SEARCH

SEARCH BY CITATION

Keywords:

  • clock-drawing;
  • cognitive function;
  • life function;
  • long-term hospitalization;
  • older adults;
  • schizophrenia

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Background:  There has been a growing need for a cognitive assessment tool that can be used for older adults with schizophrenia in clinical settings. The clock-drawing test (CDT) is a brief cognitive test that covers a wide range of cognitive function. Although it is widely used to assess patients with dementia, limited data are available on its usefulness in older patients with schizophrenia. Thus, we investigated the psychometric properties of the CDT and their relationship with life functions to examine the test's usefulness for assessing cognitive function in older adults with schizophrenia.

Methods:  Seventy-three older adults with chronic schizophrenia who had been hospitalized for over 1 year participated in the study. We adopted the executive clock-drawing task for administration and scoring of the CDT, which consists of free-drawn and copy conditions. The Mini-Mental State Examination and the Brief Assessment of Cognition in Schizophrenia were administered. Symptom severity and life functions were assessed with the Positive and Negative Syndrome Scale and the Life Skills Profile, respectively.

Results:  Both free-drawn and copy scores significantly correlated with the Mini-Mental State Examination score and the Brief Assessment of Cognition in Schizophrenia composite score. These scores also significantly correlated with symptom severity and length of current hospitalization. Stepwise regression analysis showed that only the copy score, together with symptom severity, predicted the Life Skills Profile score.

Conclusions:  The CDT can assess cognitive function in older adults with schizophrenia. Moreover, CDT performance is associated with life functions independent from other clinical variables. These results suggest that the CDT is a useful cognitive assessment tool for this population.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Older adults with schizophrenia have been reported to be impaired in several domains of cognitive function, such as executive function and visuospatial ability.1 These impairments affect functional outcomes including functional status in the community, social problem solving, and psychosocial skill acquisition,2,3 and they worsen during the course of the illness as a result of ageing,4 long-term use of antipsychotic medications,5 and the disease progression itself.1 Given that older patients with schizophrenia make up the largest group of psychiatric inpatients in Japan,6 cognitive assessment of this population is important even in clinical settings.

There have been several standardized cognitive test batteries developed for patients with schizophrenia, which consist of several neuropsychological tests, such as the Measurement and Treatment Research to Improve Cognition in Schizophrenia and the Brief Assessment of Cognition in Schizophrenia (BACS).7–10 Although these batteries can assess various aspects of cognitive function in detail, they are time-consuming (e.g. the Measurement and Treatment Research to Improve Cognition in Schizophrenia is approximately 90 and BACS is 30 min) and are difficult to administer in clinical settings. Therefore, a brief and easy assessment tool for cognitive function in clinical settings is needed.

The clock-drawing test (CDT) is a brief cognitive test that can be administered within 5 min. It is widely used to assess patients with dementia because of its easy administration and good psychometric properties.11 Several studies have tried to examine the usefulness of the CDT as a measure of cognitive function in older adults with schizophrenia. Bozikas et al. compared CDT the performance of elderly patients with schizophrenia with that of normal controls and reported poor performance in the patients.12,13 Lowery et al. also reported poor performance in elderly institutionalized patients with schizophrenia compared to an age-matched comparison group.14 They reported that CDT performance was associated with other neuropsychological tests although the sample size was small. Additionally, a recent study with a larger sample size confirmed poor CDT performance in patients with schizophrenia compared to healthy controls.15

Although these previous studies suggested the usefulness of the CDT in older adults with schizophrenia, there have been no studies that addressed the relationship between CDT performance and cognitive test batteries specifically developed for patients with schizophrenia. Moreover, the relationship between CDT performance and functional outcomes remains unclear. Thus, we sought to investigate whether CDT performance is associated with cognitive function, as evaluated by a standardized cognitive measurement for patients with schizophrenia (BACS), to confirm its validity as a cognitive assessment tool for patients. In addition, we examined the relationship between CDT performance and life functions assessed by an objective measure designed specifically to assess general levels of function in patients with schizophrenia.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Participants

Seventy-three older adults (31 men, 42 women, mean age 63.9 ± 8.2 years, mean duration of illness 36.8 ± 9.2 years) who met DSM-IV-TR criteria for schizophrenia (American Psychiatric Association, 2000) participated in this study. The participants were recruited from Gojouyama Hospital in Nara and Kitayama Hospital in Kyoto, Japan. Inclusion criteria for the patients were hospitalization for more than 1 year and over 50 years of age. Exclusion criteria were as follows: (i) history of other concomitant Axis I disorder (including dementia); (ii) mental retardation; (iii) history of head trauma; (iv) history of neurological disorder; and (v) past or present substance dependence/abuse. All patients were clinically stable on antipsychotics. Demographic data including age, gender, education, age of onset, duration of illness, duration of current hospitalization, and dose of antipsychotics were obtained from medical records. Table 1 lists the participants' demographic characteristics.

Table 1. Patient characteristics (n= 73)
CharacteristicData
Age (years)63.9 ± 8.2
Sex (n) 
 Men31
 Women42
Education, mean ± SD (years)11.6 ± 2.1
Age of onset, mean ± SD (years)27.1 ± 9.5
Duration of illness, mean ± SD (years)36.8 ± 9.2
Duration of current hospitalization, mean ± SD (years)15.1 ± 10.9
Total chlorpromazine-equivalent doses, mean ± SD (mg/day)855 ± 454
Total biperiden-equivalent doses, mg/day1.7 ± 1.8

The study protocol was approved by the ethics committees of Gojouyama Hospital and Kitayama Hospital. All participants gave written informed consent after they received a complete description of the study.

The clock-drawing test (CDT)

Of the many available scoring systems, we adopted the executive clock-drawing task (CLOX) method developed by Royall et al. because it covers a wide range of cognitive functions, including executive function and visuospatial function, by using a unique administration procedure and scoring system.16

CLOX consists of free-drawn (CLOX1) and copy (CLOX2) conditions. In CLOX1, the subject is shown the blank side of a piece of paper, on a light coloured surface, with a circle visible from the underside. On the blank side of the paper, the subject is instructed to ‘Draw a clock that says 1:45. Set the hands and numbers on the face so that a child could read them.’ Once the individual begins to draw, no further assistance is provided. In CLOX2, the examiner turns over the paper and draws a clock face in the pre-drawn circle in the lower left corner. The examiner writes 12, 6, 3 and 9, fills in the rest of the numbers and sets the hands to show 1:45. The subject is asked to observe this procedure and copy the clock in the lower right corner. The performance is rated on a 15-point scale in both CLOX1 and CLOX2. Lower CLOX scores indicate impaired cognitive function. The average administration time for CLOX is approximately 5 min.

CLOX for all of the patients were scored by one rater (A.O.). Inter-rater reliability was established by the independent rating of 10 subjects randomly sampled from the entire subject group by an additional experienced psychologist (T.I.), who participated in the same interview and scored the subjects independently. The intraclass correlation coefficients were then calculated. The inter-rater intraclass correlation coefficients of CLOX1 and CLOX2 were greater than 0.91.

Cognitive assessments

We assessed global cognitive function with the Mini-Mental State Examination,17 and we administered the Japanese version of the BACS,10 which was specifically developed to assess cognitive function of patients with schizophrenia.9 BACS consists of the six subscales including list learning (verbal memory), a digit-sequencing task (working memory), token motor task (motor speed), category fluency and letter fluency (verbal fluency), symbol coding (attention and processing speed), and the Tower of London test (executive function). A z-score was calculated from each subscale score with means and standard deviations based on the dataset of 340 healthy controls from a Japanese population, aged 20–79, and a composite score was calculated by averaging all the z-scores of the six subscale measurements. Lower z-scores indicate lower cognitive function. The time to administer the BACS is approximately 30 min. These cognitive assessments were performed by a trained psychiatrist and three psychologists.

Of the initial 73 patients, seven patients could not complete the BACS because of a lack of concentration or physical disability, although all the participants completed the CLOX. Thus, a total of 66 patients were included in statistical analyses.

Life functions

We assessed the functional status of patients with the Japanese version of the Life Skills Profile (LSP).18 The LSP is an objective 39-item measure designed specifically to assess general levels of function and disability in patients with schizophrenia.19 Each item has a four-point ordinal rating with specific anchor points from 1 (extreme difficulty) to 4 (no difficulty). These items are classified into one of five subscales (self-care, non-turbulence, social contact, communication and responsibility). Lower total LSP scores indicate lower level of life functioning. This measure was scored by the nurse who was the patient's primary caregiver.

Symptom severity

We assessed symptom severity with the Japanese version of the Positive and Negative Syndrome Scale (PANSS).20,21 It consists of 40 items, and each item has a seven-point ordinal rating with specific anchor points from 1 (absent) to 7 (extreme severe). These items are classified into one of three subscales (positive symptoms, negative symptoms and general psychopathology). Lower total PANSS scores reflect lower symptom severity.

Statistical analysis

In order to examine the relationship between each CLOX score and the clinical variables (the Mini-Mental State Examination score, z-scores on six subscales and the composite score of BACS, scores on three subscales and total of the PANSS, scores on five subscales and total of the LSP, age, education, age of onset, duration of illness, duration of current hospitalization and chlorpromazine-equivalent doses of antipsychotic agents, and biperiden-equivalent doses of anticholinegic agents), Spearman's rank correlation analyses were performed.

To examine the extent of the cognitive function effects assessed by each cognitive test on life functions, we performed multiple regression analyses using a stepwise forward selection method. The total LSP score was regarded as a dependent variable. The CLOX1, CLOX2 and BACS composite score were alternately entered as independent variables together with candidate clinical variables (i.e. total PANSS score, chlorpromazine-equivalent doses of antipsychotic agents, biperiden-equivalent doses of anticholinegic agents, age, education, age of onset, duration of illness, and duration of current hospitalization). There was no multi-collinearity as indicated by the fact that there were no components with two or more variables that showed large variance (0.50 or greater) corresponding to large condition indices.

We used SPSS 12.0J software for Windows (SPSS Japan, Tokyo, Japan) for the statistical analysis. Statistical significance was set at P < 0.05 (two-tailed).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

The results of CLOX, cognitive assessments, symptom severity, and life functions are given in Table 2. The mean score of CLOX2 was higher than that of CLOX1, which reflected the difficulty of each task.

Table 2. CDT, symptom severity, cognitive function and life functions (n= 73; mean ± SD)
  • Data were not available for the entire sample.

  • BACS, Brief Assessment of Cognition in Schizophrenia; CDT, clock-drawing test; CLOX, executive clock-drawing test; CLOX1, free-drawn CLOX; CLOX2, copied CLOX; LSP, Life Skills Profile; MMSE, Mini-Mental State Examination; PANSS, Positive and Negative Syndrome Scale.

CDT 
 CLOX19.6 ± 3.9
 CLOX212.7 ± 3.0
PANSS 
 Positive15.3 ± 5.0
 Negative19.5 ± 6.2
 General36.3 ± 7.7
Total71.2 ± 15.6
BACS (n= 66) (z-score) 
 Verbal memory−2.7 ± 1.5
 Working memory−2.3 ± 1.5
 Motor−4.2 ± 1.7
 Verbal fluency−1.9 ± 1.2
 Attention and processing speed−2.7 ± 1.3
 Executive function−2.5 ± 2.1
 Composite score−2.7 ± 1.2
MMSE23.4 ± 4.7
LSP 
 Self-care27.8 ± 5.0
 Non-turbulence42.9 ± 4.8
 Social contact13.7 ± 3.5
 Communication19.9 ± 3.4
 Responsibility16.4 ± 3.3
 Total120.7 ± 16.2

The correlation matrix of each CLOX score and clinical variable is given in Table 3. Multiple regression analysis revealed that a higher CLOX2 score (β= 0.235) and a lower total PANSS score (β=−0.409) predicted a higher total LSP score (adjusted R2= 0.254, F= 12.073, P < 0.001). When the CLOX1 or BACS composite score was entered as the independent variable instead of the CLOX2 score, only a lower total PANSS score (β=−0.476) was identified as a predictor of a higher total LSP score (adjusted R2= 0.214, F= 18.698, P < 0.001).

Table 3. CLOX scores and clinical variables (n= 66)
 CLOX1CLOX2
  1. *P < 0.05,**P < 0.01.

  2. BACS, Brief Assessment of Cognition in Schizophrenia; CDT, clock-drawing test; CLOX, executive clock-drawing test; LSP, Life Skills Profile; MMSE, Mini-Mental State Examination; PANSS, Positive and Negative Syndrome Scale.

MMSE0.342**0.508**
BACS (z-score)  
Verbal memory0.2200.214
 Working memory0.379**0.468**
 Motor0.246*0.254*
 Verbal fluency0.1710.254*
 Attention and processing speed0.1040.366**
 Executive function0.267*0.468**
 Composite score0.339**0.461**
PANSS  
 Positive symptoms−0.001−0.085
 Negative symptoms−0.182−0.256*
 General psychopathology−0.258*−0.299*
 Total−0.165−0.277*
LSP  
 Self-care0.0140.283*
 Non-turbulence−0.0650.153
 Social contact−0.0990.169
 Communication0.0400.244*
 Responsibility0.0720.237
 Total−0.0160.300*
Age−0.0940.092
Education−0.0040.031
Age of onset0.1360.062
Duration of illness−0.167−0.021
Duration of current hospitalization−0.314*−0.256*
Total chlorpromazine-equivalent doses−0.0270.118
Total biperiden-equivalent doses−0.026−0.111

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

The present study demonstrates that CDT performance scored by the CLOX method correlates with standardized cognitive batteries for patients with schizophrenia. In addition, multiple regression analysis shows that the CLOX2 score predicts life functions independently from other clinical variables. Together with good inter-rater reliability and tolerability, these results suggest that the CDT is a useful and quick measure for assessing cognitive function in older adults with schizophrenia.

With regard to the CDT's psychometric properties, several subscales of the BACS including working memory, motor speed, and executive function correlated with both the CLOX1 and CLOX2 scores. Additionally, the CLOX2 score correlated with the verbal fluency and attention subscale. Although CLOX2 is recognized as only a copy condition and reflects visuospatial function, it may be related to various cognitive functions in the present population. In performing CLOX2, a subject must pay attention to the examiner's whole clock-drawing process in order to follow it, which requires attention and working memory. Lowery et al. reported that the CDT performance scored by another scoring method significantly correlated with a wide range of cognitive functions including memory, word-list generation, and visuospatial/visuoconstruction in older adults with schizophrenia.14 Taken together with the previous report, this study suggests that the CDT covers a wide range of cognitive functions despite its being brief and easy to administer.

As for its relationship with sociodemographic and clinical variables, CDT performance significantly correlated with current length of hospitalization. This may reflect the deteriorative effect of long-term hospitalization on cognition. Harvey et al. reported that cognitive and functional decline can be detected in a subset of geriatric long-stay patients with schizophrenia.22 With regard to the CDT's relationship to symptom severity, poorer performance of the CLOX1 was associated with greater severity of general psychopathology, as measured by PANSS. In contrast, poorer performance on the CLOX2 was associated with greater severity of negative symptoms and general psychopathology. Earlier studies reported the association between CDT performance and symptom severity on the Brief Psychiatric Rating Scale and positive symptoms on the PANSS.13,23 One possible reason for the inconsistency may be attributable to the difference in CDT methods. Because the performance of CLOX requires greater attention than other methods, CDT performance of CLOX may depend on general psychopathology as represented by items on the PANSS such as attention and disturbance of volition.

With respect to the relationship with life functions, only CLOX2, not CLOX1, was identified as a predictor of life functions as assessed by LSP. This is somewhat unexpected because previous studies in other populations, such as community-dwelling older adults and people with traumatic brain injuries,24,25 reported that CLOX1 was related to life functions. The present result may relate to the fact that CLOX2 covers a wider range of cognitive functions than CLOX1 in this study. Moreover, the BACS did not predict life functions when sociodemographic and symptom severity variables were entered as independent variables at the same time. This is noteworthy because the CDT is much briefer than the BACS. The CDT would be useful for clinicians to briefly measure cognitive functions when a patient is preparing to be relocated from the hospital to long-term care facilities in order to arrange for a suitable environment. Given that visuospatial ability worsens in older adults with schizophrenia and is related to functional outcome,1,26 CLOX2's better psychometric property in visuospatial function may contribute to this result.16 However, it is still unclear how impaired visuospatial function relates to poorer life function in long-term hospitalized older adults with schizophrenia.27–29 Further studies are needed. Finally, CLOX2 was associated with LSP subscales of self-care and communication. Green at al. reported that this specific cognitive domain is related to the specific aspect of life functions.2 As shown by the correlation with BACS subscales, CLOX2 was related to attention and executive function, reflecting that CLOX2 requires a subject to pay attention to the examiner's whole clock-drawing process and follow the process based on the memory. These cognitive functions measured by CLOX2 seem to be essential to performing self-care and communication.

There are some limitations in this study. First, the study sample size was small, which may reduce the strength of our conclusions. However, the sample size was bigger than in previous studies that investigated the relationship between CDT performance and neuropsychological tests.14,25 Second, this study was conducted using a cross-sectional design. Further longitudinal studies are needed to confirm the usefulness of CLOX for predicting life functions. Third, this study's participants were long-term hospitalized older adults. Therefore, our result may not be generalized to other groups with schizophrenia.

In conclusion, the present study suggests that the CDT is a useful measure for assessing cognitive function in older adults with schizophrenia. Because CDT is quick and easy to administer, it is a feasible tool for clinical settings.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES