Psychiatry Research 228 (2015) 671–679
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Changes in verbal learning and memory in schizophrenia and non-psychotic controls in midlife: A nine-year follow-up in the Northern Finland Birth Cohort study 1966 Irina Rannikko a,e,n, Marianne Haapea b,c,1, Jouko Miettunen a,b,d,e,2, Juha Veijola a,b,3, Graham K. Murray f,g,4, Jennifer H. Barnett f,5, Anja P. Husa a,b,6, Peter B. Jones f,7, Matti Isohanni a,b,8, Erika Jääskeläinen a,b,d,e,9 a
Center for Clinical Neurosciences, Department of Psychiatry, University of Oulu, P.O. BOX 5000, FIN-90014, Finland Oulu University Hospital, Department of Psychiatry, OYS, P.O. Box 26, FIN-90029, Finland Oulu University Hospital, Department of Diagnostic Radiology, OYS, P.O. Box 50, FIN-90029, Finland d Centre for Life Course Epidemiology and Systems Medicine, University of Oulu, P.O. Box 5000, FIN-900014, Finland e Medical Research Center Oulu, University Hospital and University of Oulu, P.O. Box 5000, FIN-90014, Finland f University of Cambridge, Department of Psychiatry, Box 189 Addenbrooke’s Hospital, Cambridge CB2 0QQ, United Kingdom g University of Cambridge, Behavioural and Clinical Neuroscience Institute, Herchel Smith Building, Forvie Site, Cambridge Biomedical Campus, Cambridge CB2 0SZ, UK b c
art ic l e i nf o
a b s t r a c t
Article history: Received 9 June 2014 Received in revised form 8 April 2015 Accepted 20 April 2015 Available online 16 June 2015
Findings on longitudinal change of cognitive performance in schizophrenia are extremely variable in the case of verbal learning and memory, and it is still unclear which dimensions of verbal learning and memory exhibit possible deterioration over the long-term. Our aim was to compare the change in verbal learning and memory in individuals with schizophrenia 10–20 years after the illness onset and healthy controls during a nine-year follow-up in a general population sample. Our sample included 41 schizophrenia spectrum subjects and 73 controls from the Northern Finland Birth Cohort study 1966. The California Verbal Learning Test (CVLT) was used to estimate the degree of change in verbal learning and memory during a nine-year follow-up from age 34-years to 43- years. Both cases and controls deteriorated. There was statistically significant decline in two out of 20 CVLT items among cases and in 13 out of 20 CVLT items among controls. With the exception of two variables, the decline in verbal learning and memory over nine years was not significantly larger in cases. We conclude that during midlife verbal learning and memory in schizophrenia mostly declines in a normative fashion with aging at the same rate as the general population. & 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Population-based birth cohort study Schizophrenia Cognition California Verbal Learning Test Follow-up
n
Corresponding author. Tel.: þ 358 40 539 3199. E-mail addresses:
[email protected] (I. Rannikko), marianne.haapea@oulu.fi (M. Haapea), jouko.miettunen@oulu.fi (J. Miettunen), juha.veijola@oulu.fi (J. Veijola),
[email protected] (G.K. Murray),
[email protected] (J.H. Barnett),
[email protected].fi (A.P. Husa),
[email protected] (P.B. Jones), matti.isohanni@oulu.fi (M. Isohanni), erika.jaaskelainen@oulu.fi (E. Jääskeläinen). 1 Tel.: þ358 8 315 6910. 2 Tel.: þ358 8 315 6923. 3 Tel.: þ358 8 315 6910. 4 University of Cambridge, Department of Psychiatry, Box 189, Addenbrooke's Hospital, Cambridge CB2 2QQ UK. Tel.: þ 44 1223 764678. 5 University of Cambridge, Department of Psychiatry, Box 189, Addenbrooke's Hospital, Cambridge CB2 2QQ UK. 6 Tel.: þ3588 3156910. 7 University of Cambridge, Department of Psychiatry, Box 189, Addenbrooke's Hospital, Cambridge CB2 2QQ UK. Tel.: þ 44 1 2233 36960. 8 Tel.: +358 400 890024. 9 Tel.: þ358 40 7474376. http://dx.doi.org/10.1016/j.psychres.2015.04.048 0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction It is widely established that neurocognitive deficits are a core feature of schizophrenia, but controversy exists about whether and when further progression in neuropsychological impairment may occur (Ekerholm et al., 2012; Hedman et al., 2013). Individuals with schizophrenia show pronounced deficits in many areas of neurocognitive functioning including particularly memory functions (Ekerholm et al., 2012). When read a story or list of words they remember less than healthy controls (Saykin et al., 1991), and if the list or story is repeated, patients with schizophrenia learn much less than healthy controls. Even so, a sizeable proportion of individuals with schizophrenia have normal performance on any one test: for example, in one study, 44% of subjects with schizophrenia, schizoaffective disorder or delusional disorder had
672
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
normal learning ability as measured by the Rey Auditory Verbal Learning Test (RAVLT) (Karilampi et al., 2007). In a meta-analysis of cross-sectional first-episode studies, cognitive impairments were reliably and broadly present by the first-episode or early phase schizophrenia, and were maximal in immediate verbal memory and processing speed, compared to other neuropsychological domains (Mesholam-Gately et al., 2009). In a recent meta-analysis on change of global IQ, less prominent improvement in IQ in schizophrenia subjects compared to healthy controls was found (Hedman et al., 2013). Another metaanalysis showed large and generalized cognitive deficits in older individuals with schizophrenia, but these deficits did not decline over a 1–6 year period (Irani et al., 2010). A first-episode study followed up over 10–20 years presented that some domains of cognition might already have deteriorated by the first-episode, but do not progress over time; on the other hand other domains of cognitive functioning might be spared during the first-episode but deteriorate with time (Stirling et al., 2003). Taken together, these lines of evidence suggest that the neurocognitive impairments in general in schizophrenia do not deteriorate significantly during the midlife (Censits et al., 1997; Hoff et al., 2005). However, in the domain of the verbal learning and memory, there is some evidence of further deterioration over the long term, and there is a need for further studies on this topic (Bozikas and Andreou, 2011). Findings on longitudinal change of cognitive performance are extremely variable in the case of verbal memory and learning. Some studies show less improvement with practice, or even deterioration in cases compared to controls on some measures of verbal memory (Hill et al., 2004; Hoff et al., 2005; Albus et al., 2006) while others report no difference (Censits et al., 1997; Heaton et al., 2001; Hoff et al., 2005; Albus et al., 2006; Leeson et al., 2009) or even greater improvement in cases compared to non-psychotic controls on at least some measures of verbal memory (Addington et al., 2005). One reason for divergent findings may be differences in length of follow-ups. In short follow-ups (maximum two years) in firstepisode samples both cases and controls improved in many cognitive domains, including verbal learning and memory (Addington et al., 2005; Mayoral et al., 2008). One possible explanation for this could be practice effect and improvement in symptoms. In contrast, in studies with follow-ups from two to 10 years, cases improved less compared to controls over time in memory functions (Hoff et al., 2005; Albus et al. 2006; Leeson et al., 2009; Ekerholm et al., 2012). In studies with longer follow-up (over 10 years) cases may perform more poorly or show deterioration compared to healthy controls in different measures of verbal learning and memory (Hoff et al., 1999; Hoff et al., 2005), but in some studies do not show evidence of cognitive decline in memory functioning over time (Caspi et al., 2003). Understanding the longitudinal changes in cognitive performance is important for the etiological investigation of schizophrenia and for treatment purposes, but the number of studies in this area is small. Generally, long-term follow-up of general population-based samples with inclusion of controls and individuals with schizophrenia, especially in a later phase of the illness, are still scarce. Furthermore, it is still relatively rare to have similar cognitive measures in both cases and controls at two-time points. In this same Northern Finland Birth Cohort study 1966 (NFBC 1966) sample Kobayashi et al. (2014) found that between the ages of 34 and 43 the schizophrenia group exhibit progressive cognitive decline in executive function, but there were no differences between cases and controls in rate of change of visual object learning or change in one indicator of verbal learning. Kobayashi et al. (2014) studied only one of the several items from the California Verbal Learning Test (CVLT).
Verbal learning and memory in schizophrenia has received considerable attention by cognitive neuropsychology and severe verbal learning impairment has been established. However, it is still unclear in which dimensions of verbal learning and memory deterioration over the long follow-up occurs. The aim of our study was to compare the change in several dimensions of verbal learning and memory in individuals with schizophrenia in midlife 10–20 years after the illness onset and healthy controls during a 9-year follow-up in the general population-based NFBC 1966. We had the opportunity to study the changes in verbal learning and memory in schizophrenia and non-psychotic controls in midlife from age 34 to age 43 in large, unselected, epidemiologically sound sample. To our knowledge, there are no previous studies on this topic with this length of follow-up in this age group. Our research hypothesis was that subjects with schizophrenia would experience less improvement, or more deterioration, in measures of verbal learning and memory when compared to healthy controls.
2. Methods 2.1. Participants The NFBC 1966 is an unselected general population birth cohort ascertained during mid-pregnancy. The NFBC 1966 consist of 12, 058 live-born children in the provinces of Lapland and Oulu with an expected delivery date during 1966. There were 11 017 eligible individuals in Finland at the age of 16 years. Of these, 83 individuals did not consent to the use of their data and have been excluded. The Faculty of Medicine Ethics Committee of the University of Oulu keeps the study design of the NFBC 1966. A psychiatric field study including diagnostic interviews and the CVLT was conducted in 1999–2001 (baseline, around age 34 years), with follow-up in 2008– 2010 (around age 43 years). Participants were given a complete description of the study and had an opportunity to refuse to participate at baseline and follow-up. All participants provided written informed consent. 2.2. Case identification The nationwide Finnish Hospital Discharge Register (FHDR) covers all mental and general hospitals, as well as beds in local health centers and private hospitals nationwide. All cohort members over 16 years appearing on the FHDR until the end of 1997 for any mental disorder (i.e. ICD-8 diagnoses 290-309, ICD-9 290-316, and ICD-10 F00-F69, F99) were identified. All case records were scrutinized and diagnoses were validated for the DSM-III-R criteria. The reliability of researchers assessing schizophrenia diagnoses was good (kappa ¼0.85). Altogether 160 individuals with known psychotic episodes until the year 1999 were detected. The detection of cases and validation of diagnoses are described in Moilanen et al. (2003). Fourteen psychosis cases (8.8%) had died by the year 2001 (Alaräisänen et al., 2006). At baseline, during 1999–2001 around age 34-years all 146 living cases (62 females, 42%) and 187 controls (71 females, 38%) were invited to participate in the field study. Altogether 92 (63%) cases (73 (79%) with schizophrenia spectrum disorders) and 104 (56%) controls participated. Controls were randomly selected from the NFBC 1966 members who were living in Oulu area, and who did not have any history of psychotic disorder (Haapea et al., 2007). The Structured Clinical Interview for DSM-III-R, (SCID I; Spitzer et al., 1989) was used for diagnostic assessment, together with all available anamnestic information including individual hospital notes. SCID I was performed for both cases and controls at both baseline and follow-up assessments. At baseline, after diagnostic interviews a total of 61 cases with a lifetime diagnosis of schizophrenia and 12 cases with other schizophrenia spectrum disorder were detected. Hereafter in this paper we use term “schizophrenia” meaning schizophrenia and other schizophrenia spectrum disorders (in our study including schizophreniform disorder, schizoaffective disorder and delusional disorder). Mean duration of illness was 10 years (S.D. 4.3) at baseline. The field study in 1999–2001 is presented in detail in Haapea et al. (2007). The follow-up study took place in 2008–2010 when the participants were around age 43 years. All the participants of the baseline study were invited to participate in the follow-up. Of the individuals with schizophrenia 44 (60%) and 76 (73%) of the control subjects participated in the follow-up. The mean duration of illness until the follow-up study was 20.0 years (S.D. 4.1). The SCID I interview (First et al., 2002) was performed in the follow-up as well. The original diagnosis was upheld at follow up for all participants with schizophrenia on the basis of the SCID I interview and case note review.
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
673
Table 1 Baseline and follow-up characteristics of individuals with schizophrenia (n¼ 41) and controls (n¼ 73). Baseline
Characteristics Follow-up time, years (mean, S.D.) Females (n, %) Education (n, %)2 Basic Secondary Tertiary Social and Occupational Functional Scale (SOFAS) (mean, S.D.) Current or earlier alcohol use disorder (n, %) Current alcohol use disorder (n, %) Current non-psychotic psychiatric diagnoses, including alcohol use disorder (n, %)3 Diagnosis Schizophrenia Other schizophrenia spectrum disorder Age at onset of psychosis, years (mean, S.D.) Positive and Negative Syndrome Scale (PANSS) (mean, S.D.)4 Total Positive Negative Disorganization Excitement Emotional Remission Yes No Cumulative number of hospital treatment days (median, IQR)5 Antipsychotic medication before baseline (n, %)6 No long-term medication Previously long-term, currently at most low dose Previously long-term, currently high dose Current use (last 3 months) Antipsychotic medication (n, %)6 No current medication Typical antipsychotic Atypical antipsychotic Both typical and atypical antipsychotic Other psychiatric medication (n, %)6 Antidepressants Tranquillizers Other
Follow-up
Schizophrenia
Controls
P-value1
Schizophrenia
Controls
P-value1
19 (46%)
28 (38%)
0.433 0.034
9.1 (0.6)
8.5 (0.7)
o0.001
22 (54%) 12 (29%) 7 (17%) 51 (16.5) 9 (22%)
31 (43%) 13 (18%) 29 (40%) 86 (5.3) 1 (1%)
20 (50%) 10 (25%) 10 (25%) 53 (18.0)
27 (37%) 17 (23%) 29 (40%) 85 (9.3)
o 0.001 o 0.001
o0.001
6 (15%)
3(4%)
0.068
5 (12%) 11 (27%)
5 (7%) 9 (12%)
0.490 0.069
34 (83%) 7 (17%) 23.6 (4.4)
n/a
52.5 (19.6) 12.1 (5.6) 14.4 (8.6) 17.6 (8.5) 11.4 (3.9) 13.4 (5.0)
n/a n/a n/a n/a n/a n/a
67.1 (24.0) 16.3 (8.0) 18.7 (9.2) 24.5 (11.5) 14.4 (5.0) 18.1 (6.2)
n/a n/a n/a n/a n/a n/a
15 (37%) 26 (63%) 161 (44–753)
n/a n/a n/a
14 (35%) 26 (65%) 257 (79–932)
n/a n/a n/a
10 (27%) 15 (41%) 12 (32%)
n/a n/a n/a
13 (32%) 15 (37%) 9 (22%) 4 (10%)
0 0 0 0
8 (20%) 6 (15%) 19 (47.5%) 7 (17.5%)
0 0 1 0
7 (17%) 12 (29%) 2 (5%)
3 (4%) 1 (1%) 0 (0%)
8 (21%) 15 (38%) 10 (26%)
4 (5%) 1 (1%) 1 (1%)
(0%) (0%) (0%) (0%)
0.279
(0%) (0%) (1%) (0%)
S.D. ¼Standard Deviation, IQR ¼Inter Quartile Range. 1
Significance from χ2-test (n, %) or independent samples t-test (mean, S.D.). Level of basic education (O level, 9 years; A level, 12 years) and vocational education (low: none, course or school, currently studying; high: college, polytechnic, or university) combined as level of education (Basic: O level with low vocational education; secondary: O level with high vocational education or A level with low vocational education; and tertiary: a level with high vocational education. 3 Specific diagnoses at baseline in cases: depression (n¼1), alcohol abuse (n¼ 1), alcohol dependence (n ¼3), depression and alcohol dependence (n¼ 1); and in controls: depression (n¼1), panic disorder (n¼ 1), depression and panic disorder (n¼ 1). Specific diagnoses at follow-up in cases: anxiety disorder (n¼4), alcohol dependence (n¼ 3), depressive disorder and alcohol dependence (n¼ 2), anxiety disorder and bulimia nervosa (n ¼1), mood disorder due to general medical condition and anxiety disorder (n¼ 1); and in controls: anxiety disorder (n¼ 4), alcohol abuse (n¼ 1), alcohol dependence (n ¼2), anxiety disorder and alcohol abuse or dependence (n¼ 2). 4 PANSS was used to measure psychopathological symptoms. Subscales were determined based on van der Gaag et al. (2006). Remission was defined by Andreasen et al. (2005) without duration criteria. 5 Cumulative number of days in psychiatric hospitalization until baseline and follow-up. Information from the Finnish Hospital Discharge Register. 6 Current and lifetime use of antipsychotic medication was obtained from all participants at the time of the cognitive tests. Long term medication: at least one year of regular use preceding the baseline, low dose: r 300 mg CPZ/day, high dose: 4300 mg CPZ/day. CPZ¼Chlorpromazine equivalents. 2
The present study is based on those 41 individuals with schizophrenia and 73 controls for whom CVLT data was available in both baseline and follow-up studies. The mean follow-up time for schizophrenia was 9.1 years (S.D. 0.6) years and 8.4 (S. D. 0.7) years for the control group. The schizophrenia group included the following DSM-III-R diagnoses: schizophrenia 295.1 (n¼9); 295.3 (n¼8); 295.6 (n ¼1); 295.9 (n¼ 16); schizophreniform psychosis 295.4 (n¼ 1), schizoaffective disorder 295.7 (n¼ 5) and delusional disorder 297.1 (n¼ 1). Those who completed the baseline and follow-up CVLT in 2008–2010 did not differ from those who did not participate in the follow-up in gender, education, or summary measure of the CVLT (correct responses in trials 1–5) at baseline, analyzed separately for cases and controls. Within cases, those who completed the follow-up CVLT did not differ from those who did not participate in medication use, symptoms, onset age or number of hospitalization days. There were no statistically significant differences in alcohol use disorder diagnoses between non-completers and completers within cases, or within controls. None of the non-completing or completing cases or controls had other substance use disorder at baseline or at follow-up. Of the
completing cases 34 (83%) had a schizophrenia diagnosis, whereas of noncompleters 27 (87%) had a diagnosis of schizophrenia (p¼0.747).
2.3. Neuropsychological assessments We administered the CVLT (Delis et al., 1987) at baseline at age 34 years and in exactly the same way at the follow-up at age 43 years. The CVLT was administered using the paper–pencil Record Form. The types of verbal learning processes, strategies and error types assessed by the CVLT were scored manually. At the baseline, researcher-psychiatrists administered the CVLT. In the follow-up, the CVLT was administered by researchers (e.g. one clinical neuropsychologist and two psychologists) who were trained in neuropsychological assessments through specific courses held by two clinical neuropsychologists. The reliability of CVLT scores was not analyzed, but the training of the examiners was updated regularly throughout at the follow-up phase of the study during 2008–2011. Both the training of the
674
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
Table 2 Original values of CVLT items in individuals with schizophrenia (n¼ 41) and controls (n ¼73) at ages 34- and 43-years. Baseline
Span of apprehension Verbal learning Immediate free recall Learning slope Short-term memory Short delay free recall Short delay cued recall Long-term memory Long delay free recall Long delay cued recall Recognition discriminability Rate of forgetting* Susceptibility to interference Proactive interference Retroactive interference Organization strategies Semantic cluster Serial cluster Recall Primacy Middle Recency Recall consistency Recall errors Perseverations Free recall intrusions Cued recall intrusions n
1
Follow-up
Schizophrenia
Controls
P-value1
Schizophrenia
Controls
P-value1
6.6 (2.0)
8.4 (2.0)
o 0.001
6.4 (2.3)
7.6 (2.0)
0.006
48.0 (13.8) 1.4 (0.8)
60.1 (6.8) 1.6 (0.7)
o 0.001 0.162
45.1 (14.4) 1.2 (1.1)
55.0 (8.4) 1.6 (0.6)
o0.001 0.031
10.1 (3.9) 11.1 (3.5)
13.2 (2.1) 13.7 (1.6)
o 0.001 o 0.001
9.9 (3.8) 10.7 (3.2)
12.0 (2.6) 12.5 (2.2)
0.003 0.002
11.2 11.5 92.7 13.8
13.7 (2.0) 13.9 (1.7) 97.8 (3.0) 5.2 (11.8)
o 0.001 o 0.001 0.002 0.097
10.3 11.0 92.4 5.7
12.5 (2.5) 13.0 (2.2) 96.2 (3.6) 5.8 (14.3)
0.001 0.002 0.008 0.970
9.4 (38.3) 5.1 (12.1)
0.051 0.396
22.4 (19.1) 6.6 (16.7)
0.689 0.682
2.4 (0.7) 2.2 (1.8)
o 0.001 0.987
2.3 (0.9) 2.0 (1.5)
0.008 0.844
28.4 46.8 24.8 0.87
0.308 0.702 0.140 0.001
28.7 45.2 26.2 0.85
0.076 0.789 0.138 0.005
(3.7) (3.2) (9.6) (30.7)
23.7 (35.1) 9.1 (28.4) 1.8 (0.9) 2.2 (1.6) 27.0 46.2 26.8 0.78
(8.2) (8.5) (7.8) (0.15)
7.0 (8.0) 1.8 (2.7) 1.0 (1.8)
(3.2) (4.2) (3.1) (0.07)
4.1 (3.9) 1.2 (1.9) 0.4 (1.0)
(3.6) (3.5) (8.2) (24.4)
19.7 (40.0) 8.6 (28.7)
0.032 0.172 0.029
1.8 (0.9) 2.1 (1.6) 26.2 44.7 29.0 0.76
(8.1) (8.9) (11.7) (0.17)
6.2 (7.7) 2.2 (2.5) 2.1 (3.0)
(4.3) (5.9) (4.6) (0.09)
4.2 (4.4) 1.8 (2.3) 0.8 (1.2)
0.076 0.393 0.016
n¼40 in the baseline and n¼ 39 in the follow up in the schizophrenia. Significance from the independent samples t-test.
examiners throughout the follow-up and interpretation of the results from the CVLT at baseline and follow-up were supervised by two clinical neuropsychologists. We ensured that our control group is representative according to standardized norms for the CVLT at baseline and follow-up. We compared our control group's raw CVLT scores to normed CVLT scores of same age group described in the Manual for CVLT (Delis et al., 1987). The comparison was made separately for males and females, and the raw CVLT scores of our control group were similar to those presented in the CVLT manual. The CVLT provides a brief, individually administered assessment of multiple strategies, processes, and errors involved in learning and remembering verbal material. The test measures both recall and recognition of word lists of 16 words over a number of trials and requires memorization of a word list consisting of items from four semantic categories, four words per category. The words are presented so that a given word is never followed by another word from the same category. The test has been administered by evaluating a subject's ability to recall a word list in any order over five trials (Delis et al., 1987). 20 measures of verbal learning and memory are reported in this study. A detailed description of the measures used is presented in Table A1.
2.4. Measures of sample characteristics The age at the onset of first psychotic symptoms was based on evaluation of patient files (Penttilä et al., 2010). The lengths of psychiatric hospitalizations until baseline and follow-up were obtained using the FHDR. Information about use of antipsychotic medication before the baseline, at baseline and at follow-up from all participants was ascertained during interviews (at baseline and follow-up) by asking about psychiatric medication (antipsychotics or other psychiatric medication) use during the previous 3 months and for as long as the subject could remember. The Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987) was used to measure psychopathological symptoms from the period of 1 week before the interview. The PANSS was based on the baseline and follow-up SCID I-interviews. Higher scores indicate more symptoms. Subscales of positive, negative, disorganization, excitement and emotional scores were used (van der Gaag et al., 2006), as well as the total score of all PANSS items. Remission was based on the criteria of Andreasen et al. (2005), with the exception that, as the symptoms had been assessed only once (for the previous week), the duration of remission criterion of 6 months was not applied. The symptom criteria for remission are: maximum scores of 3 in PANSS items P1, P2, P3, N1, N4, N6, G5, and G9. Diagnosis of alcohol use disorder was based on information gained in the SCID I -interviews at baseline and follow-up. The Social and Occupational Functioning Scale (SOFAS) was used to rate social and occupational functioning during the previous week (Spitzer et al., 2000) at baseline and follow-up. Higher scores reflect better functioning.
Levels of basic education (O level, 9 years; or A level, 12 years) and vocational education (none, course or school, currently studying, college, polytechnic, or university) were ascertained in the baseline and follow-up questionnaires. The basic and vocational educations were combined as level of education: basic¼O level with low vocational education (none, course or school, or currently studying); secondary¼ O level with high vocational education (college, polytechnic, or university), or A level with low vocational education; and tertiary¼A level with high vocational education. 2.5. Statistical analyses Baseline sample characteristics were compared between cases and controls using t-tests or χ2. Cognitive performance was compared between the cases and controls at baseline and follow-up using independent samples t-test. The change from baseline to follow-up separately within cases and controls was analyzed using paired samples t-tests. To analyze the difference between cases and controls in change from baseline to follow-up linear regression adjusted for the baseline cognitive performance was used. Other covariates were gender and education at baseline. As sensitivity analyses repeated measures analysis of variance (ANOVA) was used. All analyses were twotailed, and the probability level of Po0.05 indicated statistical significance. IBM SPSS Statistics 21.0 was used to conduct the analyses.
3. Results 3.1. Characteristics of the sample Nineteen (46%) of the cases and 28 (38%) of the controls were females. The mean age of illness onset in cases was 23.6 years (S.D. 4.4). Thirty-four (89%) cases had schizophrenia and 7 (11%) cases had other schizophrenia spectrum disorder. Twenty-eight (68%) cases had current antipsychotic medication at baseline, and 32 (80%) at followup. A more detailed description of the sample is presented in Table 1. 3.2. Verbal learning and memory in cases and controls at ages 34and 43-years Cases had lower scores than controls in many of the CVLT variables at both baseline and follow-up (Table 2). Statistically
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
675
Table 3 The change of CVLT within and between individuals with schizophrenia (n¼ 41) and controls (n ¼73). Difference between the baseline and follow-up1 Schizophrenia
Span of apprehension Verbal learning Immediate free recall Learning slope Short-term memory Short delay free recall Short delay cued recall Long-term memory Long delay free recall Long delay cued recall Recognition discriminability Rate of forgetting Susceptibility to interference Proactive interference Retroactive interference Organization strategies Semantic cluster Serial cluster Recall Primacy Middle Recency Recall consistency Recall errors Perseveration Free recall intrusions Cued recall intrusions 1 2 3
Linear regression analysis2
Repeated measures ANOVA3
Controls
Mean (S.D.)
P-value
Mean (S.D.)
P-value
B (SE)
Beta
P-value
P-value
0.20 (2.4)
0.599
0.82 (2.3)
0.004
0.55 (0.43)
0.11
0.206
0.117
2.9 (13.0) 0.26 (1.3)
0.167 0.214
5.1 (6.8) 0.04 (0.81)
o 0.001 0.640
2.60 (2.05) 0.40 (0.17)
0.13 0.19
0.208 0.018
0.406 0.131
0.27 (3.2) 0.37 (2.9)
0.592 0.419
1.2 (2.0) 1.2 (2.0)
o 0.001 o 0.001
0.03 (0.53) 0.37 (0.47)
0.01 0.08
0.956 0.434
0.101 0.202
0.90 (3.1) 0.49 (2.7) 0.33 (8.9) 7.0 (27.7)
0.070 0.259 0.812 0.123
1.2 (2.1) 0.95 (1.7) 1.6 (4.1) 0.62 (18.4)
o 0.001 o 0.001 0.001 0.774
0.56 (0.51) 0.22 (0.46) 1.46 (1.20) 2.62 (3.35)
0.12 0.05 0.11 0.06
0.271 0.637 0.184 0.435
0.693 0.367 0.232 0.095
4.0 (47.8) 0.71 (31.1)
0.596 0.886
13.0 (42.7) 1.5 (19.1)
0.011 0.513
3.58 (5.76) 0.68 (4.15)
0.04 0.01
0.660 0.870
0.082 0.414
0.00 (0.88) 0.13 (2.2)
0.991 0.709
0.16 (0.78) 0.20 (1.9)
0.084 0.388
0.15 (0.16) 0.21 (0.29)
0.09 0.05
0.344 0.473
0.644 0.545
0.81 (11.0) 1.5 (12.7) 2.3 (12.7) 0.02 (0.20)
0.637 0.464 0.258 0.492
0.28 (4.2) 1.6 (5.4) 1.3 (5.1) 0.03 (0.09)
0.574 0.014 0.031 0.014
2.00 (1.18) 0.77 (1.39) 2.53 (1.59) 0.06 (0.03)
0.13 0.04 0.14 0.19
0.091 0.579 0.115 0.033
0.619 0.864 0.557 0.718
0.78 (4.6) 0.44 (2.6) 1.0 (2.9)
0.282 0.286 0.028
0.14 (5.4) 0.70 (2.7) 0.47 (1.4)
0.829 0.031 0.007
0.38 (0.95) 0.06 (0.45) 0.77 (0.40)
0.04 0.01 0.18
0.690 0.902 0.057
0.517 0.546 0.270
Unadjusted comparison between baseline and follow-up, p-value from paired samples t-test. Analyses are adjusted for baseline cognition, gender, and education at baseline; statistically significant results are in bold. Analyses are adjusted for gender and education at baseline; p-value of the interaction effect of follow-up and schizophrenia.
significant differences occurred in several CVLT variables from different dimensions of verbal memory and learning.
and controls, except that cases had more decline in the measures of the learning slope and recall consistency. This difference was relatively small based on beta coefficients.
3.3. Change in verbal learning and memory in cases and controls in 9-year follow-up 4.2. Comparison with earlier studies In within group analyses performed separately for schizophrenia and controls there was statistically significant decline in 2 out of 20 CVLT items among cases and in 13 out of 20 CVLT items among controls (Table 3). Out of 20 measures of CVLT, schizophrenia cases declined more than controls in two variables: cases had a statistically significantly larger decrease in the learning slope (B(SE)¼ 0.40 (0.17), Beta¼ 0.19, Sig.¼0.018) and recall consistency (B(SE)¼ 0.06(0.03), Beta¼ 0.19, Sig.¼0.033) (Table 3). Additionally, cases declined almost statistically significantly in intrusions in cued recall (B(SE)¼0.77(0.40), Beta¼0.18, Sig.¼0.057). In repeated measures ANOVA, interaction of follow-up and group (cases vs. controls) was not significant in any of the CVLT variables. The difference in CVLT variables at 34-years, at 43-years and the change of verbal learning and memory in cases and controls is illustrated in Fig. 1.
4. Discussion 4.1. Main findings Compared to non-psychotic controls, subjects with schizophrenia had poorer cognitive performance in many different dimensions of verbal learning and memory in cross sectional analyses at age 34 and 43 years. Both cases and controls declined in performance in the 9year follow-up. However, there was no major difference in the amount of change of verbal learning and memory over time in cases
In our study, the neuropsychological deficits in schizophrenia appeared to remain relatively stable over time suggesting that the neurocognitive impairments in schizophrenia (at least in case of verbal learning and memory) do not deteriorate significantly during middle age, and that midlife schizophrenia may not be a progressing degenerative process (Censits et al., 1997; Hoff et al., 2005; Bora and Murray, 2013). However, in the study from this same sample (Kobayashi et al., 2014) the cases showed grater cognitive deterioration in area of executive function (abstraction with memory), but the deterioration did not occur in several dimensions of verbal learning and memory in our analyses, suggesting that different dimensions of cognition may act differently during the illness. The discrepancy in findings on change of verbal learning and memory might be attributable to differences in length of longitudinal evaluations. Studies that retested their participants for relatively short intervals after the first-episode usually report similar changes in cognition over time in patients and controls (Censits et al., 1997; Albus et al., 2006; Crespo-Facorro et al., 2009; Leeson et al., 2009), or even greater improvement in individuals with schizophrenia compared to healthy controls (Addington et al., 2005; Mayoral et al., 2008), which might result from the practice effects and improvement in symptoms. Impairment in verbal memory already exists in the early stages of illness (Hoff et al., 2005; Leeson et al., 2009; Ekerholm et al., 2012) and findings on
676
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
Fig. 1. The difference in CVLT variables between individuals with schizophrenia and controls at 34-years, at 43-years and the change of verbal learning and memory in 9-year follow-up. Axis indicates baseline values in controls.
longitudinal change of this cognitive domain are extremely variable (Hoff et al., 1999: Caspi et al., 2003; Hoff et al., 2005). Based on the results of the previous studies it is difficult to know why across all 20 memory domains, exactly the learning slope and recall consistency show statistically significant deteriorations, and the cued recall intrusions nearly significant deteriorations, during the follow-up time in cases compared to controls. In addition to the possibility of being chance findings, it is possible that learning slope, recall consistency and intrusion errors may have a true role in the course of illness and neurocognition in schizophrenia, or at least in a subgroup of individuals with schizophrenia. Our findings of longitudinal change extend the findings that were reported by Roofeh et al. (2006) in adolescents with schizophrenia spectrum disorders and controls in a cross-sectional study. In earlier studies lower rate of improvement (learning slope) and poorer recall organization (recall consistency) have been associated with an encoding deficit or less sophisticated (serial recall) strategy for memory encoding in
schizophrenia (Roofeh et al., 2006); intrusion errors have been associated with poorer social functioning and independent living skills (Stip et al., 2007), and greater disorganized symptoms and lower self-reflectivity (Fridberg et al., 2010). To our knowledge, there are no earlier studies analyzing the change of these dimensions of verbal learning and memory. Another essential finding of our study is the decline in verbal learning and memory in the non-psychotic controls from age 34 to age 43 years. Since studies reporting significant decrements have relatively long testing intervals, it has been suggested that long retest intervals may be needed to observe reliable change in cognition. In a study by Zelinski and Burnight (1997), the authors concluded in their 16-year longitudinal general population study, that six years interval of follow-up seems to be the minimal retest interval to assure that longitudinal decline in memory can be reliably assessed. In earlier studies, among the general population, memory decline from the age of 53 onwards has been observed in seven studies with six years or longer follow-up (Colsher and
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
Wallace, 1991; Schaie, 1994; McDonald-Miszczak et al., 1995; Zelinski and Burnight, 1997; Richards et al., 2004; Tomaszewski Farias et al., 2011; Hahn and Lachman, 2015). Despite this trend, there is still a gap in evaluation of the longitudinal course of verbal learning and memory in early midlife among non-psychotic adults. We extended these previous works by indicating memory decline over a long follow-up time interval between age 34 to age 43 years in non-psychotic individuals from the general population. 4.3. Strengths and limitations One of the strengths of the present study is a long (nine-year) follow-up interval in midlife schizophrenia from age 34 to age 43 years. We also minimized a potential sample selection bias by utilizing an unselected non-clinical, general population-based birth cohort sample. The inclusion of an age-matched, general population, non-psychotic, but not selected (e.g. not exceptionally healthy) comparison group is a major strength in studies of the cognitive changes in schizophrenia. Without the presence of a control group, it cannot be concluded whether a decline in patient
677
performance reflects a neurodegenerative process rather than normal age-related changes in cognition. Moreover, we maximized the test validity by using at baseline and in the follow-up the identical and widely used and accepted measure of verbal learning and memory, the CVLT, which is used frequently in schizophrenia (Censits et al., 1997; Hill et al., 2004; Hoff et al., 2005; Albus et al., 2006). In our sample, 32% did not use antipsychotics at baseline and 20% did not use antipsychotics at follow-up, and this may be marker of less severe illness. On the other hand, about one third had used long term antipsychotic medication before the baseline and used currently high dose antipsychotic medication at baseline. Additionally, 37% of the sample was in remission at baseline and 35% at follow-up, which is a similar proportion to many earlier studies (Emsley et al., 2011; Haro et al., 2011). Thus, we consider that our sample is heterogeneous, including also those with good prognosis. Our results are generalizable to persons having schizophrenia in general, but not to all clinical samples with more severe patient populations. A limitation of this study is that the number of cases is relatively small. Owing to the small sample size the power of the
Table A1 Interpretation of variables obtained in the California Verbal Learning Test (CVLT) test. Variable
Interpretation
Span of apprehension Verbal learning Immediate free recall Learning slope
Performance (correct responses) on List A Trial 1 provides a gauge of initial learning/attention span. Performance (correct responses) on List A provides a sum of trials 1 through 5. The rate of improvement from first to final trial indicates the amount of new learning per trial, i.e. reflects the increment in words recalled per trial over Trials 1–5.
Short-term memory Short delay free recall
After interference List B (a second list containing 16 items, presented for one trial) the subject is asked to recall the items of the List A in any order. Short delay cued recall After interference List B (a second list containing 16 items, presented for one trial) the examinee is asked to recall the items of the List A by category. Long-term memory Long delay free recall The number of correct responses on List A in any order after 15–20 min interval (in which the examinee is occupied with other tests to minimize interference) reflects the examinee´s ability to retain verbal information over time. Long delay cued recall The number of correct responses on List A by category after 15–20 min interval (in which the examinee is occupied with other tests to minimize interference) reflects the examinee´s ability to retain verbal information over time. Recognition Auditory recognition task designed to assess long-term retention and the ability to discriminate target words from distractor items, i.e. discriminability examinee is tested for recognition of the List A items by list which contains all the List A items and 28 distracters (8 List B items, 8 phonetic distracters, 4distracters prototypical of the semantic categories and 8 unrelated distracters). Rate of forgetting To assess the degree of forgetting over the 20-min delay period a difference score is calculated for the number of words recalled during Short Delay and Long Delay Free Recall. Susceptibility to interference Proactive interference Reflects the degree to which learning List A interfered with the subsequent learning of List B, i.e. the decremental effect of prior learning on the retention of subsequently learned material. Retroactive Reflects the degree that learning the second list interfered with later recall of the original list, i.e. the decremental effect of subsequent learning interference on the retention of previously learned material. Organization strategies Semantic cluster Consecutive recall of List A words grouped by semantic category is the ratio of correct responses followed by another correct response from the same category, relative to the expected clustering by chance. Indicates the degree to which the examinee uses the active learning strategy of reorganizing the target words into categorical groups. Serial cluster Recalling words according to their order in the word list denotes the ratio of word pairs recalled in the same succession as the stimulus list, relative to chance expected serial ordering. Indicates the degree to which the examinee recalls targets words in the same order as they are presented. Recall The serial position effect is divided among Primacy, Middle and Recency regions, related to the position of words in the list. Examinees typically recall a larger percentage of the words that are in the primacy and recency regions of a list than of the words in the middle regions. Primacy The percentage of correct responses from primary (first 4 words on the target list) from Trials 1–5 of List A immediate Free Recall. Poor recall of primacy-region words in conjuction with considerably better recall of recency-region words may indicate a passive learning style. Middle The percentage of correct responses from middle (8 words on the target list) from Trials 1–5 of List A immediate Free Recall. Recency The percentage of correct responses from recency (last 4 words on the target list) from Trials 1–5 of List A immediate Free Recall. Recall consistency The ability to recall consistently the same words across repeated presentations of the same list. This index measures the percentage of target words recalled on one of the first four trials that are also recalled on the very next trial. Perseverations The type of recall errors defined as a repetition of any response on the same trial. Free recall intrusions The type of recall errors, which are responses not on the target list on short and long delay. Cued recall intrusions The type of recall errors, which are responses not on the target list on short and long delay. Interpretation of CVLT variables has been composed by utilizing the following references: Delis et al. (1987). Roofeh et al. (2006).
678
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
study is also decreased and may affect the interpretation of the non-significant results. The cases have had the disorder for a long period of time (approximately 10 years) and as cognitive tests were not administered at the initial stage of the disorder, it is not possible to consider the temporal characteristics of changes in cognition in the early stages of the illness.
5. Conclusions In summary we found no evidence of significantly greater decline in verbal learning and memory over nine years in middle-aged individuals with schizophrenia compared to nonpsychotic controls. Compared to controls, cases had lower cognitive ability at the baseline, but the decline during the follow-up was similar in both cases and controls, with the exception of learning slope and recall consistency that had larger decline in cases. These results imply that between ages 34- and 43-years verbal learning and memory in schizophrenia declines mostly in a normative fashion with aging at the same rate as in the general population.
Acknowledgments This work was supported in part by grants from the Academy of Finland (#132071, #268336, #278286), the Sigrid Jusélius Foundation, the Northern Finland Health Care Support Foundation, the Jalmari and Rauha Ahokas Foundation, the NARSAD: Brain and Behavior Research Fund, and the UK Medical Research Council.
Appendix A See Table A1. References Addington, J., Saeedi, H., Addington, D., 2005. The course of cognitive functioning in first episode psychosis: changes over time and impact on outcome. Schizophrenia Research 78, 35–43. Alaräisänen, A., Miettunen, J., Lauronen, E., Rasanen, P., Isohanni, M., 2006. Good school performance is a risk factor of suicide in psychoses: a 35-year follow-up of the Northern Finland 1966 Birth Cohort. Acta Psychiatrica Scandinavica 114, 357–362. Albus, M., Hubmann, W., Mohr, F., Hecht, S., Hinterberger-Weber, P., Seitz, N.-N., Küchenhoff, H., 2006. Neurocognitive functioning in patients with first-episode schizophrenia: results of a prospective 5-year follow-up study. European Archives of Psychiatry and Clinical Neuroscience 256, 442–451. Andreasen, N.C., Carpenter, W.T., Kane, J.M., Lasser, R.A., Marder, S.R., Weinberger, D.R., 2005. Remission in schizophrenia: proposed criteria and rationale for consensus. American Journal of Psychiatry 162, 441–449. Bora, E., Murray, R.M., 2013. Meta-analysis of cognitive deficits in ultra-high risk to psychosis and first-episode psychosis: do the cognitive deficits progress over, or after, the onset of psychosis? Schizophrenia Bulletin 40 (4), 744–755. Bozikas, V.P., Andreou, C., 2011. Longitudinal studies of cognition in first episode psychosis: a systematic review of the literature. Australian and New Zealand Journal of Psychiatry 45, 93–108. Caspi, A., Reichenberg, A., Weiser, M., Rabinowitz, J., Kaplan, Z., Knobler, H., Davidson-Sagi, N., Davidson, M., 2003. Cognitive performance in schizophrenia patients assessed before and following the first psychotic episode. Schizophrenia Research 65, 87–94. Censits, D.M., Ragland, J.D., Gur, R.C., Gur, R.E., 1997. Neuropsychological evidence supporting a neurodevelopmental model of schizophrenia: a longitudinal study. Schizophrenia Research 24 (3), 289–298. Colsher, P.L., Wallace, R.B., 1991. Longitudinal application of cognitive function measures in a defined population of community-dwelling elders. Annual of Epidemiology 1, 215–230. Crespo-Facorro, B., Rodriguez-Sanchez, J.M., Perez-Iglesias, R., Mata, I., Ayesa, R., Ramirez-Bonilla, M., Martínez-Garcia, O., Vázquez-Barquero, J.L., 2009. Neurocognitive effectiveness of haloperidol, risperidone, and olanzapine in firstepisode psychosis: a randomized, controlled 1-year follow-up comparison. Journal of Clinical Psychiatry 70, 717–729. Delis, D.C., Kramer, J.H., Kaplan, E., Ober, B.A., 1987. California Verbal Learning Test. Psychological Corporation, New York.
Ekerholm, M., Waltersson, S.F., Fagerberg, T., Söderman, E., Terenius, L., Agartz, I., Jönsson, E.G., Nyman, H., 2012. Neurocognitive function in long-term treated schizophrenia: a five-year follow-up study. Psychiatry Research 200, 144–152. Emsley, R., Chiliza, B., Asmal, L., Lehloenya, K., 2011. The concepts of remission and recovery in schizophrenia. Current Opinion in Psychiatry 24 (2), 114–121. First, M., Spitzer, R., Gibbon, M., Williams, J., 2002. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, research Version, Patient Edition With Psychotic Screen (SCID-I/P W/PSYSCREEN). Biometrics Research, New York State Psychiatric Institute, New York. Fridberg, D.J., Brenner, A., Lysaker, P.H., 2010. Verbal memory intrusions in schizophrenia: associations with self-reflectivity, symptomatology, and neurocognition. Psychiatry Research 179, 6–11. Haapea, M., Miettunen, J., Veijola, J., Lauronen, E., Tanskanen, P., Isohanni, M., 2007. Non-participation may bias the results of a psychiatric survey: an analysis from the survey including magnetic resonance imaging within the Northern Finland 1966 Birth Cohort. Social Psychiatry and Psychiatric Epidemiology 42 (5), 403–409. Hahn, E.A., Lachman, M.E., 2015. Everyday experiences of memory problems and control: the adaptive role of selective optimization with compensation in the context of memory decline. Aging, Neuropsychology, and Cognition 22 (1), 25–41. Haro, J.M., Novick, D., Bertsch, J., Karagianis, J., Dossenbach, M., Jones, P.B., 2011. Cross-national clinical and functional remission rates: Worldwide Schizophrenia Outpatient Health Outcomes (W-SOHO) study. British Journal of Psychiatry 199 (3), 194–201. Heaton, R.K., Gladsjo, J.A., Palmer, B.W., Kuck, J., Marcotte, T.D., Jeste, D.V., 2001. Stability and course of neuropsychological deficits in schizophrenia. Archives of General Psychiatry 58 (1), 24–32. Hedman, A.M., van Haren, N.E.M., van Baal, C.G.M., Kahn, R.S., Hulshoff Pol, H.E., 2013. IQ change over time in schizophrenia and healthy individuals: a metaanalysis. Schizophrenia Research 146, 201–208. Hill, S.K., Schuepbach, D., Herbener, E.S., Keshavan, M.S., Sweeney, J.A., 2004. Pretreatment and longitudinal studies of neuropsychological deficits in antipsychotic-naive patients with schizophrenia. Schizophrenia Research 68, 49–63. Hoff, A.L., Sakuma, M., Wieneke, M., Horon, R., Kushner, M., DeLisi, L.E., 1999. Longitudinal neuropsychological follow-up study of patients with first-episode schizophrenia. American Journal of Psychiatry 156, 1336–1341. Hoff, A.L., Svetina, C., Shields, G., Stewart, J., DeLisi, L.E., 2005. Ten year longitudinal study of neuropsychological functioning subsequent to a first episode of schizophrenia. Schizophrenia Research 78, 27–34. Irani, F., Kalkstein, S., Moberg, E.A., Moberg, P.J., 2010. Neuropsychological performance in older patients with schizophrenia: a meta-analysis of cross-sectional ad longitudinal studies. Schizophrenia Bulletin 37 (6), 1318–1326. Karilampi, U., Helldin, L., Hjärthag, F., Norlander, T., Archer, T., 2007. Verbal learning in schizopsychotic outpatients and healthy volunteers as a function of cognitive performance levels. Archives of Clinical Neuropsychology 22 (2), 161–174. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia Bulletin 13 (2), 261–276. Kobayashi, H., Isohanni, M., Jääskeläinen, E., Miettunen, J., Veijola, J., Haapea, M., Järvinen, M.-R., Jones, P.B., Murray, G., 2014. Linking the developmental and degenerative theories of schizophrenia: association between infant development and adult cognitive decline. Schizophrenia Bulletin 40 (6), 1319–1327. Leeson, V.C., Robbins, T.W., Matheson, E., Hutton, S.B., Ron, M.A., Barnes, T.R., Joyce, E.M., 2009. Discrimination learning, reversal, and set-shifting in first-episode schizophrenia: stability over six years and specific associations with medication type and disorganization syndrome. Biological Psychiatry 66, 586–593. Mayoral, M., Bombín, I., Zabala, A., Robles, O., Moreno, D., Parellada, M., RuizSancho, A., Arango, C., 2008. Neurological soft signs in adolescents with first episode psychosis: two-year follow-up. Psychiatry Research 161 (3), 344–348. McDonald-Miszczak, L., Hertzog, C., Hultsch, D.F., 1995. Stability and accuracy of metamemory in adulthood and aging: a longitudinal analysis. Psychology and Aging 10 (4), 553–564. Mesholam-Gately, R.I., Giuliao, A.J., Goff, K.P., Faraone, S.V., Seidman, L.J., 2009. Neurocognition in first-episode schizophrenia: a meta-analytic review. Neuropsychology 23 (3), 315–336. Moilanen, K., Veijola, J., Läksy, K., Mäkikyrö, T., Miettunen, J., Kantojärvi, L., Kokkonen, P., Karvonen, J., Herva, A., Joukamaa, M., Järvelin, M.-R., Moring, J., Jones, P., Isohanni, M., 2003. Reasons for the diagnostic discordance between clinicians and researchers in schizophrenia in the Northern Finland 1966 Birth Cohort. Social Psychiatry and Psychiatric Epidemiology 38, 305–310. Penttilä, M., Jääskeläinen, E., Haapea, M., Tanskanen, P., Veijola, J., Ridler, K., Murray, G.K., Barnes, A., Jones, P.B., Isohanni, M., Koponen, H., Miettunen, J., 2010. Association between duration of untreated psychosis and brain morphology in schizophrenia within the Northern Finland 1966 Birth Cohort. Schizophrenia Research 123, 145–152. Richards, M., Shipley, B., Fuhrer, R., Wadsworth, M.E., 2004. Cognitive ability in childhood and cognitive decline in mid-life: longitudinal birth cohort study. British Medical Journal 328 (7439), 552. Roofeh, D., Cottone, J., Burdick, K.E., Lencz, T., Gyato, K., Cervellione, K.L., Napolitano, B., Kester, H., Anderson, B., Kumra, S., 2006. Deficits in memory strategy use are related to verbal memory impairments in adolescents with schizophreniaspectrum disorders. Schizophrenia Research 85, 201–212. Saykin, J., Gur, R.C., Gur, R.E., Mozley, P.D., Mozley, L.H., Resnic, S.M., Kester, D.B., Stafiniak, P., 1991. Neuropsychological function in schizophrenia. Selective
I. Rannikko et al. / Psychiatry Research 228 (2015) 671–679
impairment in memory and learning. Archives of General Psychiatry 48, 618–624. Schaie, K.W., 1994. The course of adult intellectual development. American Psychologist 49 (4), 304–313. Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1989. Structured Clinical Interview for DSM-III-R -Patient Edition (SCID-P, 9/1/89 Version) New York. Biometrics Research Department New York State Psychiatric Institute, NY. Spitzer, R.L., Gibbon, M., Endicott, J., 2000. Global assessment scale (GAS), global assessment of functioning (GAF) scale, social and occupationa l functioning assessment scale (SOFAS), American Psychiatric Association. Handbook of Psychiatric Measures. American Psychiatric Association, Washington. Stip, E., Corbire, M., Boulay, L., Lesage, A., Lecomte, T., Leclerc, C., Ricard, N., Cyr, M., Guillem, F., 2007. Intrusion errors in explicit memory: their differential relationship with clinical and social outcome in chronic schizophrenia. Cognitive Neuropsychiatry 12, 112–127.
679
Stirling, J., White, C., Lewis, S., Hopkins, R., Tantam, D., Huddy, A., Montaquw, L., 2003. Neurocognitive function and outcome in first-episode schizophrenia: a 10-year follow-up of an epidemiological cohort. Schizophrenia Research 65 (2– 3), 75–86. Tomaszewski Farias, S., Mungas, D., Hinton, L., Haan, M., 2011. Demographic, neuropsychological and functional predictors of rate of longitudinal cognitive decline in hispanic older adults. The American Journal of Geriatric Psychiatry 19 (5), 440–450. van der Gaag, M., Hoffman, T., Remijsen, M., Hijman, R., de Haan, L., van Meijel, B., van Harten, P.N., Valmaggia, L., de Hert, M., Cuijpers, A., Wiersma, D., 2006. The five-factor model of the positive and negative syndrome scale II: a ten-fold cross-validation of a revised model. Schizophrenia Research 85 (1–3), 280–287. Zelinski, E.M., Burnight, K.P., 1997. Sixteen-year longitudinal and time lag changes in memory and cognition in older adults. Psychology and Aging 12 (3), 503–513.
