Cognitive functions in schizophrenia : a comparative study in an Indian population

Background Significant cognitive impairment has been considered a core characteristic of schizophrenia. Though many studies have reviewed the phenomenon, only a few Indian studies have investigated this aspect. Aims To compare the functioning of six cognitive domains of patients with chronic schizophrenia with an age, sex and education matched control group and to assess the association of socio-demographic and clinical correlates with cognitive functioning. Methods Nine different neuropsychological tests were administered to assess the different cognitive domains. These were administered to 47 patients with chronic schizophrenia and 29 healthy controls. Results Patients with schizophrenia performed poorly in the areas of processing speed, verbal learning, verbal working memory and visual working memory. We also found that negative symptoms predict poor performance in speed of processing and verbal working memory. Increased age was a predictor of better verbal fluency. Family history of schizophrenia in first degree relatives is a predictor of poor verbal working memory. Conclusions The findings suggest that patients with chronic schizophrenia show selective cognitive impairment. The study failed to show any impairment in the area of concentration and executive function, though patients needed more time and intermittent verbal prompts to reach a level of performance equivalent to controls. Negative symptoms were found to affect cognitive performance while age was associated with better verbal fluency performance.


Introduction
Symptoms of schizophrenia can be categorized into the three domains of positive symptoms, negative symptoms and cognitive dysfunctions.Significant cognitive impairment has been considered a core characteristic of schizophrenia and is not caused by chronicity of illness, treatment or institutionalization (1).Whether the cognitive deficits in schizophrenia are better characterized as part of the generalized dysfunction or reflect relatively independent domain specific deficits is not yet known.Most studies report significant impairment in individual domains such as verbal memory and executive function, vigilance, memory and concept formation, working memory, learning, verbal memory and executive function (2,3,4).
Impaired performance on neurocognitive tests correlates with deficits in everyday living and social outcomes across a number of neuropsychiatric conditions, particularly schizophrenia (5,6,7,8,9).There has been some debate about whether the relationship between impairments in different cognitive domains and aspects of disability are specific or more generalized.For instance, in a study of people with schizophrenia, attentional performance was shown to predict early occupational success while sustained work performance was better predicted by episodic memory performance (10).In contrast, other studies have found correlations of essentially similar magnitude between multiple components of a cognitive assessment battery and various aspects of everyday functional outcomes (11,12).If the relationship between multiple cognitive impairments and different elements of functional disability are similar, then a detailed and lengthy neuropsychological assessment may not really be necessary.Based on this premise, several abbreviated cognitive assessment procedures have been developed (13,14).
Only a few studies have examined cognitive impairment in Indian patients with schizophrenia.A study by Sabhesan and Parthasarthy in 2005 found varying degrees of executive function deficits in schizophrenia (15).Another study found that patients with stable schizophrenia performed poorly compared to euthymic bipolar affective disorder patients in the Wisconsin Card Sorting test, Continuous Performance Test and Spatial Working Memory Test (16).Two other studies have assessed the relationship between impaired cognitive function and functional disability of patients with schizophrenia (17,18).
There is no brief neuropsychological assessment battery for schizophrenia which has been standardized for Indian patients.Indian studies conducted upto now tested specific cognitive domains in patients with schizophrenia using individual domain-specific tests and compared the performance with that of healthy controls.
In this study different test were assembled and used to measure different cognitive domains.These were based on the MCCB (MATRICS consensus cognitive

Study population
The study population consisted of consecutive patients diagnosed with schizophrenia, attending outpatient Psychiatry Department of Medical College and Hospital, Kolkata, India.Inclusion criteria were duration of illness more than 2 years, age between 15-60 years, education above fifth standard and diagnosis made according to DSM IV TR using SCID I (Structured Clinical Interview for DSM VI TR).Age, sex and education matched healthy volunteers who didn't have any personal or family history of psychiatric illness were recruited as controls.Both cases and control were free from any major medical disorder such as diabetes, thyroid disorder.Sixty one patients were approached and 47 patients consented to take part in the study.Thirty two were males.Study was commenced after ethical clearance by the Institutional Review Board.Written informed consent was obtained using the informed consent form of ICMR from both patients and controls (27).Five patients refused to complete the digit cancellation test and appropriate statistical steps were taken to neutralize the deficiency.

Statistical analysis
To compare the performance in six cognitive domains in patients with chronic schizophrenia with those of an age, sex and education matched control group a preliminary Multivariate Analysis of Variance (MANOVA) analysis was done.Cognitive dysfunction measurements (CDM) of both cases and controls were used as dependent variables.The predictor variables included the demographic variables presence or absence of disease as well as the measurements of psychopathology of schizophrenia e.g.positive symptoms, negative symptoms, depressive symptoms.There were a total of twenty predictors.As the sample size was only 47 in the treatment group regression analysis was performed for each of the response variables in three steps.First a regression was done using all the predictor variables.Second a best subsets regression analysis was performed.This looks at all the possible models and the model with the smallest Cp value was selected as the "best model".Finally, a regression is carried out on the model selected by the Mallows' Cp criterion.
To assess the correlation between demographic and clinical variables and different cognitive domains in patients with schizophrenia, we carried out regression analyses for each of the response variables using only the data of patient group.battery).We have mostly used tests that have been already validated in Indian patients.

Methods
Objectives of the study were; 1.To compare functioning of six cognitive domains (speed of processing, attention/vigilance, verbal working memory, verbal learning, visual working memory, reasoning and problem solving) in patients with chronic schizophrenia with that of an age, sex and education matched control group.

To assess the correlation between demographic variables and clinical variables with functioning of different cognitive domains in patients with
schizophrenia.(19,20,21,22).

Outcome measures
The Adult Porteus Maze Test which is validated for Indian populations was selected instead of the Neuropsychological Assessment Battery (NAB) mazes test to test problem solving and reasoning.The Digit Cancellation Test (DCT) was selected to test attention instead of the Continuous Performance Test as some patients were not comfortable with using computers (23).This test has also been validated in adult Indians (24).
The final list of tests we used for assessing our subjects was as follows:

Demographic characteristics of study population
The mean age of cases and controls were 35.70 years (SD 11.0) and 38.10 years (SD 11.25) respectively.The difference was not statistically significant (p =0 .364)(Table 1).Among cases 68.1% were male and among controls 55.17% were males.T. The mean number of years of education was 9.77 (SD 2.56 ) in cases and in controls it was 10.38 (SD 2.55).Between group difference were not significant (p = 0.314).

Cognitive functions
Cognitive functions (except "Digid cancellationwrong" and "Adult Porteus Mazes-error") were significantly poorer in patients with schizophrenia.
Patients needed a longer time (213.7 seconds ) to complete the digit cancellation task compared to controls (172.6 seconds ).The mean time needed to complete the Adult Porteus Mazes was higher in patients with schizophrenia (40.7 seconds) compared to controls (30 seconds).Among the cognitive function measurements, the most difference between the patient and control groups were for CCB immediate recall (t=4.87) and WMS (t=5.043).Table 2 summarises the cognitive outcomes of the two groups.

Identifying factors which predict cognitive functions:
Stepwise regression analysis was carried out to find the relationship between socio-demographic variables and cognitive functions.Presence of disease was considered as one variable.The results are shown in table 3.

Discussion
The main finding of the study was that patients with schizophrenia displayed significantly poorer performance in four cognitive areas but the difference was most distinct for immediate recall (verbal memory) and spatial span (visual working memory).Significant impairment in working and verbal memory in patients with schizophrenia has been reported previously.In particular the symbol coding task yields showed most impairment (28).Wexler et al. noted a general deficit in verbal memory despite normal motivation, attention and general perceptual function in a group of patients with schizophrenia.In the same study verbal working memory tasks were more impaired than nonverbal working memory tasks in patients with schizophrenia (29).This study, however, found greater deficits in nonverbal (visual) working memory.Similar findings were reported by Chey et al. who found that both immediate and visual memory task scores were poor in patients with schizophrenia (30).It thus appears that impairment in working memory in schizophrenia is a general deficit and not just confined to the verbal domain.
We found that increasing age was associated with better verbal fluency (semantic and categorical).This can be explained if schizophrenia is viewed as a neurodevelopmental disorder which actually remains stable or may even improve with time and this has been shown in several studies (31,32).A meta-analysis of longitudinal studies of cognition in schizophrenia shows that semantic verbal fluency remains stable with age both in healthy controls and in patients with schizophrenia whereas other cognitive areas improve with age (33).Another meta-analysis shows that individuals with youth-onset schizophrenia have larger deficits in all areas of cognition whereas those with later onset disease has some relatively preserved cognitive function including vocabulary.This could either mean that patients with youth-onset schizophrenia are affected worse in terms of cognitive damage compared to the late onset group, or that increased age is associated with a richer vocabulary which provides some kind of compensation in cognitive testing (34).
This study failed to elicit any impairment in performances in the DCT (in its number of errors) in patients with schizophrenia.It has been reported that people with schizophrenia show greater distraction, slower response time, reduced vigilance and that they have problems with tasks requiring continual processing.They also have difficulties in distinguishing important stimuli from irrelevant stimuli and in multi-tasking as well as in maintaining or changing the location of attentional resources.In general, these attentional deficits come to the fore when tasks are very complex, as these require a higher degree of information processing (35).The absence of any impairment in this study may be due to the fact that the DCT task was an easier one.It was however noted that the patients needed a longer time (213.7 seconds) to complete the DCT task compared to controls (172.6 seconds).This is in keeping with findings reported by other authors (36).
This study showed no significant differences in the performances of the Adult Porteus Maze test between patients with schizophrenia and healthy participants as far as number of errors were concerned.
8. Visual retention subtest of PGI-BBD-(This is a neuropsychological battery developed in Post Graduate Institute, Chandigarh, India).It tests visual memory.9. Digit cancellation test.This assesses attention.
Four original tests of the MCCB were replaced by similar tests already standardized for Indian populations.These were 1.Digit symbol subtest of WAPIS (Wechsler's adult performance intelligence scale) instead of BACS symbol coding 2. Verbal fluency subtest of KCSB (Kolkata Cognitive Screening Battery) for category fluency 3.Wordlist memory task subtest of KCSB in place of Hopkins verbal memory test 4. Visual retention subtest of PGI BBD (Post Graduate institute of India-Battery of Brain Dysfunction) for brief visuospatial memory test

Table 1 -
Socio demographic profile of cases and controls -the model selected included disease (p<0.001),age (p=0.014) and education (p=0.067).Schizophrenia was associated with a significantly worse performance.Increased age was associated with better outcome for both groups.Dependent variable digit symbol subtest of WAPISthe selected model included presence of disease (p<0.001),age (p=0.005) and education (p=0.003).

Table 2 -
Comparison of cognitive functioning between cases and controls Dependent variable CCB-verbal-the selected model included disease (p=0.01),age (p=0.007),sex (p=0.125),education(p=0.158) and occupation (0.11).Presence of disease and age were significant predictors.Increasing age was associated with better performance in the CCB verbal test.Dependent variable digit cancellation-the selected model included education (p=0.104) and occupation (p=0.05).R-square and adjusted R-square values were low (9.3% and 6.6%).Dependent variable WAPIS digit symbol-the selected model included only sex (p=0.014).Dependent variable CCB immediate recall -Selected variables were positive symptom score(p=0.01),education (p=0.027) and depression (0.105).Higher scores on positive symptoms and lesser years of education were associated with poorer performance.

Table 3 -
Association between socio-demographic variables and cognitive functions (42)xpanding the concept of risk for pathophysiological dysfunction beyond overt psychosis(41).Risk was defined by familial status and presence or absence of SSPD (schizophrenia spectrum personality disorder).SWM was assessed in patients with schizophrenia, SSPD relatives of patients with schizophrenia, non SSPD relatives of patients with schizophrenia, SSPD community members with no family history of psychosis and non SSPD community members.As expected, maximum deficits were found in the schizophrenia group..Significant deficit were seen in the former two groups but not in patients with SSPD without a family history of psychosis(41).Ioana et al. also found working memory deficit in patients with a first episode of schizophrenia and their healthy siblings(42).This study found that a family history of schizophrenia was associated with a poorer verbal working memory among patients with schizophrenia; this is in keeping with findings in the existing literature.Speed of processing, verbal learning, visual learning and working memory were impaired in patients with chronic schizophrenia when compared with the control group.The areas most impaired were immediate recall and visual working memory.Negative symptoms predicted worse performances in speed of processing and verbal working memory whereas increased age was associated with better verbal fluency.This study however, failed to analyse the patients' performance in terms of categorical and phonemic fluency separately, and instead considered verbal fluency as a single domain.As these two cognitive domains are controlled by different neural networks and appear to be affected differentially in patients with schizophrenia, this is a limitation of the study.Family history of schizophrenia was a predictor of greater impairment in verbal working memory.There was no impairment in concentration (digit cancellation test) and executive function (Adult Porteus Maze), though patients needed more time and intermittent verbal prompts to reach a level of performance equivalent to controls.
(40)ion et al. in 2009found that depression, but not negative symptoms were associated with decreased processing speed in patients with schizophrenia(40).Working memory has often been considered to be an endophenotype of schizophrenia.Saperstein et al. conducted a study in 2006 to examine visuospatial working memory (SWM) as a cognitive endophenotype of schizophrenia