Abstract
Introduction:
Cognitive impairment is one of the fundamental features among patients with schizophrenia. The relationship between schizophrenia symptoms, insight and cognitive domains remains controversial. We aimed to study these relations in a sample of Egyptian patients with schizophrenia.
Methods:
A total of 109 patients with schizophrenia were assessed using Structured Clinical Interview for DSM-IV (Diagnostic and Statistical Manual of Mental Disorders (4th ed.)) Axis I diagnosis (SCID-I), Positive and Negative Syndrome Scale (PANSS) and Scale to Assess Unawareness of Medical Disorder (SUMD). Cognitive functions were assessed using the Wechsler Adult Intelligence Scale (WAIS), the Wisconsin Card Sorting Test (WCST) and the Wechsler Memory Scale (WMS). The cognitive functions would be distributed to cover six cognitive domains: attention/vigilance speed of processing, verbal learning, visual learning, working memory and reasoning/problem solving.
Results:
There was a significant correlation between all cognitive domains (except attention) and PANSS subscales. PANSS negative and general psychopathology subscales were significantly correlated with five cognitive domains: speed of processing, verbal learning, visual learning, working memory and reasoning/problem solving. PANSS negative subscale was significantly correlated with verbal learning (verbal paired association 1) and visual learning (visual paired association 1). There was a significant correlation between all cognitive domains and SUMD, except verbal and visual learning domains assessed by verbal and visual paired association 1 subtests, as well as attention assessed by failure to maintain set subtest. Only visual learning (trials administered), working memory (percentage error), and processing speed (perseverative responses, and trials to complete first category) were significantly negatively correlated to SUMD.
Conclusion:
Cognitive impairment in patients with schizophrenia is most likely to underlie negative symptoms, general psychopathology symptoms and poor insight, suggesting that treatment strategies minimizing these symptoms would improve cognitive impairment.
Introduction
Schizophrenia is a heterogeneous disorder characterized by unrealistic and abnormal sensory involvements, withdrawal from social activities, limited or improper expression of emotions and disarranged or catatonic behavior and negative symptoms (Keefe & Harvey, 2012).
Cognitive impairment in patients with schizophrenia is defined as a decline in execution on neuropsychological test batteries by 1–2 standard deviation compared with healthy controls (Palmer et al., 2009). Significant cognitive impairment is common in schizophrenia, touching up to 75% of the patients (Talreja et al., 2013).
Cognitive impairments are frequently moderate to a serious degree, seem to present across the life span, stable over time, obvious at the time of the first episode of illness and most likely originate before the disorder (Aich et al., 2016). It has been found that the cognitive impairments most commonly described in patients with schizophrenia comprise processing speed, attention, visual and verbal learning, working memory, abstract thinking, problem solving and social cognition (Kaneko, 2018; Marder, 2006).
The interplay between cognitive abilities and schizophrenia symptoms is exceptionally critical, as these deficiencies seem, by all accounts, to be identified with the functional consequences of patients with schizophrenia—this was evidenced by Kaneko (2018) in regard to negative symptoms and cognitive impairment in patients with schizophrenia and revealed that negative symptoms are lasting, and contrarily sway the social functioning over the span of the disease with negative outcomes in social, conjugal and work matters.
De Gracia Dominguez et al. (2009) stated that negative symptoms were most strongly correlated with verbal fluency, verbal learning, memory and intelligence quotient (IQ). On the contrary, disorganization was most strongly linked to attention, visual memory, learning and IQ. In the meantime, no significant associations were found between positive and depressive symptoms and cognitive domains (de Gracia Dominguez et al., 2009).
Kuo et al. (2019) investigated the validity of the MATRICS Consensus Cognitive Battery (MCCB) across 100 patients with schizophrenia and 113 patients with autism spectrum disorder. The authors found similar measures of the following cognitive aspects: speed of processing, attention, working memory, and learning. The authors concluded that MCCB could be beneficial in treatment evaluation of cognitive outcomes in both patients with schizophrenia and autism spectrum.
Almost all patients with schizophrenia are undermined in their capacity to function in the community because of cognitive impairment (Aich et al., 2016; Bowie & Harvey, 2006). Therapeutic modalities for cognitive impairment include pharmacological treatments and cognitive remediation. Application of cognitive remediation strategy for patients with schizophrenia had led to improvement in wide areas of cognitive functioning and enhanced neuroplasticity (Matsuda et al., 2019).
The present study uses a series of established neuropsychological tests to explore the relationship between cognitive impairment and both symptom domains and insight in patients with schizophrenia. This study was based on the hypothesis that patients with schizophrenia relied upon to perform more regrettable in the cognitive domains of verbal learning, visual learning, working memory, processing speed, and psychomotor speed, as well as in reasoning/problem solving, and it was hypothesized that schizophrenia symptoms and insight would correlate with specific cognitive deficits.
Methods
Participants and procedure
A total of 109 newly admitted patients with schizophrenia were recruited from the Institute of Psychiatry, Ain Shams University Hospitals (ASAP). The sample size was calculated by reviewing the existing literature of similar studies, using (Epi Info) program assuming that the prevalence of schizophrenia is 1% (Warner & De Girolamo, 1995) and the percentage of compliance is 50%–60% (Lacro et al., 2002). Considering the dropout, the recruitment continued until 137 patients with schizophrenia were enrolled. Of the 137 patients, 28 of them (20.4% dropout rate) were dropped out of the study, one patient died, and 27 patients did not fulfill the inclusion criteria.
The diagnosis of schizophrenia was made according to the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994). Subsequently, participants had a series of clinical and neuropsychological tests in a fixed order.
Male and female patients with schizophrenia, with age ranged from 18 to 65, were included in the study after giving informed consent. Patients with IQ score below 90, co-existing neuropsychiatric disorders, any disorder that clearly interferes with cognition such as central nervous system lesions, severe hepatic or renal impairment, significant uncontrolled chronic medical conditions, and receiving electroconvulsive therapy sessions before performing the cognitive function tests were excluded.
Patients were receiving their prescribed antipsychotic medications whether typical or atypical with a dose range of 642.3 ± 501.7 mg/day chlorpromazine equivalent. Dose of anti-Parkinson drugs was 2.4 ± 2.4 mg/day biperiden equivalent. Eighty-five patients had been prescribed the same antipsychotic regimen. Meanwhile, 24 patients had changes in their antipsychotic regimen.
Neuropsychological assessment
Clinical assessment
Patients were interviewed and demographic characteristics were collected. Then, the neuropsychological test battery was conducted by an experienced clinical psychologist after the second week of patient admission to the ASUIP hospital, as this will allow more stabilization of the patients.
Structured Clinical Interview for DSM-IV Axis I diagnosis
Structured Clinical Interview for DSM-IV Axis I diagnosis (SCID-I) is a semi-structured diagnostic interview to confirm the diagnosis of schizophrenia and to exclude any comorbid neuropsychiatric disorder (First et al., 1995). The instrument is administered by an experienced trained researcher and was used in the Arabic version (El Missiry et al., 2004).
The Positive and Negative Syndrome Scale
The Positive and Negative Syndrome Scale (PANSS) is a semi-structured clinical interview, which is well defined and standardized for clinical assessment of schizophrenic symptoms (Kay et al., 1987). The PANSS is carried out by a clinician because considerable clinical judgment is required. The PANSS is a reliable and valid tool for estimating the severity of schizophrenia symptoms. It incorporates 30 items covering three symptom domains: positive symptoms (e.g., delusions and hallucinations; seven items), negative symptoms (e.g., social withdrawal, flat affect, lack of motivation; seven items) and general psychopathology (e.g., anxiety, depression; 16 items). The test has high interrater reliability (.80). The split-half reliability of the general psychopathology subscale is .80. The scale also demonstrates excellent criterion-related validity and constructs validity.
Insight: Scale to Assess Unawareness of Medical Disorder
The Scale to Assess Unawareness of Medical Disorder (SUMD) is viewed as the most commonly used scale for surveying the awareness of the disorder in schizophrenia (Amador et al., 1993). The SUMD is a standardized scale that necessitates direct patient interview, and consequently, ratings are made on this basis. The SUMD constitutes general items (six) and subscales (four) from which 10 summary scores can be determined. All scores extend from 1 to 5, with higher scores demonstrating poorer awareness or attribution. These items cover three aspects of insight (global awareness of the mental disorder, awareness of the desired medication effects, and awareness of the impact of mental disorder on the social life). The test has good intraclass reliability coefficients between investigators ranged from .55 to .97 for the general items to assess awareness of mental disorder and .56 to .98 for the symptom items.
Assessment of cognitive functions
The clinical psychologist did not have difficulties while applying the neuropsychological test batteries because the significant level of education of patients seemed to enhance their understanding and performance of these tests, and also the delay in applying these tests after 2 weeks of admission until patients were stabilized.
Assessment of cognitive functions has been agreed currently to be of use for the MCCB (the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) Consensus Cognitive Battery), using 10 tests to assess seven neurocognitive domains (Nuechterlein & Green, 2006). The seven cognitive domains include speed of processing, attention/vigilance, working memory, verbal learning, visual learning, reasoning/problem solving and social cognition. Accordingly, the psychometric tests done in the current study would be distributed to cover these cognitive domains (except social cognition as it would not be covered by any of the psychometric tests done).
Wechsler Adult Intelligence Scale
The Wechsler Adult Intelligence Scale (WAIS) is considered a common test used for the assessment of general intelligence and broad cognitive function measurement for adult individuals (Wechsler, 1981). It is administered for individuals with age range from 16 to 89 years. It is a valid and reliable measure of general intelligence, measuring human intelligence in two domains: verbal and performance abilities. In the current study, we used the Arabic version by Louis Kamel Melika (1996). This test is considered a valid and reliable measure of general intelligence. The reliability coefficient is .93 for the performance IQ, .97 for the verbal IQ, and .97 for the whole scale.
Wechsler Memory Scale
The Wechsler Memory Scale–Revised (WMS-R) was announced in 1987 and is viewed as a standout among the most widely used tools to evaluate memory functions in adults as it provides normative information from ages ranging from 16 to 74 years (Wechsler, 1987). It incorporates data and orientation questions, eight short-term memory tasks and four delayed recall trials, all of which take around 45 minutes to 1 hour to complete. In our study, we used the following subtests of the WMS: information and orientation, digit span backwards, digit span forwards, visual memory span backwards, visual memory span forwards, visual paired association 1 and visual paired association 2, verbal paired association 1 and verbal paired association 2 to cover the different aspects of memory (immediate and delayed recall, visual and auditory, attention and concentration).
Wisconsin Card Sorting Test: The computerized version
The Wisconsin Card Sorting Test (WCST) is developed as a tool to measure executive function that requires planning strategies, searching in an organized manner, the ability to shift cognitive sets, goal-targeted behavior and the ability to modulate the response to impulsive stimuli (Heaton, 2003). It gives information on a wide exhibit of items of problem-solving behavior beyond basic indices of task success or failure, for example, the quantity of perseverative errors, the failure to maintain set, and the number of categories accomplished.
Ethical consideration
The agreement for this study was obtained from the authority of Ain Shams University Ethical and Research Committee, and an informed oral and written consent was obtained from all patients enrolled in the study after a very clear explanation providing information about the purpose of the study.
Statistical analysis
Data were analyzed using Statistical Program for Social Science (SPSS) version 17.0. Data were expressed as the M ± SD. Independent two-sample t test was used for comparison between means of the different groups of patients. Pearson’s correlation coefficient was used to assess the existing correlations. The p value was used to indicate the level of significance where p ⩽ .05 is considered significant, p ⩽ .01 is highly significant, p ⩽ .001 is very highly significant.
Results
Sociodemographic characteristics of the participants
All participants were aged between 16 and 59 years with an M ± SD of 32.2 ± 9.0 years; there were 77 men (70.6%) and 32 women (29.4%) and the majority of patients were single (68.8%), and 24.8% were married and 6.4% were divorced. The majority of patients received school education (67.5%) (primary, preparatory, secondary, and technical schools), and 32.5% were university graduates. Approximately 57% (62/109) of the patients were unemployed while about 43% (47/109) were employed. Regarding residency, most of the patients (82.6%) were living in urban areas while only 17.4% were living in a rural area (Table 1).
Sociodemographic characteristics of the sample.
Clinical characteristics of the participants
The severity of symptoms of the studied sample was assessed using PANSS; it revealed that the mean total scores of PANSS was 86.0 ± 10.7, the mean positive subscales was 32.7 ± 6.8, the mean negative subscales was 24.9 ± 6.82 and the general psychopathology subscale was 28.4 ± 3.3 (Table 2).
Insight was assessed through SUMD; the total score of SUMD was 1.2294 ± 0.42236 (Table 2).
Cognitive assessment: Patients were assessed using WAIS, WMS and WCST scales. Results of these subtests were distributed to cover six main cognitive domains: verbal learning (WAIS: verbal IQ, WMS: verbal paired association 1, WCST: total number correct); visual learning (WMS: visual paired association 1, WCST: trials administered); working memory (WMS: digit span backward, visual memory span, WCST: percentage errors); attention (WCST: conceptual-level response, failure to maintain set); reasoning/problem solving (WCST: categories completed, WMS: information and orientation); processing speed (WCST: perseverative responses, trial to complete first category, WAIS: performance IQ) (Table 2).
PANSS, SUMD, and cognitive profile of the sample.
PANSS: positive and negative scale for schizophrenia; SUMD: Scale to Assess Unawareness of Medical Disorder; WAIS: Wechsler Adult Intelligence Scale; IQ: intelligence quotient; WMS: Wechsler Memory Scale; WCST: Wisconsin Card Sorting Test.
Correlation of cognitive domains to schizophrenia symptoms (PANSS)
Verbal learning: Verbal paired association 1 was significantly negatively correlated to PANSS negative subscale, while total correct was significantly negatively correlated to PANSS negative and general psychopathology domains.
Visual learning: Visual paired association 1 was significantly negatively correlated to PANSS negative subscale (r = −.166, p = .84), while trials administered was significantly correlated to PANSS negative and general psychopathology domains (p = .007 and .041, respectively).
Working memory: There was a significant negative correlation between digit span backward and visual memory span and PANSS negative and general psychopathology subscales. Meanwhile, visual memory span had the strongest negative correlation to negative symptoms (r = −.409, p = .000). Percentage error was significantly positively correlated to PANSS negative and general psychopathology domains. Percentage error had the weakest positive correlation to general psychopathology domain (r = .191, p = .047).
Attention: There was no significant correlation between conceptual-level response, failure to maintain set and all PANSS subscales.
Reasoning/problem solving: Categories completed and information and orientation were significantly negatively correlated to PANSS negative and general psychopathology domains.
Processing speed: Perseverative responses were significantly positively correlated to PANSS negative and general psychopathology domains. Performance IQ had significant negative correlation with all PANSS domains. Performance IQ had the weakest correlation to positive symptoms (r = −.195, p = .042; Table 3).
Pearson’s correlation of PANSS domains to cognitive domains—Significant (two-tailed).
PANSS: positive and negative symptoms scale; WAIS: Wechsler Adult Intelligence Scale; WMS: Wechsler Memory Scale; WCST: Wisconsin Card Sorting Test. Boldfaced values = significant values.
Correlation of cognitive domains to insight
All cognitive domains were significantly correlated to SUMD, except verbal and visual learning domains assessed by verbal and visual paired association 1 subtests, as well as attention assessed by failure to maintain set subtest. Only visual learning (trials administered), working memory (percentage error), and processing speed (perseverative responses, and trials to complete first category) were significantly negatively correlated to SUMD (Table 4).
Pearson’s correlation of SUMD to cognitive domains—Significant (two-tailed).
SUMD: Scale to Assess Unawareness of Mental Disorder; WAIS: Wechsler Adult Intelligence Scale; IQ: intelligence quotient; WMS: Wechsler Memory Scale; WCST: Wisconsin Card Sorting Test. Boldfaced values = significant values.
Discussion and conclusion
Cognitive impairment is a vital core feature of schizophrenia. Patients with schizophrenia experience impaired cognitive functioning across a broad range of cognitive domains: processing speed, working memory, verbal and visual learning (de la Torre et al., 2016). These findings were reported in first-episode schizophrenia (FES) and in people with a chronic course of disease (Heinrichs & Zakzanis, 1998; Mesholam-Gately et al., 2009).
Profile of neurocognitive deficits in relation to schizophrenia symptoms
Our results indicated that there was a significant correlation between all cognitive domains (except attention) and PANSS subscales.
The findings reported by Daban et al. (2006) were in agreement with our findings; the authors studied two categories of patients with schizophrenia with different severity of symptoms and found that there was no statistically significant difference regarding attention performance, and concluded that psychotic symptoms, including negative symptoms, would not influence attention performance.
Also, our findings were consistent with the results of Berman et al. (1997) who reported that patients with PANSS negative symptoms would perform worse on processing speed (perseverative responses), working memory (percentage error) and reasoning/problem solving (categories completed). Nieuwenstein et al. (2001) also found that more PANSS negative symptoms were significantly associated with worse performance on WCST scores (Nieuwenstein et al., 2001).
Contrary to our results, the study carried by Bozikas et al. (2004) and Lindsberg et al. (2009) reported that the severe degree of negative symptoms was associated with lowered performance of the digit span task and audible attention problem (Bozikas et al., 2004; Lindsberg et al., 2009). Another study by Villalta-Gil et al. (2006) reported that patients with schizophrenia who had severe negative symptoms expressed poor performance in tasks concerning verbal memory, attention, and working memory (Villalta-Gil et al., 2006).
Our findings pointed to the significant correlation between PANSS subscales and specific cognitive deficits. PANSS negative and general psychopathology subscales were significantly correlated with five cognitive domains: speed of processing, verbal learning, visual learning, working memory and reasoning/problem solving. PANSS negative subscale was significantly correlated with verbal learning (verbal paired association 1) and visual learning (visual paired association 1). PANSS positive subscale was significantly correlated with verbal learning (verbal IQ) and processing speed (performance IQ).
Our findings were in line with the studies that reported significant association between specific cognitive deficits and negative symptoms. In a recent study conducted on 61 stable patients with schizophrenia, Tanaka et al. (2012) concluded that negative symptoms correlated significantly with verbal memory, working memory, speed of processing information and verbal fluency (Tanaka et al., 2012). Also, the findings of Rahimi et al. (2011) concluded that greater negative symptom severity was associated with more perseverative errors.
Comparable outcomes were obtained by Banno et al. (2012) who found in their study on 131 Japanese patients with schizophrenia that on performing Spearman’s rank correlation between WCST 5 main scores (categories achieved, difficulties of maintaining set, perseverative errors in Milner, perseverative errors in Nelson, and total errors), the Spearman’s correlation between negative symptoms and positive symptoms was high (.74; Banno et al., 2012).
However, our findings seemed not to be consistent with the study by Berman et al. (1997) who concluded that patients with negative and positive symptoms were associated with cognitive deficits. Patients with severe negative symptoms experienced more difficulties on processing speed (trail making test) and verbal ability and executive control (verbal fluency test). Meanwhile, patients with positive symptoms experienced poor performance on the digit span forward intended to measure attention. Also, the study by Ilonen et al. (2000), which was not in line with our findings, concluded that greater positive or depressive symptoms were associated with poorer performance on WCST scores.
In sum, our findings are in agreement with the studies examining the correlation between cognitive functions and schizophrenia symptoms, suggesting that patients with negative symptoms performed poorly on cognitive test batteries measuring verbal learning, visual learning, working memory, reasoning/problem solving and processing speed (Addington et al., 1991). Meanwhile, positive symptoms did not correlate with cognitive impairment in patients with schizophrenia (Addington et al., 1991; Davidson et al., 1995; Mohamed et al., 1999). Interestingly, our study pointed to the significant correlation between general psychopathology symptoms and cognitive dysfunction in patients with schizophrenia. To date, there were few studies illustrating the impact of general psychopathology symptoms on cognitive profile of patients with schizophrenia (Mutsatsa et al., 2006; Seghers & Docherty, 2009).
Profile of neurocognitive deficits in relation to insight
Our participants were assessed using SUMD to measure clinical insight; the results correlated with neuropsychological test batteries and revealed a significant correlation between all cognitive domains and SUMD, except verbal and visual learning domains assessed by verbal and visual paired association 1 subtests, as well as attention assessed by failure to maintain set subtest. Only visual learning (trials administered), working memory (percentage error), and processing speed (perseverative responses, and trials to complete first category) were significantly negatively correlated with SUMD.
The researches investigating the association between insight and cognitive functions showed contrasting results. Goodman et al. (2005) reported that insight was significantly correlated with performance in working memory (verbal domain), object learning (visual domain) and identification of facial emotions (Goodman et al., 2005). Many studies reported a significant correlation between poor insight and impairment of executive functions (Braff et al., 1991; Drake & Lewis, 2003; Kenny & Meltzer, 1991; Lysaker & Bell, 1994; Lysaker et al., 1998), memory (Keshavan et al., 2004; Mutsatsa et al., 2006; Smith et al., 2000) and attention (Lysaker & Bell, 1995). Aleman et al. (2006) conducted a meta-analysis and concluded that neuropsychological dysfunction, specifically impairment of set-shifting and error monitoring, contributes to poor insight (Aleman et al., 2006).
Future directions for research should take into consideration that our findings of specific cognitive impairment in individuals with negative symptoms, general psychopathology symptoms and poor insight deserve further assessment via longitudinal research. Furthermore, it would be interesting to speculate on possible mechanisms for why negative symptoms, general psychopathology symptoms and poor insight affect cognitive functioning. Finally, we hope future researches could find out new strategies to overcome negative symptoms and poor insight as both would be reflected on the social and adaptive functioning of patients with schizophrenia.
Research Data
Datainbrief_template – Supplemental material for Correlating cognitive functions to symptom domains and insight in Egyptian patients with schizophrenia
Datainbrief_template for Correlating cognitive functions to symptom domains and insight in Egyptian patients with schizophrenia by Afaf Hamed Khalil, Marwa Abd el-Meguid, Mostafa Bastawy, Samah Rabei, Ramy Ali and Mohamed Hossam Eldin abd elmoneam in International Journal of Social Psychiatry
Footnotes
Acknowledgements
The authors are thankful to Dr. Adbdel Gawad Khalifa for his participation in performing the cognitive battery assessment, including the Wechsler Adult Intelligence Scale (WAIS), Wechsler Memory Scale (WMS), and Wisconsin Card Sorting Test (WCST) for the participants.
Conflict of interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
