Relation between remission status and attention in patients with schizophrenia

Authors

  • Motoyuki Fukumoto MD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Core Research for Evolutionary Science and Technology (CREST) of Japan Science and Technology Agency (JST), Saitama, Japan
    Search for more papers by this author
  • Ryota Hashimoto MD, PhD,

    Corresponding author
    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Molecular Research Center for Children's Mental Development, United Graduate School of Child Development, Osaka University, Kanazawa University, Hamamatsu University, Chiba University and Fukui University School of Medicine, Osaka, Japan
    3. Core Research for Evolutionary Science and Technology (CREST) of Japan Science and Technology Agency (JST), Saitama, Japan
    • Correspondence: Ryota Hashimoto, MD, PhD, Molecular Research Center for Children's Mental Development, United Graduate School of Child Development, Osaka University, Kanazawa University, Hamamatsu University, Chiba University and Fukui University School of Medicine, D3, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan. Email: hashimor@psy.med.osaka-u.ac.jp

    Search for more papers by this author
  • Kazutaka Ohi MD, PhD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Core Research for Evolutionary Science and Technology (CREST) of Japan Science and Technology Agency (JST), Saitama, Japan
    3. National Hospital Organization, Yamato Mental-Medical Center, Nara, Japan
    Search for more papers by this author
  • Yuka Yasuda MD, PhD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Core Research for Evolutionary Science and Technology (CREST) of Japan Science and Technology Agency (JST), Saitama, Japan
    Search for more papers by this author
  • Hidenaga Yamamori MD, PhD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Department of Molecular Neuropsychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    3. Core Research for Evolutionary Science and Technology (CREST) of Japan Science and Technology Agency (JST), Saitama, Japan
    Search for more papers by this author
  • Satomi Umeda-Yano PhD,

    1. Department of Molecular Neuropsychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    Search for more papers by this author
  • Masao Iwase MD, PhD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    Search for more papers by this author
  • Hiroaki Kazui MD, PhD,

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    Search for more papers by this author
  • Masatoshi Takeda MD, PhD

    1. Department of Psychiatry, Osaka University Graduate School of Medicine, Osaka, Japan
    2. Molecular Research Center for Children's Mental Development, United Graduate School of Child Development, Osaka University, Kanazawa University, Hamamatsu University, Chiba University and Fukui University School of Medicine, Osaka, Japan
    Search for more papers by this author

Abstract

Aim

Patients with schizophrenia in remission have shown significantly higher levels of neurocognitive function than patients not in remission. However, previous studies have mainly examined the association between neurocognitive function and the remission status of schizophrenia without considering the time component of the definition for remission using cross-sectional methods. The purpose of this study was to investigate the relations between remission status with considering time components and three cognitive functions of intellectual ability, memory and attention, which were examined before fulfilling the remission criteria, using longitudinal methods.

Methods

We assessed the remission status using the Positive and Negative Syndrome Scale (PANSS) on the same patients twice: at recruitment and at 6 months after the first PANSS assessment. Cognitive tests were performed within 3 months after recruitment. At recruitment, 337 patients were enrolled. Of the patients, 63 patients were followed up and completedthe first and second PANSS assessments and three cognitive tests at the end of study.

Results

Of the patients, 33 patients fulfilled the remission criteria, while 30 patients did not fulfill the criteria. Patients in remission showed significantly higher levels of 2-digit (P = 0.020) and 3-digit (P = 0.015) Continuous Performance Test scores, attention/concentration in the Wechsler Memory Scale-Revised (P = 0.034) and processing speeds in the Wechsler Adult Intelligence Scale-III (P = 0.047) than patients not in remission. Additionally, these cognitive scores were positively correlated with each other (P < 0.05).

Conclusion

Our findings suggest that patients who achieve remission may demonstrate a pre-existing higher level of attention than patients who do not achieve remission.

Schizophrenia is a psychiatric disorder that develops positive and negative symptoms as well as cognitive impairments, such as the dysfunction of intellectual ability, memory and attention. Although these neurocognitive impairments are often correlated with the positive and negative symptoms, some neurocognitive impairments are independent of these symptoms.[1] These neurocognitive impairments are considered to be a core component of schizophrenia, are increasingly under investigation as potential treatment targets,[2] and have been shown to exert a greater influence on social and vocational functioning than the presence or severity of positive and negative symptoms.[3]

Schizophrenia has been treated with neuroleptics since the 1950s and has been assessed with standardized tools for the treatment of this disease since the 1960s. However, no goals for treatment were established until the introduction of the remission concept.[4, 5] The definition of remission for schizophrenia proposed by US[4] and European groups[5] has become increasingly used to evaluate symptoms. To formalize this concept, the Positive and Negative Syndrome Scale (PANSS) was used.[6] In this concept, remission is determined according to a threshold of symptom severity (mild or less: ≤3) on eight items of the PANSS. These eight items include the following: P1, delusions; P2, conceptual disorganization; P3, hallucinatory behavior; N1, blunted affect; N4, social withdrawal; N6, lack of spontaneity; G5, mannerisms/posturing; and G9, unusual thought content. In addition to the severity component, 6 months was identified as the minimum period that a patient needs to sustain this low level of symptomatology (mild or less: ≤ 3) to be considered in remission.

To our knowledge, there have been four studies that have investigated the association between remission status and neurocognitive functions.[7-10] Of these four studies, three have reported a significant relation between remission status and neurocognitive functioning.[7-9] Helldin et al. found that patients in remission demonstrated high cognitive abilities in vigilance, working memory, long-term memory executive functioning, learning performance, visuomotor speed/efficacy, cognitive flexibility and premorbid functioning and that patients with greater cognitive abilities may have a higher likelihood of achieving remission.[7] Hofer et al. demonstrated that remitted patients show significantly higher levels of frontal lobe functioning, including verbal fluency, alertness and optical vigilance, as compared to non-remitted patients.[8] Yun da et al. showed that remitted patients with schizophrenia had a significantly higher level of executive function based on more rapid stop signal reaction times (SSRT) for the Stop Signal Test (SST) subtests of the Cambridge Neuropsychological Test Automated Battery (CANTAB) than non-remitted patients.[9] These studies considered remission status, according to the criteria proposed by Andreasen et al.,[4] as scores no higher than 3 on any of the eight items included in the PANSS. However, the remission status was evaluated retrospectively using cross-sectional methods in these studies and the factor of time was not considered; that is, the low level of symptomatology must sustain for more than 6 months. In addition, cognitive tests were performed after evaluating remission status in all previous studies. Keefe and Fenton suggested that the improvement of symptoms is not associated with the recovery of cognitive functioning.[11] We therefore hypothesized that cognitive deficits prior to fulfilling remission criteria may contribute to remission status and remitted patients would also show a superior performance profile on cognitive tests administered prior to fulfilling the remission criteria. This study was a longitudinal study conducted in patients with schizophrenia. Remission status was determined according to the severity and time components of the definition of remission. We investigated the potential relation between remission status and three cognitive functions measured before the patients fulfilled the definition of remission.

Methods

Subjects and assessments

A longitudinal study was conducted in patients with schizophrenia. All subjects were recruited from outpatients at the Osaka University Hospital. Written informed consent was obtained from all subjects after the procedures had been fully explained. The study was carried out in accordance with the World Medical Association's Declaration of Helsinki and was approved by the Research Ethics Committee of Osaka University. A total of 337 patients agreed to participate in the study from 1 January 2004 to 30 June 2011 (Fig. 1). The subjects were Japanese and all biologically unrelated. Each subject with schizophrenia had been diagnosed by at least two trained psychiatrists according to the criteria of the DSM-IV, based on the Structured Clinical Interview for DSM-IV. When the diagnoses of the two trained psychiatrists were discordant, these psychiatrists discussed the proper diagnosis. In cases where the diagnostic disputes were resolved and the patient was diagnosed as schizophrenic, we included the patient. When the diagnostic disputes were not resolved by discussion or the patient was not diagnosed as schizophrenic, we excluded the patient. Additionally, patients were excluded from this study if they had neurological or medical conditions that could have potentially affected their central nervous system, such as atypical headaches, head trauma with a loss of consciousness, chronic lung disease, kidney disease, chronic hepatic disease, thyroid disease, active cancer, cerebrovascular disease, epilepsy, seizures, substance-related disorders or mental retardation.

Figure 1.

Flowchart of patients with schizophrenia included in the current study. A total of 337 patients with schizophrenia were included in this study. The first Positive and Negative Syndrome Scale (PANSS) was assessed on the day the patient gave consent to participate in the study. Three cognitive functions, including IQ, memory and attention, were assessed within 3 months after the first PANSS assessment. The second PANSS assessment was performed 6 months after the first PANSS. When patient scores showed mild severity at both the first and second PANSS assessments, according to the definition of remission, we considered the patients as having achieved remission status. Thirty-three patients fulfilled the definition of remission, while 30 patients did not fulfill the definition of remission.

The first PANSS was assessed on the day the patient consented to participate in the study. At the first PANSS assessment, 147 subjects scored mild or less (≤3) on all eight PANSS items of remission (P1, P2, P3, N1, N4, N6, G5 and G9). In contrast, 190 subjects scored moderate or more (≥4) on any of the eight items. The PANSS scores were assessed by the same two psychiatrists who had diagnosed the schizophrenia.

The tests for cognitive function were selected from core domains known to be impaired in patients with schizophrenia.[2] Three cognitive functions, including IQ, memory and attention, were assessed within 3 months after the first PANSS assessment. Intellectual ability was evaluated using the Wechsler Adult Intelligence Scale – Third Edition (WAIS-III).[12] Memory was evaluated using the Wechsler Memory Scale – Revised (WMS-R).[13, 14] Attention was evaluated using a Continuous Performance Test (CPT), as described in previous studies.[15, 16] To estimate premorbid IQ, the Japanese Adult Reading Test (JART: a Japanese version of the National Adult Reading Test) was performed.[17] These cognitive functions were evaluated by trained clinical psychologists. Ninety-six subjects completed these tests within 3 months after the first PANSS assessment, while 51 subjects did not complete these tests. The average days and standard deviations of three cognitive tests (WAIS-III, WMS-R and CPT) performed within 3 months after the first PANSS assessment were 19.4 ± 20.0 days (range 0–71 days), 14.8 ± 15.5 days (range 0–56 days) and 17.0 ± 16.3 days (range 0–63 days), respectively.

Among those subjects who completed the cognitive tests, 63 subjects underwent a second PANSS assessment 6 months after the first PANSS assessment. Thirty-three patients fulfilled the remission criteria for schizophrenia according to the severity and time components proposed by Andreasen et al.[4] Thus, there were 33 patients with a condition of mild severity who continued to the second PANSS test 6 months later, and these patients were defined as the remission group. Thirty patients did not fulfill the definition of remission (Fig. 1). The remission group included all subjects who achieved remission status at any stage in the study, regardless of whether they were able to maintain this status for longer than the 6 months designated as the study end-point. The differences in characteristics between the remitted (n = 33) and non-remitted (n = 30) patients at the first and second PANSS assessments are shown in Tables 1 and 2, respectively.

Table 1. Demographic and clinical characteristics of patients with schizophrenia at the first PANSS assessment
 RemissionNo remissionP-values (z)
n = 33n = 30
  1. Means ± SD and P-values are shown.
  2. aχ2-test. Significant P-values are shown as bold face.
  3. CPZ-eq, chlorpromazine equivalent; GAF, global assessment of functioning; PANSS, Positive and Negative Syndrome Scale.
Age (years)37.3 ± 11.834.3 ± 13.80.10 (1.63)
Age at onset (years)26.9 ± 11.224.5 ± 11.40.32 (−0.99)
Sex (male/female)17/1615/150.90 (0.01)a
Education (years)14.6 ± 2.113.6 ± 3.00.06 (−1.87)
Duration of untreated psychosis (years)1.4 ± 2.12.2 ± 4.20.40 (0.85)
Duration of illness (years)10.4 ± 8.99.8 ± 7.70.89 (−0.15)
Typical antipsychotics, CPZ-eq (mg/day)66.9 ± 168.781.3 ± 169.80.66 (0.45)
Atypical antipsychotics, CPZ-eq (mg/day)322.0 ± 291.9470.8 ± 379.30.07 (1.82)
Premorbid IQ103.4 ± 9.2101.6 ± 12.20.57 (−0.57)
PANSS, Total score67.4 ± 15.572.9 ± 15.60.12 (1.55)
PANSS, Positive symptoms14.8 ± 3.016.1 ± 3.50.10 (1.63)
PANSS, Negative symptoms15.7 ± 3.417.6 ± 4.50.021 (2.30)
PANSS, General Psychopathology36.9 ± 10.839.2 ± 9.70.24 (1.12)
GAF50.5 ± 8.947.1 ± 5.00.21 (−1.26)
Table 2. Differences in the clinical symptoms of patients with schizophrenia at the second PANSS assessment
 RemissionNo remissionP-value (z)
  1. Means ± SD are shown. Significant P-values are shown as bold face.
  2. GAF, global assessment of functioning; PANSS, Positive and Negative Syndrome Scale.
PANSS, total score64.7 ± 11.682.0 ± 18.9<0.001 (3.96)
PANSS, positive symptoms14.2 ± 2.719.1 ± 3.8<0.001 (4.75)
PANSS, negative symptoms15.1 ± 2.922.5 ± 9.0<0.001 (5.60)
PANSS, general psychopathology35.4 ± 7.340.3 ± 11.90.12 (1.52)
GAF52.4 ± 7.336.1 ± 6.1<0.001 (−6.27)

Statistical analyses

Based on the assumption that demographic variables, such as age and education years, and cognitive variables might not be fitted to a normal distribution, we used the non-parametric analyses, Mann–Whitney U and Spearman's rank correlation tests, arbitrarily to assess these variables, respectively. The differences in clinical characteristics between the remission and no remission groups were analyzed with the χ2-test for categorical variables and the Mann–Whitney U-test for continuous variables at the first and second PANSS assessments, using pasw 18.0 (spss Japan, Tokyo, Japan). Cognitive functions are known to be affected by age, sex and years of education,[18] but the scores for the WAIS and WMS indices had been previously corrected for age. A significant difference in negative symptoms between the two groups was observed at the first PANSS assessment (P = 0.021, Table 1), and to control for these confounding factors, the differences between the remission and the no remission groups regarding IQ, memory and attention were analyzed by a one-way ancova with sex, years of education and negative symptoms for WAIS-III and WMS, and with age, sex, years of education and negative symptoms for CPT as covariates. To assess the relation among significant cognitive scores, we examined correlations among these cognitions using Spearman's rank correlation coefficient. All reported P-values were two-tailed. Statistical significance was defined as P < 0.05.

Results

Of the 337 patients with schizophrenia who were included in this study, 63 patients were followed up and completed the first and second PANSS assessments and three cognitive tests. Of these 63 patients, 33 patients fulfilled the definition of remission (52.4%), while 30 patients did not fulfill this definition (47.6%) (Fig. 1). At the first PANSS assessment, there were no significant differences between the remission and no remission groups regarding age, age at onset, sex, years of education, duration of untreated psychosis, duration of illness, dose of antipsychotic agent, estimated premorbid IQ, the global assessment of functioning (GAF) or PANSS scores, with the exception of those for negative symptoms (Table 1). At the second PANSS assessment, there were significant differences in the PANSS total, positive symptoms, negative symptoms and GAF scores between the groups but no significant differences in general psychopathology (Table 2). Patients who fulfilled the definition of remission had significantly lower scores of PANSS and higher scores of GAF than patients who did not fulfill the remission criteria.

We next investigated differences in IQ, memory and attention, which were assessed within 3 months after the first PANSS assessment, between the remitted and non-remitted patients (Table 3). The results indicated significant differences in the processing speeds of the WAIS-III (P = 0.047, F1,58 = 4.10), the attention/concentration of the WMS-R (P = 0.034, F1,58 = 4.73) and the 2-digit D′ (P = 0.020, F1,57 = 5.72) and 3-digit D′ (P = 0.015, F1,57 = 6.25) of the CPT between the remitted and non-remitted patients (Table 3); all of these scores in remitted patients were significantly higher than those in non-remitted patients. However, there was no significant difference in any other score between the groups (P > 0.05). Next, we examined the correlations among these significant scores and found significant positive correlations among all scores (processing speed and attention/concentration [r = 0.42, P = 5.90 × 10−4], D′2 [r = 0.36, P = 0.0033] and D′3 [r = 0.36, P = 0.0034], attention/concentration and D′3 [r = 0.27, P = 0.032], and D′2 and D′3 [r = 0.61, P = 1.31 × 10−7]), except for a correlation between attention/concentration and D′2 (r = 0.17, P = 0.20).

Table 3. Differences in neurocognition between the remitted and non-remitted patient groups
 RemissionNo remissionP-value (F)
  1. Means ± SD are shown. Significant P-values are shown as bold face.
  2. CPT, Continuous Performance Test; WAIS-III, Wechsler Adult Intelligence Scale – Third Edition; WMS-R, Wechsler Memory Scale-Revised.
WAIS-III   
Full scale IQ89.5 ± 16.482.7 ± 18.80.61 (0.26)
Verbal IQ94.2 ± 15.988.3 ± 17.50.70 (0.15)
Performance IQ85.7 ± 17.579.3 ± 18.60.59 (0.29)
Verbal Comprehension95.2 ± 14.093.3 ± 19.00.49 (0.48)
Perceptual Organization90.0 ± 18.183.7 ± 19.30.62 (0.25)
Working Memory92.7 ± 15.483.1 ± 16.20.07 (3.30)
Processing Speed83.0 ± 13.073.8 ± 13.00.047 (4.10)
WMS-R   
Verbal Memory87.6 ± 16.783.1 ± 19.00.57 (0.34)
Visual Memory90.4 ± 17.282.9 ± 15.80.05 (3.99)
General Memory86.2 ± 17.683.1 ± 17.00.33 (0.98)
Attention/Concentration98.0 ± 12.690.1 ± 14.30.034 (4.73)
Delayed Recall83.2 ± 16.081.0 ± 14.10.15 (2.12)
CPT   
Digit2 D3.6 ± 0.73.0 ± 0.90.020 (5.72)
Digit3 D3.0 ± 0.92.4 ± 1.00.015 (6.25)
Digit4 D1.6 ± 0.81.5 ± 0.90.51 (4.39)

Discussion

This longitudinal study was conducted in patients with schizophrenia to investigate a potential relation between remission status and three cognitive functions, which were measured before fulfilling the definition of remission. The remission status was determined according to the severity as well as the time components of the remission criteria. Of the 63 patients who completed the first and second PANSS assessments and cognitive tests, 33 patients fulfilled the remission criteria, whereas 30 patients did not fulfill these criteria. We investigated the differences in the cognitive functions between the remitted and non-remitted patients and found significant differences in the processing speed of the WAIS-III, the attention/concentration of the WMS-R and the 2-digit D′ and 3-digit D′ of the CPT, almost all of which were correlated with the other scores. Moreover, the scores of attention in remitted patients were significantly higher than those in non-remitted patients.

There have been four studies that have compared neurocognitive test results between remitted and non-remitted patients with schizophrenia.[7-10] These studies showed that remitted patients with schizophrenia demonstrated significantly higher performance in attention, executive functioning and IQ than non-remitted patients with schizophrenia. All previous studies conducted used a cross-sectional design and evaluated the neurocognitive tests after evaluating remission, although differences in neurocognitive functions may contribute to differences in clinical symptoms (PANSS and GAF scores) between groups at the time of the assessment of the neurocognitive functions. To reduce this possibility, we compared cognitive functions between the remission and no remission groups before they fulfilled the remission criteria. At the first PANSS assessment, there were no differences in the PANSS or GAF scores, with the exception of negative symptoms. We found that remitted patients with schizophrenia had a significantly higher performance related to attention than non-remitted patients with schizophrenia, although remitted patients did not exhibit a significantly higher performance of IQ, memory or executive functioning. These findings suggest that the impairment of attention in non-remitted patients may precede the appearance of other cognitive deficits and that attentional performance might be a possible predictor of remission in patients with schizophrenia.

Attentional impairments are one of the most important types of neurocognitive impairment in schizophrenia, as a number of studies have indicated that patients with schizophrenia have attentional impairments, as measured by the processing speed of the WAIS-III, the attention/concentration of the WMS-R or the D′ of the CPT.[19-21] Attentional functions can generally be divided into several functions, such as selective attention, sustained attention(vigilance and alertness) and divided attention.[22] The D′ of the CPT has commonly been used to measure the vigilance and alertness of attentional functions,[23] whereas processing speed in the WAIS, which comprises two subscales of digit symbols and symbol searches, has been used to measure sustained and selective attentions.[24] In addition, the attention/concentration index of the WMS-R, which comprises three subscales of mental control, digit span and visual memory span, has been used to measure complex attentional function, including working memory.[14] We showed significant correlations among these three cognitive tests. Of these tests, the most significantly associated attentional impairment with remission status was vigilance/alertness, as measured by the CPT, and this vigilance/alertness measurement of the CPT is a simpler and more attentional domain than the processing speed of the WAIS and the attention/concentration factor of the WMS. Thus, cognitive impairments related to remission status may be caused by deficits in the vigilance/alertness of the attentional functions.

However, there are several limitations to the interpretation of our results. First, a number of patients (274/337; 81.3%) did not complete the first and second PANSS assessments and the three cognitive tests. Because this was a longitudinal study, it was difficult to follow patients and ensure their completion of these tests. Although we compared only patients (63/337; 18.7%) who completed these tests, those patients who did not complete the tests may have demonstrated alternative factors involved in remission status. Thus, an examination of these factors in the patients who withdrew from our study may be instructive. Second, several demographic variables, such as age, sex and years of education as well as negative symptoms and taking medication, may have influenced our outcomes. During this longitudinal study, we did not change taking medication of patients from first PANSS to second PANSS assessments. There was no correlation between first PANSS scores (positive or negative symptoms) and CPZ-eq of typical and/or atypical antipsychotics (P > 0.05). To control for these confounding factors, we additionally performed ancova with the listed confounding factors, including CPZ-eq of typical and/or atypical antipsychotics as covariates. The differences of D′2 (typical; P = 0.021, F1,56 = 5.61, atypical: P = 0.016, F1,56 = 6.21, both; P = 0.017, F1,56 = 6.00) and D′3 (typical; P = 0.017, F1,56 = 6.11, atypical: P = 0.012, F1,56 = 6.76, both; P = 0.013, F1,56 = 6.62) between the remission and the no remission groups remained significant after controlling the CPZ-eq. However, differences of the processing speeds and the attention/concentration between the groups did not remain significant after controlling the CPZ-eq (P > 0.05). Our findings might be affected by antipsychotics. In addition, these cognitive differences might have been changed at the end of study although we have administered cognitive tests prior to fulfilling the remission criteria. Third, we did not apply the Bonferroni correction for the three neurocognitive tests. Therefore, our results likely include some false positives. As a number of statistical analyses, including the WAIS-III (seven tests), the WMS-R (five tests) and the CPT (three tests), were performed, a correction for multiple testing should be considered. However, we showed correlations among these tests, which indicated that they were not independent. Furthermore, as our results were based on a relatively small number of patients with and without remission, a future replication study using larger sample sizes is needed to confirm our findings. In conclusion, we found that preceding attentional impairment is related to schizophrenia remission status.

Acknowledgments

We thank all of the individuals who participated in this study. This work was supported by research grants from the Japanese Ministry of Health, Labor and Welfare (H22-seishin-ippan-001); the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) KAKENHI (22390225-Grant-in-Aid for Scientific Research (B) and 23659565-Grant-in-Aid for Challenging Exploratory Research); the CREST of JST; and the Japan Foundation for Neuroscience and Mental Health. We declare no conflicts of interest that may be inherent to this study.

Ancillary