Toward incorporating genetic risk scores into symptom networks of psychosis

Network analysis of comorbid aggressive behavior and testosterone among bipolar disorder patients: a cross-sectional study

Testosterone has complex effects on psychological traits and behavior; it is associated with social dominance and competition and is a potential human sex pheromone. This study aimed to investigate the associations between testosterone levels, aggressive behavior, and manic symptoms using a network analysis among bipolar disorder (BD) patients in psychiatric emergency departments (PED). Data from January 2021 and March 2022 BD patients in PED were analyzed. Manic symptoms were assessed using the Young Mania Rating Scale (YMRS). Aggression was assessed with subscale of the PANSS scale (PANSS-AG). The undirected network structures of testosterone levels, aggressive behavior, and manic symptoms were estimated, and centrality and bridge centrality indices were examined. Network stability was examined using the case-dropping procedure. The Network Comparison Test (NCT) was conducted to evaluate whether network characteristics differed by gender. We recruited a total of 898 BD patients, with the mean YMRS score as 13.30 ± 9.58. The prevalence of level II aggression was 35.6% (95%CI = 32.5%-38.7%), level III aggression was 29.5% (95%CI = 26.3%-32.6%), and level VI aggression was 7.0% (95%CI = 5.4%-8.8%). The male participants had a mean testosterone level of 391.71 (Standard Deviation (SD):223.39) compared to 36.90 (SD:30.50) for female participants in the whole sample. Through network analysis, "Increased motor activity-energy" emerged as the central symptom, with the highest centrality expected influence, followed by "Emotional Instability" and "Disruptive/aggression behavior". Notably, "Emotional Instability" appeared to be the bridge symptom linking manic symptoms to aggressive behavior. Within the flow network model, "Speech rate and amount" exhibited the strongest positive correlation with testosterone levels, followed closely by "Disruptive/aggression behavior". The constructed network model demonstrated robust stability, with gender showing no significant impact on the structure. In this study, "Increased motor activity-energy" stood out as the most influential symptom, and "Speech rate and amount" acted as the main bridge symptom linking testosterone levels, aggressive behavior, and manic symptoms. Targeting the central and bridge symptoms may improve the outcomes of aggression interventions implemented among BD patients in psychiatric emergency care.

Networks of Adversity in Childhood and Adolescence and Their Relationship to Adult Mental Health

Adverse experiences before the age of eighteen are common and include diverse events ranging from sexual abuse to parental divorce. These stressful experiences have been linked to physical and mental health issues. Previous research has focused mainly on childhood adversity, such as experiences in the family environment. Little consideration has been given to adversities that may be particularly harmful in adolescence. To understand adolescents' adverse experiences, this project used data from the Avon Longitudinal Study of Parents and Children (ALSPAC, total N = 14,901, N ≈ 1,200 - 10,000 per measure). We modelled interrelations of adversities in childhood (1-11 years) and adolescence (11-23 years) and examined adversity clusters using network analysis. We found two similar clusters in the childhood and adolescence networks: (1) direct abuse and (2) adverse family factors. We identified a third cluster of (3) educational and social adversities for adolescence. For both age groups, emotional abuse in the family environment was closely linked to mental health in early adulthood and most adversities were linked with depression in early adulthood. In adolescence, housing and academic issues and abuse by a romantic partner were particularly central to the network of adversities. Thus, we found commonalities and differences in the relevance of adverse experiences at different developmental stages. These findings highlight the need to develop age-dependent frameworks for adversity research and policymaking.

Symptom Network Analysis Tools for Applied Researchers With Cross-Sectional and Panel Data - A Brief Overview and Multiverse Analysis

In recent years, there has been a growing interest in utilizing symptom-network models to study psychopathology and relevant risk factors, such as cognitive and physical health. Various methodological approaches can be employed by researchers analyzing cross-sectional and panel data (i.e., several time points over an extended period). This paper provides an overview of some commonly used analytical tools, including moderated network models, network comparison tests, cross-lagged network analysis, and panel graphical vector-autoregression (VAR) models. Using an easily accessible dataset (easySHARE), this study demonstrates the use of different analytical approaches when investigating (a) the association between mental health and cognitive functioning, and (b) the role of chronic disease in mediating or moderating this association. This multiverse analysis showcases both converging and diverging evidence from different analytical avenues. These findings underscore the importance of multiverse investigations to increase transparency and communicate the extent to which conclusions depend on analytical choices.

A network analysis of depressive symptoms and metabolomics

BACKGROUND

Depression is associated with metabolic alterations including lipid dysregulation, whereby associations may vary across individual symptoms. Evaluating these associations using a network perspective yields a more complete insight than single outcome-single predictor models.

METHODS

We used data from the Netherlands Study of Depression and Anxiety (N = 2498) and leveraged networks capturing associations between 30 depressive symptoms (Inventory of Depressive Symptomatology) and 46 metabolites. Analyses involved 4 steps: creating a network with Mixed Graphical Models; calculating centrality measures; bootstrapping for stability testing; validating central, stable associations by extra covariate-adjustment; and validation using another data wave collected 6 years later.

RESULTS

The network yielded 28 symptom-metabolite associations. There were 15 highly-central variables (8 symptoms, 7 metabolites), and 3 stable links involving the symptoms Low energy (fatigue), and Hypersomnia. Specifically, fatigue showed consistent associations with higher mean diameter for VLDL particles and lower estimated degree of (fatty acid) unsaturation. These remained present after adjustment for lifestyle and health-related factors and using another data wave.

CONCLUSIONS

The somatic symptoms Fatigue and Hypersomnia and cholesterol and fatty acid measures showed central, stable, and consistent relationships in our network. The present analyses showed how metabolic alterations are more consistently linked to specific symptom profiles.

A network analysis of depressive symptoms, psychosocial factors, and suicidal ideation in 8686 adolescents aged 12-20 years

Suicide has aroused global concern, and a better understanding of the complex interactions between suicide ideation and various psychopathological features is critical. We aimed to explore the complex interplays among adolescents. This study adopted a multistage stratified cluster sampling method and recruited adolescents aged 12 to 20 year-old from 8 middle and high schools between December 2020 and September 2021 in Guangzhou, China. We assessed loneliness, social support, bullying victimization, depressive symptoms, and suicidal ideation. We used network analysis to examine the network structure of the correlates of suicidal ideation and identify central symptoms and bridge symptoms. We used case-drop bootstrapping and correlation stability coefficients to examine the stability of the network. Among 8686 adolescents, 347 (4 %) adolescents reported suicidal ideation in the past two weeks. Network analyses identified 'hopeless', 'psychomotor', and 'failure' were the three strongest edges linked to suicidal ideation. The most central nodes were identified as 'hopeless' being the most central node, followed by loneliness and verbal bullying victimization, while sexual bullying victimization, sex, and relational bullying were the strongest bridging symptoms. The findings shed light on the complexity of associations of suicidal ideation and could provide insight into school-based suicide risk assessment and prevention for adolescents.

Network analysis of comorbid depression, suicidality and biomarkers on HPA axis among mood disorder patients to psychiatric emergency services

Rapid assessment and intervention of suicide risk are common and challenging in psychiatric emergency departments (PED). It is unclear whether distinct pathophysiological processes exist among depressive patients with suicidality. This study examined the network structures of biomarkers on Hypothalamic-Pituitary-Adrenal (HPA) axis, such as Adrenocorticotropic hormone (ACTH) and Corticosterone (Cort), as well as suicidality and depressive symptoms in mood disorder patients in PED. Mood disorder patients in PED were assessed with the measurements of suicidality and depressive symptoms, respectively. A network analysis was performed to identify central symptoms and bridge symptoms of this network and their links to ACTH and Cort. Network stability was examined using the case-dropping procedure. The Network Comparison Test (NCT) was conducted to evaluate whether network characteristics differed by gender. A total of 1815 mood disorder patients were recruited. The prevalence of SI was 31.2% (95% CI: 28.15-34.21%), SP was 30.4% (95% CI: 27.39-33.41%), SA was 30.62% (95% CI: 27.61-33.64%) among psychiatric outpatients. The mean score of HAMD-24 was 13.87 ± 8.02. Network analysis revealed that 'Somatic anxiety' had the highest expected centrality, followed by 'Hopelessness' and 'Suicide attempt'. 'Corticosterone' and 'Retardation' may be the main bridge symptoms between depressive symptoms and the suicidality community. The network model showed a high degree of stability. Gender did not significantly influence the network structure. The central symptoms and key bridge symptoms identified could be potential targets for interventions of the HPA axis, which is designed for regular screening of a range of suicidal activity. In the light of this, timely treatment should be provided for psychiatric emergency care.

Internet addiction and residual depressive symptoms among clinically stable adolescents with major psychiatric disorders during the COVID-19 pandemic: a network analysis perspective

To assess the inter-relationships between residual depressive symptoms (RDS) and Internet addiction (IA) using network analysis among clinically stable adolescents with major psychiatric disorders during the COVID-19 pandemic. RDS and IA were assessed using the Patient Health Questionnaire-9 (PHQ-9) and the Internet Addiction Test (IAT), respectively. Central symptoms and bridge symptoms in the network model were examined. A total of 1,454 adolescents met the study criteria and were included in the analyses. The prevalence of IA was 31.2% (95% CI: 28.8%-33.6%). In the network analysis, the nodes IAT15 ("Preoccupation with the Internet"), PHQ2 ("Sad mood"), and PHQ1 ("Anhedonia") were the most central symptoms in the IA-RDS network model. Bridge symptoms included IAT10 ("Sooth disturbing about your Internet use"), PHQ9 ("Suicide ideation"), and IAT3 ("Prefer the excitement online to the time with others"). Additionally, PHQ2 ("Sad mood") was the main node linking "Anhedonia" to other IA clusters. Internet addiction was common among clinically stable adolescents with major psychiatric disorders during the COVID-19 pandemic. Core and bridge symptoms identified in this study could be prioritized as targets for the prevention and treatment of IA in this population.

Using network analysis to detect associations between suspected painful health conditions and behaviour in dogs

Pain associated with chronic health conditions in non-human animals is an important animal welfare issue. To identify animals in pain and develop an understanding of the mechanisms by which pain affects behaviour, it is therefore important to establish the direct behavioural effects of painful health conditions. We reanalyse data from a cross-sectional survey that considered the presence or absence of a painful condition in dogs and quantified their affective predispositions using the Positive and Negative Activation Scale (PANAS). By applying ideas from network theory, we conceptualise pain as a stressor that exerts direct effects on a network of interacting behavioural variables, and subsequently estimated a network model of conditional dependence relations. Painful health conditions were positively conditionally associated with age (posterior mean partial correlation, ρ = 0.34; standard deviation [SD]=0.05), and negatively conditionally associated with the item 'your dog is full of energy' (ρ = -0.14; SD=0.06). In turn, the energy item was conditionally associated with other PANAS items which were marginally associated with pain, such as items representing ease of excitability and persistence in play. This suggests these marginal effects might be indirectly mediated via the energy item. Further, utilising the posterior predictive distribution we estimated that the median conditional probability (95% credible interval) of a painful health condition given an answer of 'strongly agree' on the energy item was 0.08 (0.05, 0.11), which increased to 0.32 (0.09, 0.58), given a response of 'strongly disagree'. This provides a potentially clinically useful interpretation of the conditional dependencies detected in the network.

Suicide ideation and anhedonia among clinically stable adolescents with the recurrent depressive disorder during the COVID-19 pandemic: A network perspective

BACKGROUND

Anhedonia is a suicide risk factor among adolescent patients with recurrent depressive disorder (depression hereafter). This study examined associations between suicidal ideation (SI) and residual depressive symptoms (RSD), including anhedonia, among clinically stable adolescents with depression.

METHOD

A network analysis was performed to examine the association between RDS and SI among adolescents with depression. Node-specific predictive betweenness was computed to examine short paths between anhedonia and SI. Additionally, a Network Comparison Test (NCT) was conducted to examine gender differences in derived network model characteristics.

RESULTS

The network analysis identified close associations of PHQ9 ("Suicide ideation") with PHQ1 ("Anhedonia") as well as some other RDS including PHQ6 ("Guilt"), PHQ2 ("Sad mood") and PHQ8 ("Motor disturbances"). Additionally, PHQ2 ("Sad mood") and PHQ4 ("Fatigue") were the main bridge nodes linking anhedonia and SI. Comparisons of network models did not find significant differences in network global strength or edge weights.

LIMITATION

Causal relations between anhedonia and SI could not be determined due to the cross-sectional study design.

CONCLUSIONS

SI was directly related to Anhedonia in addition to Guilt, Sad mood and Motor disturbances. Sad mood and Fatigue were the main bridge nodes linking Anhedonia and SI. To reduce the risk of SI among clinically stable adolescents with depression during the COVID-19 pandemic, specific RDS including Anhedonia, Guilt, Sad mood, Motor disturbances and Fatigue should be targeted in interventions.

Inter-relationships of depressive and anxiety symptoms with suicidality among adolescents: A network perspective

BACKGROUND

Persons with suicidality including suicidal ideation (SI), suicide plans (SP) and/or suicide attempts (SA) are at higher risk for future suicide than those without suicidality. To reduce the risk of future suicide, it is important to understand symptoms of emotional distress that have the strongest links with SI, SP and SA. This network analysis examined item-level relations of depressive and anxiety symptoms with suicidality among adolescents during the COVID-19 pandemic.

METHODS

Adolescents between 12 and 20 years of age were assessed with the Patient Health Questionnaire (PHQ-9), Generalized Anxiety Disorder Scale (GAD-7), and individual binary reponse (no/yes) items assessing SI, SP, and SA during the pandemic. The structure of depressive symptoms, anxiety symptoms and suicidality was characterized using "Expected Influence" and "Bridge Expected Influence" as centrality indices in the symptom network. Network stability was tested using a case-dropping bootstrap procedure. Node-specific predictive betweenness was computed to examine short paths of anhedonia, other depressive symptoms and anxiety symptoms with suicidality. A Network Comparison Test (NCT) was conducted to examine whether network characteristics differed based on gender.

RESULTS

Prevalence rates of depressive symptoms, anxiety symptoms, and suicidality were 44.60 % (95% confidence interval (CI) = 41.53-47.67 %), 31.12 % (95%CI = 28.26-33.98 %), and 16.95 % (95%CI = 14.63-19.26 %), respectively, in the study sample. The network analysis identified GAD3 ("Worry too much") as the most central symptom, followed by GAD6 ("Irritability") and PHQ6 ("Guilt") in the sample. Additionally, PHQ6 ("Guilt"), GAD6 ("Irritability"), and PHQ2 ("Sad mood") were bridge nodes linking depressive and anxiety symptoms with suicidality. A flow network indicated that the connection between S ("Suicidality") and PHQ6 ("Guilt") reflected the strongest connection, followed by connections of S ("Suicidality") with GAD2 ("Uncontrollable worrying"), and S ("Suicidality") with PHQ2 ("Sad mood"). Finally, PHQ2 ("Sad mood") was the main bridge node linking anhedonia with other depressive and anxiety symptoms and suicidality in the sample.

CONCLUSIONS

Findings highlight the potential importance of reducing specific depressive and anxiety symptoms as possible means of reducing suicidality among adolescents during the pandemic. Central symptoms and key bridge symptoms identified in this study should be targeted in suicide prevention for at-risk adolescents.

Do the core symptoms play key roles in the development of problematic smartphone use symptoms

Aims

Previous research determined the core symptoms (loss of control and being caught in the loop) of problematic smartphone use (PSU), which are of great importance to understand the structure and potential intervention targets of PSU. However, the cross-sectional design fails to reveal causality between symptoms and usually conflates the between- and within-subjects effects of PSU symptoms. This study aims to determine whether the core symptoms of PSU, indeed, dominate the future development of PSU symptoms from longitudinal between- and within-subjects levels.

Materials and methods

In this study, 2191 adolescents were surveyed for 3 years for PSU symptoms. A cross-lagged panel model (CLPM) was used to explore longitudinal between-subjects causal relationships between symptoms, and a graphic vector autoregressive model (GVAR) was used to separate the between- and within-subjects effects and detect the longitudinal effect at the within-subject level.

Results

The results of CLPM indicated that the core symptoms (both loss of control and being caught in the loop) of PSU, indeed, dominate the future development of PSU symptoms at a longitudinal between-subjects level. From T1 to T2, the cross-lagged model showed that both the loss of control (out-prediction = 0.042) and being caught in the loop (out-prediction = 0.053) at T1 have the highest out-prediction over other symptoms at T2. From T2 to T3, the loss of control (out-prediction = 0.027) and being caught in the loop (out-prediction = 0.037) at T2 also have the highest out-prediction over other symptoms of PSU at T3. While, after separating the between- and within-subjects effects, only being caught in the loop at T1 played a key role in promoting the development of other PSU symptoms at T3 at the within-subjects level. The contemporaneous network showed intensive connection, while the cross-sectional between-subjects network is very sparse.

Conclusion

These findings not only confirm and extend the key roles of core symptoms in the dynamic aspect of PSU symptoms and PSU itself but also suggest that interventions should consider the core symptoms of PSU, individual- and group-level effects and that individualized intervention programs are needed in future.

Bridging the phenomenological gap between predictive basic-symptoms and attenuated positive symptoms: a cross-sectional network analysis

Attenuated positive symptoms (APS), transient psychotic-like symptoms (brief, limited intermittent psychotic symptoms, BLIPS), and predictive cognitive-perceptive basic-symptoms (BS) criteria can help identify a help-seeking population of young people at clinical high-risk of a first episode psychosis (CHRp). Phenomenological, there are substantial differences between BS and APS or BLIPS. BS do not feature psychotic content as delusion or hallucinations, and reality testing is preserved. One fundamental problem in the psychopathology of CHRp is to understand how the non-psychotic BS are related to APS. To explore the interrelationship of APS and predictive BS, we fitted a network analysis to a dataset of 231 patients at CHRp, aged 24.4 years (SD = 5.3) with 65% male. Particular emphasis was placed on points of interaction (bridge symptoms) between the two criteria sets. The BS ‘unstable ideas of reference’ and “inability to discriminate between imagination and reality” interacted with attenuated delusional ideation. Perceptual BS were linked to perceptual APS. Albeit central for the network, predictive cognitive basic BS were relatively isolated from APS. Our analysis provides empirical support for existing theoretical accounts that interaction between the distinct phenomenological domains of BS and APS is characterized by impairments in source monitoring and perspective-taking. Identifying bridge symptoms between the symptom domains holds the potential to empirically advance the etiological understanding of psychosis and pave the way for tailored clinical interventions.

A network analysis on the relationship between loneliness and schizotypy

BACKGROUND

Schizotypy is a multidimensional personality trait related to the heightened risk for the development of schizophrenia spectrum disorders. While it has been suggested that loneliness may be associated with schizotypy in general, whether it relates to the specific schizotypal traits differentially remains unknown. Besides, as loneliness often co-occurs with depression and anxiety, it is important to delineate its relationship with schizotypy in consideration of these co-occurring emotional disturbances.

METHODS

A demographically diverse sample of young people was obtained from multiple sources. The validated sample consisted of 2089 participants (68.4% female, age range: 18-30). The structural relationship between loneliness and schizotypy was modelled using a network analytic approach. The Gaussian graphical model with loneliness and nine schizotypal traits as nodes was first estimated without, and then with adjustment for the levels of depressive and anxiety symptoms. Edges were estimated as unique associations between nodes.

RESULTS

'Suspiciousness', 'odd beliefs or magical thinking', 'no close friends', 'constricted affect' and 'excessive social anxiety' were linked to loneliness directly. Loneliness was found to be more strongly associated with 'suspiciousness' and 'no close friends' than other schizotypal traits. After adjustment for the levels of depressive and anxiety symptoms, the above direct edges remained robust.

LIMITATIONS

The use of cross-sectional data indicated only undirected associations between variables.

CONCLUSIONS

Loneliness was more strongly linked to some schizotypal traits than others, with the relationships maintaining above and beyond the effects of anxiety and depression. These findings warrant further investigation of the specific relationships between loneliness and individual schizotypal traits.

Evidence, and replication thereof, that molecular-genetic and environmental risks for psychosis impact through an affective pathway

BACKGROUND

There is evidence that environmental and genetic risk factors for schizophrenia spectrum disorders are transdiagnostic and mediated in part through a generic pathway of affective dysregulation.

METHODS

We analysed to what degree the impact of schizophrenia polygenic risk (PRS-SZ) and childhood adversity (CA) on psychosis outcomes was contingent on co-presence of affective dysregulation, defined as significant depressive symptoms, in (i) NEMESIS-2 (n = 6646), a representative general population sample, interviewed four times over nine years and (ii) EUGEI (n = 4068) a sample of patients with schizophrenia spectrum disorder, the siblings of these patients and controls.

RESULTS

The impact of PRS-SZ on psychosis showed significant dependence on co-presence of affective dysregulation in NEMESIS-2 [relative excess risk due to interaction (RERI): 1.01, p = 0.037] and in EUGEI (RERI = 3.39, p = 0.048). This was particularly evident for delusional ideation (NEMESIS-2: RERI = 1.74, p = 0.003; EUGEI: RERI = 4.16, p = 0.019) and not for hallucinatory experiences (NEMESIS-2: RERI = 0.65, p = 0.284; EUGEI: -0.37, p = 0.547). A similar and stronger pattern of results was evident for CA (RERI delusions and hallucinations: NEMESIS-2: 3.02, p < 0.001; EUGEI: 6.44, p < 0.001; RERI delusional ideation: NEMESIS-2: 3.79, p < 0.001; EUGEI: 5.43, p = 0.001; RERI hallucinatory experiences: NEMESIS-2: 2.46, p < 0.001; EUGEI: 0.54, p = 0.465).

CONCLUSIONS

The results, and internal replication, suggest that the effects of known genetic and non-genetic risk factors for psychosis are mediated in part through an affective pathway, from which early states of delusional meaning may arise.

Role of polygenic and environmental factors in the co-occurrence of depression and psychosis symptoms: a network analysis

Depression and psychosis are often comorbid; they also have overlapping genetic and environmental risk factors, including trauma and area-level exposures. The present study aimed to advance understanding of this comorbidity via a network approach, by (1) identifying bridge nodes that connect clusters of lifetime depression and psychosis symptoms and (2) evaluating the influence of polygenic and environmental risk factors in these symptoms. This study included data from European ancestry participants in UK Biobank, a large population-based sample (N = 77,650). In Step 1, a network model identified bridge nodes between lifetime symptoms of depression and psychosis and functional impairment. In Step 2, genetic and environmental risk factors were incorporated to examine the degree to which symptoms associated with polygenic risk scores for depression and schizophrenia, lifetime exposure to trauma and area-level factors (including deprivation, air pollution and greenspace). Feelings of worthlessness, beliefs in unreal conspiracy against oneself, depression impairment and psychosis impairment emerged as bridges between depression and psychosis symptoms. Polygenic risk scores for depression and schizophrenia were predominantly linked with depression and psychosis impairment, respectively, rather than with specific symptoms. Cumulative trauma emerged as a bridge node associating deprivation with feelings of worthlessness and beliefs in unreal conspiracy, indicating that the experience of trauma is prominently linked with the co-occurrence of depression and psychosis symptoms related to negative views of oneself and others. These key symptoms and risk factors provide insights into the lifetime co-occurrence of depression and psychosis.

A network approach to relationships between cannabis use characteristics and psychopathology in the general population

Cannabis use characteristics, such as earlier initiation and frequent use, have been associated with an increased risk for developing psychotic experiences and psychotic disorders. However, little is known how these characteristics relate to specific aspects of sub-clinical psychopathology in the general population. Here, we explore the relationships between cannabis use characteristics and psychopathology in a large general population sample (N = 2,544, mean age 29.2 years, 47% women) by employing a network approach. This allows for the identification of unique associations between two cannabis use characteristics (lifetime cumulative frequency of cannabis use, age of cannabis use initiation), and specific psychotic experiences and affective symptoms, while controlling for early risk factors (childhood trauma, urban upbringing). We found particularly pronounced unique positive associations between frequency of cannabis use and specific delusional experiences (persecutory delusions and thought broadcasting). Age of cannabis use initiation was negatively related to visual hallucinatory experiences and irritability, implying that these experiences become more likely the earlier use is initiated. Earlier initiation, but not lifetime frequency of cannabis use, was related to early risk factors. These findings suggest that cannabis use characteristics may contribute differentially to risk for specific psychotic experiences and affective symptoms in the general population.

Expanding the system: A brief psychosocial complex systems model of internalising disorder

BACKGROUND

The aetiology of internalising disorders remains poorly understood. Recently, a bottom-up network perspective has suggested mental disorders are best conceptualised as emergent systems, and may be explained by mapping systems of symptoms embedded within a complex biopsychosocial environment. Under this framework, the complex system in which internalising disorders are embedded remains poorly understood. The present research outlines a brief psychosocial system of internalising disorders as a basis for future research.

METHODS

A Mixed Graphical Model was fitted on 15 psychosocial variables (including depression and anxiety) collected during the Christchurch Health and Development Study, a representative population birth cohort of 1265 people born in 1977 in Christchurch, New Zealand.

RESULTS

The model demonstrates that psychosocial risk factors for internalising disorders tend to be inter-related. The psychosocial system accounted for 19.9% of the variance in the diagnostic depression variable, and 5.0% of the variance in diagnostic anxiety. Most variables (10/13) were associated with depression and anxiety either directly or indirectly.

LIMITATIONS

First, the estimated model is undirected, so causal directions are unspecified except for longitudinal relationships. Second, binary diagnostic variables were used for depression and anxiety, meaning the model does include symptom-level complexity. Third, the model does not account for within-person effects.

CONCLUSIONS

This exploratory model may serve as a basis for the mapping of greater (bio) psychosocial complexity around internalising disorders. The model concisely demonstrates the need for researchers to "embrace complexity", but also underscores the conceptual scope that is required to do so on a broader (bio) psychosocial level.

Meta-analytic Gaussian Network Aggregation

A growing number of publications focus on estimating Gaussian graphical models (GGM, networks of partial correlation coefficients). At the same time, generalizibility and replicability of these highly parameterized models are debated, and sample sizes typically found in datasets may not be sufficient for estimating the underlying network structure. In addition, while recent work emerged that aims to compare networks based on different samples, these studies do not take potential cross-study heterogeneity into account. To this end, this paper introduces methods for estimating GGMs by aggregating over multiple datasets. We first introduce a general maximum likelihood estimation modeling framework in which all discussed models are embedded. This modeling framework is subsequently used to introduce meta-analytic Gaussian network aggregation (MAGNA). We discuss two variants: fixed-effects MAGNA, in which heterogeneity across studies is not taken into account, and random-effects MAGNA, which models sample correlations and takes heterogeneity into account. We assess the performance of MAGNA in large-scale simulation studies. Finally, we exemplify the method using four datasets of post-traumatic stress disorder (PTSD) symptoms, and summarize findings from a larger meta-analysis of PTSD symptom.

Unconsidered issues of measurement noninvariance in biological psychiatry: A focus on biological phenotypes of psychopathology

There is increasing appreciation that certain biological processes may not be equally related to all psychiatric symptoms in a given diagnostic category. Research on the biological phenotyping of psychopathology has begun examining the etiological and treatment implications of identified biotypes; however, little attention has been paid to a critical methodological implication of these results: measurement noninvariance. Measurement invariance is the ability of an instrument to measure the same construct, the same way, across different people, or across different time points for the same individual. If what a measure quantifies differs across different people (e.g., those with or without a particular biotype) or time points, then it is invalid to directly compare means on that measure. Using a running example of inflammatory phenotypes of depression, we first describe the biological phenotyping of psychopathology. Second, we discuss three types of measurement invariance. Third, we demonstrate how differential biology-symptom associations invariably creates measurement noninvariance using a theoretical example and simulated data (for which code is provided). We also show how this issue can lead to false conclusions about the broader diagnostic construct. Finally, we provide several suggestions for addressing these important issues to help advance the field of biological psychiatry.

Autistic Symptoms and Social Functioning in Psychosis: A Network Approach

Psychotic and autistic symptoms are related to social functioning in individuals with psychotic disorders (PD). The present study used a network approach to (1) evaluate the interactions between autistic symptoms, psychotic symptoms, and social functioning, and (2) investigate whether relations are similar in individuals with and without PD. We estimated an undirected network model in a sample of 504 PD, 572 familial risk for psychosis (FR), and 337 typical comparisons (TC), with a mean age of 34.9 years. Symptoms were assessed with the Autism Spectrum Quotient (AQ; 5 nodes) and the Community Assessment of Psychic Experiences (CAPE; 9 nodes). Social functioning was measured with the Social Functioning Scale (SFS; 7 nodes). We identified statistically significant differences between the FR and PD samples in global strength (P < .001) and network structure (P < .001). Our results show autistic symptoms (social interaction nodes) are negatively and more closely related to social functioning (withdrawal, interpersonal behavior) than psychotic symptoms. More and stronger connections between nodes were observed for the PD network than for FR and TC networks, while the latter 2 were similar in density (P = .11) and network structure (P = .19). The most central items in strength for PD were bizarre experiences, social skills, and paranoia. In conclusion, specific autistic symptoms are negatively associated with social functioning across the psychosis spectrum, but in the PD network symptoms may reinforce each other more easily. These findings emphasize the need for increased clinical awareness of comorbid autistic symptoms in psychotic individuals.

Understanding the Association Between Intolerance of Uncertainty and Problematic Smartphone Use: A Network Analysis

BACKGROUND

Existing research has demonstrated that intolerance of uncertainty (IU) is associated with problematic smartphone use (PSU). However, little is known about how different IU components such as uncertainty-related beliefs, emotions, and behaviors may impact on different PSU symptoms.

METHODS

Extending previous research, the current study examined the specific associations between IU components and PSU symptoms via a symptom-level network approach. A regularized partial correlation network consisting of different IU components and PSU symptoms was estimated among 1,849 Chinese university students. We examined pathways and influential nodes (i.e. central components/symptoms and bridge components/symptoms) within the IU-PSU network.

RESULTS

The strongest pathway linking IU and PSU was between emotional reactions to uncertainty and coping-motivated smartphone use. Importantly, emotional reactions toward not having enough information (a reflection of emotional reactions to uncertainty) may act as both a central and a bridge component in maintaining the whole IU-PSU network.

CONCLUSIONS

The results are in line with the I-PACE model and highlight that PSU may be a coping response for negative emotions derived from uncertainty. Finally, the current findings highlight the potential of interventions targeting intolerance of uncertainty for reducing PSU.

The relations between different components of intolerance of uncertainty and symptoms of generalized anxiety disorder: a network analysis

BACKGROUND

Intolerance of uncertainty (IU) is considered as a specific risk factor in the development and maintenance of generalized anxiety disorder (GAD). Yet, researches have investigated the relations between IU and GAD (or worry) using total scores on self-report measures. This ignores that there are different components exist in IU and the heterogeneity of GAD symptoms. In the present study, we explored the relations among different components of IU and symptoms of GAD.

METHODS

A dimensional approach which take individual differences into consideration in different components of IU along a full range of normal to abnormal symptom severity levels of GAD were used in this study. Components of IU were measured by 12-item Intolerance of Uncertainty Scale and symptoms of GAD were measured by Generalized Anxiety Disorder 7-Item Questionnaire. Regularized partial-correlation network was estimated using cross-sectional data from 624 university students.

RESULTS

Four strongest edges are between components of IU, like "Unforeseen events upset me greatly" and "It frustrates me not having all the information I need". Two strongest edges are between symptoms of GAD, like "Being so restless that it is hard to sit still" and "Feeling afraid as if something awful might happen". Symptom "Worrying too much about different things" and component "It frustrates me not having all the information I need" have the highest expected influences in the present network. In the community of IU, component "It frustrates me not having all the information I need" has the highest bridge expected influence. And in the community of GAD, symptoms "Worrying too much about different things" and "Not being able to stop or control worrying" have the highest bridge expected influence.

CONCLUSIONS

This study reveals potential pathways between different components of IU and various symptoms of GAD. Understanding how putative risk factors such as different components of IU are related to symptoms of GAD may provide some references for related preventions and interventions, such as targeting component "It frustrates me not having all the information I need" may be more effective at reducing symptoms of GAD than targeting other components of IU.

Can network analysis of self-reported psychopathology shed light on the core phenomenology of bipolar disorders in adolescents and young adults?

OBJECTIVES

Network analysis is increasingly applied to psychopathology research. We used it to examine the core phenomenology of emerging bipolar disorder (BD I and II) and 'at risk' presentations (major depression with a family history of BD).

METHODOLOGY

The study sample comprised a community cohort of 1867 twin and nontwin siblings (57% female; mean age ~26) who had completed self-report ratings of (i) depression-like, hypomanic-like and psychotic-like experiences; (ii) family history of BD; and (iii) were assessed for mood and psychotic syndromes using the Composite International Diagnostic Interview (CIDI). Symptom networks were compared for recent onset BD versus other cohort members and then for individuals at risk of BD (depression with/without a family history of BD).

RESULTS

The four key symptoms that differentiated recent onset BD from other cohort members were: anergia, psychomotor speed, hypersomnia and (less) loss of confidence. The four key symptoms that differentiated individuals at high risk of BD from unipolar depression were anergia, psychomotor speed, impaired concentration and hopelessness. However, the latter network was less stable and more error prone.

CONCLUSIONS

We are encouraged by the overlaps between our findings and those from two recent publications reporting network analyses of BD psychopathology, especially as the studies recruited from different populations and employed different network models. However, the advantages of applying network analysis to youth mental health cohorts (which include many individuals with multimorbidity) must be weighed against the disadvantages including basic issues such as judgements regarding the selection of items for inclusion in network models.

Polygenic risk for immuno-metabolic markers and specific depressive symptoms: A multi-sample network analysis study

BACKGROUND

About every fourth patient with major depressive disorder (MDD) shows evidence of systemic inflammation. Previous studies have shown inflammation-depression associations of multiple serum inflammatory markers and multiple specific depressive symptoms. It remains unclear, however, if these associations extend to genetic/lifetime predisposition to higher inflammatory marker levels and what role metabolic factors such as Body Mass Index (BMI) play. It is also unclear whether inflammation-symptom associations reflect direct or indirect associations, which can be disentangled using network analysis.

METHODS

This study examined associations of polygenic risk scores (PRSs) for immuno-metabolic markers (C-reactive protein [CRP], interleukin [IL]-6, IL-10, tumour necrosis factor [TNF]-α, BMI) with seven depressive symptoms in one general population sample, the UK Biobank study (n = 110,010), and two patient samples, the Munich Antidepressant Response Signature (MARS, n = 1058) and Sequenced Treatment Alternatives to Relieve Depression (STAR*D, n = 1143) studies. Network analysis was applied jointly for these samples using fused graphical least absolute shrinkage and selection operator (FGL) estimation as primary analysis and, individually, using unregularized model search estimation. Stability of results was assessed using bootstrapping and three consistency criteria were defined to appraise robustness and replicability of results across estimation methods, network bootstrapping, and samples.

RESULTS

Network analysis results displayed to-be-expected PRS-PRS and symptom-symptom associations (termed edges), respectively, that were mostly positive. Using FGL estimation, results further suggested 28, 29, and six PRS-symptom edges in MARS, STAR*D, and UK Biobank samples, respectively. Unregularized model search estimation suggested three PRS-symptom edges in the UK Biobank sample. Applying our consistency criteria to these associations indicated that only the association of higher CRP PRS with greater changes in appetite fulfilled all three criteria. Four additional associations fulfilled at least two consistency criteria; specifically, higher CRP PRS was associated with greater fatigue and reduced anhedonia, higher TNF-α PRS was associated with greater fatigue, and higher BMI PRS with greater changes in appetite and anhedonia. Associations of the BMI PRS with anhedonia, however, showed an inconsistent valence across estimation methods.

CONCLUSIONS

Genetic predisposition to higher systemic inflammatory markers are primarily associated with somatic/neurovegetative symptoms of depression such as changes in appetite and fatigue, consistent with previous studies based on circulating levels of inflammatory markers. We extend these findings by providing evidence that associations are direct (using network analysis) and extend to genetic predisposition to immuno-metabolic markers (using PRSs). Our findings can inform selection of patients with inflammation-related symptoms into clinical trials of immune-modulating drugs for MDD.

Network structure of psychotic-like experiences in adolescents: Links with risk and protective factors

AIM

The main goal was to analyse the network structure of psychotic-like experiences (PLEs) in a large sample of adolescents. In addition, the network structure between PLEs and putative risk (mental health difficulties, suicidal behaviour, depression symptoms) and protective factors (prosocial behaviour, subjective well-being, self-esteem) for psychosis was analysed.

METHODS

The sample compromised a total of 1790 adolescents (M=15.7 years; SD=1.26), 816 men (45.6%), selected by stratified random cluster sampling. Various tools were used to measure PLEs, general psychopathology, suicide ideation and behaviour, depression symptoms, prosocial behaviour, subjective well-being, and self-esteem. The Gaussian graphical model for continuous variables and Ising model for binary variables were used for network estimation.

RESULTS

The PLEs estimated network was strongly interconnected. Unusual perceptual experiences were among the most central nodes. The average predictability of this network was 16.41%. The PLEs and risk and protective factors estimated network showed a high degree of interconnectedness between PLEs and psychopathology domains. PLEs, behavioural problems, and emotional symptoms were among the most central nodes. The mean predictability of this network was 43.46%. The results of the stability and accuracy analysis indicated that networks were accurately estimated.

CONCLUSIONS

At population level, extended psychosis phenotype can be conceptualized as a network of interacting cognitive, emotional, and behavioural features. The network model allows us to understand psychosis risk, at the same time opening new lines of study in the mental health arena.

El análisis de redes en la investigación clínica

En el número 1 del volumen 69 de la presente revista se publicó un estudio que evaluó los niveles del burnout en residentes de medicina en Colombia y que evidenció el impacto que puede causar este síndrome en quienes lo padecen.1 El burnout es una respuesta a los problemas emocionales e interpersonales que se presentan en el trabajo y en la cual intervienen sentimientos de agotamiento, actitud indiferente, percepción de incompetencia ante la falta de recursos para afrontar las responsabilidades, insatisfacción y baja autoestima.

The Mental Health Ecosystem: Extending Symptom Networks With Risk and Protective Factors

Inspired by modeling approaches from the ecosystems literature, in this paper, we expand the network approach to psychopathology with risk and protective factors to arrive at an integrated analysis of resilience. We take a complexity approach to investigate the multifactorial nature of resilience and present a system in which a network of interacting psychiatric symptoms is targeted by risk and protective factors. These risk and protective factors influence symptom development patterns and thereby increase or decrease the probability that the symptom network is pulled toward a healthy or disorder state. In this way, risk and protective factors influence the resilience of the network. We take a step forward in formalizing the proposed system by implementing it in a statistical model and translating different influences from risk and protective factors to specific targets on the node and edge parameters of the symptom network. To analyze the behavior of the system under different targets, we present two novel network resilience metrics: Expected Symptom Activity (ESA, which indicates how many symptoms are active or inactive) and Symptom Activity Stability (SAS, which indicates how stable the symptom activity patterns are). These metrics follow standard practices in the resilience literature, combined with ideas from ecology and physics, and characterize resilience in terms of the stability of the system's healthy state. By discussing the advantages and limitations of our proposed system and metrics, we provide concrete suggestions for the further development of a comprehensive modeling approach to study the complex relationship between risk and protective factors and resilience.

Obsessive-Compulsive Symptoms and Other Symptoms of the At-risk Mental State for Psychosis: A Network Perspective

BACKGROUND

The high prevalence of obsessive-compulsive symptoms (OCS) among subjects at Ultra-High Risk (UHR) for psychosis is well documented. However, the network structure spanning the relations between OCS and symptoms of the at risk mental state for psychosis as assessed with the Comprehensive Assessment of At Risk Mental States (CAARMS) has not yet been investigated. This article aimed to use a network approach to investigate the associations between OCS and CAARMS symptoms in a large sample of individuals with different levels of risk for psychosis.

METHOD

Three hundred and forty-one UHR and 66 healthy participants were included, who participated in the EU-GEI study. Data analysis consisted of constructing a network of CAARMS symptoms, investigating central items in the network, and identifying the shortest pathways between OCS and positive symptoms.

RESULTS

Strong associations between OCS and anxiety, social isolation and blunted affect were identified. Depression was the most central symptom in terms of the number of connections, and anxiety was a key item in bridging OCS to other symptoms. Shortest paths between OCS and positive symptoms revealed that unusual thought content and perceptual abnormalities were connected mainly via anxiety, while disorganized speech was connected via blunted affect and cognitive change.

CONCLUSIONS

Findings provide valuable insight into the central role of depression and the potential connective component of anxiety between OCS and other symptoms of the network. Interventions specifically aimed to reduce affective symptoms might be crucial for the development and prospective course of symptom co-occurrence.

Toward a Complex Network of Risks for Psychosis: Combining Trauma, Cognitive Biases, Depression, and Psychotic-like Experiences on a Large Sample of Young Adults

Although the linkage between traumatic life events and psychotic-like experiences (PLEs) is well established, the knowledge of potential mechanisms of this relationship is scarce. The aim of the present study was to better understand the structure of connections between traumatic life events and PLEs by considering at the same time the role of cognitive biases and depressive symptoms in the population of young adults (18-35 years of age, M = 26.52, SD = 4.74, n = 6772). Our study was conducted within a framework of network analysis. PLEs were measured with the Prodromal Questionnaire (PQ-16), cognitive biases were measured with nine items from the Davos Assessment of Cognitive Biases Scale-18 (DACOBS-18), depressive symptoms were assessed with the Center for Epidemiologic Studies-Depression Scale (CESD-R) and exposure to traumatic life events was measured with a combination of Childhood Experience of Care and Abuse Questionnaire (CECA.Q) and Traumatic Experience Checklist (TEC). The results present a network of all nodes being interconnected within and between domains, with no isolated factors. Exposures to sexual trauma were the most central node in the network. Pathways were identified from trauma to PLEs via cognitive biases and depressive symptoms. However, the shortest pathway between the most central traumatic life event and PLEs was through other traumatic life events, without cognitive biases or depressive symptoms along the way. Our findings suggest the importance of environmental adversities as well as dysfunctional information processing and depression in the network of psychosis risks.

The relations among worry, meta-worry, intolerance of uncertainty and attentional bias for threat in men at high risk for generalized anxiety disorder: a network analysis

BACKGROUND

Improving the psychotherapies for generalized anxiety disorder (GAD) is dependent on a deeper understanding of the relations between GAD and its associated cognitive factors. In the present study, we investigate how the core feature of GAD (i.e., worry) and its associated cognitive factors, such as meta-worry, intolerance of uncertainty, and attention bias towards threat, relate to each other in men at high risk for GAD.

METHODS

We used network analysis to explore the relations among these variables in a cross-sectional sample of 122 men at high risk for generalized anxiety disorder. Specifically, we computed the expected influence and predictability of each variable.

RESULTS

In the final network, we found that worry and meta-worry had the highest expected influence and predictability. In contrast, attention bias towards threat showed the lowest expected influence and predictability. The estimates of the expected influence of the nodes were stable (correlation stability coefficient = 0.52).

CONCLUSIONS

The present study is the first to investigate the relations among worry, meta-worry, intolerance of uncertainty, and attention bias towards threat in men at high risk for generalized anxiety disorder. These findings indicate that worry and meta-worry may play important roles in the present network. The implications for clinical interventions and future studies are discussed.

Psychometric network models from time-series and panel data

Researchers in the field of network psychometrics often focus on the estimation of Gaussian graphical models (GGMs)-an undirected network model of partial correlations-between observed variables of cross-sectional data or single-subject time-series data. This assumes that all variables are measured without measurement error, which may be implausible. In addition, cross-sectional data cannot distinguish between within-subject and between-subject effects. This paper provides a general framework that extends GGM modeling with latent variables, including relationships over time. These relationships can be estimated from time-series data or panel data featuring at least three waves of measurement. The model takes the form of a graphical vector-autoregression model between latent variables and is termed the ts-lvgvar when estimated from time-series data and the panel-lvgvar when estimated from panel data. These methods have been implemented in the software package psychonetrics, which is exemplified in two empirical examples, one using time-series data and one using panel data, and evaluated in two large-scale simulation studies. The paper concludes with a discussion on ergodicity and generalizability. Although within-subject effects may in principle be separated from between-subject effects, the interpretation of these results rests on the intensity and the time interval of measurement and on the plausibility of the assumption of stationarity.

Unravelling the complex nature of resilience factors and their changes between early and later adolescence

BACKGROUND

Childhood adversity (CA) is strongly associated with mental health problems. Resilience factors (RFs) reduce mental health problems following CA. Yet, knowledge on the nature of RFs is scarce. Therefore, we examined RF mean levels, RF interrelations, RF-distress pathways, and their changes between early (age 14) and later adolescence (age 17).

METHODS

We studied 10 empirically supported RFs in adolescents with (CA+; n = 631) and without CA (CA-; n = 499), using network psychometrics.

RESULTS

All inter-personal RFs (e.g. friendships) showed stable mean levels between age 14 and 17, and three of seven intra-personal RFs (e.g. distress tolerance) changed in a similar manner in the two groups. The CA+ group had lower RFs and higher distress at both ages. Thus, CA does not seem to inhibit RF changes, but to increase the risk of persistently lower RFs. At age 14, but not 17, the RF network of the CA+ group was less positively connected, suggesting that RFs are less likely to enhance each other than in the CA- group. Those findings underpin the notion that CA has a predominantly strong proximal effect. RF-distress pathways did not differ in strength between the CA+ and the CA- group, which suggests that RFs have a similarly protective strength in the two groups. Yet, as RFs are lower and distress is higher, RF-distress pathways may overall be less advantageous in the CA+ group. Most RF interrelations and RF-distress pathways were stable between age 14 and 17, which may help explain why exposure to CA is frequently found to have a lasting impact on mental health.

CONCLUSIONS

Our findings not only shed light on the nature and changes of RFs between early and later adolescence, but also offer some accounts for why exposure to CA has stronger proximal effects and is often found to have a lasting impact on mental health.

Polygenic Risk Scores for Psychiatric Disorders Reveal Novel Clues About the Genetics of Disordered Gambling

Disordered gambling (DG) is a rare but serious condition that results in considerable financial and interpersonal harms. Twin studies indicate that DG is heritable but are silent with respect to specific genes or pathways involved. Existing genomewide association studies (GWAS) of DG have been substantially underpowered. Larger GWAS of other psychiatric disorders now permit calculation of polygenic risk scores (PRSs) that reflect the aggregated effects of common genetic variants contributing risk for the target condition. The current study investigated whether gambling and DG are associated with PRSs for four psychiatric conditions found to be comorbid with DG in epidemiologic surveys: major depressive disorder (MDD), attention-deficit hyperactivity disorder (ADHD), bipolar disorder (BD) and schizophrenia (SCZ). Genotype data and survey responses were analyzed from the Wave IV assessment (conducted in 2008) of the National Longitudinal Study of Adolescent to Adult Health, a representative sample of adolescents recruited in 1994-1995 and followed into adulthood. Among participants classified as having European ancestry based on genetic analysis (N = 5215), 78.4% reported ever having gambled, and 1.3% reported lifetime DG. Polygenic risk for BD was associated with decreased odds of lifetime gambling, OR = 0.93 [0.87, 0.99], p = .045, pseudo-R2(%) = .12. The SCZ PRS was associated with increased odds of DG, OR = 1.54 [1.07, 2.21], p = .02, pseudo-R2(%) = .85. Polygenic risk scores for MDD and ADHD were not related to either gambling outcome. Investigating features common to both SCZ and DG might generate valuable clues about the genetically influenced liabilities to DG.

Izmir Mental Health Cohort for Gene-Environment Interaction in Psychosis (TürkSch): Assessment of the Extended and Transdiagnostic Psychosis Phenotype and Analysis of Attrition in a 6-Year Follow-Up of a Community-Based Sample

Objective: TürkSch is a prospective, longitudinal study in a representative community sample (İzmir, Turkey), consisting of several data collection stages, to screen and follow-up mental health outcomes, with a special focus on the extended and transdiagnostic psychosis phenotype. The aim of the present paper is to describe the research methodology, data collection results, and associations with noncontact and refusal in the longitudinal arm. Methods: Households were contacted in a multistage clustered probability sampling frame, covering 11 districts and 302 neighborhoods at baseline (n = 4,011) and at 6-year follow-up (n = 2,185). Both at baseline and at follow-up, participants were interviewed with the Composite International Diagnostic Interview. Participants with probable psychotic disorder were reinterviewed with the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID)-I either at the hospital or at the participant's residence. Relevant neighborhood-level measures were assessed in a separate sample (n = 5,124) in addition to individual-level measures. Candidate gene-by-environment interactions were investigated using two nested case-control studies. Results: Individuals with a mental health problem had lower refusal rates. Older and lower educated individuals had a lower probability of noncontact. Discussion: The TürkSch study has an advanced design to meet the challenges of evaluating the multidimensional etiological and phenomenological nature of the extended and transdiagnostic psychosis phenotype.