A Comparative Analysis of Neurocognitive Function in Community- and Hospital-Based Patients With Schizophrenia
Article information
Abstract
Objective
The study compared differences in various neurocognitive characteristics across treatment modalities among schizophrenia patients recruited in one city between July 2020 and June 2023 who were on regular medication.
Methods
Of the 151 participants, 113 were included in the analysis. Participants were divided into community-based and hospital-based groups, and their demographics and clinical characteristics, including insight, quality of life, positive and negative symptoms, and personal and social functioning, were examined. In addition, several aspects of cognition were assessed using neurocognitive assessments such as the Trail Making Test (TMT), Stroop test, and Wisconsin Card Sorting Test (WCST). After adjusting for age differences between groups, the final analysis included data from 42 participants in the community-based group and 33 participants in the hospital-based group.
Results
Hospital-based group participants completed Stroop-W more rapidly, excelled in Rey-Osterrieth Complex Figure Test recall and recognition, and incurred fewer TMT Part B (TMT-B) and Stroop-C errors. Additionally, they outperformed in WCST total, non-persistent errors, and categories completed. Contrastingly, community-based group participants showed superior outcomes in WCST persistent responses and errors, suggesting specific neurocognitive strengths.
Conclusion
We found differences in neurocognitive characteristics between the two groups. These differences were consistent across a range of cognitive domains, including attention, visual discrimination, memory, and executive functioning. Further large-scale study is needed to generalize cognitive characteristics across treatment modalities.
INTRODUCTION
Schizophrenia is a mental illness that usually occurs in adolescence and early adulthood and can last a lifetime. The longer treatment intervention is delayed after the onset of schizophrenia, the greater the individual suffering and societal costs, and the poorer the prognosis [1,2]. Schizophrenia is diagnosed when there is a marked decline in social functioning based on positive symptoms such as delusions and hallucinations [3]. In addition, it is difficult to ignore the fact that a wide range of cognitive deficits are found before and after the onset of schizophrenia [4]. Understanding the cognitive features of schizophrenia is crucial, given that the negative symptoms and cognitive problems of schizophrenia patients lead to problems with interpersonal relationships, work, and adjustment in the real world [5].
Cognitive function refers to a wide range of human mental processes and can be categorized into neurocognitive and social cognitive functions [6]. Neurocognitive functions are represented by intelligence and include psychomotor speed, attention, memory, inhibition, executive function, etc [7]. Social cognition is a set of cognitive activities that perceive, understand, and process social information. It includes concepts such as theory of mind, emotional processing, social cognition and perception, and attribution. Currently, human cognition is understood to be an interplay of neurocognitive and social cognition that affects an individual’s life, and cognitive deficits in schizophrenia are strongly linked to occupational and functional impairment. In this study, we aimed to explore the neurocognitive functioning of schizophrenia patients [6].
People with schizophrenia have various life trajectories due to the lifelong nature of the disease. The course of schizophrenia and treatment outcomes are known to be variable and difficult to explain by a single predictor [2]. While some patients recover completely, the majority of people diagnosed with schizophrenia require formal or informal daily support [3]. As a result, patients choose different forms of treatment, including outpatient treatment, inpatient treatment, day hospitals, and psychiatric rehabilitation facilities [8]. These differences will be determined by a number of factors, including the patient’s daily lifestyle, access to care, and cost. Hospital-based treatment is typically intensive and provides specialized care. However, it can be expensive and inaccessible to all patients. Conversely, community-based care can help patients maintain social functioning because they are treated while maintaining as much of their daily routine as possible. It also has the advantage of providing comprehensive support, including housing, employment, and social services in addition to mental health support [9,10].
Numerous studies have previously reported that treatment course, medication adherence [11], or duration of illness [12] may contribute to clinical differences and differences in neurocognitive functioning. However, differences in objective cognitive functioning across schizophrenia treatment modalities have rarely been systematically studied. We predicted that differences in clinical and neurocognitive characteristics of individuals across treatment modalities would be observed in patients with schizophrenia who were continuously managed with pharmacotherapy, and assessed the functioning of various clinical assessments and domains of neurocognitive functioning.
METHODS
Participants
From July 2020 to June 2023, patients with schizophrenia registered at community mental health centers, social rehabilitation facilities, a university hospital, and a local psychiatric clinic in Incheon, South Korea, were recruited. Participants were recruited for a study of behavioral activation treatment for improving negative symptoms in the schizophrenia population. Both groups of patients were outpatients who continued to receive regular monitoring and medication prescriptions without being hospitalized. A total of 151 subjects were recruited, of which 2 refused to participate in the study and 36 were excluded based on the study inclusion and exclusion criteria, and the remaining 113 were assigned to the study analysis (Figure 1).
Flow diagram of participants’ recruitment.
The inclusion criteria were as follows: 1) adults aged 18 to 65 years, 2) with a primary diagnosis of schizophrenia, and 3) stable on prescribed antipsychotic medication for at least 6 months. Exclusion criteria were 1) a history of organic brain syndrome, traumatic brain injury, or epilepsy, 2) mental retardation, 3) comorbid disorders such as current alcohol or other drug dependence, 4) a score of 5 or higher on at least two of the Positive and Negative Syndrome Scale (PANSS) positive symptom items, 5) a diagnosis of major depressive disorder, and 6) a risk of self-harm or harm to others. Of these, participants were categorized into two groups: hospital-based and community-based. Both groups were receiving regular medication, but participants in the community-based group were receiving additional services provided by the psychiatric rehabilitation facility, such as individual or group counseling, vocational rehabilitation programs, life skills training, and family support programs.
Clinical assessments were conducted at seven community mental health centers, a university hospital, and a psychiatric clinic in Incheon, South Korea. Most of the study participants were assessed at the psychiatric clinic and the hospital. For those who had difficulty accessing healthcare, evaluators visited each center. The assessment included a diagnostic interview with a psychiatrist (approximately 30 minutes), a psychiatric interview assessment that included levels of psychosocial functioning (approximately 50 minutes), neurocognitive testing (approximately 60 minutes), and a self-report survey assessing depression, anxiety, and insight (approximately 20 minutes). The assessments were administered under the supervision of one psychiatrist and one clinical psychologist.
All participants were fully informed about the study and provided written informed consent prior to participation. This study protocol was approved by the Institutional Review Board (INHA 2021-01-026).
Measures
Clinical scales
Korean version of the Revised Insight Scale for Psychosis
The Insight Scale for Psychosis (ISP) is a self-report measure of multifaceted insight into psychosis developed by Marková et al. [13] Participants respond yes or no to 30 items, with higher scores indicating greater insight. In this study, we used the standardized Korean version of the Revised Insight Scale for Psychosis (KISP) [14].
Schizophrenia Quality of Life Scale
The Schizophrenia Quality of Life Scale (SQLS) is a tool to assess subjective quality of life in people with schizophrenia. It is a self-report measure that consists of disease-specific items that reflect the daily life of a person with schizophrenia. There are 33 items in total, and a higher score indicates a lower quality of life. In this study, we used the standardized Korean version of the SQLS [15].
PANSS
The PANSS is a clinician-rated scale developed to comprehensively assess the severity of psychotic symptoms [16]. It consists of 30 questions (7 questions each for positive and negative symptoms and 16 questions for general psychopathology) and is rated on a 7-point Likert scale according to the severity of symptoms. In this study, we used the Korean version of the PANSS, which was standardized for Korean schizophrenia patients [17].
Clinical Assessment for Negative Symptoms
The Clinical Assessment for Negative Symptoms (CAINS) was developed to assess the severity of negative symptoms in patients with schizophrenia by overcoming the limitations of existing scales and reflecting new trends in negative symptoms [18]. It consists of a total of 13 items, and the subscales consist of 9 items for ‘motivation and pleasure’ and 4 items for ‘expression’. This study used the Korean version of the CAINS adapted by Jung et al. [19].
Personal and Social Performance scale
The Personal and Social Performance scale (PSP) was developed to assess the level of daily social functioning of patients, compensating for the shortcomings of existing scales which did not account for functional status unrelated to psychological stress [20]. It evaluates socially valuable behaviors including occupational activities and learning, interpersonal and social relationships, self-care, and the absence of disruptive or aggressive behaviors. These domains are divided into four subscales, A, B, C, and D, each rated on a 6-point scale. Subsequently, the evaluator synthesizes the degree of impairment in each domain to score a final result between 1 and 100, with higher scores indicating better functioning. This study employed the standardized Korean version of the Personal and Social Performance scale (K-PSP) [21].
Neurocognitive assessments
Trail Making Test
The Trail Making Test (TMT) is included in the Halstead-Reitan Neuropsychological Battery and is comprised of two parts, Part A and Part B, each assessing different cognitive functions [22]. Part A (TMT-A) measures sustained attention and psychomotor speed, while Part B (TMT-B) is used to evaluate executive functions associated with the frontal lobe, specifically mental shifting [22]. In this study, TMT-A was utilized to assess attention and working memory, and TMT-B served as an index for the evaluation of executive functions.
Stroop test
The Stroop test assesses one aspect of frontal function, which involves the suppression of habitual responses (e.g., reading words) while selectively attending to stimuli required for the task, specifically inhibition and selective attention [23]. In this study, the word reading condition (Stroop-W) was used to assess attention, and the color-word reading condition (Stroop-C) was used to evaluate frontal executive functions. Raw scores for correct responses and errors for each condition were included in the data analysis.
Rey-Osterrieth Complex Figure Test
The Rey-Osterrieth Complex Figure Test (RCFT) is designed to evaluate visuospatial constructional ability and visual memory [24]. The test is administered in the sequence of copy, immediately recall, delayed recall, and recognition. The copy phase assesses visuoperception, spatial organizational ability, and the capacity for organization. The immediately recall phase evaluates the encoding ability of memory, and delayed recall assesses the retention and retrieval abilities of memory. This study utilized the original scoring criteria proposed by Meyers and Meyers [24] for the assessment and scoring of each phase, and these raw scores were incorporated into the data analysis.
Controlled Oral Word Association test
The Controlled Oral Word Association test is a measure of verbal fluency that includes subtests for semantic or category fluency and phonemic or letter fluency [25]. Phonemic fluency is more closely associated with frontal lobe function [26], while semantic fluency outcomes are known to be related to temporal lobe function [27].
Wisconsin Card Sorting Test
The Wisconsin Card Sorting Test (WCST) serves as a representative tool for measuring frontal lobe function, assessing complex abilities such as the acquisition of abstract information for concept formation, mental set shifting, and perseveration. In this study, the computerized version of the WCST, as outlined by Heaton et al. [28], was administered, and the raw scores of each item were utilized for data analysis.
Statistical analyses
All data collected in this study were analyzed using IBM SPSS 23.0 (IBM Corp., Armonk, NY, USA). Inter-rater reliability and internal consistency of self-report measures were analyzed using Cronbach’s α. Descriptive statistics were used to determine the demographic characteristics of the study population.
To analyze differences between groups, independent t-tests were used for continuous variables and chi-square tests for categorical variables. Analysis of covariance (ANCOVA) was performed to analyze differences in the results of each cognitive function test. Ranked ANCOVA was used for variables that did not meet the assumption of equal variance. For the analysis of the 20–40 age group, we conducted an independent t-test without covariates. This decision was based on the lack of correlation between sex and cognitive test results found in the Pearson correlation analysis. Previous studies have also shown no significant difference in cognitive function between sex in the schizophrenia population [29].
Two-tailed analyses were performed, and p-values less than 0.05 were considered statistically significant.
RESULTS
Demographic characteristics
Table 1 shows the basic characteristics of the study participants. The mean age of the community-based patient group was 39.00 years (standard deviation [SD]=10.90), which was significantly lower than the mean age of the hospital-based patient group (mean=44.90 years, SD=11.44). In terms of sex, the community-based patients were 34 males (64.2%) and 19 females (35.8%), which was significantly different from the hospital-based patients (21 males: 35.0%, 39 females: 65.0%). In terms of marital status, both groups were predominantly single (49 community-based patients: 92.5%, 30 hospital-based patients: 50.0%), but the difference was statistically significant (Table 1).
Demographic characteristics in all participants (N=113)
It was considered that the age difference between the two groups might affect clinical variables, including cognitive function. Therefore, re-analysis was done taking into account the age variable, and the results of 42 community-based patients and 33 hospital-based patients were included in the analysis. No significant differences were observed between the two groups, except that the proportion of single marital status was 40 (95.2%) in the community-based patient group and 23 (69.7%) in the hospital-based patient group (Table 2).
Demographic characteristics in 20–40 years (N=75)
Clinical characteristics
Table 3 shows a comparison of clinical features between the two groups. Patients in the community-based group reported a significantly higher quality of life score with a SQLS of 48.76 (SD=20.46) than patients in the hospital-based group (37.14, SD=23.16). On the PANSS, the community-based group had a positive symptom scale of 13.62 (SD=4.24), which was significantly higher than the hospital-based group (11.39, SD=2.98). However, no significant differences were observed between the two groups on the PANSS total score, negative symptom score, or general psychopathology score. The total score of K-PSP was significantly lower in the community-based group at 46.90 (SD=12.39) compared to the hospital-based group (54.24, SD=13.00). However, on the K-PSP personal and social relationships (B) subscale, patients in the community-based group scored significantly higher at 4.17 (SD=0.76) compared to the hospital-based group (3.70, SD=0.77) (Table 3).
Comparison of clinical data in 20–40 years (N=75)
Neurocognitive assessment
Table 4 compares neurocognitive functions between the two groups. On the Stroop-W, there was a significant difference in time to complete the test, with the hospital-based group being faster. On the RCFT, the hospital-based group performed better on the immediate recall and recognition subscales. On the TMT-B, significant differences were found in the number of errors, with fewer errors in the hospital-based group. Significant difference in errors with the hospital-based group making fewer errors on the Stroop-C. The statistical significance observed for the total number of errors, non-persistent errors, and number of categories completed on the WCST indicates that the hospital-based group performed better. However, on the persistent responses and persistent errors subscales of the WCST, the scores for the community-based group were significantly lower, indicating higher performance (Table 4).
Comparison of neurocognitive assessments in 20–40 years (N=75)
DISCUSSION
It is well established that negative symptoms and cognitive functioning are more closely related to the course and prognosis of schizophrenia than positive symptoms [30-32]. Patients with schizophrenia show a complex cognitive decline [33-35], which is also associated with prognosis [33]. However, compared to positive symptoms, cognitive functioning is difficult to measure, and there are limitations to the use of uniform measurement tools. At the “Deconstructing Psychosis,” [36] there was discussion of including cognitive functioning in the diagnostic criteria for schizophrenia, as well as the Measurement and Treatment Research to Improve Cognition in Schizophrenia [37] initiative to systematically address cognitive functioning and negative symptoms. Ultimately, these discussions were not reflected in the current Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, but understanding cognitive functioning in people with schizophrenia will play an important role in predicting course and prognosis and improving the quality of life for patients.
We wanted to determine whether there were specific differences between patients using relatively accessible health facilities in the community (community-based group) and those using tertiary hospitals (hospital-based group). However, it was confirmed that there was no significant difference in educational attainment or premorbid intellectual level other than age or marital status. Therefore, we wanted to standardize age and examine whether there were differences in neurocognitive function characteristics within each group.
In our study, the community-based patient group rated their quality of life lower than the hospital-based patient group. The community-based group was observed to have more positive symptoms and lower functional status in daily living as assessed by the K-PSP compared to the hospital-based group. In a previous study of inpatient and outpatient schizophrenia patients in Portugal [38], the authors found that, although outpatients showed higher PSP total scores as compared with inpatients, these differences were not statistically significant. In our study, the subscale personal and social relationships (B) of the K-PSP was higher in the community-based patient group, but this was contrary to the total score, which is a composite assessment of the degree of impairment in the four subscales. This suggests that community-based group patients may score higher on interpersonal and social relationships because they live in a psychiatric residential facility or have more frequent exposure to group therapy, but this does not reflect their level of personal functioning. On the PANSS and CAINS scales, which measure the severity of illness-specific symptoms of schizophrenia, there were no statistically significant differences between the groups except for the PANSS positive symptom scale. The community-based group scored higher on the PANSS positive symptom scale than the hospital-based group. A study by Okasha et al. [38] compared clinical characteristics and neurocognitive function between long-term inpatient and outpatient groups for schizophrenia, and found that PANSS positive symptoms and general symptoms were significantly correlated with cognitive function. In our study, no statistical significance was observed between the two groups in the PANSS total score, negative symptom score, and general psychopathology score, but the community-based group showed significantly higher scores in PANSS positive symptoms. There is a possibility that the difference in the level of positive symptoms between the two groups affected neurocognitive function, but different results were also confirmed for each item of neurocognitive assessments. Our results are partly consistent with previous research, but partly inconsistent with them.
There were no significant differences in negative symptoms between the two groups on either the PANSS or CAINS scales. It has been previously reported that schizophrenia patients with severe negative symptoms may perform worse on cognitive tasks, particularly executive function, memory, and attention [5,39]. The authors speculate that this may be because voice symptoms affect the motivation and attention required to perform cognitive tasks. However, statistical significance was found on neurocognitive assessments despite the fact that the two groups compared in our study did not differ in levels of negative symptoms. The hospital-based group was rated as more alert on the Stroop-W time items compared to the community-based group. This is related to attention and information processing speed, suggesting that patients in the hospital-based group are better able to process and suppress cognitive interference than those in the community-based group. Previous studies have also shown increased Stroop interference in people with schizophrenia [40]. The Stroop interference effect refers to the relative increase in reaction time observed when word meaning and stimulus color are incongruent, and increased Stroop interference predicts difficulty processing everyday behavioral information and deficits in behavioral control. However, the results are not consistent across studies, as some studies have shown that the Stroop effect is not increased in people with schizophrenia [41]. In addition, the immediate recall and recognition items of the RCFT were significantly higher in the hospital-based group. The immediate recall item of the RCFT estimates an individual’s visual memory and visual organization ability, and the recognition item assesses visual discrimination and attention by evaluating how accurately the individual recognizes and distinguishes shapes. It can be interpreted that patients in the hospital-based group had better attention span compared to the community-based group, which is consistent with the results of the Stroop-W. The TMT-B and Stroop-C, which were administered to assess executive function, also showed better functioning in the hospital-based group on some items. The error items on the TMT-B and the error items on the Stroop-C reflect deficits in attention and executive functioning, suggesting that the community-based group is less able to concentrate and adapt effectively to new information or situations in daily life. On the WCST, another tool used to assess executive function, the hospital-based group performed better on the total number of errors, non-perseverative errors, and number of categories completed items. However, patients in the community-based group performed better on perseverative responses and perseverative errors. In a previous study assessing cognitive characteristics of patients with schizophrenia using the WCST [42], patients with schizophrenia succeeded on fewer categories, committed more perseverative errors, made more perseverative responses, and gave significantly lower conceptual level responses than the controls. However, on retest, patients with schizophrenia committed significantly fewer perseverative errors. The authors interpreted these results as suggesting that even patients with schizophrenia who have reduced frontal lobe function can partially overcome this deficit with verbal reinforcement. In our study, the community-based group was relatively consistent with these previous findings.
This is contrary to our overall findings, as the various cognitive rehabilitation services that individuals receive in community programs aim to improve attention, memory, and executive function and are known to be effective in improving cognitive function [43,44]. However, in a quantitative analysis of cognitive rehabilitation programs for schizophrenia, Krabbendam and Aleman [45] found no statistically significant effectiveness. The community-based patients in our study were more likely to be living in a psychiatric rehabilitation residence for longer periods of time and to have more time devoted to education about their mental illness. In contrast, hospital-based patients were likely to have more time spent on more routine activities outside of regular hospital care. Therefore, it is possible that the experience of more realistic forms of daily living adaptation may have been a stimulus for maintaining attention and executive function. Conversely, however, it is also possible that individuals with greater ability to adapt to daily living may have opted for regular outpatient visits for simple medication rather than psychiatric rehabilitation. Davidson et al.’s study [46] suggests that cognitive impairment is a cause of long-term hospitalization, rather than the notion that cognitive impairment itself is a secondary consequence of long-term hospitalization.
The strength of this study is that it assessed different cognitive characteristics between groups of people with schizophrenia across treatment modalities using a variety of objective tools. It is important to understand the environmental influence of treatment modality on cognitive functioning in people with schizophrenia as it directly affects their daily functioning, independence, and quality of life. It is important to note that our study was age-adjusted to identify treatment modality-specific cognitive characteristics in younger patients with schizophrenia, in their 20s and 30s, given that previous studies [47,48] have observed greater impairments in cognitive functioning, especially abstract thinking, as patients with schizophrenia age. The 20s and 30s are the prime years for the onset of schizophrenia, and early intervention based on an understanding of cognitive functioning during this period may contribute to long-term improvements in patients’ symptoms and social and occupational functioning.
This study has several limitations. First, we did not ensure homogeneity between groups. We categorized the groups into those receiving medication through outpatient treatment and those receiving other ancillary psychiatric rehabilitation services, but even within the community-based group, there was a mix of treatment modalities, including residential psychiatric rehabilitation, day hospitals, and case management with local public services. This study was also limited by the small number of participants, so future studies should analyze the treatment type in detail to observe more clear group characteristics. Second, because the subjects were recruited from patients who visited one university hospital and one local psychiatric clinic in one region of Korea and were prescribed medication, the study population was small and cannot reflect the characteristics of the entire schizophrenia patient group. Therefore, future studies should be conducted with larger populations in different regions to see if the same results are replicated. Third, some patients failed to perform the TMT-B screening tool, and subjects who failed to perform the test were excluded from the analysis. There were two dropouts in the hospital-based group, compared to eight dropouts in the community-based group. It is possible that subjects with relatively high cognitive deficits were excluded from the assessment, which is a limitation of the study that makes it difficult to characterize the overall schizophrenia population. Lastly, recent studies have reported associations between insight level and neurocognitive deficits [49,50]. In our study, we found no significant difference in the level of insight between the two groups, which differed in a range of neurocognitive functions. This can be attributed to the limitation that the KISP scale relies on self-report questionnaires, which can be influenced by the patient’s subjective interpretation and current mental state.
Despite these limitations, our study is meaningful because it compares differences in cognitive characteristics across treatment modalities in young schizophrenia patients using a variety of neurocognitive measures. Future large-scale studies examining neurocognitive characteristics by treatment modality and analyzing their correlations will expand our understanding of cognitive characteristics of schizophrenia and contribute to the improvement of social and daily functioning in people with schizophrenia.
Notes
Availability of Data and Material
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
Conflicts of Interest
The authors have no potential conflicts of interest to disclose.
Author Contributions
Conceptualization: all authors. Data curation: Seri Maeng, Hee Seon Kim, Tae Joo Lee, Won-Hyoung Kim. Formal analysis: all authors. Funding acquisition: Tae Joo Lee, Hoon Jung Koo, Won-Hyoung Kim. Investigation: all authors. Methodology: Seri Maeng, Hee Seon Kim, Won-Hyoung Kim. Project administration: all authors. Resources: Seri Maeng, Tae Joo Lee, Won-Hyoung Kim. Software: Hee Seon Kim, Tae Joo Lee. Supervision: Tae Joo Lee, Won-Hyoung Kim. Validation: Tae Joo Lee, Hoon Jung Koo, Won-Hyoung Kim. Visualization: Seri Maeng, Hee Seon Kim, Won- Hyoung Kim. Writing—original draft: Seri Maeng, Hee Seon Kim, Won- Hyoung Kim. Writing—review & editing: Seri Maeng, Won-Hyoung Kim.
Funding Statement
This work was supported by the National Research Foundation of Korea (NRF) granted funded by the Ministry of Science, ICT & Future Planning (Grant number: NRF 2021R1G1A101146313). This work was supported by an INHA University Research Grant.
Acknowledgements
None