Structure and stability of symptoms in first episode psychosis: a longitudinal network approach

The longitudinal structure of negative symptoms in treatment resistant schizophrenia

BACKGROUND AND HYPOTHESIS

The negative symptoms of schizophrenia are strong prognostic factors but remain poorly understood and treated. Five negative symptom domains are frequently clustered into the motivation and pleasure (MAP) and emotional expression (EE) 'dimensions', but whether this structure remains stable and behaves as a single entity or not remains unclear.

STUDY DESIGN

We examined a cohort of 153 patients taking clozapine for treatment-resistant schizophrenia in a regional mental health clinic. Patients were assessed longitudinally over a mean period of 45 months using validated scales for positive, negative and mood symptoms. Network analyses were performed to identify symptom 'communities' and their stability over time. The influence of common causes of secondary negative symptoms as well as centrality measures were also examined.

STUDY RESULTS

Across patients at baseline, two distinct communities matching the clinical domains of MAP and EE were found. These communities remained highly stable and independent over time. The communities remained stabled when considering psychosis, depression, and sedation severity, and these causes of secondary negative symptoms were clustered into the MAP community. Centrality measures also remained stable over time, with similar centrality measures across symptoms.

CONCLUSIONS

Our results suggest that MAP and EE are independent dimensions that remain highly stable over time in chronic schizophrenia patients treated with clozapine. Common causes of secondary negative symptoms mapped onto the MAP dimension. Our results emphasise the need for clinical trials to address either MAP or EE, and that treating causes of secondary negative symptoms may improve MAP.

Structure and replicability of oral health-related quality of life networks across patients with schizophrenia and the general community

OBJECTIVES

Schizophrenia is a disabling mental disorder associated with severe social dysfunction. Individuals with long-term mental conditions have poorer Oral Health-Related Quality of Life (OHRQoL) compared to the general population, but little is known about the measurement properties of OHRQoL instruments in this group of patients. This study aimed to examine the replicability of OHRQoL networks across samples of the general community (GC) and patients with schizophrenia (PWS).

METHODS

Data were obtained from 603 community-dwelling participants and 627 patients with schizophrenia. OHRQoL was measured using the short form of the Oral Health Impact Profile (OHIP-14) questionnaire. A regularized partial correlation network was estimated for each sample. The number of dimensions and structural stability were assessed using Exploratory Graph Analysis. Global strength, edge weights and centrality estimates were compared. Network replicability was examined fitting the PWS data to the GC network structure.

RESULTS

A single OHIP-14 dimension was identified in the GC sample, whereas three dimensions were detected in the PWS sample. Structural consistency was perfect in the network of GC participants (1), and considerably low in at least two dimensions of the PWS network (0.28; 0.65; 0.16). A moderate correlation for node strength estimates was observed (τ: 0.43; 95% CI: 0.13, 0.72), although edge weights were not correlated (τ: 0.025; 95% CI: -0.11, 0.16). The fit of the PWS data to the GC network structure was deemed unacceptable.

CONCLUSION

Network models of OHRQoL did not replicate across samples of the general community and outpatients with schizophrenia. Prudent use of OHIP-14 to compare measures of OHRQoL between groups with significant cognitive impartment and the general population is recommended.

The structure stability of negative symptoms: longitudinal network analysis of the Brief Negative Symptom Scale in people with schizophrenia

BACKGROUND

The structure of negative symptoms of schizophrenia is still a matter of controversy. Although a two-dimensional model (comprising the expressive deficit dimension and the motivation and pleasure dimension) has gained a large consensus, it has been questioned by recent investigations.

AIMS

To investigate the latent structure of negative symptoms and its stability over time in people with schizophrenia using network analysis.

METHOD

Negative symptoms were assessed in 612 people with schizophrenia using the Brief Negative Symptom Scale (BNSS) at baseline and at 4-year follow-up. A network invariance analysis was conducted to investigate changes in the network structure and strength of connections between the two time points.

RESULTS

The network analysis carried out at baseline and follow-up, supported by community detection analysis, indicated that the BNSS's items aggregate to form four or five distinct domains (avolition/asociality, anhedonia, blunted affect and alogia). The network invariance test indicated that the network structure remained unchanged over time (network invariance test score 0.13; P = 0.169), although its overall strength decreased (6.28 at baseline, 5.79 at follow-up; global strength invariance test score 0.48; P = 0.016).

CONCLUSIONS

The results lend support to a four- or five-factor model of negative symptoms and indicate overall stability over time. These data have implications for the study of pathophysiological mechanisms and the development of targeted treatments for negative symptoms.

Inter-relationships among psychopathology, cognition, and real-life functioning in early and late phase schizophrenia: A network analysis approach

Many illness-related factors contribute to the reduction of the real-life functioning observed in people with schizophrenia (SZ). These include the psychopathological dimensions of the disorder such as positive, negative, disorganization, and depressive symptoms as well as impairment in neurocognition, social cognition, and metacognition. The associations between some of these variables change with the duration of illness (DOI), but this aspect was not explored with a network approach. This study aimed at describing and comparing the inter-relationships between psychopathological, cognitive, and functioning variables in early (DOI ≤ 5 years) and late (DOI > 5 years) phase SZ with network analyses and at assessing which variables were more strictly and directly associated with the real-life functioning. A network representation of the relationships between variables and the calculation of centrality indices were performed within each group. The two groups were compared with a network comparison test. Seventy-five patients with early and ninety-two with late phase SZ were included. No differences in the global network structure and strength were found between the two groups. In both groups, visual learning and disorganization exhibited high centrality indices and disorganization, negative symptoms, and metacognition were directly and strongly associated with real-life functioning. In conclusion, regardless of the DOI, a rehabilitation aimed at improving visual learning and disorganization (i.e., the most central variables) might reduce the strength of the associations that compose the network and therefore indirectly facilitate functional recovery. Simultaneously, therapeutic interventions targeting disorganization and metacognition might directly improve real-life functioning.

First episode psychosis and beyond: much progress made but much more needed

First episode psychosis remains one of the most critical research areas in psychiatry. Much progress has been made, but more progress is required to translate the ideas and promises into reality. In this Editorial, we provide the context and invite contributions for our BMC Psychiatry Collection on First Episode Psychosis.

Psychotic-like experiences as a co-occurring psychopathological indicator of multi-dimensional affective symptoms: Findings from a cross-sectional survey among college students

BACKGROUND

Although connections between psychotic-like experiences (PLEs) and a series of non-psychotic disorders have been widely explored in previous research, it is unclear whether PLEs could act as a co-occurring psychopathological indicator of multi-dimensional affective symptoms.

METHODS

A total of 4761 students took part in an online survey which assessed the frequency of PLEs and three types of affective symptoms over lifetime. Binary logistic regression models were used to examine associations between PLEs and each type of affective symptom. Network analysis was conducted to explore the relationship among three subtypes of PLEs - persecutory ideation (PI), bizarre experiences (BEs) and perceptual abnormalities (PAs), and different types of affective symptoms.

RESULTS

The results showed that compared with the non-PLEs group, the PLEs group suffered significantly higher risk of experiencing three types of affective symptoms, including manic symptoms [adjusted odds ratio (aOR) 11.50, 95 % confidence interval (CI) (4.59-28.81)], depressive symptoms [aOR 7.28, 95 % CI (4.98-10.66)] and anxiety symptoms [aOR 4.47, 95 % CI (3.10-6.43)]. In the network model, bizarre experiences were the most critical central symptom. Both depressive and anxiety symptoms were most strongly associated with persecutory ideation while manic symptoms were most closely related to bizarre experiences.

LIMITATIONS

Cross-sectional data and self-reported symptoms.

CONCLUSIONS

These findings suggest that PLEs are a vital co-occurring indicator of multi-dimensional affective symptoms and show its enormous potential as a target for a host of mental health problems. Further investigation may shed light on the aetiology of the relationship between different subtypes of PLEs and affective symptoms.

Depressive Features in Individuals with First Episode Psychosis: Psychopathological and Treatment Considerations from A 2-Year Follow-Up Study

Objective

Comorbid depression is quite common in early psychosis and specifically related to suicidal behavior and poor long-term outcomes. However, Depressive Symptoms (DS) are often neglected in both research and treatment, especially at the psychosis onset. The goals of this investigation were: (a) to longitudinally explore DS levels in patients with First Episode Psychosis (FEP) during 24 months of follow-up, and (b) to investigate the associations of DS with psychopathology and intervention components of an "Early Intervention in Psychosis" (EIP) program across the follow-up period.

Method

The Global Assessment of Functioning (GAF) and the Positive And Negative Syndrome Scale (PANSS) were completed by 266 FEP subjects. A linear regression analysis with DS as the dependent parameter and psychopathological and treatment characteristics as independent variables was performed (both at baseline and across the follow-up period).

Results

DS had enduring associations with PANSS "Positive Symptoms" and "Negative Symptoms" subscores. During the investigation, FEP subjects significantly improved their DS severity levels. This was related to the number of individual psychotherapy meetings supplied within the EIP protocol, as well as to a higher antidepressant dose and a lower antipsychotic dose prescribed during the follow-up.

Conclusions

DS are quite prominent in FEP, even at the recruitment time in EIP services. Nevertheless, DS severity tends to diminish overtime, especially with the provision of specialized EIP treatments.

Longitudinal Network Analysis Reveals Interactive Change of Schizophrenia Symptoms During Acute Antipsychotic Treatment

BACKGROUND AND HYPOTHESIS

Complex schizophrenia symptoms were recently conceptualized as interactive symptoms within a network system. However, it remains unknown how a schizophrenia network changed during acute antipsychotic treatment. The present study aimed to evaluate the interactive change of schizophrenia symptoms under seven antipsychotics from individual time series.

STUDY DESIGN

Data on 3030 schizophrenia patients were taken from a multicenter randomized clinical trial and used to estimate the partial correlation cross-sectional networks and longitudinal random slope networks based on multivariate multilevel model. Thirty symptoms assessed by The Positive and Negative Syndrome Scale clustered the networks.

STUDY RESULTS

Five stable communities were detected in cross-sectional networks and random slope networks that describe symptoms change over time. Delusions, emotional withdrawal, and lack of spontaneity and flow of conversation featured as central symptoms, and conceptual disorganization, hostility, uncooperativeness, and difficulty in abstract thinking featured as bridge symptoms, all showing high centrality in the random slope network. Acute antipsychotic treatment changed the network structure (M-test = 0.116, P < .001) compared to baseline, and responsive subjects showed lower global strength after treatment (11.68 vs 14.18, S-test = 2.503, P < .001) compared to resistant subjects. Central symptoms and bridge symptoms kept higher centrality across random slope networks of different antipsychotics. Quetiapine treatment network showed improvement in excitement symptoms, the one featured as both central and bridge symptom.

CONCLUSION

Our findings revealed the central symptoms, bridge symptoms, cochanging features, and individualized features under different antipsychotics of schizophrenia. This brings implications for future targeted drug development and search for pathophysiological mechanisms.

Relationship between adjustment disorder symptoms and probable diagnosis before and after second lockdown in Israel: longitudinal symptom network analysis

BACKGROUND

There is cumulative evidence of the importance of exploring the change of dynamics between symptoms over time as reflective of consolidation of psychopathology.

AIMS

To explore the interactions between symptoms of ICD-11 adjustment disorder before and after the second lockdown of the COVID-19 pandemic in Israel and identify the most central symptoms and their concurrent and prospective associations with probable adjustment disorder.

METHOD

This is a population-based study drawn from a probability-based internet panel. A representative sample of the adult Israeli population was assessed at two time points (T1, pre-second lockdown, n = 1029, response rate 76.17%; T2, post-second lockdown, n = 764, response rate 74.24%). Symptoms of adjustment disorder were assessed by the International Adjustment Disorder Questionnaire (IADQ).

RESULTS

Although the overall strength of associations at the two measurement points was similar and two same communities were found, there was a significant change in their structure, with a more consolidated network at T2. The most central item was 'difficult to relax' in both networks. Cross-sectionally, all symptoms of failure to adapt significantly predicted adjustment disorder. 'Worry a lot more' (preoccupation) and 'difficult to adapt to life' (failure to adapt) at T1 significantly predicted this diagnosis at T2.

CONCLUSIONS

Adjustment disorder symptoms consolidated during the second lockdown of the pandemic. In line with the ICD-11 conceptualisation of adjustment disorder, both preoccupation and failure-to-adapt symptoms have prognostic validity. This highlights the importance of identifying and targeting adjustment disorder symptoms during a period of stress such as the COVID-19 pandemic.

Searching for bridges between psychopathology and real-world functioning in first-episode psychosis: A network analysis from the OPTiMiSE trial

BACKGROUND

Network analysis has been used to explore the interplay between psychopathology and functioning in psychosis, but no study has used dedicated statistical techniques to focus on the bridge symptoms connecting these domains. The current study aims to estimate the network of depressive, negative, and positive symptoms, general psychopathology, and real-world functioning in people with first-episode schizophrenia or schizophreniform disorder, focusing on bridge nodes.

METHODS

Baseline data from the OPTiMiSE trial were analyzed. The sample included 446 participants (age 40.0 ± 10.9 years, 70% males). The network was estimated with a Gaussian graphical model, using scores on individual items of the positive and negative syndrome scale (PANSS), the Calgary depression scale for schizophrenia, and the personal and social performance scale. Stability, strength centrality, expected influence (EI), predictability, and bridge centrality statistics were computed. The top 20% scoring nodes on bridge strength were selected as bridge nodes.

RESULTS

Nodes from different rating scales assessing similar psychopathological and functioning constructs tended to cluster together in the estimated network. The most central nodes (EI) were Delusions, Emotional Withdrawal, Depression, and Depressed Mood. Bridge nodes included Depression, Conceptual Disorganization, Active Social Avoidance, Delusions, Stereotyped Thinking, Poor Impulse Control, Guilty Feelings, Unusual Thought Content, and Hostility. Most of the bridge nodes belonged to the general psychopathology subscale of the PANSS. Depression (G6) was the bridge node with the highest value.

CONCLUSIONS

The current study provides novel insights for understanding the complex phenotype of psychotic disorders and the mechanisms underlying the development and maintenance of comorbidity and functional impairment after psychosis onset.