showing an average 30 premorbid deficit equal to 8 IQ points (0.5 Standard Deviation (SD)) (3,4). Cognitive deficits in adults 31 diagnosed with schizophrenia are more pronounced, with meta-analyses reporting a 14-point IQ 32 deficit (0.90 SD) in first-episode schizophrenia patients (5) and 15-to 21-point IQ deficits (1.0 to 1. 5 33 SD) in chronic schizophrenia patients (1,6,7). In line with cross-sectional evidence, longitudinal 34 studies of cognitive change in schizophrenia from before to after illness onset have shown evidence 35 for cognitive decline (8). Three population-based studies have reported cognitive declines ranging 36 from 6 to 12 IQ points (0.4 -0.8 SD) between childhood and adulthood in individuals later diagnosed 37 with schizophrenia (8)(9)(10). 38 Despite evidence for cognitive decline from before to after illness onset, the course of cognitive 39 decline in schizophrenia remains unclear. While it is widely believed that cognitive impairments 40 stabilize after illness onset (11)(12)(13), at least until older adult life (12,14), few longitudinal studies 41 have examined cognitive change from illness onset through to a decade later (sTable 1), and findings 42 across studies and cognitive domains are mixed. Studies have reported a stabilization of the 43 cognitive deficits, cognitive decline, as well as amelioration of cognitive functioning (sTable 1 and ref 44 # (15)). 45 Previous studies have been unable to comprehensively chart the course of cognitive deficits for 46 several reasons. First, the majority of studies have used clinical samples, which may not be fully 47 representative of the population of individuals with schizophrenia (8). Second, most studies followed 48 participants for only 1 to 3 years from illness onset (sTable 1). We previously reported a slow, 49 gradual increase in premorbid cognitive deficits, with losses equal to between 0.5 and 1 IQ point per 50 year (16). Studies with short follow-ups, therefore, may be underpowered to capture decline. Third, 51 few studies have included comparison groups, and therefore have not considered the potential 52 impact of normative age-associated changes in cognitive functioning, which is necessary to 53 rigorously test for cognitive change. Since, brain maturation continues into the third decade of life 54 (17), previous estimates of the magnitude of cognitive decline may be biased. Finally, few studies 55 have examined the effect of medication on cognitive functioning, and yet recent findings suggest 56 that antipsychotic medications may contribute to the severity of cognitive decline (18). 57 In a previous report on this population-based, case-control study, we provided evidence for an IQ 58 deficit, as well as varying degrees of impairment across individual cognitive domains following the 59 first psychiatric diagnosis of schizophrenia (19). Study participants have since been followed-up and 60 underwent neuropsychological testing a second time. Using identical neuropsychological measures 61 at first assessment and follow-up, we were able to directly examine change in IQ and in individual 62 cognitive functions after the first episode. To provide an accurate estimate of cognitive change over 63 time, we compared patients to the healthy comparison subjects in the study followed during the 64 same period. We tested three hypotheses. First, we examined the "IQ decline" hypothesis to 65 establish whether schizophrenia patients exhibit a static IQ deficit or IQ decline. Second, we tested 66 the "generalized decline" hypothesis to determine whether decline occurs across multiple cognitive 67 domains, namely verbal knowledge, memory, language, processing speed, executive 68 function/working memory and visuospatial ability. Finally, we tested the "specificity" hypothesis to 69 establish whether any cognitive decline is specific to schizophrenia or common to other psychoses 70 by examining cognitive change in individuals with psychotic disorders other than schizophrenia. 71 72 defined by ICD-10), and IQ<50. A random sample of control subjects with no past or present 86 psychotic disorder were recruited using a sampling method that matched cases and controls by area 87 of residence. Hereafter, data collected at this phase of the AESOP study is referred to as 'baseline'. 88

AESOP Study
At baseline, detailed information was collected to enable patients to be traced, re-contacted and re-89 interviewed approximately 10 years later ('follow-up'). At follow-up, patients currently in contact 90 with mental health services were invited to participate through their clinical teams. Letters of 91 invitation were sent to last known addresses of those not in contact with services. Non-responders 92 were sent a second letter two to three weeks later. If patients were thought to have moved, contact 93 was sought through their GP. Control subjects also provided contact details at baseline. Letters of 94 invitation were sent and were followed-up with phone calls if no reply had been received within 2 95 weeks. If no reply had been received after 4 weeks, or where telephone numbers could not be 96 obtained, in-person visits were made to the subject's address. A detailed overview of the AESOP 97 study design and methods, as well as the follow-up has been published elsewhere (20,21). 98

Analytic Cohort 99
Derivation of the sample included in the present analysis is illustrated in Figure 1. The analytic 100 cohort consisted of healthy comparison subjects and subjects who had a consensus ICD-10 diagnosis 101 at last follow-up of schizophrenia (F20), bipolar disorder or mania (F30.2, F31.2, F31.5), depressive 102 psychoses (F32.3, F33.3) or other psychotic disorders including persistent delusional disorders and 103 psychosis NOS (F22, F23, F28, F29). Both case and comparison subjects were required to be native 104 English speakers or to have migrated to the UK by age 11. The latter ensured that all participants had 105 a good command of English, even as a non-native language, by verifying that participants had 106 completed at least their secondary education in the UK. Thus, this minimized the effect of linguistic 107 or cultural biases on cognitive performance in a multiethnic sample. 108

Neuropsychological assessment 113
At baseline and follow-up, participants underwent cognitive testing with a neuropsychological 114 battery, which assessed general intellectual ability (IQ), as well as specific cognitive functions. 115 Administration and scoring followed standard procedures. Full-scale IQ was estimated using the 116 vocabulary, comprehension, digit symbol coding and block design subtests of the WAIS-R (22). Short 117 forms of the WAIS-R have been shown to produce accurate estimates of full-scale IQ (23,24). 118 Specific functions were assessed using the following neuropsychological tests: Memory using the Rey 119 Auditory-Verbal-Learning Test (RAVLT) trials 1 to 7 (learning, immediate and delayed verbal recall) 120 (25), and the Visual Reproduction subtest of the Wechsler Memory Scale -Revised (WMS-R) (26); 121 Verbal knowledge using the Vocabulary and Comprehension subtests of the WAIS-R (22); Processing 122 speed using the WAIS-R digit symbol coding and the Trails-Making-Test Part A (27); Executive 123 function/working memory using Trails-Making-Test -Part B (27), and Letter-Number Span (28); 124 Language using Category (semantic) and Letter Fluency (categories: 'body parts'; 'fruits'; 'animals', 125 letters: F; A; S) (29), and Visuospatial ability using the WAIS-R Block Design subtest. 126

Diagnostic Assessment 127
Clinical data were collected using the Schedules for Clinical Assessment in Neuropsychiatry 128 (SCAN) (30). The SCAN incorporates the Present State Examination, Version 10, to elicit symptom-129 related data at time of presentation. Ratings on the SCAN are based on clinical interview, case note 130 review, and information from informants (e.g. health professionals, close relatives). Researchers 131 were trained on the SCAN with a World Health Organization-approved course and reliability was 132 established prior to commencement of the study using independent ratings of videotaped 133 interviews. Rater agreement was evaluated using Kappa statistics, which ranged from 1.0 for 134 psychosis as a category to between 0.6 and 0.8 for individual diagnoses. ICD-10 diagnoses were 135 determined using SCAN data through consensus meetings with one of the PIs and other team 136 members. Symptom severity was classified based on the SCAN Symptom Severity Rating Scale 2 as: 0 137 = Absent, 1 = Mild, 2 = Moderate and 3 = Severe (21). 138

Covariates and medication information 139
Age was collected at baseline and follow-up. Sex, ethnicity, and level of education were collected at 140 baseline. Treatment history with typical and/or atypical antipsychotic medication was ascertained 141 for all patients from interview data and record review at follow-up. 142

Creating Norms for Neuropsychological Tests 143
A regression-based approach was used to create normative standards for the neuropsychological 144 tests. Age at assessment, sex, ethnicity, and education were regressed on each of the 145 neuropsychological measures in the healthy comparison sample at baseline and follow-up. Next, 146 scores were adjusted on the basis of the regression results, and standard scores (i.e., z-scores) were 147 created. The same adjustment and standardization procedure were applied to the patient groups, 148 using the normative standards from the healthy comparison group. 149

Statistical analysis 150
Demographic and clinical characteristics of the baseline and follow-up cohorts were compared using 151 summary statistics. For descriptive purposes, we compared patients with schizophrenia or other 152 psychoses (including bipolar disorder, mania, depressive psychoses and other psychotic disorders) to 153 the comparison group on normative-adjusted IQ and specific neuropsychological tests at baseline 154 and follow-up using analysis of variance (ANOVA) models. 155 To examine the "IQ decline" "generalized decline" and "specificity" hypotheses, we compared the 156 schizophrenia and other psychoses groups to the comparison group on change in normative-157 adjusted IQ and specific neuropsychological tests from baseline to follow-up. Change scores were 158 calculated by subtracting follow-up test scores from baseline test scores, so that positive scores 159 indicate cognitive amelioration and negative scores indicate cognitive decline. ANCOVA models with 160 planned orthogonal comparisons of each psychosis group to the comparison group, adjusting for 161 time from baseline assessment and baseline test score were used. Adjustment for baseline 162 performance is common in studies on cognitive change (31,32). For the "IQ decline" hypothesis, the 163 significance level was set at p=0.05 (two-sided). For the "generalized decline" hypothesis, the 164 significance level was set at a Bonferroni-corrected level of 0.0038 (0.05/13). All analyses were 165 conducted using IBM SPSS Statistics version 24. 166 167

168
Demographic characteristics of the baseline cohort and the cohort assessed at follow-up are 169 presented in Table 1

Cognitive impairment in schizophrenia and other psychoses at baseline and follow-up 176
As we have previously shown in the AESOP study cohort (19), patients with schizophrenia and 177 patients with other psychoses showed deficits in IQ and individual neuropsychological tests at 178 baseline. Figure 2 illustrates that that schizophrenia patients exhibited widespread, persistent, 179 cognitive impairment, performing significantly worse than comparison subjects at both baseline and 180 follow-up on 11 out of the 14 measures. Patients with other psychoses also showed widespread 181 impairments, but these were generally of smaller magnitude than schizophrenia patients (Figure 2). 182 (sTable2 presents the non-adjusted performance in IQ and specific neuropsychological tests at 183 baseline and follow-up) 184

Cognitive change in schizophrenia and other psychoses 194
Next, we compared cognitive change over time in each of the psychoses groups (schizophrenia and 195 other psychoses) to cognitive change in controls to test the "IQ decline", "generalized decline" and 196 "specificity" hypotheses.  We examined the potential moderating effect of antipsychotic medication on IQ decline in the 233 schizophrenia group. There was no statistically significant difference in IQ decline (p=0.23) between 234 patients with a history of treatment with typical antipsychotics only (45% of sample) and those with 235 a history of treatment with both typical and atypical antipsychotics (55% of sample). Duration of 236 antipsychotic medication (mean = 323±192 weeks) did not attenuate IQ decline in schizophrenia 237 (F=7.30, p=0.008 vs. F=7.20, p=0.009 for ANCOVA models with vs. without duration of treatment as a 238 covariate). 239

Symptom severity 240
Since illness severity might influence cognition, we also examined the association between baseline 241 symptom severity and change in cognitive functioning, as well as change in symptom severity 242 between baseline and follow up and change in cognitive functioning. Schizophrenia patients with 243 severe symptoms at baseline showed statistically significantly greater cognitive decline than patients 244 with mild or moderate symptoms across multiple tests in the memory domain ( Figure 4). However, 245 there was no association between change in symptom severity and change in cognitive functioning 246 (sTable 2 and sFigure 1), and no evidence for a dose-response relationship across levels of severity 247 ( Figure 4). In the other psychoses group there was no evidence for an association between symptom 248 severity, or change in symptom severity, and change in cognitive functioning (Figure 4, sFigure1). 249

Sensitivity analyses 259
We also examined the potential impact of attrition by applying linear mixed models which permit 260 varying numbers of measurements per person and time point, while adjusting for within-individual 261 (i.e. between measures) variation. Similar results were obtained in models that included only cases 262 and controls with data from both assessment time points, and in models that also included cases 263 and controls with data from a single assessment, indicating results were not biased by attrition. 264 As a further comparison, we examined IQ change in controls with lower IQ (IQ<90 at baseline, equal 265 to 1SD below the control group mean, N=17, 16.5% of sample). These individuals are of interest 266 because, like schizophrenia patients, they also exhibit lower IQ, and yet they did not develop 267 psychosis. In contrast to patients with schizophrenia, individuals with lower IQ did not show 268 evidence of IQ decline, neither in absolute terms nor relative to controls without a cognitive 269 impairment since mean IQ at baseline was 84.9, and at follow up was 89. 8

272
Using a population-based, case-control sample followed prospectively from the first psychotic 273 episode we provide evidence for cognitive decline after illness onset in patients with schizophrenia. 274 These findings advance knowledge in three important ways. First, the results lend support to the "IQ 275 decline" hypothesis. As a group, schizophrenia patients showed IQ decline between baseline and 276 follow up assessments, with an effect size of small magnitude (ES=0.28). This finding is in contrast 277 with earlier studies reporting stabilization of cognitive deficits after the onset of psychosis (15). 278 However, previous studies had important methodological limitations, including a short follow-up 279 period of patients, and lack of a comparison group that is similarly followed-up. The finding of IQ 280 decline is in line with findings from neuroimaging studies of greater age-associated brain volume loss 281 (34), as well as deviated gyrification trajectories in schizophrenia patients in adulthood (35). 282 Moreover, reduction in cortical volume has been associated with IQ decline in schizophrenia 283 patients (36). 284 Second, the current findings do not support the "generalized decline" hypothesis. Decline was not 285 ubiquitous and varied across cognitive domains. The schizophrenia group exhibited declines in verbal 286 knowledge and memory. In contrast, processing speed, executive functions and visuospatial ability 287 did not decline. These contrasts can be generally viewed as reflecting differences between the 288 impact of the illness on crystalized (verbal knowledge) vs. fluid (processing speed, executive 289 functions, visuospatial) abilities. Our findings of decreasing crystalized abilities and memory scores 290 between baseline and follow-up is in line with previous evidence (37) and suggest that increasing 291 deficits in these domains may reflect actual loss of ability, rather than abnormal cognitive 292 development (i.e. "lag") (16). Alternatively, our findings may reflect difficulties with the maintenance 293 and acquisition of new verbal knowledge due to substantial and increasing memory deficits. While 294 most cognitive abilities in the general population start to show stabilization or even decline in early 295 adulthood, crystalized abilities may peak much later (38)(39)(40). In our study, measures of fluid abilities 296 showed a large deficit already at the first episode, which remained static thereafter. While previous 297 longitudinal epidemiological studies have shown cognitive decline in schizophrenia from the 298 premorbid period in childhood to the chronic stage in mid-adulthood (8-10), they were unable to 299 determine when this decline occurred. Our findings suggest that most of the decline in fluid abilities 300 occurs before the first episode, while crystalized abilities may continue to decline after onset. 301 Importantly, the decline in IQ after onset is likely to be due to the decline seen in crystallized 302

abilities. 303
Third, the current findings do not support the "specificity" hypothesis since patients with 304 schizophrenia, but also other psychoses, experienced cognitive decline. However, while patients 305 with schizophrenia showed decline in IQ, memory and verbal knowledge, patients with other 306 psychoses showed decline only in certain memory functions. Moreover, in line with previous reports 307 (41, 42), the other psychoses group showed an overall impairment profile that was qualitatively 308 similar, yet quantitatively smaller than the schizophrenia group. Thus, our findings suggest that 309 cognitive decline is not specific to schizophrenia, but also evident in other psychoses. However, 310 large, widespread, cognitive decline may still be specific to schizophrenia, since the other psychoses 311 group showed a smaller and less generalized cognitive decline. Interestingly, there was no evidence 312 of decline in a key comparison group, namely individuals with lower IQ who did not develop 313 psychosis. This group may in fact experience a different process of regression-to-the-mean. 314 The current findings should be viewed in the context of certain limitations. First, although we found 315 evidence for cognitive decline after illness onset, we could not fully map the course of deficits and 316 cognitive functions may vary in the timing of decline following the first episode. Second, group sizes 317 did not allow for an analysis of the heterogeneity of cognitive course and also limited our ability to 318 investigate more specific diagnostic sub-groups, such as bipolar/mania. Third, we ruled out two 319 explanations for the observed cognitive decline, namely, type or duration of antipsychotic 320 treatment. Unfortunately, we did not have information to examine other potential moderators of 321 cognitive decline, such as social isolation, smoking and illicit drug abuse, victimization, or physical 322 health problems such as obesity, diabetes and hypertensions. Moreover, despite the fact that we 323 adjusted for education in all our analyses, poor education in the schizophrenia group after the first 324 psychotic episode could still partly explain some of the group differences. 325 There is conflicting evidence regarding the relationship between change in symptoms and cognitive 326 functioning (43, 44). In our study, change in severity of psychosis was only minimally associated with 327 cognitive change. These results are consistent with cross-sectional findings of only a weak 328 association between positive symptoms and cognitive impairment (45). Longitudinal evidence also 329 suggests a minimal association between change in positive as well as negative symptoms, and 330 change in cognition (43, 44, 46). Interestingly, in our study, schizophrenia patients with severe 331 symptoms at baseline showed greater cognitive decline than patients with mild or moderate 332 symptoms. While this group was small (21% of overall group), the magnitude of decline in the 333 memory domain was large. Thus, this finding points to a potential subgroup of schizophrenia 334 patients that may greatly benefit from being specifically targeted for cognitive remediation. 335 Our findings have important implications for understanding the nature and course of cognitive 336 impairment in schizophrenia, as well as other psychoses. Integrating the current findings with those 337 of previous studies (16) suggests that cognitive dysfunction in schizophrenia may result from a 338 complex interplay between an early, static neuropathology (47, 48) and dynamic age-related 339 processes (49, 50). As such, cognitive functions that develop and peak relatively early in life, such as 340 processing speed and visuospatial abilities (39) may show aberrant development, resulting in slowed 341 growth prior to the onset of schizophrenia (16), but relative stabilization throughout the illness 342 course. On the other hand, cognitive functions that continue to evolve through adult life, such as 343 language (39), may show further deterioration throughout the course of schizophrenia. Finally, 344 functions sensitive to age-related cognitive decline, such as memory, may begin to decline in middle 345 adulthood before normative aging becomes apparent (40). 346 In conclusion, the present study demonstrates that while a substantial proportion of the cognitive 347 impairment seen in adult patients with schizophrenia, as well as other psychoses, is present already 348 at the first episode, these patients continue to experience cognitive decline after illness onset.   Table 1. Demographic characteristics of first-episode psychosis patients and healthy comparison subjects from the Aetiology and Ethnicity in Schizophrenia and Other Psychoses (AESOP) Project. Presented are baseline characteristics of the baseline cohort and of the cohort available at the 10-Year Follow Up.    The table lists the partial correlation coefficients between severity of psychotic symptoms at baseline and change in cognitive functioning between baseline and follow up, and the correlation coefficients between change in severity of psychotic symptoms and change in cognitive functioning from baseline to follow up for each group. The partial correlations were calculated between severity of psychotic symptoms at baseline and change in severity of symptoms and change in normativeadjusted IQ and specific neuropsychological tests, controlling for time from baseline assessment and baseline test score. Correlation coefficients of 0.10, 0.30, and 0.50 reflect small, medium, and large effect sizes, respectively (33) a -A positive coefficient indicates more severe symptoms at baseline are associated with increasing cognitive impairment from baseline to follow up. b -A positive coefficient indicates that increasing severity of symptoms from baseline to follow up is associated with increasing cognitive impairment. * p<0.05 sFigure 1.