The Self-Assessment Scale of Cognitive Complaints in Schizophrenia: A validation study in Tunisian population
© Johnson et al. 2009
Received: 22 April 2009
Accepted: 8 October 2009
Published: 8 October 2009
Skip to main content
© Johnson et al. 2009
Received: 22 April 2009
Accepted: 8 October 2009
Published: 8 October 2009
Despite a huge well-documented literature on cognitive deficits in schizophrenia, little is known about the own perception of patients regarding their cognitive functioning. The purpose of our study was to create a scale to collect subjective cognitive complaints of patients suffering from schizophrenia with Tunisian Arabic dialect as mother tongue and to proceed to a validation study of this scale.
The authors constructed the Self-Assessment Scale of Cognitive Complaints in Schizophrenia (SASCCS) based on a questionnaire covering five cognitive domains which are the most frequently reported in the literature to be impaired in schizophrenia. The scale consisted of 21 likert-type questions dealing with memory, attention, executive functions, language and praxia. In a second time, the authors proceeded to the study of psychometric qualities of the scale among 105 patients suffering from schizophrenia spectrum disorders (based on DSM- IV criteria). Patients were evaluated using the Positive and Negative Syndrome Scale (PANSS), the Global Assessment Functioning Scale (GAF scale) and the Calgary Depression Scale (CDS).
The scale's reliability was proven to be good through Cronbach alpha coefficient equal to 0.85 and showing its good internal consistency. The intra-class correlation coefficient at 11 weeks was equal to 0.77 suggesting a good stability over time. Principal component analysis with Oblimin rotation was performed and yielded to six factors accounting for 58.28% of the total variance of the scale.
Given the good psychometric properties that have been revealed in this study, the SASCCS seems to be reliable to measure schizophrenic patients' perception of their own cognitive impairment. This kind of evaluation can't substitute for objective measures of cognitive performances in schizophrenia. The purpose of such an evaluation is to permit to the patient to express his own well-being and satisfaction of quality of life.
It is now well proven that schizophrenia is associated with multiple cognitive deficits [1–3] which can be profound and devastating . Patients with chronic schizophrenia demonstrate impairments that range between one and a half to two standard deviations below healthy controls on several key dimensions of cognition , especially verbal memory, working memory, motor speed, attention, executive functions and verbal fluency .
These deficits are thought to be a core feature of schizophrenia and not simply the result of the symptoms or the current treatments of the illness [7, 8]. Moreover, they seem to have an impact on functional outcome  as they are correlated with poor functional abilities including skills acquisition, problem solving, and community living [10–12]. Furthermore, neurocognitive deficits are believed to be the single strongest correlate of real-world functioning .
The number of publications on cognitive deficits in schizophrenia has grown vastly over the past two decades. At the same time, an increasing number of sophisticated laboratory tasks has been developed for a better assessment of cognition . However, little is known about how patients suffering from schizophrenia perceive their own cognition. Are they aware of their eventual cognitive impairments? Do they realize that their social functioning is highly influenced by these deteriorations? Do they complain about their memory problems to their doctor and do they demand specific treatments for them?
Traditionally, the study of subjective symptoms of schizophrenic patients has been limited to delusions and hallucinations . Nowadays, abnormal subjective experiences concerning fields other than delusions and hallucinations are becoming more investigated since they are believed to be important in understanding and treating schizophrenia [16, 17]. From a historical point of view, the first author who described a patient's subjective experiences in schizophrenia was Huber [18, 19]. This German author introduced the term of "basic symptoms" to designate the first symptoms of schizophrenia that constitute the basis on which the others symptoms develop. These symptoms do not include behavioural abnormalities or verbal impairments that can be assessed objectively by clinicians. In fact, they are only reported by patients that describe them as subjective experiences of deficits including loss of energy, motor dysfunctions, abnormal corporeal sensations, altered cognitive processes, difficulties to feel emotions and vulnerability to stress . The basic symptoms were targeted by a multitude of scales comprising the Bonn Scale , the Frankfurt Complaint Scale , the Subjective Experience of Deficit Scale , the Interview on Subjective Experience , the Subjective Deficit Syndrome Scale  and the Ependorff Inventory of Schizophrenia . What is significant is that these scales dealt with different aspects of subjective experiences in schizophrenia including cognitive dysfunctions but didn't focus specifically on the latter. Only one scale, the SSTICS or Subjective Scale To Investigate Cognition in Schizophrenia , assessed specifically the cognitive subjective symptoms in schizophrenia. The psychometric properties of this scale were evaluated within a population of 114 French speaking schizophrenic patients. Validation study of the SSTICS was shown to be successful proving that cognitive complaints in schizophrenia can be reliably assessed.
To our knowledge, no similar instrument has been published and validated in the Arabic language. Consequently, the purpose of our study was to create a scale to collect subjective cognitive complaints of patients suffering from schizophrenia whose mother tongue is Tunisian Arabic.
The authors constructed the Self-Assessment Scale of Cognitive Complaints in Schizophrenia (SASCCS) based on a questionnaire covering five cognitive domains which are the most frequently reported in the literature to be impaired in schizophrenia [6, 26]. The scale consisted of 21 questions dealing with memory, attention, executive functions, language and praxia. Memory was evaluated through its components: working memory (item 1&2), episodic memory (item 3 though 9) and semantic memory (item 10&11). Attention was investigated through its components: distractibility (item 12), alertness (item13), selective attention (item14), divided attention (item15) and sustained attention (item16). Executive functions were explored through their components: planning (item17), organisation (item18) and flexibility (item19). Finally, language was examined through item 20 and praxia through item 21. The scale was made to be as clear, simple and easy to use by patients suffering from schizophrenia. It was written in Tunisian Arabic dialect. 'See additional file 1: Tunisian version of the SASCCS'. 'See additional file 2: English version of the SASCSS'.
The questionnaire was first administered to a reduced sample of 38 patients (35 men, 3 women) meeting the DSM-IV diagnostic criteria for schizophrenia (n = 35) or schizoaffective disorder (n = 3) . The aim of this preliminary work was to collect comments from both patients and investigators in order to better formulate the items and furthermore, to add examples to the questions that closely suit the patient's daily life. Mean age of the patients was 34 ± 8.9 years and time elapsed since onset of the disease was 10.3 years (SD = 6.89). 'See additional file 3: Table S1: demographic characteristics and psychiatry history of pre-test sample'.
Mean total score of the PANSS was 61 ± 16 .
Accordingly to this purpose, item 8 was modified in a way to provide examples corresponding to both men and women in their daily activities. The wording of items 10, 14 and 15 was reviewed in a way to be clarified. This pre-test also served to harmonize the modalities of the scale's administration and the instructions given each time to the patients.
The SASCCS is a self-rated questionnaire administered during a structured interview in which the investigator explains to the patient the way he should answer to the 21 Likert-type questions of the scale. The patient is asked to read each of the items in which problems of memory or concentration of daily life are presented and may have been experienced by him self. He is then asked to estimate the frequency of occurrence of such situations in his own life. For that purpose, he must circle the number that best corresponds to his experienced life. (4-very often; 3-often; 2-sometimes; 1-rarely; 0-never). The SASCCS total score is calculated by adding each item score together. The more the patient complained about cognitive impairments, the higher was the scale's total score.
The approximate time to completion was 15 minutes on average. The questionnaire was administered at the outpatient clinic. The same trained psychiatrist proceeded to the administration of the scale among all participants. The investigator should remain on site until the patient is done with the questionnaire. He could provide explanations to some questions or even examples to clarify the meaning of items especially item 13, 15, 18 and 19. 'See additional file 4: examples for items 13, 15, 18 and 19 of the SASCCS'.
have a minimum educational level of 5 years,
no evidence of mental retardation,
being at the time of testing under unchanged medication dosage for the last 4 weeks.
never undergone electroconvulsive therapy,
no evidence of organic brain pathology including cerebral tumor, epilepsy, systemic disease, history of cranial trauma, brain surgery
no history of substance abuse or dependence, and consumption of psychoactive.
Demographic sample characteristics and psychiatric history
Age (years; mean, SD)
Years of education (mean,SD)
Marital status (n)
Duration of illness (years; mean, SD)
Number of hospitalisations (mean, min-max)
Total period of hospital stay (weeks; mean, min-max)
Chlorpromazine equivalent of antipsychotic dosage (mean, SD)
Psychopathological symptoms were evaluated using the PANSS , the Calgary depression scale (CDS)  and the Global Assessment Functioning scale (GAF scale) . PANSS, CDS and EGF were administered by the same trained psychiatrist for all participants. Mean scores on these clinical scales were as follows: 52.84 (SD = 9.64) for the PANSS total score, 1.35 [min = 0; max = 5] for the CDS and 62.58 (SD = 13.88) for the GAF scale. Mean scores for the PANSS subscales were as follows: 10.05 (SD = 2.5) for the positive symptoms, 16.32 (SD = 4.49) for the negative symptoms and 26.4 (SD = 5) for general psychopathology. Mean score for the item G12 of the PANSS assessing insight was 2.32 (SD = 1.15).
Using the 5-factor model of the PANSS as identified by Lindenmayer et al. , we calculated the cognitive factor and the depression factor which had respectively a score of 10.14 (SD = 2.49) and 5.93 (SD = 1.99).
We conducted an exploratory principal component analysis (PCA) on the correlation matrix of the 21 items of the SASCCS. Several guidelines were used to select the number of factors: the Kaiser criteria and the interpretability of the factors. Oblimin rotation was then performed.
Construct validity and reliability were evaluated by calculating Cronbach's alpha coefficient and the average of correlations between each item and the total score.
Correlation analyses were performed using the Pearson coefficient when data had normal distribution; elsewhere, Spearman rank correlation was calculated.
Statistical significance level was set atp = 0.01 (two-tailed).
Statistical analyses were performed using SPSS software in his 12th version.
This research has been undergone in a psychiatric university department in RAZI hospital. It has been approved by the local ethic committee. Patients have signed a written and informed consent.
The SASCCS global score mean was 24.98 (SD = 14.83; min = 0, max = 109; median = 24).
It was evaluated by calculating Cronbach's alpha coefficient  which was equal to 0.85 proving a good internal consistency of the scale but furthermore, a satisfactory reliability of its measure.
Its was assessed within a subgroup of 39 patients examined by the same investigator at a mean interval of 80 days (SD = 33). Intra-class correlation coefficient was equal to 0.77 (p = 0.00) suggesting a good stability over time.
Principal component analysis: Total variance explained
% of variance
Subjective cognitive domains of complaints
5, 12, 16, 20, 21
8, 9, 15, 18
4, 6, 7
3, 14, 17, 19
We examined whether correlations existed between scores derived from the scale and positive, negative and disorganisation factors derived from the factorial analyses studies of the PANSS . We also considered the item G 12 assessing insight as well as the Calgary Depression Scale total score.
The SASCCS total score wasn't correlated to any of the PANSS. A weak negative correlation between the SASCCS total score and PANSS insight score was found (r = -.21) but didn't reach the statistical significance (p = .03). Calgary score was correlated with the SASCCS total score (r = .33;p = .001).
The cognitive factor of the 5-factor model of the PANSS wasn't correlated to the SASCCS total score or sub-scores.
The depression factor was correlated to the SASCCS total score (r = .20) although this correlation didn't reach the statistical significance (p = .03).
The aim of this study was to construct and to validate a scale to measure the subjectivity of patients with schizophrenia regarding their cognition. The SASCCS, which was easy to administer in less than 15 minutes, had good reliability and stability over time. No cut-off has been determined for this scale. In fact, the SASCCS total score is used to estimate a patient's level of complaining.
The composition of subjective cognitive domains as derived from factor analysis was slightly different from that of the initial theoretical model which has been the basis of the scale's construction. Actually, the scale's items have been distributed after PCA differently from the original structure of the scale leading to a neo-construct of the instrument. These differences were not surprising since the questionnaire was based on the neuropsychological theoretical conception of cognition whereas factor analysis of the scale reflected the patient's own perception of his cognition. Stip et al., using the Subjective Scale To Investigate Cognition in Schizophrenia (SSTICS), have also found a difference between the distribution of the items in the initial model and in the neo-construct of their scale . It could be that the selected items did not exactly measure what they were supposed to. Also, their specificity might be imperfect as it refers to several overlapping dimensions.
These findings point to the complex representation of schizophrenic patients of their own cognition. And even though the latter does not correspond to the theoretical construct of cognition, the scale remains reliable because of both its good internal consistency and stability over time.
During this study, no other instrument evaluating cognitive functions was administered simultaneously to our population. Therefore, convergent validity was unnecessary. However, when reviewing the literature, no positive correlation was found between objective and subjective scores of cognition. Using the SSTICS, Prouteau et al. found that cognitive nature of subjective complaints did not strictly match with that of impaired objective performances . Chan et al. assessed prospective memory in patients with schizophrenia and did not find a correlation between objective performances and subjective measures of this cognitive function .
These results suggest that subjective evaluation of cognition could be an independent dimension from its objective assessment in patients with schizophrenia.
In fact, in our study, no correlation has been found between the SASCCS scores and the PANSS cognitive factor which could also point to the fact that self-assessment of cognition is a totally independent aspect from clinical evaluation of the cognitive functions.
The correlation of insight with subjective perception of cognition in schizophrenia is an aspect that deserves to be considered and analyzed.
In fact, awareness of one's own cognitive deficits could be highly influenced by consciousness of one's whole condition as a mentally ill person. Schizophrenia is generally accompanied by a lack of insight meaning an impaired awareness of one's psychiatric condition and life situation . Therefore, low scoring at the SASCCS could be due to a lack of insight.
In our study, a weak negative correlation between PANSS insight score and SASCCS score has been found (r = -.21,p = .03) but didn't reach the significance level set at 0.01. However, it should be noticed that our study included a majority of subjects scoring no more than 4 on the PANSS insight item. Only one patient had a score of 5.
Since insight could influence one's subjective perception of cognition, it is recommended to evaluate patient's insight while using the SASCCS.
In our study, there was a positive correlation between SASCCS total score and CDS score meaning that the more depressive symptomatology is severe, the more the patient reports cognitive troubles. Although it was not statistically significant, we also did find a correlation between SASCCS total score and the depression score of the 5-factor model of the PANSS suggesting the influence that could exert depression on self assessment of cognition by emphasizing cognitive complaints when being more depressed.
Lecardeur et al. found in their study using the SSTICS a correlation between the scale total score and the PANSS depression score. It could be suggested that subjective complaints of cognitive deficits may influence a patient's objective depressive state as rated by the clinician .
Considering the influence of depressive traits on subjective perception toward cognition, we recommend measuring the patient's mood state when using the SASCCS.
We present here a self-assessment scale to evaluate cognitive deficits as perceived by patients suffering from schizophrenia in domains of memory, attention and executive functions. Given the good psychometric properties that have been revealed in this study, the SASCCS seems to be reliable to measure schizophrenic patients' perception of their own cognitive impairment. This kind of evaluation can not replace objective measures of cognitive performances in schizophrenia. Actually, the purpose of such an evaluation is to allow the patient to express his own well-being and satisfaction of quality of life. Furthermore, subjective evaluation of cognitive functions could provide a more complete picture of the cognitive profile of an individual. Therefore, better therapeutic targets could be adapted to his condition during cognitive rehabilitation programs.
Self-Assessment Scale of Cognitive Complaints in Schizophrenia
Principal Component Analysis
Subjective Scale To Investigate Cognition in Schizophrenia.
The authors would like to thank Marie-Chantal Bourdel for her help in statistical analyses.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.